Microbolometer with filtering function

Abstract

The present disclosure relates to a microbolometer comprising an array of pixels, each pixel comprising one or more detection units, each detection unit comprising an absorption layer (530) forming a quarter-wavelength resonant cavity (533) having a height (h) between 1.5 and 5 μm, wherein the spacing of the detection units on at least one axis in the plane of the pixel array is in the range of 2.4h to 3.6h.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This patent application claims priority to French patent application filed on June 5, 2019 and assigned application number FR1905965, the contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates generally to the field of infrared imaging, and more particularly to a microbolometer and a method of manufacturing a microbolometer. Background Art

[0004] A microbolometer is an uncooled infrared (IR) camera used to capture thermal images of an image scene. Such an IR camera typically comprises an arrangement of infrared sensitive detectors forming a pixel array. Each pixel of the pixel array converts the temperature measured at the pixel into a corresponding electrical signal, typically a voltage, which in turn is converted by an ADC (analog-to-digital converter) into a digital output signal.

[0005] Each pixel of the microbolometer comprises a membrane suspended on a substrate. The membrane comprises an absorbing layer that absorbs energy from the infrared light irradiating the pixel, so that its temperature rises according to the intensity of the infrared light. For example, the membrane also comprises a thermal layer having the property of changing its resistance by this temperature rise, and the pixel can therefore be read out by detecting the change in resistance of the thermal layer, which is thermally connected to the absorbing layer.

[0006] It is often desirable for a microbolometer to have relatively high sensitivity in the wavelength range of interest, usually implying high absorptivity. It is also desirable for the device to be relatively compact and relatively low cost.

[0007] However, there are technical problems in reducing the size and / or cost of microbolometers without reducing their sensitivity. In fact, the smaller the pixel pitch of a microbolometer, the lower the amount of absorbed power and the more difficult it is to design arms to support an absorption layer with a relatively high thermal resistance. Summary of the invention

[0008] An object of the embodiments of the present disclosure is to at least partially solve one or more problems in the prior art.

[0009] According to one embodiment, a microbolometer is provided that includes a pixel array, each pixel including one or more detection units, each detection unit including an absorption layer, wherein: the spacing of the detection units in at least one direction in the plane of the pixel array is between 5 and 11 μm; the pixel filling factor FF of the absorption layer of the one or more detection units in each pixel is in the range of 0.10 to 0.50; and the sheet resistance Rs of the absorption layer of each detection unit is between 16 and 189 ohm / sq.

[0010] According to one embodiment, the ratio Rs / FF for each pixel of the array is between 200 and 600 ohm / sq.

[0011] According to one embodiment, each pixel has a pixel fill factor in the range of 0.10 to 0.40.

[0012] According to one embodiment, each pixel has a pixel fill factor in the range of 0.20 to 0.40.

[0013] According to one embodiment, the pitch of the detection units in at least one direction in the plane of the pixel array is between 8 and 9 μm.

[0014] According to one embodiment, each pixel of the array has:

[0015] - the pixel fill factor FF is equal to or greater than 0.40 and less than 0.50, and the sheet resistance Rs of the absorption layer is at least 75 ohm / sq; or

[0016] - the pixel fill factor FF is equal to or greater than 0.30 and less than 0.40, and the sheet resistance Rs of the absorption layer is at least 50 ohm / sq; or

[0017] - the pixel fill factor FF is equal to or greater than 0.20 and less than 0.30, and the sheet resistance Rs of the absorption layer is at least 25 ohm / sq; or

[0018] - The pixel fill factor FF is equal to or greater than 0.10 and less than 0.20, and the sheet resistance Rs of the absorption layer is at least 16 ohm / sq.

[0019] According to one embodiment, the ratio Rs / FF is in the range of 377 ohm / sq ± 20%.

[0020] According to one embodiment, the absorption layer of each detection unit is a metal layer.

[0021] According to one embodiment, the absorber layer is formed of TiN and has a thickness between 10 and 115 nm.

[0022] According to one embodiment, a pixel array includes a substrate, and each pixel of the pixel array includes a reflective layer formed on the substrate and a membrane suspended on the reflective layer, a quarter-wavelength resonant cavity is formed between the membrane and the reflective layer in each pixel, and the membrane includes an absorption layer and a thermal layer.

[0023] According to one embodiment, the surface area of ​​the absorbent layer is less than 75% of the surface area of ​​the membrane.

[0024] According to one embodiment, the height of the quarter-wavelength resonant cavity is in the range of 1.5 to 3.5 μm.

[0025] According to one embodiment, the pitch of the detection units in at least one direction in the plane of the pixel array is smaller than four times the height of the quarter-wavelength resonant cavity.

[0026] According to another aspect, a method for manufacturing a microbolometer array is provided, the method comprising: forming an array of pixels, each pixel having one or more detection units, wherein forming the array comprises: forming the detection units having a pitch of 5 to 11 μm on at least one axis of a pixel array plane; and forming each detection unit to include an absorption layer having a pixel fill factor FF in the range of 0.10 to 0.50 and a sheet resistance Rs between 16 and 189 ohm / sq.

[0027] According to yet another aspect, a microbolometer is provided, comprising an array of pixels, each pixel comprising one or more detection units, each detection unit comprising an absorption layer forming a quarter-wavelength resonant cavity with a height h between 1.5 and 5 μm, wherein the spacing of the detection units on at least one axis in the plane of the pixel array is in the range of 2.4h to 3.6h.

[0028] According to one embodiment, the pixel filling factor FF of the absorption layer of one or more detection units in each pixel is in the range of 0.20 to 0.70.

[0029] According to one embodiment, the pixel filling factor FF of the absorption layer of one or more detection units in each pixel is in the range of 0.10 to 0.50.

[0030] According to one embodiment, the pitch of the detection units is in the range of 4 to 15 μm.

[0031] According to one embodiment, the pitch of the detection units is in the range of 5 to 11 μm.

[0032] According to one embodiment, the absorber layer is a metal layer having a sheet resistance of 189 ohm / sq or less.

[0033] According to one embodiment, the absorber layer is a metal layer having a sheet resistance of 126 ohm / sq or less.

[0034] According to one embodiment, the absorber layer is formed of titanium nitride.

[0035] According to one embodiment, the resonant cavity height is between 1.5 and 3.5 μm.

[0036] According to one embodiment, each pixel of the array has:

[0037] - the pixel fill factor FF is equal to or greater than 0.40 and less than 0.50, and the sheet resistance Rs of the absorption layer is at least 75 ohm / sq; or

[0038] - the pixel fill factor FF is equal to or greater than 0.30 and less than 0.40, and the sheet resistance Rs of the absorption layer is at least 50 ohm / sq; or

[0039] - the pixel fill factor FF is equal to or greater than 0.20 and less than 0.30, and the sheet resistance Rs of the absorption layer is at least 25 ohm / sq; or

[0040] - The pixel fill factor FF is equal to or greater than 0.10 and less than 0.20, and the sheet resistance Rs of the absorption layer is at least 16 ohm / sq.

[0041] According to one embodiment, the ratio Rs / FF for each pixel of the array is between 200 and 600 ohm / sq.

[0042] According to one embodiment, the ratio Rs / FF for each pixel of the array is within 20% of 377 ohm / sq.

[0043] According to one embodiment, each detection unit comprises a membrane comprising an absorbent layer, a thermal layer and a dielectric layer.

[0044] According to another aspect, a method of manufacturing a microbolometer is provided, the method comprising forming an array of pixels, each pixel comprising one or more detection units, wherein forming the array comprises: forming each detection unit to include an absorption layer, the absorption layer forming a quarter-wavelength resonant cavity having a height h between 1.5 and 5 μm; and forming the detection units to have a spacing in the range of 2.4h to 3.6h on at least one axis of the plane of the pixel array. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The above features and advantages and other features and advantages will be described in detail in the following description of specific embodiments given by way of illustration and without limitation with reference to the accompanying drawings, in which:

[0046] Figure 1schematically illustrates an image capture circuit of an infrared camera according to an example embodiment;

[0047] Figure 2 is a cross-sectional view of a pixel of a microbolometer according to an example embodiment;

[0048] Figure 3 is a plan view of a portion of a microbolometer array according to an example embodiment;

[0049] Figure 4 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer with a pixel pitch of 17 μm, a Fabry-Perot cavity height of 2.5 μm and a light wavelength of 10 μm;

[0050] Figure 5 is a perspective view of a pixel of a microbolometer according to an exemplary embodiment of the present disclosure;

[0051] Figure 6 According to an exemplary embodiment of the present disclosure Figure 5 a cross-sectional view of a portion of a pixel;

[0052] Figure 7 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 8.5 μm, the Fabry-Perot cavity height is 2.5 μm and the wavelength of light is 10 μm;

[0053] Figure 8 is a plan view of a portion of a microbolometer array according to an exemplary embodiment of the present disclosure;

[0054] Fig. 9 is a plan view of a portion of a microbolometer array according to another example embodiment of the present disclosure;

[0055] Fig.10 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for a number of different sheet resistances of the absorber layer with a pixel pitch of 8.5 μm, a Fabry-Perot cavity height of 2.5 μm and a light wavelength of 10 μm;

[0056] Fig.11 is a graph showing the absorptivity gain of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for a number of different sheet resistances of the absorber layer with a pixel pitch of 8.5 μm, a Fabry-Perot cavity height of 2.5 μm and a light wavelength of 10 μm;

[0057] Fig.12 is shown in more detail Fig.11 A graph of a region of the graph corresponding to a gain of 20% or more;

[0058] Fig.13 is a more detailed representation Fig.11 A graph of a region of the graph corresponding to a gain of 50% or more;

[0059] Fig.14 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the sheet resistance / fill factor ratio for different fill factors;

[0060] Fig.15 is shown in more detail Fig.14 A graph of a region of the graph corresponding to an absorbance above 0.9;

[0061] Fig.16 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 11 μm, the Fabry-Perot cavity height is 3.5 μm and the light wavelength is 13 μm;

[0062] Fig.17 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 5 μm, the Fabry-Perot cavity height is 1.5 μm and the light wavelength is 6 μm;

[0063] Fig.18 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 5 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0064] Fig.19 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 6 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0065] Fig. 20 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 7 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0066] Fig.21 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 8 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0067] Fig. 22 is a graph showing the absorptivity of the absorption layer of a microbolometer pixel as a function of the pixel fill factor for eight different thicknesses of the absorption layer when the pixel pitch is 9 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0068] Fig.23 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 10 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0069] Fig.24 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 11 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0070] Fig.25 is a graph showing the absorptivity of the absorber layer of a microbolometer pixel as a function of the pixel fill factor for eight different sheet resistances of the absorber layer when the pixel pitch is 12 μm, the Fabry-Perot cavity height is 2.5 μm and the light wavelength is 10 μm;

[0071] Fig.26A is a graph showing the absorptivity of the absorption layer of a microbolometer pixel as a function of the wavelength of light for a pixel pitch of 8.5 μm and a Fabry-Perot cavity height of 2.5 μm for seven different pixel fill factors;

[0072] Figures 26B to 26G is a graph showing the absorbance of the absorber layer of a microbolometer pixel as a function of the wavelength of light for seven different pixel fill factors, where Fig.26B and 26C corresponds to a Fabry-Perot cavity height of 2.5 μm and to pixel pitches of 6 μm and 9 μm, respectively; Fig.26D and 26E corresponding to a Fabry-Perot cavity height of 1.5 μm and to pixel pitches of 3.6 μm and 5.4 μm, respectively; Fig.26F and26G corresponds to a Fabry-Perot cavity height of 5 μm and to pixel pitches of 12 μm and 18 μm, respectively;

[0073] Figures 26H to 26R is a graph showing the absorptivity of an absorber layer of a microbolometer pixel having a 2.5 μm Fabry-Perot cavity height as a function of the wavelength of light for seven different pixel fill factors, wherein the graphs show examples of pixel pitches of 5 μm, 5.5 μm, 6 μm, 6.5 μm, 7 μm, 7.5 μm, 8 μm, 8.5 μm, 9 μm, 9.5 μm and 10 μm, respectively;

[0074] Fig. 27 is a graph showing the ratio of non-absorbed diffracted power as a function of light wavelength for seven different pixel fill factors, a pixel pitch of 8.5 μm, and a Fabry-Perot cavity height of 2.5 μm;

[0075] Fig.28 and 29 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel with an absorption layer having a thickness of 6 nm, a pixel filling factor of 0.3, and for a light wavelength of 6 μm, respectively;

[0076] Fig.30 and Fig.31 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel with an absorption layer having a thickness of 6 nm, a pixel filling factor of 0.3, and for a light wavelength of 8 μm, respectively;

[0077] Fig.32 and 33 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel for an absorption layer with a thickness of 6 nm, a pixel fill factor of 0.3, and for a light wavelength of 10 μm, respectively;

[0078] Fig.34 and Fig.35 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel with an absorption layer having a thickness of 6 nm, a pixel filling factor of 0.3, and for a light wavelength of 12 μm, respectively;

[0079] Fig.36 and Fig.37 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel with an absorption layer having a thickness of 6 nm, a pixel filling factor of 0.3, and for a light wavelength of 14 μm, respectively;

[0080] Fig.38 and 39are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel for an absorption layer with a thickness of 22 nm, a pixel fill factor of 0.3, and for a light wavelength of 6 μm, respectively;

[0081] Fig.40 and 41 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel for an absorption layer with a thickness of 22 nm, a pixel fill factor of 0.3, and for a light wavelength of 8 μm, respectively;

[0082] Fig.42 and Fig.43 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel for an absorption layer with a thickness of 22 nm, a pixel fill factor of 0.3, and for a light wavelength of 10 μm, respectively;

[0083] Fig.44 and Fig.45 are diagrams showing the spatial distribution of electric and magnetic fields in a microbolometer pixel for an absorption layer with a thickness of 22 nm, a pixel fill factor of 0.3, and for a light wavelength of 12 μm, respectively; and

[0084] Fig.46 and 47 are diagrams showing the spatial distribution of the electric and magnetic fields in a microbolometer pixel with an absorption layer having a thickness of 22 nm, a pixel fill factor of 0.3, and for a light wavelength of 14 μm, respectively. DETAILED DESCRIPTION

[0085] Similar features have been designated by similar references in the various figures. In particular, common structural and / or functional features in the various embodiments may have the same references and may be configured with the same structure, dimensions, and material properties.

[0086] For the sake of clarity, only the operations and elements that are helpful for understanding the embodiments described herein are described in detail and described. In particular, the circuits for measuring the pixel resistance in the microbolometer array are not described in detail, and the methods for processing the captured pixel data to generate thermal images are not described in detail.

[0087] In the following disclosure, unless otherwise indicated, when referring to absolute position qualifiers, such as the terms "front", "rear", "up", "down", "left", "right", etc., or to relative position qualifiers, such as the terms "above", "below", "higher than", "lower than", etc., or to orientation qualifiers, such as "horizontal", "vertical", etc., reference is made to the direction shown in the figures, or to the microbolometer as it is oriented during normal use.

[0088] Unless otherwise specified, "about," "approximately," "substantially," and "approximately" mean within 10%, and preferably within 5%.

[0089] In the following description, the following terms will be deemed to have the following definitions.

[0090] "Absorption layer": A layer that absorbs energy from infrared light in a pixel of a microbolometer. In an embodiment of the present disclosure, this layer is formed of a metal such as titanium nitride (TiN), titanium (Ti), or platinum (Pt).

[0091] "Pixel fill factor": the ratio between the surface area of ​​the absorber layer and the surface area of ​​the pixel. In this disclosure, the value of the pixel fill factor is set with two decimal places of precision.

[0092] "Pixel pitch" or "detection element pitch": the spacing at which pixels / detection elements are formed in a microbolometer array. The pixel pitch may correspond to the width of each pixel or detection element in the x or y direction, or the distance from the edge of one pixel or detection element to the corresponding edge of an adjacent pixel or detection element. In the present disclosure, the pixel and detection element pitches are set in micrometers with an accuracy of one decimal place, and in some cases two decimal places.

[0093] In this disclosure, the values ​​of sheet resistance are assumed to be accurate to within 5%, and the thicknesses expressed in nanometers are assumed to be accurate to the nearest nanometer.

[0094] Figure 1 An image capture device 100 of an infrared camera according to an exemplary embodiment is schematically shown. The device 100 includes an array 102 of pixels 104 forming a microbolometer capable of operating at ambient temperature. For example, the array includes N columns and M rows, where N and M are, for example, respectively equal to 2 or greater, and can be equal to more than 1000. The output of the array 102 is coupled to a readout circuit 106, which is, for example, a readout integrated circuit (ROIC). For example, the circuit 106 includes one or more analog-to-digital converters for converting the signal captured in the microbolometer array 102 into a digital signal S. For example, the digital signal S is provided to a processing device (P) 108, which performs processing of the raw image data to generate a thermal image I. For example, the device 100 is capable of capturing a still thermal image or forming a video stream of thermal images.

[0095] Figure 2 yes Figure 11 is a more detailed cross-sectional view of one of the pixels 104 of the microbolometer array 102. The pixel 104 includes a membrane 202 suspended on a reflective surface 204 formed on a substrate 206. The membrane 202 is supported by arms 208, which also provide thermal insulation between the substrate 206 and the membrane 202, for example.

[0096] For example, the film 202 includes across at least a portion of its surface area an absorbent layer 210 , a thermal layer 212 , and dielectric layers 214 , one of which isolates the absorbent and thermal layers 210 , 212 from one another and the other two of which sandwich the layers 210 , 212 .

[0097] Absorption layer 210 is formed, for example, of a metal, such as titanium nitride, titanium, platinum, or other metal. Absorption layer 210 absorbs energy from infrared light striking the pixel and is thereby heated. This heat is transferred to thermal layer 212, which is formed of a material having a resistance that varies with temperature. Contacts C1 and C2 near the edge of thermal layer 212 allow for the readout circuit ( Figure 2 The resistance of the thermal layer of each pixel is measured via arm 208 (not shown).

[0098] The thickness Em of the film 202 is typically in the range of 100 nm to 1 μm.

[0099] The space between the surface of the film 202 and the reflective layer 204 defines a quarter-wavelength resonant cavity 216, also known as a Fabry-Perot cavity. Typically, the resonant cavity 216 is filled with air or is under a partial vacuum. The height h of the resonant cavity is selected to achieve a relatively high absorptivity of infrared light at a desired wavelength range by the absorption layer 210 of the film 202. In particular, the height h of the resonant cavity is selected to be equal to λtg / 4, where λtg is the target wavelength.

[0100] The height h is, for example, in the range of 0.50 to 5.0 μm, and in some embodiments in the range of 1.5 to 3.5 μm. In the following, unless otherwise stated, the examples and simulations are based on a cavity height of 2.5 μm, which provides relatively high absorption for light wavelengths centered around 10 μm. In fact, the spectrum near this wavelength is usually of most interest for thermal infrared applications. However, in alternative embodiments, different wavelengths of light may be targeted.

[0101] Figure 3 According to an example embodiment, it is shown in more detail Figure 1 A portion of the microbolometer array 102. Figure 3 In FIG. 1 , the dashed grid represents the boundaries of the pixels of the array, and the solid rectangle represents the surface area of ​​the absorption layer 210 for each pixel.

[0102] Figure 3The x-axis in is defined as the direction corresponding to the pixel rows in the plane of the microbolometer array, and Figure 3 The y-axis in is defined as the direction corresponding to the pixel columns in the plane of the microbolometer array.

[0103] For example, the pixels in the microbolometer array have a pixel pitch Px in the x-direction and a pixel pitch Py in the y-direction, and the pixel pitches Px and Py are, for example, substantially the same. In the following, when reference is made to a specific value of the pixel pitch, it is assumed that this is the pitch in the x and y directions unless otherwise stated. The area Apix of the pixel is equal to Px·Py. The absorption layer 210 has, for example, a width Wx in the x-direction that is shorter than the pitch Px, and a width Wy in the y-direction that is shorter than the pitch Py. Therefore, the surface area Aabs of the absorption layer is equal to Wx·Wy. The pixel fill factor FF of the absorption layer within each pixel can be defined as Aabs / Apix.

[0104] First aspect

[0105] In order to provide a relatively high absorptivity of the microbolometer, and therefore a high sensitivity, the fill factor FF is usually chosen to be as close to 1 as possible, as now referred to Figure 4 described.

[0106] Figure 4 is a graph showing the performance of a microbolometer pixel (such as Figures 1 to 3 A graph showing the functional relationship between the absorption rate (ABSORPTION RATE) of the pixel 104 and the fill factor (FF). Figure 4 The sheet resistance of curves 401 to 408 and the thickness of the titanium nitride absorption layer are shown in the following table:

[0107] [Table 1]

[0108]

[0109]

[0110] from Figure 4 It can be seen that using a relatively thin absorber layer with a thickness of 5 or 10 nm and using a fill factor above 0.65, an absorptivity above 0.9 can be achieved. The best absorptivity is obtained using an absorber layer with a thickness of 5 nm and a fill factor close to 0.90. However, for pixels with a pitch of 17.0 μm, the achievable fill factor is typically around 0.60.

[0111] In order to reduce the size of the microbolometer array while maintaining the number of pixels, it is desirable to reduce the pixel pitch. However, it is difficult to achieve a reduction in pixel pitch without significantly reducing the fill factor. In addition, Figure 4 The curves indicate that reducing the fill factor will result in a significant decrease in absorption, which is undesirable.

[0112] The inventors have unexpectedly discovered that when the pixel pitch on at least one axis (and preferably both axes) is reduced to a value in the range of 5 to 11 μm, and preferably in the range of 8 to 9 μm, higher absorption rates can still be achieved by using a relatively low fill factor between 0.10 and 0.50 and selecting an absorption layer 210 with a sheet resistance between 16 and 189 ohm / sq and in some embodiments between 16 and 130 ohm / sq.

[0113] Lowering the sheet resistance of the absorber layer means increasing its thickness. For example, for an absorber layer of titanium nitride, to achieve a sheet resistance of less than 200 ohm / sq, its thickness is at least 10 nm, and to achieve a sheet resistance of less than 130 ohm / sq, its thickness is at least 15 nm.

[0114] This increase in absorber layer thickness increases the thermal conduction between the absorber layer and the substrate, leading to poor performance. In particular, Ref. Figure 2 For pixels having a thickness of 210, increasing the thickness of layer 210 will increase the heat conduction through arm 208 to substrate 206, resulting in reduced sensitivity. Figure 5 A modified pixel having a relatively low fill factor and relatively low thermal conduction to the substrate is described in more detail.

[0115] Figure 5 is a perspective view of a pixel 500 of a microbolometer according to an example embodiment of the present disclosure.

[0116] Pixel 500 , for example, includes a membrane 502 suspended over a reflective layer 504 formed on a substrate 506 .

[0117] The membrane 502 is supported, for example, by a pair of arms 518, 520, which are anchored to two posts 508 formed at opposite corners of the pixel, respectively. Each post 508 includes a base portion 510 from which a post 512 extends. The posts 512 of the two posts pass through the respective ends 514, 516 of the arms 518, 520 and enter the cover 522. The arms 518, 520 connect the posts to the membrane 502 and, in particular, provide the function of mechanically supporting the membrane 502, providing a connection between the membrane 502 and, for example, the ROIC ( Figure 5The arms 518 and 520 are provided with the functions of providing electrical connection between the film 502 and the substrate 506 (not shown), and providing thermal insulation between the film 502 and the substrate 506. In order to provide good thermal insulation, the arms 518 and 520 are relatively long, for example, to provide relatively high thermal resistance. Figure 5 In the example of FIG. 5 , arms 518, 520 extend parallel to and separate from opposite edges of film 502 and are attached to opposite corners of film 502 via connecting portions 524. Thus, arms 518, 520, for example, substantially extend the length of one edge of film 502. In some embodiments, each of arms 518, 520 has a length that is at least 50% of the pixel pitch.

[0118] For example, the film 502 includes, in a portion of its surface area, a stack 526 including a thermal layer 528 and an absorber layer 530, which are composed, for example, of a dielectric layer ( Figure 5 518, 520). For example, stack 526 occupies less than 75% of the surface area of ​​film 502, including the surface of arms 518, 520. The area of ​​film 502 surrounding stack 526 forms a support layer for stack 526 and also provides electrical connection between points near the edge of thermal layer 528 and arms 518, 520 so that the resistance of thermal layer 528 can be measured by a readout circuit. For example, Figure 5 The dashed lines in the figure represent schematic diagrams of electrical connections 531, 532 extending from the arms 518, 520, respectively, to opposite edges of the thermal layer 528. These electrical connections 531, 532 and the arms 518, 520 are formed, for example, by metal layers of titanium nitride, titanium, platinum or other metals, which are sandwiched between dielectric layers formed, for example, of silicon nitride, silicon dioxide, silicon oxynitride or other electrically insulating materials.

[0119] Although in Figure 5 In the embodiment of FIG. 5 , the absorber layer 530 is formed on the thermal layer 528 , but in alternative embodiments the order may be reversed, for example, the absorber layer 530 is formed on the bottom surface of the membrane 502 .

[0120] The resonant cavity 533 between the reflective layer 504 and the membrane 502 forms a quarter-wavelength resonant cavity having a height h. In one example, the height h is 2.5 μm to target a wavelength of light of about 10 μm, although in alternative embodiments, different heights, such as in the range of 1.5 to 3.5 μm, may be used to target different wavelengths of light.

[0121] exist Figure 5 In the example of FIG. 5 , the membrane 502 is substantially square, but includes cutouts 534, 536 at opposite corners to leave space for the posts 508. The stack 526 is formed, for example, in a portion of the membrane 502 between these cutouts, extending to almost the entire length of the membrane 502.

[0122] Of course, despite Figure 5 An example of an implementation of a microbolometer having an absorption layer with relatively low sheet resistance is provided, but it will be apparent to those skilled in the art that many different implementations may be implemented, such as omitting the base portion 510 and / or the cover 522, and / or using different forms of arms 518, 520 and electrical connections 531, 532.

[0123] Figure 6 It is along Figure 5 A cross-sectional view taken along line AA′ passing through membrane 502 and arms 518 , 520 .

[0124] like Figure 6 As shown, the membrane 502 is formed, for example, by a dielectric layer 602, wherein electrical connections 531, 532 are formed, for example, by metal deposition. The thermal layer 528 and the absorber layer 530 are electrically insulated from each other, for example, by a dielectric layer 604, and the area near the bottom edge of the thermal layer 528, for example, contacts the top surface of the electrical connections 531, 532, and is otherwise covered by a dielectric material. In one embodiment, the absorber layer 530 is a titanium nitride layer having a thickness of at least 10 nm and in some cases at least 15 nm. For example, the thermal layer 528 has a thickness of approximately 100 nm.

[0125] It should be noted that although the electrical connections 531, 532 increase the absorptive surface area of ​​the pixel, given the relatively low thickness and high sheet resistance of these metal layers, the inventors have found that their effect on the effective optical fill factor of the pixel is relatively small and can therefore be ignored.

[0126] Figure 7 is a graph showing the absorption rate (ABSORPTIONRATE) of the absorption layer and the microbolometer pixel (such as Figure 5 A graph showing the functional relationship between the fill factor (FF) of the pixel 500) and the pitch of the pixel 50 ... Figure 4 As in the example, Figure 7 Based on a quarter-wavelength resonant cavity with a height of 2.5 μm and a light wavelength of 10 μm. Assuming that the absorption layer 530 is formed of titanium nitride, and Figure 7 The sheet resistances of the absorption layer 530 corresponding to the curves 701 to 708 of Table 1 are the same as those of the curves 401 to 408 of Table 1, respectively.

[0127] from Figure 7 It can be seen that in the case of curves 702 to 708, the optimum absorption rate can be achieved when the fill factor is equal to 0.50 or lower, and in the case of curves 703 to 708, it can be achieved to 0.40 or lower.

[0128] from Figure 7It can be seen that below a fill factor of 0.20, the absorptivity drops off rapidly, even for the thickest absorber layer. Thus, in some embodiments, the fill factor may be selected to be equal to or greater than 0.20.

[0129] Therefore, from Figure 7 It can be seen that by reducing the fill factor of the absorption layer of each pixel to the range of 0.1 to 0.5, and also reducing the sheet resistance of the absorption layer to less than 189 ohm / sq, and in some embodiments less than 130 ohm / sq, a microbolometer with relatively low pixel pitch and high absorptivity can be achieved.

[0130] Now refer to Figure 8 and Fig. 9 Examples of microbolometer arrays having a fill factor of 0.5 or less are described in more detail.

[0131] Figure 8 A portion of a microbolometer array 800 is shown according to an example embodiment. Figure 3 Similarly, the dashed grid represents the boundaries of the pixels 804 of the array, and the solid rectangles represent the surface area of ​​each absorber layer 530. Figure 3 Example, the pixel filling factor FF is Figure 8 has been reduced in the array, for example, to a value in the range of 0.10 to 0.50. In addition, Figure 8 The absorption layer 530 in the embodiment is not a square but a rectangle, and the width Wx on the x-axis is less than the width Wy on the y-axis. The pixels are, for example, square, and the pixel pitch Px in the x-direction is equal to the pixel pitch Py in the y-direction, although these pitches may be different in some embodiments. For example, the pitches Px and Py are each in the range of 5.0 to 11.0 μm.

[0132] exist Figure 8 In the example of FIG. 5 , each pixel 804 of the array corresponds to a detection unit including a single absorption layer 530. In alternative embodiments, each pixel may include more than one detection unit, as will now be described in conjunction with FIG. Fig. 9 described.

[0133] Fig. 9 FIG. 9 is a diagram showing a portion of a microbolometer array 900 of microbolometers according to another example embodiment. Fig. 9 In the example of FIG. 5 , the dashed grid represents the boundaries of the pixels 904 of the array, and the solid rectangle represents the surface area of ​​the absorption layer 530. Each pixel 904 includes more than one detection unit, each detection unit having a corresponding absorption layer 530. Fig. 9In the example of , each pixel 904 includes four detection units arranged two by two. The detection units of each pixel 904 are, for example, coupled together so that they generate a single pixel value, and thus each group of detection units can be considered to form a single pixel of the array.

[0134] exist Fig. 9 In this case, the pixel fill factor FF becomes equal to the ratio Aabs / Apix, where Apix is ​​the surface area of ​​each pixel and Aabs is the combined surface area of ​​the absorption layer 530 in each pixel.

[0135] exist Fig. 9 In this case, the relevant pitch is no longer the pixel pitch, but the pitch of the detection units, that is, the pitch of the absorption layer 530. Fig. 9 In the embodiment of the present invention, these spacings are marked as Px and Py in the x and y directions, respectively. For example, these spacings Px, Py are measured from the edge of one absorption layer 530 to the corresponding edge of an adjacent absorption layer 530, respectively.

[0136] Fig.10 is similar to Figure 7 , for the same type of microbolometer pixel, but further showing the curve for sheet resistance down to 16 ohm / sq. Fig.10 Curves 1001 to 1008 of Table 1 represent absorber layers having the same sheet resistances as those of curves 401 to 408 of Table 1. Curves 1009 to 1023 correspond to the sheet resistances and titanium nitride thicknesses shown in the following table.

[0137] [Table 2]

[0138] Reference curve Titanium nitride thickness (nm) Sheet resistance (ohm / sq) 1009 45 42 1010 50 38 1011 55 34 1012 60 31 1013 65 29 1014 70 27 1015 75 25 1016 80 24 1017 85 22 1018 90 21 1019 95 20 1020 100 19 1021 105 18 1022 110 17 1023 115 16

[0139] from Fig.10 It can be seen that providing microbolometer pixels with fill factors between 0.10 and 0.50 and sheet resistances between 16 and 189 ohm / sq can result in a significant reduction in pixel pitch (in Fig.10 5 μm) while achieving an absorptivity of at least 0.3, or, for a sheet resistance between 42 and 126 ohm / sq, at least 0.45.

[0140] Furthermore, while lower sheet resistance levels of less than 50 ohm / sq may not achieve an absorber ratio approaching 1.0, for fill factors between 0.10 and 0.30, they can provide significant gains over using an absorber layer having a sheet resistance approaching 377 ohm / sq corresponding to curve 1001, which will now be referenced. Figures 11 to 13 Explain in more detail.

[0141] Fig.11 is a graph showing curves 1102 to 1135, which represent the relative Fig.10 The absorption gain of the absorption layer represented by the curve 1001 in FIG. Fig.11 The curves 1102 to 1123 represent the corresponding Fig.10 The curves 1002 to 1023 are the gain of the absorption layer. Fig.11 Curves 1124 to 1135 in FIG. 1 represent an absorber layer having a thickness increasing in 5 nm increments from 120 to 190 nm. Curves 1126 to 1134 are not shown in FIG. Fig.11 , but can be easily identified by the sequence between curves 1125 and 1135.

[0142] from Fig.11 It can be seen that detectable gain can be achieved in either of the following cases:

[0143] - Pixel fill factor equal to or greater than 0.40 and less than 0.50, and sheet resistance between 75 and 189 ohm / sq; or

[0144] - A pixel fill factor equal to or greater than 0.30 and less than 0.40 and a sheet resistance between 47 and 189 ohm / sq; or

[0145] - A pixel fill factor equal to or greater than 0.20 and less than 0.30, and a sheet resistance between 25 and 189 ohm / sq; or

[0146] -Pixel fill factor is equal to or greater than 0.10 and less than 0.20, and sheet resistance is between 16 and 189 ohm / square.

[0147] Fig.12 Describes in more detail Fig.11 curves 1102 to 1125, and in particular gains of 20% or more. Fig.12 It can be seen that gains of more than 20% can be achieved in any of the following situations:

[0148] - Pixel fill factor equal to or greater than 0.30 and less than 0.40, and sheet resistance between 94 and 189 ohm / sq; or

[0149] - A pixel fill factor equal to or greater than 0.20 and less than 0.30 and a sheet resistance between 38 and 189 ohm / sq; or

[0150] -Pixel fill factor is equal to or greater than 0.10 and less than 0.25, and sheet resistance is between 22 and 189 ohm / sq.

[0151] Fig.13Describes in more detail Fig.11 curves 1102 to 1115, and in particular gains of 50% or more. Fig.13 It can be seen that when the pixel fill factor is equal to or greater than 0.10 and less than 0.24, and the sheet resistance is between 34 and 189 ohm / sq, a gain of more than 50% can be achieved.

[0152] Fig.14 is for fill factors of 0.10 (curve 1401) and 0.20 to 0.90 (curves 1402 to 1409, Fig.14 1402 and 1409), a graph of the absorptivity as a function of the ratio of sheet resistance (Rs) to fill factor (FF). As in the previous example, Fig.14 The curve corresponds to an absorption layer formed of titanium nitride, with a pixel pitch of 8.5 μm, a quarter-wavelength resonant cavity height of 2.5 μm, and a light wavelength of 10 μm.

[0153] Fig.15 More details on the absorption rate greater than or equal to 0.90 Fig.14 1402 to 1409 in FIG. 1403. It can be seen that when the fill factor is between 0.10 and 0.50, and the ratio Rs / FF is in the range of 200 to 600 ohm / sq, an absorption rate exceeding 0.90 can be achieved. Fig.15 It can also be noted that the curve is substantially centered around a ratio of 377 ohm / sq, where 377 ohm corresponds to the impedance of free space Z0. In fact, for example, a high absorption rate of more than 0.93 can be observed when the ratio Rs / FF is equal to 377±40%, and even a higher absorption rate of more than 0.95 can be observed when the ratio Rs / FF is equal to 377±20%.

[0154] The above examples are based on a quarter-wavelength resonant cavity height of 2.5 μm. The principles described in connection with these examples are equally applicable to different quarter-wavelength resonant cavity heights, for example for quarter-wavelength resonant cavity heights in the range of 1.5 to 3.5 μm, which will now be referred to Fig.16 and Fig.17 Provide explanation.

[0155] Fig.16 ABSORPTION RATE vs. microbolometer pixel (such as Figure 5Assume that the absorption layer 530 is formed of titanium nitride and the sheet resistance of the absorption layer 530 (corresponding to Fig.16 The sheet resistances of curves 1601 to 1608 of Table 1 are respectively the same as those of curves 401 to 408 of Table 1.

[0156] from Fig.16 It can be seen that the results are Figure 7 The results were similar.

[0157] Fig.17 ABSORPTION RATE vs. microbolometer pixel (such as Figure 5 Assume that the absorption layer 530 is formed of titanium nitride and the sheet resistance of the absorption layer 530 (corresponding to Fig.17 The sheet resistances of curves 1701 to 1708 of Table 1 are respectively the same as those of curves 401 to 408 of Table 1.

[0158] from Fig.17 It can be seen that the gain shown by curves 1702 to 1708 with respect to curve 1701 is more significant for fill factors in the range of 0.10 to 0.50, but the best results are obtained when the sheet resistance is less than 75 ohm / sq.

[0159] More generally, the principles described herein are applicable, for example, to any pixel pitch in the range of 5 to 11 μm, where the quarter-wavelength resonant cavity height is, for example, equal to λtg / 4, where λtg is the wavelength of light of interest (target wavelength), and the pitch is less than λtg. Thus, with a quarter-wavelength resonant cavity height of 1.5 μm, the pixel pitch is, for example, less than 6 μm, and with a quarter-wavelength resonant cavity height of 3.5 μm, the pixel pitch is, for example, anywhere in the range of 5 to 11 μm.

[0160] Figure 7 and Figures 10 to 13 The results are based on pixels with a pitch of 8.5 μm. The principles described with respect to these examples can be applied to pixels with a pitch anywhere in the range of 5 to 11 μm, as will now be referred to. Figures 18 to 25 Described in more detail.

[0161] Figures 18 to 25, which shows that with a quarter-wavelength resonant cavity height of 2.5 μm and a light wavelength of 10 μm, for eight different sheet resistances of the absorption layer 530, the absorption rate (ABSORPTION RATE) and the microbolometer pixel (such as Figure 5 Assume that the absorption layer 530 is formed of titanium nitride and corresponds to Figures 18 to 25 The sheet resistances of the absorption layer 530 of the curves i01 to i08 (for i from 18 to 25) are the same as the sheet resistances of the curves 401 to 408 of Table 1, respectively.

[0162] Fig.18 An example with a pixel pitch of 5 μm is shown, Fig.19 For example, the pixel pitch is 6 μm. Fig. 20 For example, the pixel pitch is 7 μm. Fig.21 For example, the pixel pitch is 8 μm. Fig. 22 For example, the pixel pitch is 9 μm. Fig.23 For example, the pixel pitch is 10 μm. Fig.24 is an example of a pixel pitch of 11 μm, and Fig.25 This is an example of a pixel pitch of 12 μm.

[0163] like Figures 18 to 22 As shown, for a pixel pitch of 5 to 6 μm and a fill factor of 0.10 to 0.40, the best results are obtained when the sheet resistance is between 94 and 189 ohm / sq; for a pixel pitch of 6 to 7 μm and a fill factor from 0.10 to 0.50, the best results are obtained when the sheet resistance is between 94 and 189 ohm / sq, and for a pixel pitch of 7 to 9 μm and a fill factor from 0.10 to 0.50, the best results are obtained when the sheet resistance is between 75 and 189 ohm / sq.

[0164] like Figures 22 to 25 As shown, for a pixel pitch of 9 to 10 μm and a fill factor from 0.10 to 0.50, the best results are obtained when the sheet resistance is between 75 and 189 ohm / sq, and for a pixel pitch of 10 to less than 12 μm and a fill factor from 0.10 to 0.50, the best results are obtained when the sheet resistance is between 94 and 189 ohm / sq.

[0165] Second aspect

[0166] Another difficulty with microbolometers is filtering out light of wavelengths outside the target range. In some embodiments, surface treatments and / or coatings can be applied to one or more optical elements between the microbolometer array and the image scene to filter the received light. However, these techniques increase cost.

[0167] The inventors have found that a certain choice of pixel pitch or detection unit pitch can result in a filtering function, as now referred to Fig.26A and Fig. 27 described.

[0168] Fig.26A is shown for a pixel pitch / detection unit pitch of 8.5 μm, and for a microbolometer pixel such as Figure 5 Graph of absorption as a function of light wavelength for seven different pixel fill factors for a pixel 500). For each fill factor, the thickness of the absorption layer is selected so that the ratio Rs / FF is substantially equal to 377 ohm / sq. In particular, Fig.26A Curves 2601 to 2607 in FIG. 2 are based on an absorber layer formed from titanium nitride having a fill factor and sheet resistance as defined in the following table.

[0169] [Table 3]

[0170]

[0171] from Fig.26A It can be seen that, depending on the filling factor, a filter function with a cut-off frequency λr (defined as an absorbance less than 0.5) is obtained over a wavelength range (RANGEλr) of about 6 to 8 μm. The inventors have found that this filter function is a result of diffraction that occurs when the spacing of the pixels or detection units in either or both of the x and y directions approaches the lower cut-off wavelength defined by the quarter-wavelength resonant cavity of each pixel. Fig.26A In the example of , the height h of the quarter-wavelength resonant cavity is 2.5 μm, resulting in a lower cut-off wavelength λc of about 6 μm. More generally, the lower cut-off wavelength λc defined by the quarter-wavelength resonant cavity is equal to about 0.6λ0, where λ0 is equal to the target wavelength, which is equal to four times the height h of the quarter-wavelength resonant cavity.

[0172] It should be noted that below the trough at about 6 μm, the absorbance rises again, reaching a peak at a wavelength of about 4 μm. However, these lower wavelengths are removed, for example, by relatively inexpensive surface treatments or filter layers applied to the optical elements between the microbolometer array and the image scene.

[0173] More generally, the height h of the quarter-wavelength resonant cavity of each pixel (see, for example, Figure 2 and Figure 5 ) is in the range of 0.5 to 5 μm, resulting in a target light wavelength between 2 and 20 μm. However, the filtering performance is particularly pronounced for a quarter-wavelength resonant cavity of at least 1.5 μm (corresponding to a target light wavelength of at least 6 μm).

[0174] A quarter-wavelength resonant cavity with a height of 1.5 μm results in a lower cutoff wavelength λc of about 3.6 μm, and a quarter-wavelength resonant cavity with a height of 5 μm results in a lower cutoff wavelength λc of about 12 μm. In some embodiments, the height h of the quarter-wavelength resonant cavity of each pixel is in the range of 1.5 to 3.5 μm.

[0175] For example, in order to obtain the filtering function, the detection unit spacing is selected in the range of 0.9λc to 1.65λc, corresponding to the range [1.2λ0 / 2–1.8λ0 / 2]±10%, and preferably from λc to 1.5λc. In fact, this corresponds to the range in which the quarter-wave cutoff frequency effect works. The filtering performance is particularly evident when the detection unit spacing is selected in the range of 2.4h to 3.6h, where h is the quarter-wavelength resonant cavity height, and therefore for a quarter-wavelength resonant cavity height of 2.5μm, the pixel spacing is in the range of 6 to 9μm. In some embodiments, according to the second aspect, the spacing of the detection units is between 3.6μm and 18μm, and for example between 4μm and 15μm, and preferably between 5 and 11μm.

[0176] from Fig.26A As can be seen in FIG. 1 , as the fill factor decreases and the thickness of the absorber layer increases, the rejection, ie, the filtering, is enhanced. In some embodiments, the absorber layer is formed of metal and has a sheet resistance Rs of less than 189 ohm / sq.

[0177] In some embodiments, the filling factor and sheet resistance of the absorption layer are selected based on the same criteria as described above with respect to the first aspect. For example, the absorption layer has a pixel filling factor FF in the range of 0.10 to 0.50 and a sheet resistance Rs in the range of 16 to 189 ohm / sq. In addition, in order to obtain a relatively high absorption gain, the absorption layer is selected such that:

[0178] - Pixel fill factor equal to or greater than 0.40 and less than 0.50, and sheet resistance between 75 and 189 ohm / sq; or

[0179] - A pixel fill factor equal to or greater than 0.30 and less than 0.40 and a sheet resistance between 47 and 189 ohm / sq; or

[0180] - A pixel fill factor equal to or greater than 0.20 and less than 0.30, and a sheet resistance between 25 and 189 ohm / sq; or

[0181] -Pixel fill factor is equal to or greater than 0.10 and less than 0.20, and sheet resistance is between 16 and 189 ohm / sq.

[0182] Furthermore, in some embodiments, the ratio Rs / FF is in the range of 200 to 600 ohm / sq, or in the range of 377±40%, or even in the range of 377±20%.

[0183] Fig. 27 is a graph showing the ratio of non-absorbed diffracted power as a function of the wavelength of light. Fig. 27 The seven curves 2701 to 2707 correspond to Fig.26A The same absorption layer is shown in curves 2601 to 2607. The non-absorbed diffracted power is calculated as the input power minus the absorbed power minus the reflected power.

[0184] Although Fig. 27 The graph of shows some artifacts, such as negative power near the 5 μm wavelength, but the graph still represents the general behavior of the absorber layer. Fig. 27 The graph of confirms in particular the power filtered out at wavelengths of 6 to 8 μm, ie around the lower cut-off wavelength λc of the quarter-wave resonator, equal to about 7 μm.

[0185] also, Fig. 27 The graph confirms that the filtering effect increases with decreasing fill factor and increasing absorber thickness.

[0186] Figures 26B to 26G It shows that for Fig.26A Graph of the absorbance of the absorber layer of a microbolometer pixel as a function of the wavelength of light for the same seven different pixel fill factors and absorber layer thicknesses, where the corresponding curve is plotted with Fig.26A Same reference numerals.

[0187] Fig.26B and 26C corresponds to a Fabry-Perot cavity height of 2.5 μm and to pixel pitches of 6 μm and 9 μm, respectively.

[0188] Fig.26D and 26E corresponds to a Fabry-Perot cavity height of 1.5 μm and to pixel pitches of 3.6 μm and 5.4 μm, respectively.

[0189] Fig.26F and 26G corresponds to a Fabry-Perot cavity height of 5 μm and to pixel pitches of 12 μm and 18 μm, respectively.

[0190] Figures 26H to 26R It shows that for Fig.26AGraph of the absorbance of the absorber layer as a function of the wavelength of light for a microbolometer pixel with a Fabry-Perot cavity height of 2.5 μm for the same seven different pixel fill factors and absorber layer thicknesses, where the corresponding curves are plotted with Fig.26A The same reference numerals are used in the drawings. Figures 26H to 26R Examples with pixel pitches of 5μm, 5.5μm, 6μm, 6.5μm, 7μm, 7.5μm, 8μm, 8.5μm, 9μm, 9.5μm and 10μm are shown respectively, and it is demonstrated that when the pixel pitch is in the range of 2.4h to 3.6h, corresponding to the range of 6μm to 9μm, and in particular the fill factor is at least 0.2, the filtering function is particularly obvious.

[0191] It should be noted that if the curve is plotted with Figures 26H to 26R Similarly, if the quarter-wavelength resonant cavity height is 1.5 μm, 5 μm, or anywhere in between, a similar curve will be obtained, but with the x-axis scaled for the wavelength of interest.

[0192] Common aspects

[0193] Figures 28 to 47 is a diagram showing the spatial distribution of electric and magnetic fields in a pixel (such as the above-described pixel 104 or 500) when the pixel pitch is 8.5 μm and the quarter-wavelength resonant cavity height h is 2.5 μm. Figure 2 ,exist Figures 28 to 47 In the vertical scale shown in the figure of , the reflective layer 204 is positioned at approximately 0 μm, and the absorbing layer 210 is positioned at approximately 2.5 μm.

[0194] For example, the electric field is represented by contour lines corresponding to the electric field strength represented in V / m in the figure. For example, the magnetic field is represented by contour lines associated with the magnetic field strength represented in A / m in the figure.

[0195] Figures 28 to 37 The graphs of show the spatial distribution of the electric and magnetic fields in a pixel of a microbolometer having an absorption layer thickness of 6 nm and a pixel fill factor of 0.30. In these figures, Fig.28 and 29 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 6 μm, Fig.30 and 31 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 8 μm, Fig.32 and 33 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 10 μm, Fig.34 and Fig.35 represent the electric and magnetic field distributions in the case of light with a wavelength of 12 μm, and Fig.36and Fig.37 The electric field and magnetic field distributions are shown respectively in the case of light with a wavelength of 14 μm.

[0196] Figures 38 to 47 The graphs of show the spatial distribution of the electric and magnetic fields in a pixel of a microbolometer having a 22 nm thick absorption layer and a pixel fill factor of 0.30. In these figures, Fig.38 and 39 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 6 μm, Fig.40 and 41 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 8 μm, Fig.42 and Fig.43 They represent the electric field and magnetic field distribution in the case of light with a wavelength of 10 μm, Fig.44 and Fig.45 represent the electric and magnetic field distributions in the case of light with a wavelength of 12 μm, and Fig.46 and Fig.47 The electric field and magnetic field distributions are shown respectively in the case of light with a wavelength of 14 μm.

[0197] It can be seen that at a wavelength of 6 μm ( Fig.28 , 29 , 38 and 39), the Fabry-Perot cavity effect dominates, while in the case of relatively thick absorption layers ( Fig.38 and 39 ), there is a diffraction effect, which leads to a highly dispersed field strength.

[0198] For light with a wavelength of 8 μm ( Fig.30 , 31 , 40 and 41), the diffraction effect is dominant, especially for relatively thick absorption layers ( Fig.40 and 41 ). This phenomenon is also amplified when the fill factor decreases. In addition, the field strength is strongly amplified for relatively thick absorption layers.

[0199] For light with wavelengths of 10 μm, 12 μm, and 14 μm, when the thickness of the absorption layer is relatively thin ( Figures 32 to 37 ), Fabry-Perot coupling dominates, while for thicker absorbing layers ( Figures 42 to 47 ), Fabry-Perot coupling is replaced by phenomena characterized by electric field concentration at the edges of the absorbing layer and magnetic field confinement beneath the absorbing layer.

[0200] An advantage of the embodiments described herein in relation to the first aspect is that relatively high absorptivity can be achieved while using a relatively compact microbolometer array.

[0201] An advantage of the embodiments described herein in relation to the second aspect is that a filtering functionality may be obtained, thereby relaxing the constraints on the optical filter of the microbolometer.

[0202] Furthermore, based on simulations performed for absorber layers with thicknesses of 6 and 18 nm, the inventors have found that a decrease in pixel fill factor to a range of 0.10 to 0.50 and an increase in absorber layer thickness does not result in an increase in crosstalk between pixels.

[0203] Various embodiments and variations have been described. Those skilled in the art will appreciate that certain features of these embodiments may be combined, and that the art will readily generate ideas for other variations. For example, it will be apparent to those skilled in the art that while examples of absorber layers formed of titanium nitride have been described in detail, the principles described herein may be applied to other materials, including different metals.

Claims

1. A microbolometer comprising an array of pixels, each pixel comprising one or more detection units, each detection unit comprising an absorption layer (530), the absorption layer (530) forming a quarter-wavelength resonant cavity (533) having a height h between 1.5 and 5 μm, wherein the spacing of the detection units on at least one axis in the plane of the pixel array is in the range of 2.4h to 3.6h.

2. The microbolometer according to claim 1, wherein: The pixel filling factor FF of the absorption layer (530) of the one or more detection units in each pixel is in the range of 0.20 to 0.

70.

3. The microbolometer according to claim 1 or 2, wherein: The pixel filling factor FF of the absorption layer (530) of the one or more detection units in each pixel is in the range of 0.20 to 0.

50.

4. The microbolometer according to claim 1 or 2, wherein: The pitch of the detection units is in the range of 4 to 15 μm.

5. The microbolometer according to claim 1 or 2, wherein: The pitch of the detection units is in the range of 5 to 11 μm.

6. The microbolometer according to claim 1 or 2, wherein: The absorption layer (530) is a metal layer having a sheet resistance of 189 ohm / sq or less.

7. The microbolometer according to claim 1 or 2, wherein: The absorption layer (530) is a metal layer having a sheet resistance of 126 ohm / sq or less.

8. The microbolometer according to claim 6, wherein: The absorption layer (530) is formed of TiN.

9. The microbolometer according to claim 1 or 2, wherein: The resonant cavity height (h) is between 1.5 and 3.5 μm.

10. The microbolometer according to claim 1 or 2, wherein: Each pixel of the array has: - the pixel fill factor FF is equal to or greater than 0.40 and less than 0.50, and the sheet resistance Rs of the absorption layer (530) is at least 75 ohm / sq; or - the pixel fill factor FF is equal to or greater than 0.30 and less than 0.40, and the sheet resistance Rs of the absorption layer (530) is at least 50 ohm / sq; or - the pixel fill factor FF is equal to or greater than 0.20 and less than 0.30, and the sheet resistance Rs of the absorption layer (530) is at least 25 ohm / sq; or - The pixel fill factor FF is equal to or greater than 0.10 and less than 0.20, and the sheet resistance Rs of the absorption layer (530) is at least 16 ohm / sq.

11. The microbolometer according to claim 1 or 2, wherein: The ratio Rs / FF for each pixel of the array is between 200 and 600 ohm / sq.

12. The microbolometer according to claim 1 or 2, wherein: The ratio Rs / FF for each pixel of the array is within 20% of 377 ohm / sq.

13. The microbolometer according to claim 1 or 2, wherein: Each detection unit comprises a membrane (202, 502), the membrane (202, 502) comprising the absorption layer (210, 530), the thermal layer (212, 528) and the dielectric layer (214, 604).

14. A method of making a microbolometer, the method comprising forming an array of pixels, each pixel comprising one or more detection cells, wherein forming the array comprises: - forming each detection unit to include an absorption layer (530) forming a quarter-wavelength resonant cavity (533) having a height h between 1.5 and 5 μm; and The detection units are formed so as to have a pitch in the range of 2.4h to 3.6h on at least one axis in the plane of the pixel array.

15. The method according to claim 14, wherein: The pixel filling factor FF of the absorption layer (530) of the one or more detection units in each pixel is within the range of 0.20 to 0.70.

Images