WO2018167433A1

WO2018167433A1 - Radiation sensor with anti-glare protection

Info

Publication number: WO2018167433A1
Application number: PCT/FR2018/050618
Authority: WO
Inventors: Sébastien Becker; Denis Pelenc; Jean-Jacques Yon; Abdelkader Aliane
Original assignee: Commissariat A L'energie Atomique Et Aux Energies Alternatives
Priority date: 2017-03-15
Filing date: 2018-03-14
Publication date: 2018-09-20
Also published as: FR3064061A1

Abstract

The invention relates to a radiation sensor comprising a plurality of pixels (800) formed in and on a semiconductor substrate (101), each pixel comprising a micro-plate (103) suspended above the substrate by means of thermally insulating arms (105a, 105b), said micro-plate (103) comprising an element (601) for converting electromagnetic radiation into thermal energy. According to the invention, in each pixel (800), at least one of the thermally insulating arms (105a, 105b) of the pixel comprises a layer (801) of phase change material having, below a phase change temperature, a first thermal conductivity value and, above the phase change temperature, a second thermal conductivity value greater than the first.

Description

RADIATION SENSOR HAVING ANTI-GLARE PROTECTION

The present patent application claims the priority of the French patent application FR17 / 52110 which will be considered as an integral part of the present description.

Field

The present application relates to the field of radiation sensors, and more particularly relates to sensors of the type comprising a plurality of elementary micro-detectors, or pixels, arranged in and on a semiconductor substrate, each pixel comprising a radiation conversion element. electromagnetic signal into an electrical signal, and a signal reading circuit provided by the conversion element. The present application relates more particularly to the protection of such a sensor against glare likely to damage its pixels. The embodiments described below are particularly advantageous in the case where the pixel conversion elements are microbolometers.

Presentation of the prior art

A bolometer conventionally comprises an absorber adapted to transform electromagnetic radiation to which it is subjected, generally located in the infrared, into thermal energy, and a thermometer thermally coupled to the absorber and adapted to provide an electrical signal representative of the temperature of the absorber. The thermometer generally comprises a thermistor, and a circuit for reading the electrical resistance of the thermistor.

It has already been proposed, for example in the French patent application No. 2796148 filed July 8, 1999 or in the French patent application No. 2822541 filed March 21, 2001, a thermal radiation sensor comprising a plurality of pixels arranged in and on a semiconductor substrate, each pixel comprising a microbolometer and an electronic circuit for controlling and reading the microbolometer.

A problem is that when they are subjected to strong radiation, for example laser radiation in the case of a malicious attack of the sensor, or solar radiation, the pixels of such a sensor undergo a heating up to several hundred degrees, which can damage them temporarily or permanently.

More generally, this problem of critical heating of the pixels under the effect of the radiation to be detected can arise in other types of electromagnetic radiation sensors, in particular sensors in which the measurement of the incident radiation is based on a radiation conversion. in thermal energy within the pixels of the sensor.

It would be desirable to have a means of protecting a radiation sensor against glare that could damage its pixels.

summary

Thus, an embodiment provides a radiation sensor comprising a plurality of pixels formed in and on a semiconductor substrate, each pixel comprising a microplanche suspended above the substrate by heat-insulating arms, the microplanche comprising a conversion element. from electromagnetic radiation to thermal energy,

wherein, in each pixel, at least one of the thermal insulation arms of the pixel comprises a layer of a material phase change transistor having, below a phase transition temperature, a first thermal conductivity value, and, above the phase transition temperature, a second value of thermal conductivity greater than the first value.

According to one embodiment, the layer is a metal oxide having an insulating phase below the transition temperature and a metal phase above the transition temperature.

According to one embodiment, the layer is in a vanadium oxide.

According to one embodiment, the layer is made of a titanium oxide.

According to one embodiment, the transition temperature of the layer is between 60 and 180 ° C.

According to one embodiment, each pixel further comprises a thermistor thermally coupled to the pixel conversion element.

According to one embodiment, each pixel further comprises a circuit for reading the value of the pixel's thermistor.

According to one embodiment, in each pixel, the conversion element is a layer made of an absorbent material for the radiation to be detected.

According to one embodiment, in each pixel, the conversion element is a metal layer.

According to one embodiment, in each pixel, the heat-insulating arms rest on vertical electrical connection pillars.

According to one embodiment, in each pixel, the microplate and the heat-insulating arms are arranged in a cavity closed by a transparent cover for the radiation to be detected. According to one embodiment, in each pixel, the transparent cover hermetically closes the cavity, and the cavity is at a pressure below atmospheric pressure.

Brief description of the drawings

These and other features and advantages will be set forth in detail in the following description of particular embodiments in a non-limiting manner with reference to the accompanying drawings in which:

FIGS. 1A and 1B are respectively a sectional view and a simplified top view of an example of a pixel of a radiation sensor according to a first embodiment;

Figure 2 is a simplified sectional view of another example of a pixel of a radiation sensor according to the first embodiment;

Figs. 3A, 3B, 3C, 3D, 3E and 3F are sectional views illustrating steps of an example of a method of manufacturing a pixel according to the first embodiment;

Figure 4 is a sectional view illustrating an alternative of the manufacturing method of Figures 3A, 3B, 3C, 3D, 3E and 3F;

Figs. 5A, 5B, 5C and 5D are sectional views illustrating steps of another example of a method of manufacturing a pixel according to the first embodiment;

Figure 6 is a partial and simplified sectional view of an example of a pixel of a radiation sensor according to a second embodiment;

Figs. 7A, 7B, 7C, 7D, 7E and 7F are sectional views illustrating steps of an example of a method of manufacturing a pixel according to the second embodiment;

Figure 7a is a sectional view illustrating an alternative embodiment of the method of Figures 7A, 7B, 7C, 7D, 7E and 7F;

Fig. 8 is a simplified sectional view of an example of a pixel of a radiation sensor according to a third embodiment; and Figs. 9A, 9B, 9C, 9D, 9E, 9F and 9G are sectional views illustrating steps of an example of a method of manufacturing a pixel according to the third embodiment. detailed description

The same elements have been designated by the same references in the various figures and, in addition, the various figures are not drawn to scale. For the sake of clarity, only the elements useful for understanding the described embodiments have been shown and are detailed. In particular, the circuits for controlling and reading the pixels described have not been detailed, the described embodiments being compatible with the control and reading circuits usually provided in pixels of this type, or the realization of these circuits. being within the abilities of those skilled in the art from the functional indications of the present description. Moreover, in the illustrative figures of the examples described, only one pixel of a radiation sensor is visible. In practice, the radiation sensors may comprise several identical or similar pixels arranged in and on the same semiconductor substrate, for example in a matrix or bar arrangement. The arrangement of the different pixels of the sensor, the interconnections between the pixels of the sensor, and the peripheral sensor control circuits have not been detailed, the embodiments described being compatible with the arrangements, interconnections, and peripheral control circuits. usually provided in such sensors. Furthermore, the uses that can be made of the sensors described have not been detailed, the described embodiments being compatible with the usual applications of radiation sensors. It will be noted, however, that the described embodiments are particularly advantageous for infrared imaging, thermography, gas detection applications by measuring the optical absorption in the infrared spectrum, detection or recognition of persons, objects or movements in the infrared spectrum, etc. In the description which follows, when referring to absolute position qualifiers, such as the terms "forward", "backward", "up", "down", "left", "right", etc., or relative, such as the terms "above", "below", "upper", "lower", etc., or with qualifiers for orientation, such as the terms "horizontal", "vertical", etc., reference is made to to the orientation of the sectional views of the figures, it being understood that, in practice, the devices described may be oriented differently. Unless otherwise specified, the terms "approximately", "substantially", and "of the order of" mean within 10%, preferably within 5%. Note that in the following description, the radiation sensors are intended to be illuminated or irradiated by their upper face (in the orientation of sectional views of the figures).

First embodiment - active cover protection

FIGS. 1A and 1B are respectively a simplified sectional view and a simplified top view of an example of a pixel of a radiation sensor according to a first embodiment.

The pixel 100 of FIGS. 1A and 1B is formed in and on a semiconductor substrate 101, for example made of silicon.

The pixel 100 comprises an electronic reading and control circuit 102 formed in and on the substrate 101, for example in CMOS technology. The control circuit and reading has not been detailed in the figures. Only electrical connection pads flush with the upper face of the circuit 102, intended to connect the circuit 102 to other elements of the pixel, are shown in FIG. 1A in the form of hatched rectangular zones.

The pixel 100 further comprises a microplanche 103 suspended above the circuit 102 by thermal insulation arms, two arms 105a and 105b in the example shown. More particularly, in the example shown, the substrate 101 and the circuit 102 are arranged horizontally, and the microplate 103 and the heat-insulating arms 105a and 105b are arranged in the same middle plane substantially parallel to the upper face of the circuit 102. Each of the arms 105a and 105b has a first end or end proximal contact mechanically and electrically with the microplate 103, and a second end or distal end resting on the top of a vertical conductive pillar 107a, respectively 107b, for example copper or tungsten, the base of which rests on the upper face of the Circuit 102. The pillars 107a and 107b mechanically support the microplate 103 via the arms 105a and 105b, and electrically connect the microplanche 103 to the circuit 102, also via the arms 105a and 105b. Thus, a space free of any solid material is located between the upper face of the circuit 102 and the lower face of the microplanche 103. In other words, the microplanche 103 is in mechanical contact only the arms 105a and 105b, which thermally isolate the microplanche of the remainder of the structure and in particular of the circuit 102 and the substrate 101.

In this example, the microplanche 103 is a bolometric microplanche, that is to say it comprises an absorber (not detailed in Figures 1A and 1B), for example in the form of a conductive layer, suitable for converting incidental electromagnetic radiation in thermal energy, and a thermistor (not detailed in Figures 1A and 1B) for measuring the temperature of the absorber. By way of example, the absorber is made of titanium nitride (TiN) and the thermistor is of amorphous silicon or of vanadium oxide. The two ends of the thermistor are electrically connected respectively to the conductive pillars 107a and 107b via the arms 105a and 105b.

In the example shown, the base of the support pillar 107a is in mechanical and electrical contact with a connection pad 109a of the upper face of the circuit 102, and the base of the support pillar 107b is in mechanical contact with and electrically with a connection pad 109b of the upper face of the circuit 102. The control and reading circuit 102 is thus connected to the ends of the pixel thermistor via the pads 109a and 109b and the pillars 107a and 107b of the pixel . The circuit 102 is adapted to provide an electrical signal representative of the value of the electrical resistance of the pixel thermistor.

The pixel 100 further comprises a cover 111 transparent to the radiation to be detected, resting on the upper face of the control circuit 102 and delimiting, with the upper face of the circuit 102, a cavity or hermetic enclosure 113 in which is located the suspended microplanche 103 A space left free of any solid material is located between the upper face of the microplanche 103 and the underside of the cover 111, this space communicating with the free space between the lower face of the microplanche 103 and the upper face of the circuit 102. The cavity 113 is preferably evacuated or under a pressure lower than the atmospheric pressure, so as to reinforce the thermal insulation of the microplate 103 vis-à-vis the rest of the sensor, limiting the thermal conduction by the 'air.

According to one aspect of the embodiment of Figures 1A and 1B, the pixel 100 comprises an optical shutter having a layer 115 of a thermochromic material coating the upper face of the transparent cover 111, facing the microplanche 103 of the pixel. The optical shutter is an active shutter, that is to say that it is electrically controllable or in an open state in which the layer 115 is substantially transparent for the radiation to be detected, that is to say in which the reflection and / or absorption coefficient of the layer 115 for the radiation to be detected is relatively low, ie in a closed state in which the reflection and / or absorption coefficient of the layer 115 for the radiation to be detected is relatively high, that is to say greater than its reflection coefficient and / or absorption in the open state. By way of example, the radiation to be detected is a thermal infrared radiation of length wavelength in the band from 8 to 14 μm, and the transmission coefficient of the layer 115 for the radiation to be detected in the closed state is less than 0.2 and preferably at least 0 , 4 to its transmission coefficient in the open state.

In the example of FIGS. 1A and 1B, the shutter comprises a heating resistor 117 thermally coupled to the layer 115 and connected to the control circuit 102 of the pixel via connection pads 119a and 119b of the circuit 102.

In the example shown, the heating resistor 117 is disposed, in plan view, at the periphery of the bolometric microplate 103, so as not to disturb the passage of the incident electromagnetic radiation towards the microplanche when the shutter is at the open state. For example, the heating resistor 117 is a metal strip forming a conductor cord surrounding, in top view, the microplanche 103. Alternatively, the heating resistor may be made of a transparent material for the radiation to be detected, for example a dielectric or semiconductor material, for example, in the case of an infrared radiation detector, vanadium dioxide (VO 2), germanium (Ge), or an α-SiGe: B alloy of amorphous silicon and germanium Boron doped. In this case, the heating resistor 117 may extend at least in part opposite the bolometric microplate 103 of the pixel.

The heating resistor 117 is for example disposed under the thermochromic layer 115, for example contiguous to the underside of the transparent cover 111. In the example shown, the heating resistor 117 is connected to the pads 119a and 119b of the control circuit 102 respectively by vertical conductive pillars 121a and 121b, for example of copper or tungsten, connecting the lower wall to the upper wall of the cavity 113.

For the sake of simplicity, only the microplanche 103, the holding arms 105a and 105b, the heating resistor 117, and the vertical connecting pillars 107a, 107b, 121a and 121b of the pixel 100 have been shown in FIG. 1B. In addition, in Figure 1A, the vertical connecting pillars 107a, 107b, 121a and 121b have been shown in a same plane. In practice, the pillars 107a, 107b, 121a and 121b, however, are not necessarily aligned.

The operation of the anti-glare protection of the pixel 100 is as follows. The reading circuit 102 of the pixel is adapted to detect a glare capable of damaging the pixel, via its connection pads 109a and 109b, for example by detecting an excessively fast and intense variation of the value of the pixel. pixel thermistor, a sign of an excessively rapid heating of the microplanche, or when the value of the thermistor reaches a predefined threshold.

By way of example, the control and reading circuit 102 is adapted, during a phase of acquisition of a value representative of the electromagnetic radiation received by the pixel, to implement a reading of the pixel by integration, in a capacitive element of the circuit 102, a current delivered by the microbolometer thermistor of the pixel for a determined DC bias voltage of the thermistor.

In a preferred embodiment, the circuit 102 is configured to integrate the current delivered by the pixel's thermistor during two successive periods of integration of distinct durations. More particularly, the circuit 102 is adapted to measure the integrated current during a short integration period, for example of duration between 1 and 5 ys, to detect a possible glare of the pixel, then during a long integration period, by example of the order of 30 to 100 ys, for example about 64 ys, for the acquisition itself of a value representative of the radiation received by the pixel

(or output value of the pixel), for example for the purpose of constructing an image. The signal from the first integration can be sampled and compared with a threshold, for example by a comparator, the shutter then being controlled according to the result of the comparison. Alternatively, prior to the implementation of the long integration period, the circuit 102 may be adapted to integrate the current delivered by the thermistor during two successive short integration periods of different durations, for example during a first period. of the order of 1 ys and for a second period of the order of 5 ys. The difference between the signals from the first and second short integration periods is then determined and compared to a threshold to decide whether or not to activate the shutter. One advantage is that this makes it possible to overcome the differences in values between the thermistors of the different pixels, related to the technological dispersions, and thus to improve the glare detection accuracy.

In another variant, the control and reading circuit 102 is adapted, during a phase of acquisition of a value representative of the electromagnetic radiation received by the pixel, to implement a reading of the pixel by measuring the voltage at the terminals of the microbolometer thermistor of the pixel for a determined DC bias current injected into the pixel thermistor. The voltage across the thermistor can then be compared to a threshold to decide whether or not to activate the shutter of the pixel.

When a glare is detected, the circuit 102 controls the application of a current in the resistor 117, so as to cause a heating of the thermochromic layer 115 to a transition temperature leading to pass the shutter of the open state (layer 115 transparent to the radiation to be detected) in the closed state (layer 115 opaque for the radiation to be detected). The incident electromagnetic radiation is then stopped or limited by the layer 115, which makes it possible to avoid or limit a degradation of the bolometric microplanche of the pixel. After a predetermined closing period or when the circuit 102 detects a return to a predetermined acceptable temperature of the bolometric microplanche, the circuit 102 interrupts the current flowing in the heating resistor 117. The temperature of the layer thermochrome 115 then drops below its transition temperature, so that the shutter reopens.

It will be noted that the electrical connection pillars 121a and 121b advantageously make it possible to thermally couple the upper part of the protective cover 111 to the substrate 101, so as to circumscribe the heating produced by the heating resistor 117, and to prevent the heat generated by the resistor 117 propagates in the encapsulation covers and / or in the thermochromic layers of the neighboring pixels. To improve the thermal coupling with the substrate 101, the number of vertical electrical connection pillars between the resistor 117 and the circuit 102 may be greater than two. By way of example, in the case of a bolometric microplate 103 of generally square or rectangular shape and a heating resistor 117 in the form of a square or rectangular annular band surrounding in top view the microplate 103, as represented in FIG. 1B, the pixel may comprise four electrical connection pillars respectively arranged, in top view, at the four corners of the conductive annular band forming the heating resistor.

Furthermore, the control and reading circuit 102 of the pixel 100 may be adapted to regulate the current injected into the heating resistor 117 as a function of the temperature of use of the sensor, so as to inject into the resistor 117 only the current required to get the shutter closed. Indeed, the operating temperature of a radiation sensor can generally vary over a wide range, for example ranging from -40 ° C to +70 ° C, and the current to be injected into the resistor 117 to obtain the transition of the thermochromic layer 115 is even higher than the temperature of use is low. For example, the sensor comprises at least one temperature probe, for example disposed in and on the semiconductor substrate 101, for example a probe based on PN junctions. The control and reading circuit 102 of the pixel 100 is then adapted, in case of glare of the pixel, to inject into the heating resistor 117 of the pixel a current chosen in function of the temperature measured by the temperature sensor. One advantage is that this makes it possible to limit the power consumption related to the glare protection, in particular when the sensor is used at high temperatures, and to avoid unnecessary heating of the encapsulation cap 111 of the pixels beyond the transition temperature of the thermochromic layer.

The material of the thermochromic layer 115 is for example a phase-change material having, below a transition temperature, a substantially transparent phase for the radiation to be detected, and, above the transition temperature, a phase reflective or absorbent for the radiation to be detected. The transition temperature of the thermochromic material is preferably greater than the maximum temperature that can reach the cover 111 of the pixel in normal operation, for example between 60 and 180 ° C. The variation of the coefficient of transmission or reflection of the thermochromic layer around the transition temperature is preferably relatively steep, for example greater than 2.5% per degree for a wavelength of 10 μm. For example, the thermochromic material is a crystallized metal oxide having a transparent insulating phase below its transition temperature, and a reflective metal phase above its transition temperature. The thermochromic material is for example crystallized vanadium dioxide (VO 2), having a transition temperature of the order of 68 ° C. Alternatively, the thermochromic material is vanadium dioxide crystallized and doped with low-valence cations, for example A13 +, Cr3 + or Ti4 +, so as to increase its transition temperature. More generally, depending on the desired transition temperature, other vanadium oxides can be used, for example V3O5. Alternatively, the thermochromic material of the layer 115 is T13O5, 12O3, or Sm103. Alternatively, the thermochromic material of the layer 115 is a rare earth nickelate of general composition RN1O3, where R denotes a rare earth or a rare earth binary alloy, for example a compound of the type Sm _x Nd _] _x Ni03 or Eu _x Sm _] _{x x} Ni03. Alternatively, the thermochromic material of layer 115 is Ag2S or FeS. Alternatively, the thermochromic material is monocrystalline germanium, which has the advantage of being relatively transparent for thermal infrared radiation at room temperature, and relatively absorbent for this radiation for temperatures above 100 ° C.

FIG. 2 is a simplified sectional view of another example of a pixel of a radiation sensor according to the first embodiment. The pixel 200 of FIG. 2 comprises elements that are in common with the pixel 100 of FIGS. 1A and 1B. These elements will not be detailed again below. In the following, only the differences between the pixels 100 and 200 will be detailed.

The pixel 200 of FIG. 2 differs from the pixel 100 of FIGS. 1A and 1B principally in that it does not comprise a heating resistor 117 distinct from the thermochromic layer 115 for controlling the transitions in the closed state or in the open state of the layer 115. In this example, the upper ends of the connection pillars 121a and 121b of the pixel are in direct electrical contact with the thermochromic layer 115.

The operation of the pixel 200 of FIG. 2 is substantially the same as that of the pixel 100 of FIGS. 1A and 1B, except that, in the example of FIG. 2, when the control and reading circuit 102 of the pixel detects a glare, it injects an electric current directly into the thermochromic layer 115, via the connection pads 119a and 119b and connecting pillars 121a and 121b. This current causes heating of the thermochromic layer 115, leading to closure of the shutter. When the current flowing in the thermochromic layer 115 is interrupted, the temperature of the layer 115 drops below its transition temperature, so that the shutter reverts to the open state. FIGS. 3A, 3B, 3C, 3D, 3E and 3F are cross-sectional views illustrating steps of an example of a method of manufacturing a radiation sensor of the type described with reference to FIG. that is, not having a heating resistor distinct from the thermochromic layer for controlling the transitions in the closed state or in the open state of the thermochromic layer. FIGS. 3A, 3B, 3C, 3D, 3E and 3F more particularly illustrate the production of a single pixel of the sensor, it being understood that, in practice, a plurality of identical or similar pixels can be formed simultaneously in and on the same substrate semiconductor 101.

FIG. 3A illustrates a step of manufacturing the control and reading circuit 102 of the pixel 200, in and on the substrate 101. The circuit 102 is for example made in CMOS technology. The manufacture of the circuit 102 is not detailed here, the realization of this circuit being within the abilities of those skilled in the art from the functional indications mentioned above. In FIG. 3A, only the electrical connection pads 109a, 109b, 119a and 119b of the circuit 102, flush with the upper face of the circuit 102, have been shown.

FIG. 3B illustrates a step of depositing a sacrificial layer 301 on and in contact with the upper face of the circuit 102. The layer 301 is for example continuously deposited on substantially the entire surface of the substrate 101. By way of example the layer 301 is made of polyimide or silicon oxide. The thickness of the layer 301 fixes the distance between the upper face of the circuit 102 and the bolometric microplate 103 of the pixel 200. By way of example, the layer 301 has a thickness of between 1 and 5 μm, for example order of 2.5 ym.

FIG. 3B furthermore illustrates the formation of the electrical connection pillars 107a and 107b of the pixel, in vias etched in the sacrificial layer 301 in line with the connection pads 109a and 109b of the circuit 102. The pillars 107a and 107b are extend vertically over substantially the entire thickness of the layer 301, from the upper face of the connection pads 109a and 109b of the circuit 102, to the upper face of the layer 301.

FIG. 3B further illustrates the formation of the bolometric microplate 103 and the holding arms 105a, 105b of the microplanche on and in contact with the upper face of the sacrificial layer 301 and the connecting pillars 107a and 107b. This step comprises depositing the constituent materials of the bolometric microplate 103 and the holding arms 105a, 105b, and the delimitation or individualization of the microplanche 103 and the arms 105a, 105b of the pixel. The formation of the bolometric microplanche 103 and the holding arms 105a, 105b is not detailed here, it can be implemented by methods known to those skilled in the art, for example processes of the type described in the French Patent Applications Nos. 2796148 and 2822541 mentioned above.

FIG. 3C illustrates a deposition step, on and in contact with the upper face of the structure obtained at the end of the steps of FIG. 3B, of a second sacrificial layer 303, preferably of the same kind as the layer 301. The layer 303 is for example deposited continuously over substantially the entire surface of the sensor. The lower face of the layer 303 is in contact with the upper face of the microplates 103 and the holding arms 105a, 105b of the pixels, and, with the upper face of the layer 301 in the regions separating, in top view, the microplates 103 and the arms 105a, 105b of the pixels. The thickness of the layer 303 sets the distance between the upper face of the microplate 103 and the upper part of the encapsulation cap of the pixel. By way of example, the layer 303 has a thickness of between 1 and 2.5 μm.

FIG. 3C also illustrates the formation of the electrical connection pillars 121a and 121b of the pixel, in vias etched in the sacrificial layers 303 and 301 in line with the connection pads 119a and 119b of the circuit 102. The pillars 121a and 121b extend vertically on substantially any the thickness of the sacrificial layers 303 and 301, from the upper face of the connection pads 119a and 119b of the circuit 102, to the upper face of the layer 303. By way of example, the connecting pillars 121a, 121b have a surface area of between 0.25 and 1 μm.

FIG. 3C further illustrates the formation of a conductive layer 305 located on the upper surface of the electrical connection pillars 121a and 121b. In particular, the layer 305 serves to prevent the diffusion of the metal of the pillars 121a, 121b, for example copper or tungsten, into the material of the encapsulation cap of the pixel. Layer 305 may also serve as an etch stop layer in a subsequent step of resuming contact on pillars 121a, 121b. By way of example, the layer 305 is made of titanium nitride (TiN). The layer 305 for example has a thickness of between 20 and 80 nm.

FIG. 3D illustrates a step subsequent to the steps of FIG. 3C, during which a vertical peripheral trench entirely surrounding, in plan view, the assembly comprising the bolometric microplanche 103, the holding arms 105a, 105b and the pillars of FIG. electrical connection 107a, 107b, 121a, 121b, is etched from the upper face of the sacrificial layer 303 and up to the upper face of the circuit 102. The trench 307 separates the elements 103, 105a, 105b, 107a, 107b, 121a, 121b of the pixel of the corresponding elements of neighboring pixels. Trench 307 is intended to receive the side walls of the encapsulation cap of the pixel. In this example, a specific trench 307 is made for each pixel of the sensor, i.e., two neighboring pixels are separated by two distinct trenches 307.

FIG. 3D further illustrates a step of deposition of a transparent layer 309 for the radiation to be detected over substantially the entire upper surface of the structure obtained after the etching of the trenches 307, to form the encapsulation covers 111 of the pixels of the sensor . The layer 309 is, for example, an amorphous silicon layer 0.5 to 1 μm thick, for example of the order of 0.8 ym thick. The layer 309 is in particular deposited on and in contact with the side walls and the bottom of the trenches 307, as well as on and in contact with the upper face of the sacrificial layer 303 or the barrier layer 305 outside the trenches 307, so that to hermetically encapsulate, in each pixel, the assembly comprising the microplanche 103, the arms 105a, 105b, and the pillars 107a, 107b, 121a, 121b of the pixel.

FIG. 3E illustrates a step of etching, in the layer 309, trenches 311 completely surrounding, in top view, each of the pixels of the sensor so as to individualize and electrically isolate the encapsulation covers 111 of the different pixels of the sensor. Indeed, in the case where the material of the layer 309 is electrically conductive, which may be the case of amorphous silicon if it is doped, it is preferable to electrically isolate the encapsulation covers 111 of the different pixels, so that that the polarization applied by the control and reading circuit 102 of a pixel via its connecting pillars 121a, 121b does not cause the circulation of a parasitic current in all of the encapsulation covers of the sensor. In the example shown, the trenches 311 are arranged, in top view, in the gaps between the trenches 307 of the neighboring pixels of the sensor. The isolation trenches 311 extend vertically over the entire thickness of the layer 309, and open on the upper face of the sacrificial layer 303.

FIG. 3E further illustrates a step of etching, in each pixel, at least one opening 313 in the layer 309, inside the zone delimited (seen from above) by the trench 307, that is, that is to say in the upper part of the encapsulation cap 111 of the pixel, for example facing a central part of the microplanche 103 of the pixel. The opening 313 is provided to allow the implementation of a subsequent step of removing the sacrificial layers 301 and 303 inside the cover 111. The opening 313 extends vertically over any the thickness of the layer 309, and opens on the upper face of the sacrificial layer 303. The width of the opening 313, in a view from above, is for example between 0.1 and 1 .mu.m.

FIG. 3E further illustrates a step of etching apertures 315 in the layer 309, located opposite the electrical connection pillars 121a, 121b so as to release access to the upper face of the barrier layer 305.

The openings 311, 313 and 315 are for example made simultaneously during the same etching step.

FIG. 3F illustrates a subsequent step of removing the sacrificial layers 303 and 301, for example by anisotropic chemical etching, so as to release the microplate 103 and the holding arms 105a, 105b of the pixel.

FIG. 3F furthermore illustrates a step subsequent to the removal of the sacrificial layers 303 and 301 from deposition of the thermochromic layer 115 on and in contact with the upper face of the encapsulation cap 111, facing the bolometric microplate 103 of the pixel. For example, the thermochromic layer 115 is deposited on the entire upper surface of the structure, then etched opposite the openings 311 so as to electrically isolate the portions of the layer 115 overcoming the different pixels of the sensor. Depending on the type of thermochromic material used, an annealing of the layer 115 may optionally be used to obtain the crystalline phase and the transition temperature of the desired material. By way of example, in the case of a vanadium dioxide (VO2) thermochromic layer, the deposition can be carried out at ambient temperature by spraying a vanadium target in an oxygen-containing atmosphere. This leads to the formation of an amorphous vanadium dioxide layer. An annealing at a temperature of the order of 350 to 400 ° C can then be used to crystallize the vanadium oxide layer and obtain the desired thermochromic properties. The thermochromic layer has for example a thickness of between 20 and 100 nm, for example between 20 and 60 nm. At the openings 315, the thermochromic layer 115 comes into contact with the upper face of the conductive layer 305 so as to electrically connect the thermochromic layer 115 to the connecting pillars 121a and 121b. In the example shown, the thermochromic material of the layer 115 additionally closes the opening 313 provided for the removal of the sacrificial layers 303 and 301, so as to seal the encapsulation cavity 113 of the pixel. The deposition of the thermochromic layer 115 is for example carried out under vacuum or at a pressure below atmospheric pressure so as to evacuate or at low pressure the encapsulation cavity of the pixel. It will be noted that germanium advantageously makes it possible to ensure the dual function of thermochromic layer and hermetic sealing material of the opening 313. Alternatively, if the thermochromic material is not suitable for sealing the opening 313, an intermediate step of depositing a material adapted to plug the opening 313, for example germanium, or a metal such as aluminum, may be provided before the deposition of the thermochromic layer 115. In the case where the layer intermediate blocking of the opening 113 is not transparent enough for the radiation to be detected, the latter can be deposited in a localized manner only opposite the opening 113, or be etched after deposit to be kept only in look at the opening 113.

At the end of the step of FIG. 3F, a pixel 200 of the type described in relation with FIG. 2 is obtained.

FIG. 4 illustrates a variant of the method described with reference to FIGS. 3A, 3B, 3C, 3D, 3E and 3F. FIG. 4 is a sectional view of the pixel obtained at the end of the process, corresponding to the sectional view of FIG. 3F.

The embodiment variant of FIG. 4 differs from the method previously described mainly by the way of isolating the encapsulation covers 111 from the different pixels of the sensor. The process of Figure 4 comprises the same initial steps as the previously described method up to the steps of Figure 3C included.

In the variant of FIG. 4, vertical trenches 307 for delimiting the pixels are etched in the sacrificial layers 303 and 301 from the upper face of the structure. These trenches are similar to what has been described in connection with FIG. 3D, except that, in the variant of FIG. 4, two neighboring pixels of the sensor are separated by a single trench 307. In other words, unlike FIG. the example of the 3D figure in which a specific annular trench 307 is formed for each pixel of the sensor, so that the trenches 307 of distinct pixels are disjoint, in the example of FIG. 4, the trenches 307 have, in view from above, the shape of a continuous grid extending over substantially the entire surface of the sensor and defining the different pixels of the sensor.

The following steps of the method of FIG. 4 are similar to what has been described in relation with FIGS. 3D and 3E, except that, in place of the trench 311 of FIG. 3E, the layer 309 is etched in for each pixel, an annular trench 411 completely surrounding, in top view, the assembly comprising the bolometric microplate 103, the holding arms 105a, 105b and the electrical connection pillars 107a, 107b, 121a, 121b of the pixel, the trench 411 being located, in plan view, within the area delimited by the trench 307 delimiting the pixel. The annular trench 411 extends vertically over the entire thickness of the encapsulation layer 309, and opens out on the upper face of the sacrificial layer 303. The trench 411 is then filled with an electrically insulating material, for example nitride of silicon (SiN) or aluminum nitride (AlN), so as to obtain an insulating frame or ring 413 entirely surrounding an upper central portion of the encapsulation cap 111 of the pixel, preventing the flow of parasitic currents between the protective covers. encapsulation of the different pixels of the sensor. The following steps are identical or similar to what has been described with reference to FIGS. 3E and 3F, the thermochromic layer 115 of each pixel being inscribed, in plan view, inside the frame or the insulating ring 413. of the pixel.

An advantage of the embodiment variant of the figure

4 is that it allows, at the cost of an additional step of forming the insulating frame 413 in the opening 411, reduce the area used to electrically isolate the encapsulation caps 111 and the thermochromic layers 115 of the different pixels of the sensor .

FIGS. 5A, 5B, 5C, 5D are cross-sectional views illustrating steps of an example of a method for manufacturing a radiation sensor of the type described with reference to FIGS. 1A and 1B, that is to say said having a heating resistor distinct from the thermochromic layer for controlling the closed or open state transitions of the thermochromic layer. FIGS. 5A, 5B, 5C and 5D are more particularly concerned with the case where the heating resistor is a metal conductive bead surrounding in top view without masking the bolometric microplate 103 of the pixel.

The method of FIGS. 5A, 5B, 5C, 5D comprises elements that are common with the process described in relation to FIGS. 3A, 3B, 3C, 3D, 3E and 3F and / or in relation with FIG. 4. In the following, only the differences with the methods of FIGS. 3A, 3B, 3C, 3D, 3E, 3F and 4 will be detailed.

The process of Figs. 5A, 5B, 5C and 5D includes the same initial steps as the previously described method up to the steps of Fig. 3C included.

FIG. 5A illustrates a step of deposition of a first electrically insulating layer 501, for example of silicon nitride (SiN) or of aluminum nitride (AlN), on the upper face of the structure obtained at the end of the steps of Figure 3C. Openings are then etched in the layer 501 above the electrical connection pillars 121a, 121b, so as to release access to the upper face of the barrier layer. 305. During this etching step, the layer 501 is conserved at the periphery and on the flanks of the portions of the layer 305 surmounting the electrical connection pillars 121a, 121b. The layer 501 is furthermore preserved over the entire surface of the structure intended to receive the heating resistor 117, so that, at the end of the process, the heating resistor 117 is entirely encapsulated in an electrically insulating sheath, except at its level. regions of electrical connection to the layer 305. The layer 501 can be removed on the rest of the surface of the structure, and in particular opposite the bolometric microplate 103 of the pixel.

FIG. 5B illustrates a subsequent step of depositing a conductive layer 503 on the upper face of the structure obtained at the end of the steps of FIG. 5A, to form the heating resistor 117 of the pixel. The layer 503 is for example a layer of aluminum, titanium, or titanium nitride. The layer 503 has for example a thickness of between 10 and 100 nm, for example a thickness of the order of 20 nm. For example, the layer 503 is deposited on the entire upper surface of the structure, then etched to retain only a conductive strip 117 disposed on and in contact with the upper face of the insulating layer 501, surrounding, in top view , the assembly comprising the bolometric microplate 103 and the connecting arms 105a, 105b of the pixel, and electrically connected to the connection pillars 121a, 121b of the pixel via the barrier layer 305. The width, seen from above, conductive ribbon 117 is for example between 0.5 and 2 μm, for example of the order of 1 μm. For example, in top view, the conductive strip 117 has the general shape of a frame or square ring of about 25 μm side.

FIG. 5C illustrates a step of depositing a second electrically insulating layer 505, for example of the same nature as the insulating layer 501, on the upper face of the structure obtained at the end of the steps of FIG. 5B. The layer 505 is arranged so as to cover the upper face and the flanks of the conductive cord forming the heating resistor 117, so as to form with the layer 501 an insulating encapsulation sheath around the resistor 117. For example, the layer 505 is first deposited on the entire upper surface of the structure and, in particular, the layer 505 can be removed in comparison with the bolometric microplate 103 of the pixel.

In a variant, the insulating layer 501 is not etched in the step of FIG. 5A, except opposite the electrical connection pillars 121a, 121b. The layers 501 and 505 are then etched simultaneously during the same step subsequent to the formation of the conductive strip 117 and the deposition of the layer 505.

The following steps of the method are substantially identical to what has been described above in relation to FIGS. 3D, 3E and 3F and / or in relation with FIG. 4. It will be noted, however, that due to the electrical insulation of the heating resistor 117 by the dielectric layers 501, 505, the electrical isolation steps of the encapsulation caps of the different pixels are optional.

Figure 5D shows the pixel obtained at the end of the process.

An advantage of the radiation sensors of the type described in connection with FIGS. 1A, 1B, 2, 3A to 3F, 4 and 5A to 5D is that each pixel of the sensor can be individually protected against glare, depending on the measured heating. its bolometric microplanche, without the need to simultaneously close all the pixels of the sensor to implement the protection.

In addition, in the open state, the entire surface of the shutter opposite the bolometric microplate is transparent for the radiation to be detected, so that the shutter does not attenuate or little incident radiation.

Embodiments have been detailed above in which each pixel comprises a bolometric microplanche 103 disposed in a cavity specific to the pixel, closed hermetically by a cap 111 specific to the pixel. Those skilled in the art, however, will be able to provide variants in which several bolometric microplates belonging to distinct pixels are arranged in the same closed cavity by a common multi-pixel encapsulation cover.

Furthermore, embodiments of embodiments in which a heating resistor 117 is disposed under the upper face of the encapsulation cap 111 of the pixel, to control the opening, have been described in relation with FIGS. 1A, 1B, and 5A to 5D. and closing the optical shutter. Alternatively, the heating resistor 117 may be disposed above the upper face of the encapsulation cap 111 of the pixel, for example on the upper face of the thermochromic layer 115 or between the encapsulation cap 111 and the layer thermochrome 115.

In addition, embodiments have been described in which the thermochromic layer 115 of the optical shutter is disposed on and in contact with the upper face or outer face of the encapsulation cap 111 of the pixel. Alternatively, the thermochromic layer 115 may be disposed within the cavity 113, in contact with the underside of the upper portion of the hood 111.

Furthermore, the sensors described in relation with FIGS. 1A, 1B, 2, 3A to 3F, 4 and 5A to 5D may comprise additional, non-detailed elements. In particular, an antireflection layer, for example a zinc sulphide (ZnS) layer, may be provided on the upper face of the pixels. In addition, a reflective layer for the electromagnetic radiation to be detected may be provided under the bolometric microplate 103, on and in contact with the upper face of the control and reading circuit 102 of the pixel, so as to define a resonant cavity for the radiation to be detected between the bolometric microplanche 103 and the upper face of the circuit 102.

In addition, there is described above, with reference to FIGS. 1A, 1B, 2, 3A to 3F, 4 and 5A to 5D, examples of wherein each pixel of the sensor comprises a suspended microplanche 103 comprising a microbolometer. Alternatively, the microbolometer may be replaced by another type of incident electromagnetic radiation conversion element into an electrical signal, for example a thermal detector or a detector based on the conversion of incident radiation into thermal energy, for example a thermistor, a pyroelectric detector, or a detector based on field-effect or PN-diode-based transistors.

Second embodiment - passive protection on microplanche

FIG. 6 is a simplified sectional view of an example of a pixel 600 of a radiation sensor according to a second embodiment. The pixel 600 of FIG. 6 comprises elements that are common with the pixels described above. In the following, only the differences from the examples of the first embodiment will be detailed.

As in the examples described above, the pixel 600 of FIG. 6 is formed in and on a semiconductor substrate 101, for example made of silicon, and comprises an electronic reading and control circuit 102 formed in and on the substrate 101. , as in the preceding examples, the pixel 600 comprises a microplanche 103 suspended above the circuit 102 by thermal insulation arms, two arms 105a and 105b in the example shown, and vertical conductive pillars 107a, 107b connected to connection pads 109a, 109b of the circuit 102. The pixel 600 may further comprise, as in the previous examples, a cover 111 transparent to the radiation to be detected, resting on the upper face of the control circuit 102 and defining, with the face upper control circuit 102, a cavity or sealed enclosure 113 in which is located the suspended microplanche 103. The cavity 113 can be evacuated or pressio n less than the atmospheric pressure so as to reinforce the thermal insulation of the microplanche 103. Unlike the examples of the first embodiment, the pixel 600 of FIG. 6 does not comprise an active optical shutter and in particular does not comprise a thermochromic layer coating one face of the transparent cover 111 facing the microplanche 103 and connected to the control circuit and reading 102 of the pixel. More particularly, in the example of FIG. 6, the pixel 600 does not include the connecting pillars 121a, 121b, the connection pads 119a, 119b, the heating resistor 117, and the thermochromic layer 115 of the examples of the first embodiment of FIG. production.

Microplane 103 of pixel 600 comprises an element for converting incident electromagnetic radiation into thermal energy. In the example shown, the microplate 103 is a bolometric microplanche, that is to say it comprises an absorber 601, for example in the form of a conductive layer adapted to convert incident electromagnetic radiation into thermal energy. , and a thermistor 603 for measuring the temperature of the absorber. The absorber 601 is for example in the form of a layer extending over substantially the entire surface of the microplanche. By way of example, the absorber is made of titanium nitride (TiN). Thermistor 603 is made of a material hereinafter thermometer material whose electrical resistivity varies significantly depending on the temperature, for example amorphous silicon or vanadium oxide. For example, the thermistor 603 is in the form of a layer coating substantially the entire surface of the microplanche 103. Two ends of the thermistor are electrically connected to the control circuit of the pixel, by electrical connections not detailed on Figure 6, passing through the arms 105a, 105b and the pillars 107a, 107b of the suspension of the microplanche.

According to one aspect of the embodiment of FIG. 6, the microplanche 103 of the pixel 600 comprises a passive optical shutter comprising a layer 605 made of a thermochromic material, coating the element for converting incident radiation into thermal energy of the pixel, namely the absorber 601 in this example. By passive shutter is meant here that the shutter is not electrically controlled by the control circuit and reading 102 of the pixel. In the embodiment of FIG. 6, the thermochromic layer 605 is thermally coupled to the incident radiation converting element into thermal energy, and it is the heat produced by the conversion element which, when it becomes too much high, directly causes a modification of the optical properties of the thermochromic layer, so as to reduce its transmission coefficient for the incident radiation. In this example, the thermochromic layer 605 is disposed between the absorber 601 and the thermistor 603. The thermochromic layer 605 is chosen to have a coefficient of reflection increasing as a function of temperature. More particularly, the thermochromic layer 605 is chosen to be substantially transparent for the radiation to be detected below a transition temperature, and to exhibit a reflection coefficient for the radiation to be detected relatively high above its transition temperature. The transition temperature of the thermochromic layer is preferably greater than the maximum temperature that can reach the microplanche 103 in normal operation, and less than the maximum temperature that can support the microplanche 103 before degradation of the pixel. For example, the transition temperature of the thermochromic layer is between 60 and 180 ° C. The variation of the reflection or transmission coefficient of the thermochromic layer around the transition temperature is preferably relatively abrupt, for example greater than 2.5% per degree for a wavelength of 10 μm. In case of dazzling of the pixel, the shutter closes automatically under the effect of one heating of the microplanche. Part of the incident electromagnetic radiation is then reflected by the layer 605, and is no longer absorbed by the pixel conversion element. This avoids or limits a degradation of the microplanche bolometric of the pixel. When the temperature of the layer 605 falls below its transition temperature, the shutter reopens.

The material of the thermochromic layer 605 is, for example, a phase-change material, for example a crystallized metal oxide having a transparent insulating phase for the radiation to be detected below its transition temperature, and a reflective metal phase for the radiation to be radiated. detect above its transition temperature. The thermochromic material is for example crystallized vanadium dioxide (VO 2), having a transition temperature of the order of 68 ° C. Alternatively, the thermochromic material is vanadium dioxide crystallized and doped with low-valence cations, for example A13 +, Cr3 + or Ti4 +, so as to increase its transition temperature. More generally, depending on the desired transition temperature, other vanadium oxides can be used, for example V3O5. Alternatively, the thermochromic material of layer 605 is T13O5, 12O3, or Sm103. Alternatively, the thermochromic material of the layer 605 is a rare earth nickelate of general composition R i03, where R denotes a rare earth or a rare earth binary alloy, for example a compound of the type Sm _x Nd _] __ _x Ni03 or Eu _x Sm _] _{x x} Ni03. Alternatively, the thermochromic material of layer 605 is Ag2S or FeS.

FIGS. 7A, 7B, 7C, 7D, 7E and 7F are cross-sectional views illustrating steps of an example of a method of manufacturing a radiation sensor of the type described in connection with FIG. 6. FIGS. 7A , 7B, 7C, 7D, 7E and 7F more particularly illustrate the production of a single pixel 600 of the sensor, it being understood that, in practice, a plurality of identical or similar pixels can be formed simultaneously in and on the same semiconductor substrate 101 .

FIG. 7A illustrates a step of manufacturing the control and reading circuit 102 of the pixel 600, in and on the substrate 101, for example in CMOS technology. On the face 7A, only the electrical connection pads 109a and 109b of the circuit 102, flush with the upper face of the circuit, have been shown.

FIG. 7A further illustrates an optional step of forming, on the upper face of the circuit 102, a reflective layer 701, for example a layer of aluminum, silver or copper. This reflector, which can also be optionally provided in the examples of FIGS. 1A, 1B, 2, 3A to 3F, 4 and 5A to 5D, makes it possible to define a resonant cavity for the radiation to be detected, between the microplanche 103 and the face upper circuit 102, so as to increase the absorption of incident radiation by the microplanche 103. For example, the layer 701 is first deposited on the entire surface of the sensor, then etched to be kept only next to the future microplanche 103 of the pixel.

FIG. 7A further illustrates a step of depositing a sacrificial layer 301 on and in contact with the upper face of the circuit 102 and, if appropriate, of the reflector 701. The layer 301 is for example deposited continuously on substantially all the surface of the sensor. By way of example, the layer 301 is made of polyimide or silicon oxide. The thickness of the layer 301 sets the distance between the upper face of the circuit 102 and / or the reflector 701 and the bolometric microplate 103 of the pixel. By way of example, the layer 301 has a thickness of between 1 and 5 μm, for example of the order of 2.5 μm.

FIG. 7A further illustrates the formation of the electrical connection pillars 107a and 107b of the pixel, in vias etched in the sacrificial layer 301 in line with the connection pads 109a and 109b of the circuit 102. The pillars 107a and 107b are extend vertically over substantially the entire thickness of the layer 301, from the upper face of the connection pads 109a and 109b of the circuit 102, to the upper face of the layer 301.

FIG. 7A further illustrates a step subsequent to the formation of the pillars 107a and 107b, of deposit, on the face upper layer of the sacrificial layer 301, a first electrically insulating layer 703, for example silicon nitride (SiN), aluminum nitride (AlN) or silicon carbide (SiC), serving as a support for the construction of the microplanche 103 and thermal insulation arms 105a, 105b of the pixel. The layer 703 is for example deposited continuously over substantially the entire surface of the sensor, and etched locally above the pillars 107a and 107b so as to form openings 704a, 704b facing a central portion of the upper face. pillars 107a, 107b respectively, allowing the resumption of an electrical contact on the upper face of the pillars.

FIG. 7B illustrates a subsequent deposition step, on the upper face of the insulating layer 703 and on the upper face of the pillars 107a, 107b disengaged at the openings 704a, 704b, of a layer 601 made of an absorbent material for the radiation to detect, to form the microbolometer absorber of the pixel. The layer 601 is for example deposited continuously over substantially the entire surface of the sensor. In particular, in the example shown, the layer 601 is in contact with the upper face of the insulating layer 703, and with a portion of the upper face of the electrical connection pillars 107a, 107b.

FIG. 7B further illustrates a step of etching a trench 705 in the layer 601, for separating the absorber into two disjoint portions 601-a and 601-b in the future bolometric microplate 103 of the pixel. Indeed, in this example, the absorber 601 is made of an electrically conductive material, for example titanium nitride, and is used not only for its absorber function, but also as an electrical conductor for electrically connecting the ends of the thermistor. from the pixel to the circuit 102, via the electrical connection pillars 107a, 107b. It is therefore necessary to separate the absorber into two disjoint portions or electrodes, one (the portion 601-a) connected to the pillar 107a and at one end of the thermistor of the pixel, and the other (the portion 601- b) connected to the pillar 107b and to a second end of the pixel thermistor. The trench 705 extends vertically from the upper face to the lower face of the layer 601, and stops on the upper face of the insulating layer 703. In a view from above, the trench 705 for example extends over the entire width of the future bolometric microplanche 103, in a central part of the microplanche.

FIG. 7B furthermore illustrates a deposition step, on the upper face of the structure obtained after the formation of the trench 705, of a second electrically insulating layer 707, for example of the same kind as the layer 703, covering the upper face of the layer 601, as well as the side walls and the bottom of the trenches 705. The layer 707 is for example deposited continuously over substantially the entire surface of the sensor.

FIG. 7C illustrates a step of depositing the thermochromic layer 605 on and in contact with the upper face of the insulating layer 707. By way of example, the thermochromic layer 605 is deposited continuously over the entire surface of the sensor, then engraved to be kept only on the microplanche 103 of each pixel. In each pixel, two localized openings 709a and 709b are further etched in the thermochromic layer 605, respectively facing the portion 601-a and facing the portion 601-b of the absorber 601, for a step subsequent resumption of electrical contact on the portions 601-a and 601-b of the absorber, for connecting the thermistor of the pixel to the read circuit 102. The openings 709a and 709b are for example disposed respectively in the vicinity of two opposite edges of the microplanche. In the example shown, the openings 709a and 709b extend over the entire thickness of the thermochromic layer 605, and open on the upper face of the insulating layer 707. Depending on the type of thermochromic material used, annealing of the layer 605 may optionally be used to obtain the crystalline phase and the transition temperature of the desired material. By way of example, in the case of a vanadium dioxide (VO2) thermochromic layer, the deposit may be carried out at room temperature by spraying a vanadium target in an oxygen-containing atmosphere, and then annealing at a temperature of the order of 350 to 400 ° C. may be used to crystallize the oxide layer vanadium and obtain the desired thermochromic properties. The etching of the thermochromic layer 605 can be performed before or after the annealing. The thermochromic layer has for example a thickness of between 20 and 100 nm, for example between 20 and 60 nm.

FIG. 7C further illustrates a step subsequent to the etching and annealing of the thermochromic layer 605, deposition of a third electrically insulating layer 711, for example of the same nature as the layer 703 and / or the layer 707, on the upper surface of the structure. The layer 711 is for example deposited continuously over substantially the entire surface of the sensor. In the example shown, the insulating layer 711 extends on and in contact with the upper face and with the sides of the thermochromic layer 605, and on and in contact with the upper face of the insulating layer 707. The insulating layer 711 further extends over and in contact with the sidewalls and the bottom of the openings 709a, 709b formed in the thermochromic layer.

FIG. 7D illustrates a subsequent step of localized etching of the insulating layers 711 and 707 at the bottom of the openings 709a, 709b, so as to release access to the upper face of the portions 601-a, 601-b of the absorber 601.

FIG. 7E illustrates a step of depositing a layer 603 of the thermometer material, for example amorphous silicon or vanadium oxide, on the upper face of the structure obtained at the end of the step of FIG. 7D , to realize the thermistor of the pixel. The layer 603 is for example deposited continuously over substantially the entire surface of the sensor, and then etched to be retained only on the microplanche 103 of the pixel. The layer 603 is deposited in particular inside the openings 709a, 709b so that the thermistor 603 is connected on the one hand (by a first end) to the pad 109a of the circuit 102 via portion 601-a of absorber 601 and connecting pillar 107a, and secondly (at a second end) to pad 109b of circuit 102 via portion 601-a. b of the absorber 601 and the connecting pillar 107b.

Figure 7E further illustrates a deposition step of a fourth electrically insulating layer 713, for example of the same nature as the layers 703, 707 and / or 711, on the upper surface of the structure. The layer 713 is for example deposited continuously over substantially the entire surface of the sensor, and in particular on and in contact with the upper face and the sides of the thermistor 603.

FIG. 7F illustrates a subsequent etching step of the stack formed by the layers 703, 601, 707, 711 and 713 so as to delimit or individualize the microplanche 103 and the arms 105a, 105b of the pixel. During this step, the stack 703-601-707-711-713 is for example removed everywhere except at the level of the microplates 103 and arms 105a, 105b of the sensor pixels.

The method can then continue in a conventional manner, either directly by the removal of the sacrificial layer 301 to release the microplate 103 and the arms 105a, 105b of the pixel, or, if it is desired to form an encapsulation cover, by the depositing a second sacrificial layer and then forming the encapsulation cap according to methods of the type described with reference to FIGS. 3C to 3F.

In the example described with reference to FIGS. 7A to 7F, when a crystallization annealing of the layer 605 is carried out to obtain the desired thermochromic properties, this annealing is carried out before the deposition of the thermometer layer 603. Thermometer properties of layer 603 are not degraded by annealing. In particular, in the case where the thermochromic layer 605 and the thermometer layer 603 are both made of vanadium oxide, this makes it possible to obtain, in the same structure, crystalline vanadium dioxide in the thermochromic layer 605, and an oxide vanadium with different crystalline properties, optimized for its thermometer properties, in thermometer layer 603.

Alternatively, in the case where the thermometer material can withstand without significant degradation the crystallization annealing of the thermochromic material, which is the case for example amorphous silicon (in the case of a thermochromic layer of vanadium dioxide), the thermochromic layer 605 may be formed above the thermometer layer 603.

Furthermore, it is possible to deposit the thermochromic material above the thermometer material, then to crystallize the thermochromic material by rapid thermal annealing RTA type, for example by means of a lamp or a laser irradiating the upper face of the assembly, so as to limit the heating of the thermometer layer during annealing. In this case, a buffer layer may be further provided between the thermometer layer and the thermochromic layer, to limit the heating of the thermometer layer during annealing.

In another embodiment, the thermochromic layer 605 may be disposed on the side of the lower face of the absorber 601, that is to say on the side of the absorber opposite to the illumination face of the pixel. This configuration, although less advantageous, also makes it possible to limit the absorption of the incident electromagnetic radiation during closure of the shutter, insofar as the shutter then limits the absorption of the flux reflected by the layer 701.

Note also that in the case where the thermochromic material is a crystalline metal oxide having an insulating phase and a metal phase, the transition between the two phases can be accompanied by a significant variation in the density of the layer 605. can induce constraints likely to destabilize the microplanche. To limit these constraints, suitable structures of the thermochromic layer 605 can be provided. For example, in each pixel of the sensor, the thermochromic layer 605 of the pixel may be a discontinuous layer consisting of a plurality of disjoint regions separated by relatively narrow trenches, for example not occupying more than 20% of the surface of the microplanche.

Preferably, to promote the thermal coupling between the absorber 601 and the thermochromic layer 605, the distance between the absorber 601 and the thermochromic layer 605 is relatively small, for example less than 20 nm. By way of example, the insulating layer 707, interfacing between the absorber 601 and the thermochromic layer 605 (the insulating layer 707 has its lower face in contact with the upper face of the absorber 601 and its upper face in contact with the the lower face of the thermochromic layer 605) has a thickness of between 1 and 20 nm, for example of the order of 10 nm.

FIG. 7bis is a sectional view illustrating an alternative embodiment of the radiation sensor described in relation with FIGS. 7A to 7F, in which the absorber 601 is in contact with the thermochromic layer 605, so as to maximize the thermal coupling between The absorber 601 and the thermochromic layer 605.

The radiation sensor of Fig. 7bis, and its manufacturing method, comprise elements common with the radiation sensor and the manufacturing method described in connection with Figs. 7A-7F. In the following, only the differences between the two exemplary embodiments will be detailed.

The manufacture of the sensor of FIG. 7bis is identical or similar to that of the sensor of FIGS. 7A to 7F, until the deposition of the absorption conductive layer 601.

As in the example of Figs. 7A-7F, a trench 705 is then etched into the layer 601 to separate the absorber 601 into two disjoint portions 601-a and 601-b. However, unlike the example of FIGS. 7A to 7F, seen from above, the portions 601-a and 601-b of the absorber are symmetrical with respect to the trench 705, in the example of FIG. 7bis. , the portions 601-a and 601-b of the absorber are asymmetrical with respect to the trench 705. More particularly, in this For example, viewed from above, the surface of the portion 601-a of the absorber is greater, for example at least three times greater, on the surface of the portion 601-b of the absorber.

As in the example of FIGS. 7A to 7F, an electrically insulating layer 707 is deposited on the upper face of the structure obtained after the formation of the trench 705, that is to say on the upper face of the layer 601, as well as on the side walls and on the bottom of the trench 705. In the example of FIG. 7bis, the layer 707 is then locally removed to expose all or part of the upper face of the portion 601-a of the absorber and / or all or part of the upper face of the portion 601-b of the absorber. In this example, the layer 707 is only preserved in the trench 705 and possibly on a part of the upper face of the portion 601-a and / or the portion 601-b of the absorber, in the immediate vicinity. from the trench 705.

The thermochromic layer 605 is then deposited identically or similarly to what has been previously described, except that, in the example of FIG. 7bis, the thermochromic layer 605 comes into contact with the upper face of the portions 601. a and 601-b of the absorber.

In addition, in the example of FIG. 7bis, to prevent the thermochromic layer 605 from bypassing the portions 601-a and 601-b of the absorber, a through trench 706 opening onto the upper face of the layer 707, is formed in the thermochromic layer 605 in line with the trench 705, so as to separate the thermochromic layer 605 into two disjoint parts 605-a and 605-b, respectively in contact with the portion 601-a of 1 absorber and with the portion 601-b of the absorber. The trench 706 is for example formed at the same time as the contact recovery openings 709a and 709b (Figure 7C).

The process may then continue in the same or similar manner as described in connection with Figs. 7A-7F. More generally, other manufacturing processes can be envisaged making it possible to bring the thermochromic layer 605 into contact with all or part of the upper face of the absorber 601. By way of example, the thermochromic layer may be in contact with a only two portions 601-a and 601-b of the absorber, preferably the larger of the two portions (portion 601-a in the example of Figure 7a) to promote thermal coupling, and be electrically isolated from the other portion of the absorber by the layer 707. This makes it possible to dispense with the separation trench 706 of the example of FIG. 7bis.

An advantage of the radiation sensors of the type described in connection with FIGS. 6, 7A to 7F, and 7bis is that each pixel of the sensor is individually protected against glare.

In addition, in the open state, the entire surface of the shutter opposite the element for converting the incident radiation into thermal energy is transparent for the radiation to be detected, so that the shutter does not attenuate or weaken the incident radiation.

Another advantage of the embodiment of FIGS. 6,

7A to 7F, and 7bis is that the anti-glare protection obtained is a passive protection, operating even in the absence of power supply of the sensor.

Embodiments of the invention in which each pixel comprises a suspended microplanche 103 disposed in a pixel-specific cavity hermetically sealed by a pixel-specific cover 111 are detailed above. Those skilled in the art, however, will be able to provide variants in which several suspended microplates belonging to distinct pixels are arranged in the same closed cavity by a common multi-pixel encapsulation cover.

In addition, the embodiment of FIGS. 6, 7A to 7F, and 7bis is compatible with a sensor in which suspended microplanes 103 are not surmounted by an encapsulation cap. In addition, embodiments relating to each pixel of the sensor comprising a suspended microplanche 103 comprising a microbolometer have been described above, in relation to FIGS. 6, 7A to 7F, and 7bis. Alternatively, the microbolometer may be replaced by another type of incident electromagnetic radiation conversion element into an electrical signal, for example a thermal detector or a detector based on the conversion of incident radiation into thermal energy, for example a thermistor, a pyroelectric detector, or a detector based on field-effect or PN-diode-based transistors.

Third Embodiment - Passive Arm Protection FIG. 8 is a simplified sectional view of an example of a pixel 800 of a radiation sensor according to a third embodiment. The pixel 800 of FIG. 8 comprises elements that are common with the pixels described above. In the following, only the differences from the examples of the first and second embodiments will be detailed.

As in the examples described above, the pixel 800 of FIG. 8 is formed in and on a semiconductor substrate 101, for example made of silicon, and comprises an electronic reading and control circuit 102 formed in and on the substrate 101. , as in the preceding examples, the pixel 800 comprises a microplanche 103 suspended above the circuit 102 by thermal insulation arms, two arms 105a and 105b in the example shown, and vertical conductive pillars 107a, 107b connected to connection pads 109a, 109b of the circuit 102. The pixel 800 may further comprise, as in the previous examples, a cover 111 transparent to the radiation to be detected, resting on the upper face of the control circuit 102 and defining, with the face upper control circuit 102, a cavity or sealed enclosure 113 in which is located the suspended microplanche 103. The cavity 113 can be evacuated or pressio not less than the atmospheric pressure so as to reinforce the thermal insulation of the microplanche 103.

Unlike the examples of the first embodiment, the pixel 800 of FIG. 8 does not comprise an active optical shutter and in particular does not comprise a thermochromic layer coating one face of the transparent cover 111 facing the microplate 103 and connected to the control circuit and reading 102 of the pixel. More particularly, in the example of FIG. 8, the pixel 800 does not include the connecting pillars 121a, 121b, the connection pads 119a, 119b, the heating resistor 117, and the thermochromic layer 115 of the examples of the first embodiment of FIG. production.

In addition, unlike the examples of the second embodiment, the pixel 800 of FIG. 8 does not include a passive optical shutter constituted by a thermochromic layer integrated in the microplanche 103.

The microplanche 103 of the pixel 800 comprises a conversion element of the incident electromagnetic radiation into thermal energy. In the example shown, the microplanche 103 is a bolometric microplate differing from the microplanche 103 of the pixel 600 of FIG. 6 mainly in that it does not include the thermochromic layer 605. Thus, the microplanche 103 of the pixel 800 of FIG. 8 comprises an absorber 601, for example identical or similar to that of the microplanche 103 of the pixel of FIG. 6, and a thermistor 603 thermally coupled to the absorber 601, for example identical or silimar to that of the microplanche 103 of the pixel of FIG. Figure 6. Two ends of the thermistor are electrically connected to the control circuit of the pixel, by electrical connections through the arms 105a, 105b and the pillars 107a, 107b suspension of the microplanche.

According to one aspect of the embodiment of FIG. 8, at least one of the thermal insulation arms of the pixel 800 comprises a layer 801 made of a phase-change material having, below a transition temperature, a relatively high thermal conductivity. low, and, above the temperature of transition, a relatively high thermal conductivity, that is to say greater than its thermal conductivity below the transition temperature. In the example shown, the layer 801 is present in the two suspension arms 105a, 105b of the pixel, and is not present on the microplate 103. The variation of the thermal conductivity of the phase-change material around the temperature transition is preferably relatively steep, for example greater than 0.08 W / mK per degree. Passive anti-glare protection of the pixel is thus obtained, the operation of which is as follows. In normal operation, the material of the layer 801 has a relatively low thermal conductivity, so that the thermal resistance between the microplate 103 and the substrate is relatively high, which is favorable to the implementation of measurements of the incident electromagnetic radiation. In the event of excessive heating of the microplate 103 linked to a glare of the pixel, the thermal insulation arms 105a, 105b of the pixel also heat up, until the transition temperature of the layer 801 is reached. The thermal resistance of the or arms comprising the phase-change material then drop abruptly, which causes the evacuation towards the substrate 101 of a part of the heat accumulated in the microplanche 103. This makes it possible to avoid or limit a degradation of the microplanche bolometric of the pixel. When the temperature of the heat-insulating arms drops below the temperature of the phase-change material, the latter regains a relatively low thermal conductivity, and the pixel can again function normally.

The transition temperature of the layer 801 is preferably greater than the maximum temperature that can reach the microplanche 103 in normal operation, and less than the maximum temperature that can support the microplanche 103 before degradation of the pixel. For example, the transition temperature of the layer 801 is between 60 and 180 ° C. The material of the layer 801 is for example a metal oxide crystallized having, below its transition temperature, an insulating phase of relatively low thermal conductivity, and, above its transition temperature, a metal phase of relatively high thermal conductivity. The material of the layer 801 is for example crystallized vanadium dioxide (VO2), having a transition temperature of the order of 68 ° C. Alternatively, the material of the layer 801 is vanadium dioxide crystallized and doped with low valence cations, for example A13 +, Cr3 + or Ti4 +, so as to increase its transition temperature. More generally, depending on the desired transition temperature, other vanadium oxides can be used, for example V3O5. By way of example, the layer 801 has a thickness of between 10 and 100 nm, for example of the order of 50 nm.

FIGS. 9A, 9B, 9C, 9D, 9E, 9F and 9G are cross-sectional views illustrating steps of an example of a method of manufacturing a radiation sensor of the type described with reference to FIG. FIGS. 9A, 9B, 9C, 9D, 9E, 9F and 9G illustrate more particularly the production of a single pixel 800 of the sensor, it being understood that, in practice, a plurality of identical or similar pixels can be formed simultaneously in and on a same semiconductor substrate 101.

FIG. 9A illustrates a step of manufacturing the control and reading circuit 102 of the pixel 800, in and on the substrate 101, for example in CMOS technology. On the face

9A, only the electrical connection pads 109a and 109b of the circuit 102, flush at the upper face of the circuit, have been shown.

FIG. 9A further illustrates an optional step of forming, on the upper face of the circuit 102, a reflective layer 701 identical or similar to the layer 701 of the example of FIG. 7A.

FIG. 9A further illustrates a step of depositing a sacrificial layer 301 on and in contact with the upper face of the circuit 102 and / or the reflector 701. The layer 301 is for example identical or similar to the layer 301 of Figure 7A.

FIG. 9A further illustrates a step of forming the electrical connection pillars 107a and 107b of the pixel, in vias etched in the layer 301 in line with the connection pads 109a and 109b of the circuit 102. The pillars 107a and 107b extend vertically over substantially the entire thickness of the layer 301, from the upper face of the connection pads 109a and 109b of the circuit 102, to the upper face of the layer 301.

FIG. 9A further illustrates a deposition step, on the upper face of the sacrificial layer 301 and on the upper face of the pillars 107a, 107b, of a first layer 801 'made of the same material as that of the future layer 801 of the pixel. The function of the layer 801 'is mainly to balance the mechanical stresses of the assembly during the production of the layer 801. In this example, the layer 801' is not annealed at the stage of FIG. 9A, and will optionally be annealed only later, at the same time as the layer 801, in the case where the layer 801 of the pixel is of a material requiring annealing to obtain the desired crystalline phase and transition temperature. By way of example, the layer 801 'is an amorphous vanadium dioxide layer deposited at ambient temperature by spraying a vanadium target in an atmosphere containing oxygen.

FIG. 9A further illustrates a deposition step, on the upper face of the layer 801 ', of a first electrically insulating layer 803, for example made of silicon nitride (SiN), aluminum nitride (AlN) or silicon carbide (SiC).

The layers 801 'and 803 are for example continuously deposited over substantially the entire surface of the sensor, then etched locally above the pillars 107a and 107b so as to form openings 804a, 804b facing a central portion of the upper face of the pillars 107a, 107b respectively, allowing the recovery of an electrical contact on the upper face of the pillars.

FIG. 9B illustrates a deposition step, on the upper face of the insulating layer 803, and on the bottom and on the flanks of the openings 804a, 804b, of a layer 601 made of an absorbent material for the radiation to be detected, to form The microbolometer absorber of the pixel. The layer 601 is for example deposited continuously over substantially the entire surface of the sensor. In particular, in the example shown, the layer 601 is in contact with the upper face of the insulating layer 803, and with a portion of the upper face of the electrical connection pillars 107a, 107b.

FIG. 9B further illustrates a deposition step, on and in contact with the upper face of the layer 601, of the layer 801 made of a phase-change material. The layer 801 is for example deposited continuously over substantially the entire surface of the sensor. The layer 801 deposited in the step of FIG. 9B is for example a layer of the same kind as the layer 801 'deposited in the step of FIG. 9A. By way of example, the layer 801 deposited in the step of FIG. 9B is an amorphous vanadium dioxide layer formed by sputtering a vanadium target at room temperature in an oxygen-containing atmosphere. The layers 801 'and 801 have for example substantially the same thickness. By way of example, each of the layers 801 'and 801 has a thickness of between 10 and 100 nm, for example of the order of 50 nm.

FIG. 9C illustrates a step of annealing the structure obtained at the end of the steps of FIG. 9B, aimed at conferring on the layer 801 the desired variable thermal conductivity properties. During this step, the layer 801 'also acquires the same properties of variable thermal conductivity as the layer 801. By way of example, in the case of layers 801' and 801 of vanadium dioxide, annealing at a temperature of 100.degree. The order of 350 to 400 ° C can be used to obtain the desired properties. Figure 9D illustrates a step of removing the layer 801 over the entire surface of the future microplanche 103, for example by etching.

FIG. 9D further illustrates a subsequent step of etching a trench 805 in the layer 601, for separating the absorber into two disjoint portions 601-a and 601-b in the future bolometric microplate 103 of the pixel. Indeed, in this example, the absorber 601 is made of an electrically conductive material, for example titanium nitride, and is used not only for its absorber function, but also for electrically connecting the ends of the pixel thermistor to the circuit 102, through the electrical connection pillars 107a, 107b. It is therefore necessary to separate the absorber into two disjoint portions, one (the portion 601-a) connected to the pillar 107a and at one end of the thermistor of the pixel, and the other (the portion 601-b) connected at the pier 107b and at a second end of the pixel thermistor. The trench 805 extends vertically from the upper face to the lower face of the layer 601, and stops on the upper face of the insulating layer 803. In a view from above, the trench 805 for example extends over the entire width of the future bolometric microplanche 103, in a central part of the microplanche.

FIG. 9D further illustrates a deposition step, on the upper face of the structure obtained after the formation of the trench 805, of a second electrically insulating layer 807, for example of the same kind as the layer 803, covering the upper face. of the layer 601, as well as the side walls and the bottom of the trenches 805. The layer 807 is for example deposited continuously over substantially the entire surface of the sensor.

In each pixel, two localized openings 809a and 809b are made in the insulating layer 807, for example by etching, respectively facing the portion 601-a and facing the portion 601-b of the absorber 601, in order to a subsequent step of resumption of electrical contact on the portions 601-a and 601-b of the absorber, for connecting the pixel thermistor to the reading circuit 102. The openings 809a and 809b are for example disposed respectively in the vicinity of two opposite edges of the microplanche. In the example shown, the openings 809a and 809b extend vertically over the entire thickness of the insulating layer 807, and open on the upper face of the layer 601.

FIG. 9E illustrates a step of depositing a layer 603 of the thermometer material, for example amorphous silicon or vanadium oxide, on the upper face of the structure obtained at the end of the step of FIG. 9D , to realize the thermistor of the pixel. The layer 603 is for example deposited continuously over substantially the entire surface of the sensor, and then etched to be retained only on the microplanche 103 of the pixel. The layer 603 is in particular deposited inside the openings 809a, 809b so that the thermistor 603 is connected on the one hand (by a first end) to the pad 109a of the circuit 102 via the portion 601-a of the the absorber 601 and the connecting pillar 107a, and secondly (by a second end) to the pad 109b of the circuit 102 through the portion 601-b of the absorber 601 and the connecting pillar 107b.

Figure 9E further illustrates a step of depositing a third electrically insulating layer 811, for example of the same nature as the layers 803 and 807, on the upper surface of the structure. The layer 811 is for example deposited continuously over substantially the entire surface of the sensor, and in particular on and in contact with the upper face and the sides of the thermistor 603.

FIG. 9F illustrates a subsequent step of removing the insulating layers 807 and 811 outside the future microplate 103 of the pixel, and in particular opposite the future thermal insulation arms 105a, 105b of the pixel. By way of example, the layers 807 and 811 are removed everywhere except with respect to the microplates 103 of the pixels of the sensor. FIG. 9G illustrates a subsequent etching step of the stack formed by the layers 801, 601, 803 and 801 'so as to delimit or individualise the microplanche 103 and the arms 105a, 105b of the pixel. During this step, the stack 801-601-803-801 'is for example removed everywhere except at the level of the microplates 103 and arms 105a, 105b of the sensor pixels.

An advantage of the radiation sensors of the type described in connection with FIGS. 8 and 9A to 9G is that each pixel of the sensor is individually protected against glare.

Another advantage is that, in the absence of glare, the protective device does not attenuate the radiation to be detected.

Another advantage of the embodiment of FIGS. 8 and 9A to 9G is that the anti-glare protection obtained is passive protection, operating even in the absence of power supply to the sensor.

Alternatively, the layer 801 'of the method described in connection with FIGS. 9A to 9G may be omitted, which makes it possible not to add an additional layer to the microplanche 103 of the pixel, and therefore not to increase the thermal capacity of the microplanche. Another advantage is that this also limits the number of layers present in the thermal insulation arms, and therefore limits their thermal conductivity, which improves the thermal sensitivity of the pixel.

Various embodiments with various variants have been described above. It will be appreciated that those skilled in the art may combine various elements of these various embodiments and variants without demonstrating inventive activity. In particular, the anti-glare protections of the first, second and third embodiments may be combined in whole or part in the same pixel of a radiation sensor. In particular, it will be possible to combine the active protection of the first embodiment with one and / or the other passive protection of the second and third embodiments, or to combine the passive protection of the second and third embodiments without active protection of the first embodiment.

Claims

A radiation sensor comprising a plurality of pixels (800) formed in and on a semiconductor substrate (101), each pixel having a microplate (103) suspended above the substrate by heat-insulating arms (105a, 105b) the microplanche (103) comprising an element (601) for converting electromagnetic radiation into thermal energy,

wherein, in each pixel (800), at least one of the thermal insulation arms (105a, 105b) of the pixel comprises a layer (801) of a phase change material having, below a transition temperature of phase, a first value of thermal conductivity, and, above the phase transition temperature, a second value of thermal conductivity greater than the first value,

and wherein said layer (801) is not present on the microplanche (103).

The sensor of claim 1, wherein said layer (801) is a metal oxide having an insulating phase below the transition temperature and a metal phase above the transition temperature.

The sensor of claim 2, wherein said layer (801) is a vanadium oxide.

The sensor of claim 2, wherein said layer (801) is a titanium oxide.

5. Sensor according to any one of claims 1 to 4, wherein the transition temperature of said layer

(801) is between 60 and 180 ° C.

The sensor of any one of claims 1 to 5, wherein each pixel (800) further comprises a thermistor (603) thermally coupled to the conversion element (601) of the pixel.

The sensor of claim 6, wherein each pixel (800) further comprises a circuit (102) for reading the value of the thermistor (603) of the pixel.

The sensor of any one of claims 1 to 7, wherein in each pixel (800), the conversion element (601) is a layer of absorbent material for the radiation to be detected.

The sensor of claim 8, wherein in each pixel (800) the conversion element (601) is a metal layer.

The sensor according to any one of claims 1 to 8, wherein in each pixel (800), the heat insulating arms (105a, 105b) rest on vertical electrical connection pillars (107a, 107b).

11. A sensor according to any one of claims 1 to 10, wherein, in each pixel (800), the microplanche (103) and the heat-insulating arms (105a, 105b) are arranged in a cavity (113) closed by a cover (111) transparent for the radiation to be detected.

The sensor of claim 11, wherein, in each pixel (800), the transparent cover (111) seals the cavity (113), and the cavity (113) is at a pressure below atmospheric pressure.