Tunable Filters

Abstract

According to a first aspect of the present invention, an optical filter is provided. The optical filter comprises a first optical element and a second optical element, and one or more expansion elements. The first optical element and the second optical element are arranged along a common axis, and each optical element terminates at a flat surface perpendicular to the axis. The flat surfaces are separated by a gap having a width d so that the flat surfaces form a Fabry-Perot standard gauge, and light can propagate along an optical path passing through the elements and the standard gauge. The one or more expansion elements are connected to the optical element, are located outside the optical path, and extend parallel to the axis by a length greater than the width of the gap. The expansion element comprises a material having a linear thermal expansion coefficient different from that of the optical element, so that the difference in expansion between the expansion element and the optical element causes relative movement of the flat surface along the axis, thereby causing a change in the width of the gap.

Description

Technical Field

[0001] The present invention relates to a filter for an optical system, and in particular to a filter with a peak transmittance wavelength that can be thermally tuned.

[0002] Index of terms in equations

[0003] (unless otherwise specified in the description of the equation)

[0004] T-Temperature

[0005] α-Linear thermal expansion coefficient (LCE)

[0006] n-refractive index

[0007] ψ-thermo-optic coefficient

[0008] q-thermal path length sensitivity, q = nα + ψ

[0009] C- expansion element length

[0010] C-cylinder length

[0011] X0 - the value of "X" at the reference temperature (T0 is the reference temperature)

[0012] Change of ΔX-X, measured from X0 Background Art

[0013] Fiber optic communications use lasers to generate optical signals in narrow wavelength bands. A single fiber optic cable can transmit information in different frequency bands (and / or different channels within a frequency band). Therefore, it is important to reduce the noise generated in other frequency bands while transmitting signals in the target frequency band.

[0014] To accomplish this, the output of the laser / modulator system can be passed through a filter that is configured to select only the frequencies of interest. An optical element that can be used as a filter to accomplish this is a Fabry-Perot (FP) gage (or interferometer). Figure 1A The optical interference pattern is shown in Figure 1 and consists of a transparent plate with two reflective surfaces. As the light bounces between the surfaces, the transmitted rays interfere with each other, producing a characteristic interference pattern that depends on the frequency and optical distance between the plates.

[0015] The frequency response of the FP standard gauge is Figure 1B It can be seen that for a high-precision FP standard gauge (i.e., a standard gauge with a sharp peak), the loss of frequencies other than the target frequency is very large, and even a small deviation from the target frequency can be filtered out.

[0016] FP standards are suitable for situations where the component emits light at a constant frequency, but in some applications it is desirable for the component to be tunable to be able to emit at different frequencies, such as on each channel within a frequency band. The frequency response of the FP standard depends on the optical distance between the plates, so a current-tunable FP standard takes advantage of this by having a material between the plates where the optical distance varies greatly with temperature (i.e., large thermal path length sensitivity q, where q = nαΨ). Therefore, by adjusting the temperature of the standard, the optical distance and frequency response of the standard can be adjusted.

[0017] However, for the filter to work adequately within the band, it must have a large enough free spectral range (FSR) so that there is only one peak within the band (i.e., the frequency response curve always slopes away from the peak within the band). Otherwise, higher order peaks in the frequency response curve can allow noise to enter the fiber. The FSR is inversely proportional to the optical distance between the mirrors of the standard gauge, so to obtain a large spectral range, the optical distance between the mirrors must be small (e.g., about 30 microns in air for C-band).

[0018] Using a smaller distance between the mirrors means that temperature variations of the material between the mirrors will only result in a smaller variation of the optical distance between the mirrors. In fact, for the C-band using silicon as the material between the mirrors, the thickness variation cannot cover the band within a reasonable temperature range (e.g. 20°C to 100°C) with a satisfactory FSR.

[0019] A full-band solution can use multiple silicon-based tunable filters, where the FSR of each filter is smaller than the band. The filters are calibrated so that at any particular setting, only one peak overlaps across all filters, and the overlapping peak can be adjusted by varying the temperature of each filter individually to control the selected frequency. However, such a system requires extreme precision in manufacturing, calibration, and control, and still results in high losses compared to a single-filter system.

[0020] A tunable filter that can operate over the entire frequency band, has high FSR and low loss is needed. Summary of the invention

[0021] According to a first aspect of the present invention, an optical filter is provided. The optical filter comprises a first optical element and a second optical element, and one or more expansion elements. The first optical element and the second optical element are arranged along a common axis, and the first optical element and the second optical element both terminate at a flat surface perpendicular to the axis. The flat surfaces are separated by a gap having a width d so that the flat surfaces form a Fabry-Perot standard gauge, and light can propagate along an optical path through the elements and the standard gauge. The one or more expansion elements are directly or indirectly connected to the optical element, are located outside the optical path, and extend parallel to the axis by a length greater than the width of the gap. The expansion element comprises a material having a linear thermal expansion coefficient different from that of the optical element, so that the difference in expansion between the expansion element and the optical element causes relative movement of the flat surface along the axis, thereby causing a change in the width of the gap.

[0022] According to another aspect, an optical device is provided, comprising a modulator and an optical filter according to the previous aspect, wherein the optical filter is arranged to receive an output beam of the modulator. Further aspects and preferred features are set out in claim 2 et seq. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1A It is a schematic diagram of a standard measuring tool;

[0024] Figure 1B The frequency response curve of the standard gage is shown;

[0025] Figure 2A and Figure 2B An exemplary structure of an optical filter is shown;

[0026] Figure 3 An alternative exemplary structure is shown;

[0027] Figure 4 Another alternative exemplary structure is shown;

[0028] Figure 5A is the transmittance curve of the fully aligned standard gage;

[0029] Figure 5B is the transmittance curve of the misaligned gage;

[0030] Figure 6 shows exemplary connections between elements of an optical filter;

[0031] Fig. 7A , Figure 7B and Figure 7C Other exemplary structures of optical filters are shown;

[0032] Fig. 8A shows a simulated transmission curve for a misaligned optical filter;

[0033] Figure 8B shows a simulated transmission curve of an optical filter taking diffraction into account;

[0034] 9A to 9D Thermal simulation of an exemplary optical filter is shown. DETAILED DESCRIPTION

[0035] As described in the background, previous tunable gages have used thermal expansion materials in the inter-mirror space to adjust the inter-mirror optical distance d of the gage. However, this results in a compromise between the FSR of the gage and the possible tuning range. Instead, a design is proposed below in which the thermal expansion material is placed outside the optical path and used to change the width of the air gap between the mirrors. This allows a wider range of thermal expansion materials to be used (i.e., the thermal expansion material does not need to be transparent) and allows the device to have greater temperature sensitivity (or equivalently, a greater frequency range within a given temperature range).

[0036] Figure 2A and Figure 2B An exemplary structure of such a filter is shown. Figure 2A It is a stereogram. Figure 2B is a plan view. The filter has two optical elements 21a, 21b with an air gap 22 between them. Each optical element has a reflective surface 25a, 25b at the end facing the air gap so that the air gap forms a FP standard (the reflective surface can be provided by any suitable treatment). The optical elements are shown as cylindrical elements formed integrally with the plate, but can be of any suitable shape, such as rectangular. The optical elements 21a, 21b are separated by expansion portions 24a, 24b.

[0037] The optical element is formed of a material that is transparent at the target filter wavelength ("optical material"). The expansion element is formed of a material that has a linear coefficient of thermal expansion (LCE) greater than the LCE of the optical material ("expansion material").

[0038] The length of the expansion element between the locations where it is connected to the optical element is given by L=L0+ΔL=L0+αL0ΔT, where L0 is the length at a reference temperature (e.g. 20°C) and ΔT is the difference between the current temperature of the expansion material and the reference temperature.

[0039] The distance d between the reflective surfaces 25a, 25b is given by L-2C, where C is the length of each cylinder (or for other configurations, the length between the reflective surface and the connection of the expansion element to the optical element).

[0040] d=L-2C=(L0+ΔL)-2(C0+ΔC)=(L0+α L L0ΔT)-2(C0+α C C0ΔT)=d0+ΔT(α L L-2α C C o )

[0041] Assume that the cylinder and the expanding material are at the same temperature, where α L and α c are the linear expansion coefficients of the expansion material and the optical material, respectively. If α L ≥α c , then it is simplified to d = d0 + Δd = d0 + α L LΔT. It can be seen that the term (α L LΔT) is proportional to the initial length L0 of the expansion element and is independent of the initial value of d, so there is no longer a trade-off between FSR (needing a smaller d) and tuning range.

[0042] The length L0 is greater than the distance d0 between the reflective surfaces and may be much greater, for example, at least an order of magnitude greater than d (ie, at least ten times greater than d). For example, the distance d0 may be 20 to 30 microns and the length L0 may be 2 to 3 millimeters.

[0043] Alternatively, the coefficient of linear expansion of the "expansion element" can be smaller than that of the optical element. In this case, expansion of the optical element will cause the distance d to decrease, and an increase in temperature will cause d to decrease. The above formula still applies. In the case where the coefficient of linear expansion of the expansion element is much smaller than the CLE of the optical element, the expansion element effectively acts as a rigid "frame" or support, and the expansion of the optical element is the main factor causing the change in d. In order to make d adjustable, the coefficients of linear expansion of the expansion element and the optical element should be different.

[0044] For example, to achieve continuous tuning over the C-band with a temperature variation of 40-50°C, the linear expansion coefficient of the expansion material may be between 7 and 12 parts per million per °C (ppm / °C) for an expansion element of 1-2 mm (typically 1.5 mm) in length. If the linear expansion coefficient of the optical material is significant (e.g., >0.5 ppm), the difference between the linear expansion coefficient of the expansion material and the linear expansion coefficient of the optical material may be between 7 and 12 parts per million per °C (ppm / °C). For longer expansion elements, the linear expansion coefficient may be proportionally reduced (for shorter expansion elements, the linear expansion coefficient may be increased).

[0045] The temperature of the expansion element (and / or optical element) can be controlled by any suitable means, for example by resistive heating (integrated into the expansion element 24a, 24b as a film combined with the expansion element, or using the expansion material itself as a resistor) or coupling to a thermoelectric cooler. The temperature range of the expansion element and / or optical element can typically be 0°C to 150°C or can be a narrower range, such as 20°C to 100°C. The temperature range of the expansion element can cause an expansion or contraction of d to be at least half a wavelength or at least one wavelength of the desired optical signal. The temperature of the device can be measured by any suitable method, such as a thermistor. Alternatively, only the applied heating / cooling power can be measured and the temperature determined from this.

[0046] The distance d depends on the absolute temperature of the expansion element (or the optical elements if they have a high linear expansion coefficient). Therefore, the filter control must take the ambient temperature into account. This can be achieved by controlling the temperature of the expansion element based on a direct temperature measurement of the element, or by providing a constant ambient temperature using, for example, a thermoelectric cooler mounted on the base on which the filter is mounted.

[0047] Figure 3 An alternative structure is shown which eliminates the effect of ambient temperature. The structure has two optical elements 31a, 31b (made of optical material), each of which is connected to a respective inner expansion element 32a-d made of a first expansion material. The inner expansion element on each side is connected to an outer expansion element 33a, 33b made of a second expansion material. The arrangement of the inner and outer expansion elements is such that expansion of the inner expansion element will result in a decrease in d and expansion of the outer expansion element will result in an increase in d. The first and second expansion materials are selected to ensure that 2(α1L1+α c C) = α2L2, where α1 is the LCE of the first expansion material, L1 is the length of each inner expansion element, α2 is the LCE of the second expansion material, and L2 is the length of each outer expansion element. As mentioned above, if α1, α2 ≥ α c , then the expansion contribution of the optical material can be neglected. The distance between optical elements is given by:

[0048] d=L2-2(L1+C)=(L20+ΔL2)-2(L10+ΔL1+C)

[0049] =(L20+α2L20ΔT2)-2(L20+α1L10ΔT1+C)

[0050] =d0+α2L20ΔT2-2α1L10ΔT1=d0+α1L10(ΔT1-ΔT2)

[0051] =d0+α1L10(T1-T2)

[0052] Where T1 is the temperature of the inner expansion element and T2 is the temperature of the outer expansion element, d therefore depends only on the temperature difference between the inner and outer expansion elements, which can be controlled relatively easily without taking the ambient temperature into account, for example by separate heaters on the inner and outer expansion elements.

[0053] Apart from Figure 2A In addition to the structure shown, the filter can also be configured to be cylindrically symmetrical around the axis of the optical element, for example Figure 4 The expansion element 44 is annular in cross section, and the optical elements 21a and 21b are partially located in the hole and connected to the expansion element 44 at both ends. A similar cylindrical symmetrical structure can be used Figure 3 The cross section shown is constructed.

[0054] The filter can be mounted on a bracket 45 where the expansion element is located. The bracket 45 can be made of an expansion material or a material with similar thermal expansion characteristics (for example, the difference in linear thermal expansion coefficient does not exceed 5%, or the difference does not exceed 1%) to avoid strain caused by different expansion of the bracket 45 and the expansion element 44. In this case, the bracket can be constructed so that it only contacts the expansion element. Alternatively, the bracket can be made of an optical material or a material with similar thermal expansion characteristics. Through testing, it was found that the former method is more effective for filters with a single expansion element (such as Figure 4 The latter method performs better for filters with multiple expansion elements (such as Figure 2A However, both approaches are improvements over using a bracket that is not matched to either material.

[0055] To reduce heat transfer from the bracket to external components, the bracket can be mounted on standoffs made of thermally insulating material (such as glass or silica). The bracket can be used with any filter structure - if multiple expansion materials (such as Figure 3 ), the stent may be made of the expanded material it contacts.

[0056] like Figure 5A and Figure 5B As shown, the losses introduced by the filter depend on the alignment of the reflective surfaces of the optical element. Figure 5A This is a graph of the transmission of a perfectly aligned standard gage. Figure 5BFigure 2 is a graph of the transmittance of a gage with the reflective surface of the optic misaligned by 0.005° (18 arc seconds, all other properties are the same in both simulations). It can be seen that the peak transmittance of the perfectly aligned gage is very close to 1, and even in the case of very small misalignments, the peak transmittance of the misaligned gage is around 0.9.

[0057] Expansion of the expansion element can cause stress in the filter. This is because, where the expansion material joins the optical material, each material expands at a different rate. This effect can be mitigated by providing an elastic layer (e.g., a low modulus resin with a modulus less than 10^8 Newtons / meter (N / M) or about 10^7 N / M) between the optical material and the expansion material to absorb the stress. Figure 6 As shown, to ensure that the filter remains aligned, non-elastic spacer elements 61 can be placed within the elastic material 62 so that they maintain the distance between the optical material 63 and the expansion material 64. Such spacer elements can be spheres with a small radius (e.g., up to 30 microns). The spacer elements maintain the distance between the optical material and the expansion material, but since they are not directly connected to either material, they do not cause stress build up.

[0058] like Fig. 7A , Figure 7B and Figure 7C As shown, stresses can be further reduced by limiting the contact area between the expansion material and the optical material, for example by providing the expansion elements 71a, 71b, 71c with a reduced cross section close to the point where they abut the optical elements 72a, 72b, 72c. In the example shown, an elastic layer 73a, 73b, 73c is again shown at the abutment point. This also allows the expansion element to flex slightly in the area of ​​reduced cross section. The elastic layer 73a, 73b, 73c can be used to fill the space created in the expansion element, as shown in Figure 7c.

[0059] As an alternative to using an elastic material, the connection between the optical part and the expansion part can be arranged to allow sliding in a direction perpendicular to the optical axis of the FP standard gauge. This can be achieved, for example, by providing a force parallel to the axis that maintains close contact on the adjacent surfaces. The force can be applied by, for example, a spring, or it can be provided by bringing the two surfaces into close contact so that the Van der Waals forces between the two surfaces will form a bond. In order for the Van der Waals forces to be strong enough, the surfaces must be close to ideal planes and the gap between them must be close to or smaller than the wavelength.

[0060] The losses caused by misalignment can be further reduced by reducing the width of the beam passing through the filter. Fig. 8AAs shown, a narrower beam will experience less loss due to the misaligned filter, Fig. 8A The simulated propagation of beams ranging from 15 to 150 micrometers in width is shown for a misalignment of 0.005° (18 arc seconds). However, narrow beams experience higher losses due to diffraction effects, which means that the beam does not remain well collimated (regardless of the mirror alignment). Figure 8B 5 to 150 micron wide beams are shown in FIG. Therefore, a balance needs to be maintained between requiring a narrower beam to avoid misalignment losses and requiring a wider beam to avoid diffraction. Depending on the alignment that can be achieved and the requirements of the filter application, a beam width between 5 and 150 microns may be appropriate. Of course, standard quality control procedures can be used to separate out well-aligned filters for wider beam applications and less well-aligned filters for narrower beam applications.

[0061] Control and calibration of the filter can be achieved in several ways. The filter can be pre-calibrated and the temperature device control configured to change the temperature of the expansion element according to this calibration. Alternatively, active calibration can be implemented by providing a detector after the FP standard gauge. This detector can be used to measure the intensity of the transmitted light and a feedback loop can then be implemented to ensure that the filter is centered on the input main mode.

[0062] One possible feedback scheme is to add modulation to the thermal control, such as adding a dither (e.g., a sine wave) to the temperature changes of small amplitude, with a frequency lower than the thermal response of the expanding material (e.g., 20 Hz). This modulation can then be used to determine the location of the transmittance (intensity) peak using standard methods.

[0063] Alternatively, a second detector may be placed before the FP standard gauge, and the readings from the two detectors compared to determine the losses through the filter. If the losses exceed a threshold, the temperature may be adjusted to return to peak transmittance. However, this approach does not allow for a determination of whether the temperature is too high or too low, and so a secondary process is required to find the peak. For example, when the losses exceed a threshold, the temperature may be adjusted upward. If this movement reduces the losses, the temperature is adjusted further upward until the peak is found. If the initial upward movement does not reduce the losses, but instead increases them, the temperature is adjusted downward until the peak is found.

[0064] As a third option, a secondary beam with a wavelength within the shoulder of the transmission peak of the FP gage can be used. The secondary beam is arranged to pass through the filter at a different angle to the primary beam (same direction as the primary beam, or "backward" compared to the primary beam), and the ratio of the intensities of the primary and secondary beams after passing through the filter is measured. Since this gives 2 points of known wavelength and intensity on the slope leading to the peak, the ratio of these two intensities can be used to find the peak transmission.

[0065] Any of the above active calibration methods may be performed continuously in response to strength dropping below a threshold or loss exceeding a threshold, or may be performed periodically.

[0066] The thermal conductivity of the expansion material and the optical material will determine how quickly the filter responds to temperature changes. Generally, the filter will not be optimally calibrated until the expansion material and the optical material have reached equilibrium, as thermal expansion of the optical material can cause small changes in the distance d. This can be mitigated by choosing an expansion material with a much larger linear thermal expansion coefficient than the optical material and an expansion material with high thermal conductivity. Particularly suitable expansion materials include ceramics, glass and metals, especially copper, aluminum or stainless steel. The optical material should be selected to have a low coefficient of thermal expansion and a high thermal conductivity. A possible example is silica. 9A to 9D Shows the Figure 4 Thermal simulation of the designed filter, which is about 3 mm long, with copper as the expansion material and silicon dioxide as the optical material. It can be seen that the expansion material reaches thermal equilibrium very quickly (within the first 0.25 seconds), but the optical material takes relatively longer (about 3 seconds).

[0067] In an alternative structure where the LCE of the expansion material is less than the LCE of the optical material, the expansion material with a low LCE and the optical material with a high LCE should be selected. Suitable materials for the expansion material in this structure include silica, Invar or Zerodur (RTM). Suitable materials for the optical material include BK7 glass. As mentioned above, both materials should have high thermal conductivity.

[0068] To prevent back reflections from the etalon or within the etalon, the surface of the optical element opposite the reflective surface can be angled to the reflective surface, and the light beam to be filtered can be provided at a slight angle relative to an axis perpendicular to the reflective surface. In each case, a small angle (e.g., less than 1 degree) is sufficient.

[0069] Therefore, embodiments of the present invention provide the use of an expansion material that will control the width of the standard gauge cavity so as to tune the filter peak over the entire frequency. The length of the expansion material exceeds the cavity width and is a structure arranged outside the resonator optical path and the resonator mirror. The temperature of the thermal expansion material and the optical resonator can be controlled. The expansion material can be a metal or a ceramic or a glass or other material with greater expansion than the material comprising the resonator structure. The resonator mirror optical component can be bonded to the thermal expansion material in a manner that maintains the parallelism of the resonator gap to obtain high fineness and low loss at the resonant frequency. The bonding can include surface-to-surface contact or an intermediate material that allows the stress between the differential expansion materials to be minimized. The contact can be designed to slide by close contact or by a force that maintains close contact between the bonding surfaces, or can be designed to include a thin layer of a low elastic modulus material that can be partially filled with particles designed to maintain a specified thickness. The resonator can be bonded to a surface having a low elastic modulus material to minimize the stress on the filter structure caused by temperature changes. The temperature of the standard gauge can be changed by a thermoelectric cooler or a resistive heater. The temperature of the master gage can be controlled by a proximity temperature sensitive device such as a thermistor.

Claims

1. An optical filter, comprising: a first optical element; The second optical element, The first optical element and the second optical element are arranged along a common axis, and each of the first and second optical elements terminates in a respective planar surface perpendicular to the axis, The respective planar surfaces are separated by a gap having a width d between the first optical element and the second optical element such that the respective planar surfaces form a Fabry-Perot etalon and light can propagate along an optical path through the first optical element, the second optical element, and the etalon; and one or more expansion elements, the one or more expansion elements being located outside the light path and extending parallel to the common axis and having a length along the common axis that is greater than the gap width, the one or more expansion elements comprising a material having a different linear coefficient of thermal expansion than the first optical element and the second optical element, such that differential expansion of the one or more expansion elements and the first optical element and the second optical element causes relative movement of the respective planar surfaces along the axis, thereby causing a change in the gap width; The one or more expansion elements include a first set of expansion elements and a second set of expansion elements, The first set of expansion elements comprises: a first expansion element; the first expansion element being connected at a first end thereof to the first optical element and at a second end thereof to the second set of expansion elements; and a second expansion element; the second expansion element being connected at a first end thereof to the second optical element and at a second end thereof to the second set of expansion elements; and The expansion of the first set of expansion elements results in a decrease in the gap width, while the expansion of the second set of expansion elements results in an increase in the gap width, Wherein the gap does not have an expansion element therein, a change in the gap width enables a change in the transmission wavelength of the optical filter.

2. The optical filter according to claim 1, wherein The optical filter further comprises a temperature control mechanism for controlling the temperature of at least one of the one or more expansion elements, the first optical element, and the second optical element.

3. The optical filter according to claim 2, wherein: The temperature control mechanism includes a resistive heater or a thermoelectric cooler.

4. The optical filter according to claim 2, wherein: The optical filter further comprises: a detector for measuring the intensity of light transmitted by the filter, The temperature control mechanism is used to control the temperature according to the information received from the detector.

5. The optical filter according to claim 2, wherein: The temperature control mechanism is used to continuously change the temperature within a specific range that causes the width of the gap to change with a distance corresponding to a wavelength transmitted by the optical filter.

6. The optical filter according to claim 1, wherein The optical filter further comprises a temperature control mechanism for independently controlling the temperature of the first group of expansion elements and the second group of expansion elements.

7. The optical filter according to claim 1, wherein The first group of expansion elements and the second group of expansion elements are arranged such that a width of the gap depends on a temperature difference between the first group of expansion elements and the second group of expansion elements and is substantially unaffected by ambient temperature.

8. The optical filter according to claim 1, wherein The one or more expansion elements, the first group of expansion elements and the second group of expansion elements are in contact with each other.

9. The optical filter according to claim 1, wherein: The optical filter further comprises: A controller is configured to cause the temperature of the optical filter to be adjusted so that the value of the specific distance is set to cause a peak transmittance of the optical filter to be at an output frequency of a modulator associated with the optical filter.

10. An optical filter, comprising: a first optical element; The second optical element, The first optical element and the second optical element are arranged along a common axis, and the first optical element and the second optical element each terminate in a respective flat surface perpendicular to the axis, The respective planar surfaces are separated by a gap between the first optical element and the second optical element such that the respective planar surfaces form a Fabry-Perot etalon; as well as one or more expansion elements extending parallel to the common axis and having a length along the common axis that is greater than the gap width, The one or more expansion elements include a material having a different linear coefficient of thermal expansion than the first optical element and the second optical element; The one or more expansion elements include a first set of expansion elements and a second set of expansion elements, The second set of expansion elements comprises: a first expansion element; the first expansion element being connected at a first end thereof to the first optical element and at a second end thereof to at least one of the first set of expansion elements; and a second expansion element; the second expansion element being connected at a first end thereof to the second optical element and at a second end thereof to at least one of the first set of expansion elements; and The expansion of the first set of expansion elements results in an increase in the gap width, while the expansion of the second set of expansion elements results in a decrease in the gap width, Wherein the gap does not have an expansion element therein, the one or more expansion elements are arranged to cause a change in the width of the gap so as to enable a change in the transmission wavelength of the optical filter.

11. The optical filter according to claim 10, wherein: The optical filter further comprises a temperature control mechanism for controlling the temperature of at least one of the one or more expansion elements, the first optical element, and the second optical element.

12. The optical filter according to claim 11, wherein The temperature control mechanism includes a resistive heater or a thermoelectric cooler.

13. The optical filter according to claim 11, wherein: The optical filter further comprises: a detector for measuring the intensity of light transmitted by the optical filter, The temperature control mechanism is used to control the temperature according to the information received from the detector.

14. The optical filter according to claim 11, wherein: The temperature control mechanism is used to continuously change the temperature within a specific range that causes the width of the gap to change with a distance corresponding to a wavelength transmitted by the optical filter.

15. The optical filter according to claim 10, wherein: The optical filter further comprises a temperature control mechanism for independently controlling the temperature of the first group of expansion elements and the second group of expansion elements.

16. The optical filter according to claim 10, wherein: The first group of expansion elements and the second group of expansion elements are arranged such that a width of the gap depends on a temperature difference between the first group of expansion elements and the second group of expansion elements and is substantially unaffected by ambient temperature.

17. The optical filter according to claim 10, wherein: The one or more expansion elements, the first group of expansion elements and the second group of expansion elements are in contact with each other.

18. The optical filter according to claim 10, wherein: The optical filter further comprises: A controller is configured to cause the temperature of the optical filter to be adjusted so that the value of the specific distance is set to cause a peak transmittance of the optical filter to be at an output frequency of a modulator associated with the optical filter.

Images