JP2011237282A - Method of manufacturing detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method manufacturing a detector by which the warpage and the cutting of a supporting member supporting a detection element above a cavity part can be suppressed.SOLUTION: In a manufacturing method for a detector, a cavity part 102 is formed in a fixing part 100, external force EF is imparted to the fixing part 100 in the same direction as the direction of stress generated in a supporting member formed before formation of a sacrificial layer 150 in the cavity part 102, the supporting member 210 is formed in the sacrificial layer 150 formed in the cavity part 102 and the external force EF is released before removal of the sacrificial layer 150.

Description

  The present invention relates to a detector manufacturing method and the like.

  As a thermal photodetection device, a pyroelectric or bolometer type infrared detection device is known. Infrared detectors use the fact that the amount of spontaneous polarization of the pyroelectric material changes (pyroelectric effect or pyroelectronic effect) depending on the amount of light (temperature) of the received infrared light (electromotive force (due to polarization) at both ends of the pyroelectric material). Infrared light is detected by generating a charge) (pyroelectric type) or changing a resistance value according to temperature (bolometer type). The pyroelectric infrared detection device has an advantage in that the detection sensitivity is excellent while the manufacturing process is complicated as compared with the bolometer infrared detection device.

  A thermal type photodetection element (a thermal type detection element in a broad sense, a detection element in a broader sense) usually has a structure without a cooling device. Therefore, by placing the device in an airtight package, placing the device in a reduced pressure environment, and further thermally separating it from the substrate and the peripheral film, the heat generated by the received light (infrared rays, etc.) is not diffused to the surroundings as much as possible. There is a need to. In order to prevent the dissipation of heat to the substrate and suppress the deterioration of the detection characteristics of the thermal detection element, for example, a structure in which a cavity for thermal separation is provided between the substrate and the thermal detection element. Employing is effective (see, for example, Patent Document 1 and Patent Document 2). Patent Document 1 shows a thermal infrared array sensor having a cavity for heat separation, and Patent Document 2 shows a pyroelectric infrared detection element having a cavity for heat separation.

JP 2000-205944 A JP 2002-214038 A

  A thermal photodetection element (detection element in a broad sense) is mounted on a membrane supported by a substrate. In a region facing the thermal detection element, a cavity is formed between the membrane and the substrate. In the manufacturing process, a sacrificial layer is embedded in the cavity of the substrate, a membrane is formed on the sacrificial layer, and a thermal photodetecting element is further formed on the membrane.

  The range of the detection element includes, for example, an ultrasonic detection element and an acceleration detection element in addition to the thermal detection element. The detection element can also have the same structure as the thermal detection element (thermal detection element in a broad sense).

  In this manufacturing process, as long as the sacrificial layer is flat, the membrane also has flatness. However, when the sacrificial layer is removed by isotropic etching in the final process, a problem has been found that the membrane is warped. In some cases, a problem that the membrane is cut was also found.

  In some aspects of the present invention, there is provided a method for manufacturing a detector that can suppress warping and cutting that occur in a member that supports a detection element.

(1) A method for manufacturing a detector according to an aspect of the present invention includes:
Forming a cavity in the fixed part,
Before forming the sacrificial layer in the cavity, an external force is applied to the fixed portion in the same direction as the stress direction generated in the formed support member,
Forming the support member on the sacrificial layer formed in the cavity,
The external force is released before the sacrificial layer is removed.

  According to one embodiment of the present invention, warping and cutting of the support member can be suppressed by releasing the external force before applying the external force to the fixed portion and then forming the support member and removing the sacrificial layer. .

  (2) In one aspect of the present invention, the external force can be applied to the fixed portion by forming a member of the same type as the support member on the surface of the fixed portion facing the sacrificial layer.

  If it carries out like this, external force can be generated by forming the same kind of member in a fixed part.

(3) A method for manufacturing a detector according to another aspect of the present invention includes:
Forming at least one recess or protrusion for forming the support member redundant length in the sacrificial layer;
Forming a support member along the at least one recess or projection;
Thereafter, the sacrificial layer is removed.

  According to another aspect of the present invention, since the redundant portion is added to the support member, even if the support member is extended due to the tensile residual stress of the support member, cutting of the support member can be suppressed.

  (4) In another aspect of the present invention, the at least one concave portion or convex portion can be formed by subjecting the sacrificial layer to chemical mechanical polishing.

  In this way, at least one concave portion or convex portion can be formed by chemical mechanical polishing.

  (5) In another aspect of the present invention, the at least one concave portion or convex portion can be formed by etching the sacrificial layer.

  If it carries out like this, at least 1 recessed part or convex part can be formed by an etching.

  (6) In another aspect of the present invention, when the depth of the at least one concave portion or convex portion is h and the radius of curvature is R, the at least one concave portion or convex portion is defined by the radius of curvature. It is in contact with a virtual circle, σ is the tensile residual stress generated in the support member, E is the Young's modulus of the fixed part of the support member, T is the thickness of the fixed part, and ν is the Poisson's ratio of the support member , Tf is the thickness of the support member, and d is the thickness of the cavity of the fixed portion, h and R can satisfy the following relational expression.

  If it carries out like this, the redundant part of a supporting member can be formed in a virtual circle.

(7) In another aspect of the present invention, the method for manufacturing a detector includes applying an external force substantially the same as the stress direction and the stress value of the tensile residual stress generated in the support member to the fixing portion of the support member. Forming a support member;
Thereafter, with the external force released, the sacrificial layer can be removed from the fixed portion to form a cavity in the fixed portion.

  If it carries out like this, the cutting | disconnection of a support member can be suppressed combining two solution methods.

1 (A) to 1 (D) are diagrams for explaining the occurrence of warping or cutting of a support member on which a thermal detection element (detection element in a broad sense) according to an embodiment of the present invention is mounted, and a mechanism for eliminating it. It is a schematic explanatory drawing. 1 is a schematic plan view of a pyroelectric infrared detection device (detection device in a broad sense) according to an embodiment of the present invention. FIG. 3 is a schematic sectional view of a pyroelectric detector (detector in a broad sense) for one cell of the pyroelectric infrared detector shown in FIG. 2. It is a schematic sectional drawing of the manufacturing process which shows the supporting member and infrared rays detection element (detection element in a broad sense) formed on a sacrificial layer. FIG. 5A to FIG. 5D are schematic explanatory views of a method for manufacturing a pyroelectric detector (detector in a broad sense) according to an embodiment of the present invention. 6 (A) to 6 (D) are modified examples of the manufacturing method of FIGS. 5 (A) to 5 (D). FIG. 7A to FIG. 7C are schematic explanatory diagrams of another manufacturing method of the pyroelectric detector (detector in a broad sense) according to the embodiment of the present invention. 8 (A) to 8 (C) are modified examples of the manufacturing method of FIGS. 7 (A) to 7 (C). 9 (A) to 9 (C) show another modification of the manufacturing method of FIGS. 7 (A) to 7 (C). It is a schematic sectional drawing for demonstrating the capacitor structure of the pyroelectric infrared detector (detector in a broad sense) which concerns on embodiment of this invention. It is a schematic sectional drawing which shows the modification which strengthened the reducing gas barrier property of wiring plug vicinity. 1 is a block diagram of an electronic apparatus including a thermal photodetector (detector in a broad sense) or a thermal photodetector (detector in a broad sense). FIGS. 13A and 13B are diagrams showing a configuration example of a pyroelectric detection device (detection device in a broad sense) in which pyroelectric detectors (detectors in a broad sense) are two-dimensionally arranged. is there.

  Hereinafter, preferred embodiments of the present invention will be described in detail. The present embodiment described below does not unduly limit the contents of the present invention described in the claims, and all the configurations described in the present embodiment are indispensable as means for solving the present invention. Not always.

1. Thermal infrared detector (detector in a broad sense)
As will be described later, the support member in which the warping and cutting that occur as shown in FIGS. 1A to 1C are eliminated as shown in FIG. 1D, and the pyroelectric detection element mounted thereon A multi-cell pyroelectric infrared detector (a thermal detector in a broad sense, a detector in a broader sense) in which each cell has a thermal detector (in a broad sense, a detector in a broader sense). However, FIG. 2 shows a pyroelectric infrared detection device (a thermal detection device in a broad sense, a detection device in a broader sense) arranged in two orthogonal axes. Note that some of the plurality of cells may include a pyroelectric detection element (a thermal detection element in a broad sense, a detection element in a broader sense) illustrated in FIG. In addition, the pyroelectric infrared detector may be configured with a pyroelectric infrared detector for only one cell. In FIG. 2, a plurality of posts 104 are erected from a base (also referred to as a fixed portion) 100. For example, a pyroelectric infrared detector 200 for one cell supported by two posts 104 is arranged in two orthogonal axes. It is arranged. The area occupied by the pyroelectric infrared detector 200 for one cell is, for example, 30 × 30 μm.

  Hereinafter, an infrared detection element (thermal photodetection element in a broad sense) 220 will be described as an example of the detection element. The structure of the pyroelectric infrared detector 200 that suppresses warping and cutting of the support member (membrane) 210 is applied to a detector including an ultrasonic detection element, an acceleration detection element, and the like having a support member that can be warped or cut. it can.

  As shown in FIG. 2, the pyroelectric infrared detector 200 includes a support member (membrane) 210 connected to two posts 104 and an infrared detection element (thermal detection element in a broad sense) 220. Contains. The area occupied by the pyroelectric infrared detection element 220 for one cell is, for example, 10 × 10 μm.

  The pyroelectric infrared detector 200 for one cell is not contacted except for being connected to the two posts 104, and a cavity 102 (see FIG. 3) is formed below the pyroelectric infrared detector 200. In addition, an opening 102A communicating with the cavity 102 is disposed around the pyroelectric infrared detector 200 in plan view. Thereby, the pyroelectric infrared detector 200 for one cell is thermally separated from the pyroelectric infrared detectors 200 of the base 100 and other cells.

  The support member 210 includes a mounting portion 210A for mounting and supporting the infrared detection element 220, and two arms 210B connected to the mounting portion 210A. The free ends of the two arms 210B are attached to the post 104. It is connected. The two arms 210 </ b> B are narrow and redundantly formed so as to thermally separate the infrared detection element 220.

  FIG. 2 is a plan view in which members above the wiring layer connected to the upper electrode are omitted. FIG. 2 shows a first electrode (lower electrode) wiring layer 222 and a second electrode connected to the infrared detecting element 220. A (upper electrode) wiring layer 224 is shown. Each of the first and second electrode wiring layers 222 and 224 extends along the arm 210 </ b> B and is connected to a circuit in the base portion 100 via the post 104. The first and second electrode wiring layers 222 and 224 are also formed to extend narrowly and redundantly in order to thermally separate the infrared detection element 220.

2. Outline of Pyroelectric Infrared Detector (Detector in a Broad Definition) FIG. 3 is a cross-sectional view of the pyroelectric infrared detector 200 shown in FIG. FIG. 4 is a partial cross-sectional view of the pyroelectric infrared detector 200 during the manufacturing process. In FIG. 4, the cavity 102 of FIG. 3 is filled with a sacrificial layer 150. The sacrificial layer 150 exists from before the formation process of the support member 210 and the pyroelectric infrared detection element 220 to after the formation process, and is removed by isotropic etching after the formation process of the pyroelectric infrared detection element 220. Is.

  As shown in FIG. 3, the base 100 includes a substrate, for example, a silicon substrate 110 and a spacer layer 120 formed of an interlayer insulating film on the silicon substrate 110. The post 104 is formed by etching the spacer layer 120. A plug 106 connected to one of the first and second electrode wiring layers 222 and 224 can be disposed on the post 104. The plug 106 is connected to a row selection circuit (row driver) provided on the silicon substrate 110 or a readout circuit that reads data from the photodetector via a column line. The cavity 102 is formed simultaneously with the post 104 by etching the spacer layer 120. The opening 102A shown in FIG. 2 is formed by pattern-etching the support member 210.

The infrared detection element 220 mounted on the support member 210 includes a capacitor 230. The capacitor 230 includes a pyroelectric body 232, a first electrode (lower electrode) 234 connected to the lower surface of the pyroelectric body 232, and a second electrode (upper electrode) 236 connected to the upper surface of the pyroelectric body 232. Contains. The first electrode 234 can include an adhesion layer 234D that enhances adhesion between the support member 210 and a first layer member (eg, SiO ₂ ).

  The capacitor 230 is covered with a reducing gas barrier layer 240 that suppresses the entry of reducing gas (hydrogen, water vapor, OH group, methyl group, etc.) into the capacitor 230 in the process after the capacitor 230 is formed. This is because the pyroelectric body (such as PZT) 232 of the capacitor 230 is an oxide, and when the oxide is reduced, oxygen vacancies are generated and the pyroelectric effect is impaired.

As shown in FIG. 4, the reducing gas barrier layer 240 includes a first barrier layer 242 and a second barrier layer 244. The first barrier layer 242 can be formed, for example, by depositing aluminum oxide Al ₂ O ₃ by sputtering. Since no reducing gas is used in the sputtering method, the capacitor 230 is not reduced. The second hydrogen barrier layer 244 can be formed, for example, by depositing aluminum oxide Al ₂ O ₃ by, for example, atomic layer chemical vapor deposition (ALCVD). A normal CVD (Chemical Vapor Deposition) method uses a reducing gas, but the capacitor 230 is isolated from the reducing gas by the first barrier layer 242.

Here, the total film thickness of the reducing gas barrier layer 240 is 50 to 70 nm, for example, 60 nm. At this time, the film thickness of the first barrier layer 242 formed by the CVD method is thicker than the second barrier layer 244 formed by the atomic layer chemical vapor deposition (ALCVD) method, and is 35 to 65 nm, for example, 40 nm even if it is thin. Become. On the other hand, the thickness of the second barrier layer 244 formed by atomic layer chemical vapor deposition (ALCVD) can be reduced. For example, aluminum oxide Al ₂ O ₃ is formed to a thickness of 5 to 30 nm, for example, 20 nm. Is done. The atomic layer chemical vapor deposition (ALCVD) method has excellent embedding characteristics as compared with the sputtering method and the like, and therefore can cope with miniaturization. In the first and second barrier layers 242 and 244, The reducing gas barrier property can be enhanced. In addition, the first barrier layer 242 formed by sputtering is less dense than the second barrier layer 244, but it is effective and lowers the heat transfer rate. Dissipation can be prevented.

  An interlayer insulating film 250 is formed on the reducing gas barrier layer 240. In general, when the source gas (TEOS) of the interlayer insulating film 250 chemically reacts, a reducing gas such as hydrogen gas or water vapor is generated. The reducing gas barrier layer 240 provided around the capacitor 230 protects the capacitor 230 from the reducing gas generated during the formation of the interlayer insulating film 250.

  On the interlayer insulating film 250, the first electrode (lower electrode) wiring layer 222 and the second electrode (upper electrode) wiring layer 224 shown in FIG. A first contact hole 252 and a second contact hole 254 are formed in the interlayer insulating film 250 in advance before the electrode wiring is formed. At that time, contact holes are similarly formed in the reducing gas barrier layer 240. The first electrode (lower electrode) 234 and the first electrode wiring layer 222 are electrically connected by the first plug 226 embedded in the first contact hole 252. Similarly, the second electrode (upper electrode) 236 and the second electrode wiring layer 224 are electrically connected by the second plug 228 embedded in the second contact hole 254.

  Here, when the interlayer insulating film 250 is not present, when the first electrode (lower electrode) wiring layer 222 and the second electrode (upper electrode) wiring layer 224 are subjected to pattern etching, the second reducing gas barrier layer 240 below the second insulating gas barrier layer 240 is subjected to pattern etching. The barrier layer 244 is etched and the barrier property is lowered. The interlayer insulating film 250 is necessary for ensuring the barrier property of the reducing gas barrier layer 240.

  Here, the interlayer insulating film 250 preferably has a low hydrogen content. Therefore, the interlayer insulating film 250 is degassed by annealing. Thus, the hydrogen content of the interlayer insulating film 250 is made lower than that of the passivation film 260 that covers the first and second electrode wiring layers 222 and 224.

  Since the reducing gas barrier layer 240 on the top surface of the capacitor 230 is closed without a contact hole when the interlayer insulating film 250 is formed, the reducing gas during the formation of the interlayer insulating film 250 does not enter the capacitor 230. However, after contact holes are formed in the reducing gas barrier layer 240, the barrier properties deteriorate. As an example for preventing this, for example, as shown in FIG. 4, the first and second plugs 226 and 228 are made of a plurality of layers 228A and 228B (only the second plug 228 is shown in FIG. 4), and the barrier metal is formed on the first layer 228A. Adopt layer. The reducing gas barrier property is secured by the barrier metal of the first layer 228A. The barrier metal of the first layer 228A is not preferably highly diffusible like titanium Ti, and titanium, aluminum, nitride TiAlN having low diffusibility and high reducing gas barrier property can be employed. As a method for cutting off the intrusion of the reducing gas from the contact hole, the reducing gas barrier layer 290 may be additionally provided so as to surround at least the second plug 228 as shown in FIG. The reducing gas barrier layer 290 may be used in combination with the barrier metal 228A of the second plug 228 or may exclude the barrier metal 228A. Note that the reducing gas barrier layer 290 may cover the first plug 226.

A SiO ₂ or SiN passivation film 260 is provided to cover the first and second electrode wiring layers 222 and 224. An infrared absorber (light absorbing member in a broad sense) 270 is provided on the passivation film 260 at least above the capacitor 230. Although the passivation film 260 is also formed of SiO ₂ or SiN, it is preferable to use a dissimilar material having a high etching selectivity with respect to the lower passivation film 260 in view of the necessity of pattern etching of the infrared absorber 270. Infrared rays are incident on the infrared absorber 270 from the direction of the arrow in FIG. 2, and the infrared absorber 270 generates heat according to the amount of absorbed infrared rays. When the heat is transferred to the pyroelectric body 232, the amount of spontaneous polarization of the capacitor 230 changes due to the heat, and infrared rays can be detected by detecting the charge due to the spontaneous polarization. The infrared absorber 270 is not limited to the one provided separately from the capacitor 230, and is not necessary when the infrared absorber 270 exists in the capacitor 230.

  Even if reducing gas is generated during the CVD formation of the passivation film 260 and the infrared absorber 270, the capacitor 230 is protected by the barrier metal in the reducing gas barrier layer 240 and the first and second plugs 226 and 228.

A reducing gas barrier layer 280 is provided so as to cover the outer surface of the infrared detector 200 including the infrared absorber 270. The reducing gas barrier layer 280 needs to be formed thinner than, for example, the reducing gas barrier layer 240 in order to increase the transmittance of infrared rays (wavelength band: 8 to 14 μm) incident on the infrared absorber 270. For this purpose, an atomic layer chemical vapor deposition (ALCVD) method capable of adjusting the film thickness at the atomic size level is employed. This is because the ordinary CVD method is too thick and the infrared transmittance deteriorates. In the present embodiment, for example, aluminum oxide Al ₂ O ₃ is formed to a thickness of 10 to 50 nm, for example, 20 nm. As described above, the atomic layer chemical vapor deposition (ALCVD) method has excellent embedding characteristics as compared with the sputtering method and the like, so that it is possible to form a dense film at the atomic level corresponding to miniaturization. Therefore, even if it is thin, the reducing gas barrier property can be enhanced.

On the base 100 side, the walls defining the cavity 102, that is, the bottom wall 110A and the sidewall 104A defining the cavity are embedded in the cavity 102 in the process of manufacturing the pyroelectric infrared detector 200. An etching stop film 130 is formed when the sacrificial layer 150 (see FIG. 4) is isotropically etched. Similarly, an etching stop film 140 is also formed on the lower surface of the support member 210 (the upper surface of the sacrificial layer 150). In this embodiment, the reducing gas barrier film 280 is formed of the same material as the etching stop films 130 and 140. That is, the etching stop films 130 and 140 also have a reducing gas barrier property. The etching stop films 130 and 140 are also formed by depositing aluminum oxide Al ₂ O ₃ to a film thickness of 20 to 50 nm by atomic layer chemical vapor deposition (ALCVD).

  Since the etching stop film 130 has a reducing gas barrier property, when the sacrificial layer 150 is isotropically etched with hydrofluoric acid in a reducing atmosphere, it is possible to prevent the reducing gas from penetrating the support member 210 and entering the capacitor 230. In addition, since the etching stop film 140 covering the base portion 100 has a reducing gas barrier property, it is possible to prevent the transistors and wirings of the circuit disposed in the base portion 100 from being reduced and deteriorated.

3. Warping and cutting of support member 3.1. Structure of Support Member FIGS. 1A to 1D are schematic explanations for explaining the mechanism of warping and cutting of the support member on which the thermal detection element according to the embodiment of the present invention is mounted and the elimination mechanism thereof. FIG. As shown in FIG. 1A, in a state where the sacrificial layer 150 is embedded in the cavity 102 of the base 100, the support member 210 is formed so that both ends thereof are joined to the base 100, and then the pyroelectric infrared is used. A detection element (detection element in a broad sense) 220 is formed. In this state, the sacrificial layer 150 is buried in the cavity 102 and then the upper surface is flattened by CMP (chemical mechanical polishing) or the like. Therefore, as long as the upper surface of the sacrificial layer 150 is flattened, the support member 210 also has flatness.

  Thereafter, the sacrificial layer 150 is removed by isotropic etching. At the moment when the sacrificial layer 150 is removed, as long as there is a residual stress in the support member 210, the support member 210 is bent by compression and warped, or the support member 210 is in tension by being pulled.

In FIG. 1B, when compressive residual stress CS (residual stress in the first direction) is present on the support member 210, a bending moment M1 acts on both ends of the support member 210. In the example of FIG. 1B, the support member 210 contracts according to the compressive residual stress CS, and at that time, both ends of the support member 210 function as hinges that fall inward by the bending moment M1, and the support member 210 has a downward convexity. The warp which becomes becomes. If the support member 110 along the downward convex comes into contact with the bottom surface of the base 100, the heat of the infrared detector 200 escapes along the solid heat conduction path between the support member 210 and the base 100. In this case, the outside line detection accuracy of the pyroelectric infrared detector deteriorates.
Alternatively, if an external force moment in the direction opposite to the bending moment M1 in FIG. 1B is applied, the contracted support member 210 warps to be upwardly convex. When the support member 210 is bent so as to be convex downward or upward, the pyroelectric body 232 in the infrared detector 200 functions as a piezoelectric element and generates an electrical signal in response to the bending, so detection based on the pyroelectric effect Accuracy deteriorates.

  On the other hand, in FIG. 1C, when the tensile residual stress TS (residual stress in the second direction opposite to the first direction) exists in the support member 210, the support member 210 extends according to the tensile residual stress TS. Since both ends of the support member 210 are fixed to the base 100, the support member 210 is in a tension state. In the worst case, as shown in FIG. 1C, the support member 210 is cut by the tensile residual stress TS.

  As shown in FIGS. 1B and 1C, the support member 210 is warped or tensed upwardly or downwardly due to residual stress generated in the forming material. In the manufacturing process in which the sacrificial layer 150 is present as shown in FIG. 1A, the base 100 is regarded as a rigid body by the sacrificial layer 150, and thus a reaction force that balances the residual stress acts and no warping or cutting occurs. As soon as the layer 150 is removed, the support member 210 is warped or tensioned.

  In this embodiment, when removing the sacrificial layer 150 in the cavity 102, the shrinkage and elongation of the support member 210 are taken into consideration in advance, and the shrinkage and elongation are canceled out. That is, for example, an external force EF that is the same as the compressive residual stress CS is applied to the base 100 to form the support member 210 (see FIG. 1D), and then the cavity 102 is formed with the external force EF released. . As a result, the shrinkage of the support member 210 due to the compressive residual stress CS is suppressed, and the warp of the support member 210 is canceled. The external force direction of the external force EF is preferably substantially equal to the stress direction of the compressive residual stress CS, and the external force value of the external force EF is multiplied by the cross-sectional area on which the compressive residual stress CS acts and the stress value of the compressive residual stress CS. Preferably, it is substantially equal to the measured value.

  In addition, when the tensile residual stress TS exists as shown in FIG. 1 (C), the external force EF shown in FIG. 1 (D) is changed in the opposite direction. That is, assuming the tensile residual stress TS, an external force (in the opposite direction to the external force EF shown in FIG. 1D) that is the same as the tensile residual stress TS is applied to the base 100 to suppress the cutting of the support member 210. be able to.

  A specific method for applying the external force EF to the base 100 as shown in FIG. 1D will be described later. When the tensile residual stress TS exists, the cutting of the support member 210 may be suppressed by a solution different from the method of applying the external force EF to the base 100, and the specific solution will be described later.

  In the embodiment shown in FIG. 4, the support member 210 on which the capacitor 230 is mounted is formed of a single layer. However, the support member 210 may be formed of a plurality of layers. In this case, the residual stress of the support member 210 as a whole may be considered.

The support member 210 can be composed of an oxide film (for example, SiO ₂ ), a nitride film (for example, Si ₃ N ₄ ), or the like. Note that since the nitride film (for example, Si ₃ N ₄ ) has a reducing gas barrier property, the supporting member 210 also has a function of blocking a reducing inhibition factor that enters the pyroelectric body 232 of the capacitor 230 from the supporting member 210 side. This will be described later.

3.2. Method for Applying External Force EF FIGS. 5A to 5D are schematic explanatory views of a method for manufacturing a pyroelectric detector (detecting element in a broad sense) according to an embodiment of the present invention. In addition, the manufacturing method of this embodiment is not limited to the schematic process of FIG. 5 (A)-FIG. 5 (D), A detailed process may be added and this schematic process may be changed. When a plurality of detection elements are manufactured, the detection element manufacturing method can also be referred to as a detection device manufacturing method.

  As shown in FIG. 5A, a member 300 of the same type as the support member 210 is formed on the back surface of the base portion 100. In the example of FIG. 1A, the member 300 in FIG. 5A is not formed on the back surface of the base portion 100. In the example of FIG. 5A, the back surface (surface of the fixed portion) of the base portion 100 faces a sacrificial layer (FIG. 5B) formed thereafter. Moreover, the thickness of the same kind of member 300 is the same as the thickness of the support member 210 (FIG. 5C) formed thereafter. When the compressive residual stress CS exists in the support member 210, the compressive residual stress CS also exists in the same type of member 300. In addition, as the external force value (magnitude) of the external force EF in FIG. 1D, the cross-sectional area on which the compressive residual stress CS of the same kind of member 300 acts is multiplied by the stress value of the compressive residual stress CS of the same kind of member 300. A value can be used.

  After forming a member 300 of the same type as the support member 210 on the back surface of the base 100, a cavity (trench) 102 is formed in the base 100. Since the base 100 without the cavity 102 is regarded as a rigid body, a reaction force that balances the residual stress acts and the base 100 does not shrink, but at the moment the cavity 102 is formed, the base 100 shrinks in the direction of the residual stress. . Subsequently, a sacrificial layer 150 is formed in the cavity 102. The upper surface of the sacrificial layer 150 is planarized by CMP or the like (see FIG. 5B). Thereafter, the support member 210 is formed on the sacrificial layer 150, and then the thermal detection element (detection element in a broad sense) 220 is formed on the support member 210. In the example of FIG. 5C, a compressive residual stress CS exists in the support member 210.

  As shown in FIG. 5D, when the same type of member 300 is removed from the back surface of the base 100, the external force EF in FIG. 1D is released. When the sacrificial layer 150 is removed in a state where the external force EF is released, warping of the support member 210 can be suppressed (see FIG. 1D). Specifically, since the base 100 having the sacrificial layer 150 is regarded as a rigid body, the base 100 does not extend even if the same type of member 300 is removed. In other words, by removing the member 300 of the same type, the contracted base 100 tries to expand, but the base 100 having the sacrificial layer 150 does not extend. However, at the instant when the cavity 102 is formed by the removal of the sacrificial layer 150, the base 100 tends to stretch. At this time, the compressive residual stress CS of the support member 210 is relaxed, and the warp of the support member 210 is suppressed.

  As shown in FIG. 5D, it is important to release the compressive residual stress CS of the same type of member 300 corresponding to the external force EF before removing the sacrificial layer 150. Even if the sacrificial layer 150 is removed in a state where the same type of member 300 is formed on the back surface of the base portion 100, the support member 210 is warped (see FIG. 1B).

  Although it is preferable to completely remove the same kind of member 300, only a part of the same kind of member 300 may be removed. When the compressive residual stress CS in the same kind of member 300 is equal to the compressive residual stress CS in the support member 210, the compressive residual stress CS of the same kind of member 300 is completely released. Warpage of the support member 210 can be reduced.

  In the example of FIG. 5A, the compressive residual stress CS is present in the same type of member 300. However, if there is a tensile residual stress in the support member 210, the tensile residual stress is also generated in the same type of member 300. Is important (see FIGS. 6A to 6D). In other words, it is important to form the same kind of member (same kind of film in a narrow sense) 300 with a material that generates a residual stress in the same direction as the residual stress in the support member 210.

  6 (A) to 6 (D) show a modification of the manufacturing method of FIGS. 5 (A) to 5 (D). When tensile residual stress TS exists in the supporting member 210, the same kind of member 300 is used. Also generates a tensile residual stress TS. As shown in FIGS. 5A to 5D and FIGS. 6A to 6D, the detector manufacturing method includes the following steps. First, the cavity 102 is formed in the fixed part 100. Second, before forming the sacrificial layer 150 in the cavity portion 102, an external force EF is applied to the fixed portion 100 in the same direction as the stress direction generated in the formed support member 210. Third, the support member 210 is formed on the sacrificial layer 150 formed in the cavity 102. Fourth, the external force EF is released before the sacrificial layer 150 is removed. In FIGS. 5A to 5D and FIGS. 6A to 6D, an external force is formed by forming a member 300 of the same type as the support member 210 on the surface of the fixed portion 100 facing the sacrificial layer 150. EF can be applied to the fixed portion 100. 5A and 6A, the same type of member 300 is added to the fixing unit 100 before the cavity 102 is formed in the fixing unit 100, but the sacrificial layer 150 is added to the fixing unit 100. If it is before, the same kind of member 300 may be added to the fixed part 100 having the cavity 102.

3.3. Other Solutions FIG. 7A to FIG. 7C are schematic explanatory diagrams of another manufacturing method of the pyroelectric detector (detecting element in a broad sense) according to the embodiment of the present invention. In addition, the manufacturing method of this embodiment is not limited to the schematic process of FIG. 7 (A)-FIG.7 (C), A detailed process may be added and this schematic process may be changed.

  A cavity (trench) 102 is formed in the base 100, and then a sacrificial layer 150 is formed in the cavity 102. The sacrificial layer 150 has a recess 152 on the top surface by CMP, etching, or the like (see FIG. 7A). In the example of FIG. 1A, the upper surface of the sacrificial layer 150 is planarized. Note that as shown in FIG. 1A, the upper surface of the sacrificial layer 150 may be once planarized, and then the recess 152 may be formed. Further, in the example of FIG. 7A, the member 300 in FIG. 5A is not formed on the back surface of the base 100, but the member 300 may be formed as will be described later.

  A support member 210 is formed on the sacrificial layer 150 with the recess 152 formed in the sacrificial layer 150, and then a thermal detection element (detection element in a broad sense) 220 is formed on the support member 210. In the example of FIG. 7B, the tensile residual stress TS exists in the support member 210.

  In the example of FIG. 7A, assuming that the depth of the concave portion 152 is h and the radius of curvature is R, the concave portion 152 contacts a virtual circle defined by the radius of curvature R. σ is the residual stress (tensile residual stress TS), E is the Young's modulus of the base 100, T is the thickness of the base 100, ν is the Poisson's ratio of the support member 210, tf is the thickness of the support member 210, and d is Assuming the thickness of the cavity 102, h and R preferably satisfy the following relational expression.

  For example, σ is 400 [MPa], E is 130.8 [GPa] in the (100) plane orientation, T is 72.6 [μm], and ν is 0.361 in the (100) plane orientation. When it is [MPa] and tf is 800 [Å], R is about 5.62 [μm]. For example, when d is 1.5 [μm] and R is about 5.62 [μm], h is about 500 [Å].

  7A to 7C, each dimension does not accurately represent an actual dimension. That is, in FIGS. 7A to 7C, each dimension is enlarged or reduced for easy understanding of the following description. Similarly, drawings other than FIGS. 7A to 7C do not necessarily represent accurate dimensions. Moreover, each shape is not limited to FIG. 7 (A)-FIG.7 (C), It can change.

  Since the redundant portion is added to the support member 210 (FIG. 7B), the support member 210 is not cut even if the sacrificial layer 150 is removed from the cavity 102 as shown in FIG. 7C. Thus, the warp that the support member 210 exhibits a downward convexity is suppressed by the extension of the support member 210 due to the tensile residual stress TS.

  As shown in FIG. 7A, it is important to form a support member 210 along the recess 152 in a state where the recess 152 is formed in the sacrificial layer 150 and add a redundant portion to the support member 210. The sacrificial layer 150 preferably has one recess 152, but a plurality of recesses may be formed.

  8A to 8C show a modification of the manufacturing method of FIGS. 7A to 7C, and the sacrificial layer 150 has three recesses 152 (FIG. 8A). . Since the supporting member 210 is formed along the three concave portions 152 (FIG. 8B), even if the sacrificial layer 150 is removed, the cutting of the supporting member 210 is suppressed as shown in FIG. 8C. Is done. In other words, when the support member 210 is formed on the sacrificial layer 150 having the three recesses 152, a redundant portion of the support member 210 as shown in FIG. 1B can be generated.

  9A to 9C show another modification of the manufacturing method shown in FIGS. 7A to 7C. When the tensile residual stress TS is present on the support member 210, the sacrificial layer 150 is used. The upper surface may have a convex portion 154 by CMP, etching, or the like (see FIG. 9A). As shown in FIG. 7A to FIG. 7C, FIG. 8A to FIG. 8C, and FIG. 9A to FIG. 9C, the detector manufacturing method includes the following steps: You may have. First, at least one concave portion 152 or convex portion 154 for forming a support member redundant length is formed in the sacrificial layer 150. Second, the support member 210 is formed along at least one recess 152 or protrusion 154. Third, the sacrificial layer 150 is then removed.

  In addition, for example, the steps of FIGS. 7A to 7C and the steps of FIGS. 6A to 6C can be combined. By modifying the example of FIG. 7A, for example, the depth h and the radius of curvature R of the recess 152 are calculated using a value half the thickness tf of the support member 210. Further, the same type of member 300 as shown in FIG. 6A is formed on the back surface (surface of the fixed portion) of the base portion 100. However, the thickness of the same kind of member 300 is half of the thickness tf of the support member 210 (FIG. 7B) formed thereafter.

4). Capacitor structure 4.1. Thermal Conductance FIG. 10 is a schematic cross-sectional view for explaining a main part of the present embodiment. As described above, the capacitor 230 includes the pyroelectric body 232 between the first electrode (lower electrode) 234 and the second electrode (upper electrode) 236. The capacitor 230 is mounted and supported on a second surface (upper surface in FIG. 10) where the support member 210 faces the first surface (lower surface in FIG. 10) facing the cavity 102. Then, infrared rays can be detected by utilizing the fact that the amount of spontaneous polarization of the pyroelectric body 232 changes depending on the amount of incident infrared light (temperature) (pyroelectric effect or pyroelectronic effect). In the present embodiment, incident infrared rays are absorbed by the infrared absorber 270 and the infrared absorber 270 generates heat, and the infrared rays are transmitted via the solid heat conduction path between the infrared absorber 270 and the pyroelectric body 232. Heat generated by the absorber 270 is transmitted.

  In the capacitor 230 of the present embodiment, the thermal conductance G1 of the first electrode (lower electrode) 234 in contact with the support member 210 is made smaller than the thermal conductance G2 of the second electrode (upper electrode) 236. In this way, the capacitor 230 can easily transfer the heat caused by the infrared rays to the pyroelectric body 232 via the second electrode (upper electrode) 236, and the heat of the pyroelectric body 232 can be transmitted to the first electrode (lower electrode) 234. This makes it difficult to escape to the support member 210 via the, thereby improving the signal sensitivity of the infrared detection element 220.

  The structure of the capacitor 230 having the above-described characteristics will be described in more detail with reference to FIG. First, the thickness T1 of the first electrode (lower electrode) 234 is thicker than the second electrode (upper electrode) 236 (T1> T2). When the thermal conductivity of the first electrode (lower electrode) 234 is λ1, the thermal conductance G1 of the first electrode (lower electrode) 234 is G1 = λ1 / T1. When the thermal conductivity of the second electrode (upper electrode) 236 is λ2, the thermal conductance G2 of the second electrode (upper electrode) 236 is G2 = λ2 / T2.

  In order to make the relationship of thermal conductance G1 <G2, for example, if the materials of the first and second electrodes 234 and 236 are both the same single material such as platinum Pt or iridium Ir, λ1 = λ2. Since T1> T2 from FIG. 10, the relationship G1 <G2 can be satisfied.

  First, consider that each of the first and second electrodes 234 and 236 is formed of the same material. In order for the capacitor 230 to align the crystal direction of the pyroelectric body 232, it is necessary to match the crystal lattice level of the interface with the first electrode 234 in the lower layer where the pyroelectric body 232 is formed. That is, the first electrode 234 functions as a crystal seed layer, but platinum Pt is preferable as the first electrode 234 because of its strong self-orientation. Iridium Ir is also suitable as a seed layer material.

  Further, the crystal orientation of the second electrode (upper electrode) 236 may be continuously connected from the first electrode 234 and the pyroelectric body 232 to the second electrode 236 without destroying the crystallinity of the pyroelectric body 232. preferable. Therefore, the second electrode 236 is preferably formed using the same material as the first electrode 234.

  Thus, when the second electrode 236 is formed of the same material as the first electrode 234, for example, a metal such as Pt or Ir, the upper surface of the second electrode 236 can be used as a reflective surface. In this case, as shown in FIG. 10, the distance L from the top surface of the infrared absorber 270 to the top surface of the second electrode 236 is preferably λ / 4 (λ is the infrared detection wavelength). In this way, since infrared light having a detection wavelength λ is multiply reflected between the top surface of the infrared absorber 270 and the top surface of the second electrode 236, the infrared light having the detection wavelength λ is efficiently absorbed by the infrared absorber 270. it can.

4.2. Next, the structure of the capacitor 230 of this embodiment shown in FIG. 10 will be described. In the capacitor 230 shown in FIG. 10, the crystal orientations of the pyroelectric body 232, the first electrode 234, and the second electrode 236 are aligned such that the preferred orientation direction is, for example, the (111) plane orientation. By being preferentially oriented in the (111) plane orientation, the orientation ratio of the (111) orientation in the other plane orientation is controlled to 90% or more, for example. In order to increase the pyroelectric coefficient, the (100) orientation rather than the (111) orientation is preferable, but the (111) orientation is used in order to easily control the polarization with respect to the applied electric field direction. However, the preferred orientation direction is not limited to this.

  The first electrode 234 includes, in order from the support member 210, an orientation control layer (eg, Ir) 234A that controls the orientation of the first electrode 234 so as to preferentially orient the (111) plane, for example, and a first reducing gas barrier layer (eg, IrOx). 234B and a preferentially oriented seed layer (eg, Pt) 234C.

  The second electrode 236 includes, in order from the pyroelectric body 232 side, an alignment layer (for example, Pt) 236A in which crystal orientation is aligned with the pyroelectric body 232, a second reducing gas barrier layer (for example, IrOx) 236B, and the second electrode 236. And a low resistance layer (for example, Ir) 236C for reducing the resistance of the joint surface with the second plug 228 connected to the second plug 228.

  In this embodiment, the reason why the first and second electrodes 234 and 236 of the capacitor 230 have a multi-layer structure is that the infrared detection element 220 with a small heat capacity is processed with low damage without reducing its capability. The crystal lattice level is matched, and the pyroelectric material (oxide) 232 is isolated from the reducing gas even when the periphery of the capacitor 230 is in a reducing atmosphere during manufacturing or use.

  For example, the pyroelectric material 232 is crystal grown by preferentially orienting, for example, PZT (Pb (Zr, Ti) O3: generic name: lead zirconate titanate) or PZTN (general name of PZT added with Nb) in the (111) orientation. I am letting. Use of PZTN is preferable in that it is difficult to be reduced even when a thin film is formed, and oxidation deficiency can be suppressed. In order to crystallize the pyroelectric material 232, it is crystallized from the formation stage of the first electrode 234 under the pyroelectric material 232.

  For this purpose, an Ir layer 234A that functions as an orientation control layer is formed on the lower electrode 234 by sputtering. As shown in FIG. 10, for example, a titanium / aluminum / nitride (TiAlN) layer or a titanium nitride (TiN) layer may be formed as the adhesion layer 234D under the orientation control layer 234A. This is because, depending on the material of the support member 210, it is difficult to ensure adhesion. Further, when the support member 210 located under the adhesion layer 234D is formed of SiO2, the support member 210 is preferably formed of a material having a grain smaller than that of polysilicon or an amorphous material. This is because the flatness of the surface on which the support member 210 mounts the capacitor 230 can be ensured. If the surface on which the orientation control layer 234A is formed is a rough surface, the rough surface unevenness is reflected during crystal growth, which is not preferable.

The IrOx layer 234B functioning as a reducing gas barrier layer in the first electrode 234 has a support member 210 (for example, a reducing gas barrier property) to isolate the pyroelectric material 232 from reducing inhibitors from below the capacitor 230. Si ₃ N ₄ ) and an etching stop film (for example, Al ₂ O ₃ ) 140 of the support member 210. For example, degassing from the base 100 during firing of the pyroelectric material (ceramic) 232 or other annealing process, or a reducing gas used in the isotropic etching process of the sacrificial layer 150 becomes a reducing inhibitor.

  In addition, the IrOx layer 234B has little crystallinity per se, but since it has a good compatibility with the Ir layer 234A in a metal-metal oxide relationship, it can have the same preferred orientation direction as the Ir layer 234A. .

  The Pt layer 234C functioning as a seed layer in the first electrode 234 serves as a seed layer of the preferential orientation of the pyroelectric body 232 and is (111) oriented. In the present embodiment, the Pt layer 234C has a two-layer structure. The first Pt layer forms the basis of the (111) orientation, and the second Pt layer forms microroughness on the surface so that the pyroelectric body 232 functions as a preferentially oriented seed layer. The pyroelectric material 232 is (111) oriented following the seed layer 234C.

  When the second electrode 236 is formed by sputtering, the interface is physically rough, trap sites are generated, and the characteristics may be deteriorated. Therefore, the first electrode 234, the pyroelectric body 232, and the second electrode 236 may be used. The crystal-level lattice matching is reconstructed so that the crystal orientations are continuously connected.

  The Pt layer 236A in the second electrode 236 is formed by sputtering, but the crystal direction of the interface becomes discontinuous immediately after sputtering. Therefore, after that, annealing treatment is performed to recrystallize the Pt layer 236A. That is, the Pt layer 236A functions as an orientation matching layer in which the crystal orientation of the pyroelectric material 232 is matched.

  The IrOx layer 236B in the second electrode 236 functions as a barrier for reducing deterioration factors from above the capacitor 230. Further, the Ir layer 236C in the second electrode 236 is used to reduce the resistance value between the IrOx layer 236B and the second plug 228 because the resistance value of the IrOx layer 236B is large. The Ir layer 236C is compatible with the IrOx layer 236B in the metal oxide-metal relationship, and can have the same preferred orientation direction as the IrOx layer 236B.

  As described above, in the present embodiment, the first and second electrodes 234 and 236 are arranged in multiple layers of Pt, IrOx, and Ir in order from the pyroelectric body 232 side, and the forming material is centered on the pyroelectric body 232. Symmetrical arrangement.

  However, the thickness of each layer of the multilayer structure forming the first and second electrodes 234 and 236 is asymmetric about the pyroelectric body 232. First, the total thickness T1 of the first electrode 234 and the total thickness T2 of the second electrode 236 satisfy the relationship (T1> T2) as described above. Here, the thermal conductivities of the Ir layer 234A, IrOx layer 234B, and Pt layer 234C of the first electrode 234 are λ1, λ2, and λ3, and the thicknesses are T11, T12, and T13. The thermal conductivities of the second electrode Ir layer 236C, IrOx layer 236B, and Pt layer 236A are λ1, λ2, and λ3 as in the first electrode 232, and the thicknesses thereof are T21, T22, and T23.

  Further, assuming that the thermal conductances of the Ir layer 234A, IrOx layer 234B, and Pt layer 234C of the first electrode 234 are G11, G12, and G13, G11 = λ1 / T11, G12 = λ2 / T12, and G13 = λ3 / C13. . When the thermal conductances of the Ir layer 236C, IrOx layer 236B, and Pt layer 236A of the second electrode 236 are G21, G22, and G23, G21 = λ1 / T21, G22 = λ2 / T22, and G13 = λ3 / T23.

Since the total thermal conductance G1 of the first electrode 234 is expressed by 1 / G1 = (1 / G11) + (1 / G12) + (1 / G13),
G1 = (G11 × G12 × G13) / (G11 + G12 + G13) (1)
Similarly, the total thermal conductance G2 of the second electrode 236 is expressed by 1 / G2 = (1 / G21) + (1 / G22) + (1 / G23).
G2 = (G21 × G22 × G23) / (G21 + G22 + G23) (2)
It becomes.

  Next, the thickness of each layer of the multilayer structure forming the first and second electrodes 234 and 236 is substantially as follows under the condition satisfying T11 + T12 + T13 = T1> T2 = T21 + T22 + T23.

Ir layer 234A, 236C T11: T21 = 1: 0.7
IrOx layers 234B, 236B T12: T22 = 0.3: 1
Pt layer 234C, 236A T13: T23 = 3: 1
The reason for this film thickness relationship is as follows. First, regarding the Ir layers 234A and 236C, the Ir layer 234A in the first electrode 234 functions as an orientation control layer. The purpose of the Ir layer of the electrode 236C is to lower the resistance, and the lower the resistance, the lower the resistance.

  Next, regarding the IrOx layers 234B and 236B, the barrier property of the reducing inhibitor from the lower side and the upper side of the capacitor 230 is that other barrier films (reducing gas barrier layer 240, etching stop film / reducing gas barrier layer 140, 280). ), The IrOx layer 234B of the first electrode 234 is thinned, but the IrOx layer 236B of the second electrode 236 is thickened in preparation for the low barrier property of the second plug 228.

  Finally, with regard to the Pt layers 234C and 236A, the Pt layer 234C in the first electrode 234 functions as a seed layer that determines the preferred orientation of the pyroelectric body 232, whereas a predetermined thickness is required. Since the purpose of the Pt layer 236A of the second electrode 236 functions as an alignment matching layer that matches the alignment of the pyroelectric body 232, it may be formed thinner than the Pt layer 234C in the first electrode 234.

  The thickness ratio of the Ir layer 234A, IrOx layer 234B, and Pt layer 234C of the first electrode 234 is, for example, T11: T12: T13 = 10: 3: 15, and the Ir layer 236C and IrOx layer 236B of the second electrode 236 are set. The thickness ratio of the Pt layer 236A is, for example, T21: T22: T23 = 7: 10: 5.

  Here, the heat transfer coefficient λ3 of Pt is 71.6 (W / m · K), and the heat transfer coefficient λ1 of Ir is approximately λ1 = 147 (W / m · K), which is approximately the heat transfer coefficient λ3 of Pt. 2 times. The thermal conductivity λ2 of IrOx varies depending on the heat degree and the oxygen / metal ratio (O / M), but does not exceed the thermal conductivity λ1 of Ir. Substituting the relationship between the film thickness and the heat transfer coefficient into the equations (1) and (2) to obtain the magnitude relationship between G1 and G2, it can be seen that G1 <G2. Thus, even if the first and second electrodes 234 and 236 are formed in a multilayer structure as in this embodiment, G1 <G2 is satisfied from the relationship between the heat transfer coefficient and the film thickness.

  Further, as described above, when the first electrode 234 has the adhesion layer 234D on the joint surface with the support member 210, the thermal conductance C1 of the first electrode 234 becomes smaller, and therefore the relationship of G1 <G2 is satisfied. It becomes easy.

  Since the etching mask of the capacitor 230 deteriorates as the etching progresses, the sidewall of the capacitor 230 becomes narrower toward the upper side and wider toward the lower side as shown in FIG. However, since the taper angle with respect to the horizontal plane is about 80 degrees, the area expansion of the first electrode 234 relative to the second electrode 236 is small considering that the total height of the capacitor 230 is on the order of nanometers. Therefore, the heat transfer amount at the first electrode 234 can be made smaller than the heat transfer amount at the second electrode 236 due to the thermal conductance relationship between the first and second electrodes 234 and 236.

4.3. Modification of Capacitor Structure As described above, the single-layer structure and the multilayer structure have been described for each of the first and second electrodes 234 and 236 of the capacitor 230. Various other combinations with G1 <G2 are possible.

  First, the Ir layer 236C of the second electrode 236 can be deleted. In this case, if, for example, Ir is used as the material of the second plug 228, the purpose of reducing the resistance can be achieved. In this case, the thermal conductance G2 of the second electrode 236 becomes larger than that in the case of FIG. 10, and thus it is easy to satisfy the relationship of G1 <G2. In this case, the reflection surface defining L = λ / 4 shown in FIG. 10 is replaced with the Pt layer 236A of the second electrode 236, but the multiple reflection surface can be secured similarly.

  Next, the thickness of the IrOx layer 236B in the second electrode 236 of FIG. 10 can be made equal to or less than the thickness of the IrOx layer 234B in the first electrode 234. As described above, the barrier property of the reducing inhibitor from the lower side and the upper side of the capacitor 230 is that other barrier films (reducing gas barrier layer 240, etching stop film / reducing gas barrier layer 140, 280) are used in combination. If the reducing gas barrier property of the second plug 228 is enhanced as shown in FIG. 11, for example, the thickness of the IrOx layer 236B in the second electrode 236 does not need to be thicker than the IrOx layer 234B in the first electrode 234. . As a result, the thermal conductance G2 of the second electrode 236 becomes larger, and the relationship of G1 <G2 is more easily established.

  Next, the IrOx layer 234B in the first electrode 234 of FIG. 10 can be deleted. Even if the IrOx layer 234B is deleted, the crystal continuity between the Ir layer 234A and the Pt layer 234C is not hindered, so there is no problem with respect to the crystal orientation. By deleting the IrOx layer 234B, the capacitor 230 does not have a barrier film against the reducing inhibitor from below. However, if the etching stop film 140 is present on the lower surface of the support member 210 and the etching stop film 140 is formed of a film having a reducing gas barrier property, the capacitor 230 ensures a barrier property against the reducing inhibitory factor from below. it can.

  Here, when the IrOx layer 234B in the first electrode 234 is deleted, the thermal conductance G1 of the first electrode 234 increases. Therefore, it may be necessary to increase the thermal conductance G2 of the second electrode 236 in order to establish the relationship of G1 <G2. In that case, for example, it is conceivable to delete the IrOx layer 236B in the second electrode 236. If the IrOx layer 236B can be deleted, the Ir layer 236C is also unnecessary. This is because the Pt layer 236A functions as a low resistance layer instead of the Ir layer 236C. The barrier property of the reducing inhibitor from above the capacitor 230 can be secured by the reducing gas barrier film 240 described above, the barrier metal 228A shown in FIG. 4, or the reducing gas barrier layer 290 shown in FIG.

  When the second electrode 234 of FIG. 10 is formed of only the Pt layer 236A as described above, the first electrode 234 is a single layer of the Pt layer 234C, two layers of the Ir layer 234A and the Pt layer 234C, or FIG. A three-layer structure including an Ir layer 234A, an IrOx layer 234B, and a Pt layer 234C can be used. In any of these cases, for example, if the thickness T11 of the Pt layer 234A of the first electrode 234 is larger than the thickness T21 of the Pt layer 236C of the second electrode 236 (T11> T21), the relationship of G1 <G2 is established. It can be easily established.

5). Electronic Device FIG. 12 shows a configuration example of an electronic device including the thermal detector or thermal detector of the present embodiment. The electronic apparatus includes an optical system 400, a sensor device (thermal detection device) 410, an image processing unit 420, a processing unit 430, a storage unit 440, an operation unit 450, and a display unit 460. Note that the electronic apparatus of the present embodiment is not limited to the configuration shown in FIG. 12, and various components such as omitting some of the components (for example, an optical system, an operation unit, a display unit, etc.) or adding other components. Can be implemented.

  The optical system 400 includes, for example, one or a plurality of lenses and a driving unit that drives these lenses. Then, an object image is formed on the sensor device 410. If necessary, focus adjustment is also performed.

  The sensor device 410 is configured by two-dimensionally arranging the thermal detectors 200 of the present embodiment described above, and is provided with a plurality of row lines (word lines, scanning lines) and a plurality of column lines (data lines). In addition to the two-dimensionally arranged photodetectors, the sensor device 410 includes a row selection circuit (row driver), a readout circuit that reads data from the photodetectors via column lines, an A / D converter, and the like. Can be included. By sequentially reading out data from each of the two-dimensionally arranged photodetectors, an object image can be captured.

  The image processing unit 420 performs various image processing such as image correction processing based on digital image data (pixel data) from the sensor device 410.

  The processing unit 430 controls the entire electronic device and controls each block in the electronic device. The processing unit 430 is realized by a CPU or the like, for example. The storage unit 440 stores various types of information, and functions as a work area for the processing unit 430 and the image processing unit 420, for example. The operation unit 450 serves as an interface for the user to operate the electronic device, and is realized by various buttons or a GUI (Graphical User Interface) screen, for example. The display unit 460 displays, for example, an image acquired by the sensor device 410, a GUI screen, and the like, and is realized by various displays such as a liquid crystal display and an organic EL display.

  As described above, the sensor device 410 is configured by two-dimensionally arranging the thermal detectors for one cell in two orthogonal directions, in addition to using the thermal detector for one cell as a sensor such as an infrared sensor. This can provide a heat (light) distribution image. The sensor device 410 can be used to configure an electronic device such as a thermography, an in-vehicle night vision, or a surveillance camera.

  Of course, it can be used in analysis equipment (measuring equipment) that analyzes (measures) physical information of objects by using a single-cell or multiple-cell thermal detector as a sensor, security equipment that detects fire and heat, factories, etc. Various electronic devices such as FA (Factory Automation) devices provided can also be configured.

  FIG. 13A shows a configuration example of the sensor device 410 of FIG. The sensor device includes a sensor array 500, a row selection circuit (row driver) 510, and a readout circuit 520. An A / D converter 530 and a control circuit 550 can be included. By using this sensor device, for example, an infrared camera used in a night vision device or the like can be realized.

  In the sensor array 500, for example, as shown in FIG. 2, a plurality of sensor cells are arranged (arranged) in the biaxial direction. A plurality of row lines (word lines, scanning lines) and a plurality of column lines (data lines) are provided. One of the row lines and the column lines may be one. For example, when there is one row line, a plurality of sensor cells are arranged in a direction (lateral direction) along the row line in FIG. On the other hand, when there is one column line, a plurality of sensor cells are arranged in a direction (vertical direction) along the column line.

  As shown in FIG. 13B, each sensor cell of the sensor array 500 is arranged (formed) at a location corresponding to the intersection position of each row line and each column line. For example, the sensor cell in FIG. 13B is arranged at a location corresponding to the intersection position of the row line WL1 and the column line DL1. The same applies to other sensor cells. The row selection circuit 510 is connected to one or more row lines. Then, each row line is selected. For example, when a sensor array 500 (focal plane array) of QVGA (320 × 240 pixels) as shown in FIG. 13B is taken as an example, row lines WL0, WL1, WL2,... WL239 are sequentially selected (scanned). Perform the action. That is, a signal (word selection signal) for selecting these row lines is output to the sensor array 500.

  The read circuit 520 is connected to one or more column lines. Then, a read operation for each column line is performed. Taking the QVGA sensor array 500 as an example, an operation of reading detection signals (detection current, detection charge) from the column lines DL0, DL1, DL2,.

  The A / D conversion unit 530 performs a process of A / D converting the detection voltage (measurement voltage, ultimate voltage) acquired in the reading circuit 520 into digital data. Then, the digital data DOUT after A / D conversion is output. Specifically, the A / D converter 530 is provided with each A / D converter corresponding to each of the plurality of column lines. Each A / D converter performs A / D conversion processing of the detection voltage acquired by the reading circuit 520 in the corresponding column line. Note that one A / D converter is provided corresponding to a plurality of column lines, and the detection voltage of the plurality of column lines can be A / D converted in a time division manner using this one A / D converter. Good.

  The control circuit 550 (timing generation circuit) generates various control signals and outputs them to the row selection circuit 510, the read circuit 520, and the A / D conversion unit 530. For example, a charge or discharge (reset) control signal is generated and output. Alternatively, a signal for controlling the timing of each circuit is generated and output.

  Although several embodiments have been described above, it is easily understood by those skilled in the art that many modifications can be made without substantially departing from the novel matters and effects of the present invention. Accordingly, all such modifications are intended to be included in the scope of the present invention. For example, a term described at least once together with a different term having a broader meaning or the same meaning in the specification or the drawings can be replaced with the different term in any part of the specification or the drawings.

  According to at least one embodiment of the present invention, for example, warpage of a support member that supports a thermal detection element can be reduced. As long as this effect can be achieved, the present invention can be widely applied to various thermal detectors (for example, thermocouple elements (thermopiles), pyroelectric elements, bolometers, etc.). The wavelength of the light to detect is not ask | required.

100 base (fixed part), 102 cavity,
130,140 reducing gas barrier layer (etching stop film),
150 sacrificial layer, 152 concave portion, 154 convex portion, 200 thermal detector,
210 supporting member, 220 infrared detecting element, 222, 224 first and second electrode wiring layers,
226, 228 first and second plugs, 228A barrier metal, 230 capacitor,
232 pyroelectric material, 234 first electrode, 234A orientation control layer,
234B first reducing gas barrier layer, 234C seed layer, 234D adhesion layer,
236 second electrode, 236A orientation matching layer, 236B second reducing gas barrier layer,
236C low resistance layer, 240 reducing gas barrier layer, 250 interlayer insulating film,
260 passivation film, 270 light absorbing member (infrared absorber),
280 reducing gas barrier layer (etching stop film), 290 reducing gas barrier layer,
300 members (same type as support members)

Claims (7)

Forming a cavity in the fixed part,
Before forming the sacrificial layer in the cavity, an external force is applied to the fixed portion in the same direction as the stress direction generated in the formed support member,
Forming the support member on the sacrificial layer formed in the cavity,
The method for manufacturing a detector, wherein the external force is released before the sacrificial layer is removed. In claim 1,
A method for manufacturing a detector, comprising: forming a member of the same type as the support member on a surface of the fixed portion facing the sacrificial layer, and applying the external force to the fixed portion. Forming at least one recess or protrusion for forming the support member redundant length in the sacrificial layer;
Forming a support member along the at least one recess or projection;
Thereafter, the sacrificial layer is removed. In claim 3,
A method of manufacturing a detector, wherein the at least one concave portion or convex portion is formed by chemical mechanical polishing the sacrificial layer. In claim 3,
Etching the sacrificial layer to form the at least one concave portion or convex portion. In any of claims 3 to 5,
When the depth of the at least one concave portion or convex portion is h and the curvature radius is R, the at least one concave portion or convex portion is in contact with a virtual circle defined by the curvature radius,
σ is the tensile residual stress generated in the support member, E is the Young's modulus of the fixed portion of the support member, T is the thickness of the fixed portion, ν is the Poisson's ratio of the support member, and tf is the thickness of the support member. A method for manufacturing a detector, wherein h and R satisfy the following relational expression, where thickness d is the thickness of the cavity of the fixed part.

In any of claims 3 to 5,
Applying substantially the same external force as the stress direction and the stress value of the tensile residual stress generated in the support member to the fixing portion of the support member to form the support member,
Thereafter, the sacrificial layer is removed from the fixed part in a state where the external force is released, and a cavity is formed in the fixed part.

Images