JP7450912B2 - Detection element and light emitting element for infrared light - Google Patents

Description

The present invention relates to an infrared light detection element and a light emitting element.

Infrared light detection elements and light emitting elements are widely used in infrared image sensors and monitoring sensors. For example, infrared rays in a long wavelength band of 2 μm or more are effective for human body detection sensors, movement monitoring sensors for heat sources, non-contact temperature sensors, gas sensors, etc. due to their thermal effects and the effect of infrared absorption by gas. There is a demand.
In view of environmental improvement, gas sensors are essential devices that can be used for monitoring and protecting the atmospheric environment, as well as for early detection of fire. In particular, in the wavelength range from 3 μm to 10 μm, there are many unique absorption bands for various gases, so there is a strong demand for gas sensors that can detect this region with a high S/N ratio and high sensitivity. There is similarly strong demand for light emitting elements for infrared light.

Patch antennas are known as structures that efficiently receive and transmit radio waves (see Patent Documents 1 and 2). Recently, metal-insulator-metal (MIM) has been developed further and is a structure that efficiently absorbs radio waves such as microwaves and infrared light. ) A patch antenna consisting of a cavity has been developed (see Non-Patent Document 1).
MIM cavities can strongly absorb light and electromagnetic waves even though they are thin structures by utilizing strong interactions with tightly confined electric fields. Furthermore, metal layers formed above and below the insulator, which is a functional material, can be used as electrodes. Based on this, a patch antenna element that exhibits high detection sensitivity for infrared rays was developed by connecting patch antennas with a one-dimensional simple straight wire (see Non-Patent Document 2).

However, there has been a desire for an infrared light detection element that has higher detection sensitivity for infrared light, can ensure a higher S/N ratio, and can be made smaller.
In addition, similar to the detection element for infrared light, there has been a desire for an infrared light emitting element that emits strong infrared light, has high directivity, and can be made compact. .

Japanese Unexamined Patent Publication No. 52-134350 Japanese Patent Application Publication No. 05-206729

Phys. Rev. Lett. , Vol. 96, p. 097401 (2006) Appl. Phys. Lett. , Vol. 104, p. 031113 (2014)

An object of the present invention is to provide a compact detection element and light emitting element for infrared light having a high S/N ratio and high photoelectric conversion efficiency. In other words, in the detection element, we provide a compact infrared light detection element with low noise level and high detection sensitivity, and in the light emitting element, we provide a compact infrared light detection element that has a low noise level and high detection sensitivity. An object of the present invention is to provide a compact infrared light emitting element that has strong emission intensity.

The configuration of the present invention is shown below.
(Configuration 1)
An element that detects or emits infrared light with a peak wavelength of λ,
The element has at least a photosensitive part that is sensitive to the infrared light and a wire part that makes an electrical connection,
The photosensitive section is composed of a plurality of photosensitive pieces having a planar shape of main size L arranged two-dimensionally with a period P,
The photosensitive piece has a layered structure in which a dielectric layer and a second conductor layer are sequentially laminated on a surface made of a first conductor layer,
The thickness T of the dielectric layer is greater than 0 and less than λ/(2n), where n is the refractive index of the dielectric layer with respect to λ,
The area of the photosensitive piece is {3λ/(8n)} ² or more and {5λ/(8n)} ² or less,
The wire portion is a wire piece having a maximum width of more than 0 and having conductivity of L/5 or less, which connects at least the second conductor layers of the adjacent photosensitive pieces with a non-linear part. Consisting of
The element has an in-plane propagation mode that propagates through the photosensitive pieces and the wire pieces in a first direction, which is the direction of the arrangement, and is polarized in a second direction, which is a direction perpendicular to the plane. death,
With respect to the infrared light incident from the second direction,
|k _a ×L+k _w ×(PL)−2×m×π|≦π/2 (m is a positive integer)
When the positive propagation constants of the photosensitive piece part and the wire piece part of the propagation mode traveling in the first direction are respectively k _a and k _w ,
The device, wherein the planar shape of the functional section consisting of the photosensitive section and the wire section has three-fold symmetry or more.
(Configuration 2)
The device according to configuration 1, wherein the dielectric layer includes a semiconductor layer.
(Configuration 3)
The semiconductor layer is made of Si, Ge, AlAs, GaAs, InAs, AlSb, GaSb, InSb, AlP, GaP, InP, AlN, GaN, InN, PbS, PbSe, ZnTe, CdTe, HgTe, ZnSe, CdSe, MgSe. The infrared light detecting element and light emitting element according to Structure 2, comprising one or more selected from the group.
(Configuration 4)
The device according to configuration 2, wherein the semiconductor layer includes a quantum well.
(Configuration 5)
The device according to configuration 4, wherein the number of quantum wells is 1 or more and 5 or less.
(Configuration 6)
5. The device according to configuration 4, wherein the number of quantum wells is one.
(Configuration 7)
The quantum wells include GaAs and AlGaAs, InGaAs and InAlAs, InGaAs and InP, InGaAs and AlAsSb, InGaAsP and InP, GaAs and GaInP, GaAs and AlInP, GaN and AlGaN, GaN and AlN, InAs and AlAsSb, and Z. nCdSe and ZnCdMgSe, InAs and GaSb, InAs and GaInSb, InAs and InAsSb, InGaAs and GaAsSb, and InAs and AlGaSb, the element according to any one of configurations 4 to 6.
(Configuration 8)
8. The element according to any one of configurations 1 to 7, wherein the first conductive layer and the dielectric layer, and the dielectric layer and the second conductive layer are both in ohmic contact.
(Configuration 9)
The element according to configuration 1, wherein the dielectric layer is made of an insulator.
(Configuration 10)
The dielectric layer is made of diamond, Si, Ge, MgO, Al ₂ O ₃ , SiO ₂ , BeO, CaO, TiO ₂ , Ti ₂ O ₃ , TiO, V ₂ O ₅ , Cr ₂ O ₃ , MnO, Cu ₂ O, CuO, ZnO, _Sc2O3 , _Y2O3 , _{ZrO2 , Ga2O3} , _{Nb2O5 ,} SnO2 _{, CeO2} , _{HfO2 , Ta2O5} , LiF, NaF _{, MgF2 ,} AlF ₃ , CaF ₂ , SrF ₂ , YF ₃ , CsF, BaF ₂ , LaF ₃ , CeF 3 , NdF ₃ , GdF ₃ , YbF ₃ , PbF _{2 ,} ThF ₄ , ZnS, ZnSe, ZnTe, CdSe, CdTe, AgGaS ₂ , AgGaSe ₂ , Zn ₃ P ₂ , ZnGeP ₂ , As ₂ S ₃ , As ₂ Se ₃ , AlAs, AlSb, GaP, GaAs, GaSb, GaSe, GaTe, InP, SiC, BN, AlN, GaN, NaCl, KCl, Consisting of one or more selected from the group consisting of AgCl, CsCl, KBr, RbBr, AgBr, CsBr, KI, RbI, AgI, CsI, TlI, TlBr, CaCo ₃ , SrTiO ₃ , LiTaO ₃ , ZrSiO ₄ , KNbO ₃ 1. The element described in 1.
(Configuration 11)
The first conductor layer and the second conductor layer are made of Ti, Cr, Ni, Au, Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er. , Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, Al, one or more metals selected from the group consisting of, Ti, Cr, Ni, Au , Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh , Re, Ir, In, an alloy containing one or more metals selected from the group consisting of Al, Ti, Cr, Ni, Au, Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, A compound containing one or more metals selected from the group consisting of Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, and Al. , and any one of ITO, AZO, GZO, IZO, IGZO, ATO, FTO, FZO, and TiN, according to any one of Structures 1 to 10.
(Configuration 12)
12. The element according to any one of configurations 1 to 11, wherein the planar shape of the photosensitive piece is selected from the group consisting of a regular triangle, a square, a regular hexagon, a regular octagon, and a perfect circle.
(Configuration 13)
The planar shape of the photosensitive piece is a square with a size of one piece L,
L is 3λ/(8n _eff ) or more and 5λ/(8n _eff ) or less, where n _eff is the effective refractive index of the lowest order in-plane propagation mode polarized in the second direction of the photosensitive piece. and
From configuration 1, the effective refractive index n _eff is n·(1+2δ/T) ^1/2 , where δ is the average skin depth of the first conductor layer and the second conductor layer. 11. The device according to any one of 11.
(Configuration 14)
14. The element according to any one of configurations 1 to 13, wherein the planar shape of the wire piece includes a right-angled portion.
(Configuration 15)
15. The element according to any one of configurations 1 to 14, wherein the symmetry is 4-fold symmetry.
(Configuration 16)
16. The element according to any one of configurations 1 to 15, wherein the arrangement is matrix-like.
(Configuration 17)
17. The device according to any one of configurations 1 to 16, wherein the second conductor layer of at least a plurality of the photosensitive pieces is formed above one of the first conductor layers.
(Configuration 18)
18. The device according to any one of configurations 1 to 17, wherein the second conductive layer of at least a plurality of the photosensitive pieces is formed on one of the dielectric layers.
(Configuration 19)
Outside the infrared light detection element and light emitting element according to any one of claims 1 to 18,
a second photosensitive piece in which the dielectric layer and the second conductor layer are sequentially laminated on a surface made of the first conductor layer;
A second wire having a maximum width of more than 0 and having a conductivity of L/5 or less, which connects at least the second conductor layers of the adjacent second photosensitive pieces with a non-linear part. The pieces are arranged in a two-dimensional manner,
The wire length of the second wire piece is configured to vary depending on the distance in one dimension, two dimensions, or a radius vector of polar coordinates with the center of gravity of the array of the photosensitive pieces as the origin. The device according to any one of 1 to 16.
(Configuration 20)
20. The element according to configuration 19, wherein the second wire pieces are arranged with the period P.
(Configuration 21)
21. The element of configuration 19 or 20, wherein the change in wire length is monotonic.

The present invention provides a compact detection element and light emitting element for infrared light having a high S/N ratio and high photoelectric conversion efficiency. In other words, for the detection element, a compact infrared light detection element with a low noise level and high detection sensitivity is provided, and for the light emitting element, a compact infrared light detection element is provided that has a strong emission intensity with little noise emission. A light emitting element for light is provided.

FIG. 2 is a bird's-eye view showing the general configuration of the element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is an explanatory diagram showing the band structure of a dielectric material having quantum wells used in the present invention. FIG. 2 is a plan view showing a schematic configuration of an element. FIG. 2 is an explanatory diagram of light absorption characteristics showing a comparison between the structure of an element and light absorption spectral characteristics. It is a SEM photograph of the element produced in Example. It is a SEM photograph of the element produced in Example. FIG. 2 is a characteristic diagram showing the photoresponsive characteristics of the device produced in Examples. FIG. 2 is a characteristic diagram showing the photoresponsive characteristics of the device produced in Examples. FIG. 2 is a characteristic diagram showing the photoresponsive characteristics of the device produced in Examples. FIG. 2 is a characteristic diagram showing the photoresponsive characteristics of the device produced in Examples. FIG. 2 is a bird's-eye view showing the structure of an element according to an example. FIG. 3 is a characteristic diagram showing the photoresponsive characteristics of the device of the example.

The present invention will be explained in detail below. Although the detailed description of the present invention described below may be based on representative aspects, embodiments, and examples, these are merely illustrative, and the present invention is not limited to such aspects, embodiments, and examples. and is not limited to the examples.
Note that "A to B" indicates A or more and B or less.

(Embodiment 1)
<Element structure>
As shown in FIG. 1, the element 101 of the present invention is a patch antenna having a thin dielectric layer 22 sandwiched between two conductive layers, a first conductive layer 21 and a second conductive layer 23. It consists of a photosensitive section 11 in which a large number of photosensitive pieces 12 functioning as a so-called array antenna are regularly arranged, and a wire section 13 having a conductive layer that electrically connects the photosensitive pieces.
The wire portion 13 is made up of a plurality of wire pieces 14, and each wire piece 14 is a wire that electrically connects at least the second conductor layers 23 of adjacent photosensitive pieces 12.

In the configuration of the present invention, since the photosensitive pieces are connected by the conductive layer, a bias voltage can be applied to the dielectric layer, and a current can be extracted by photoelectric conversion. Furthermore, light can be emitted by applying a voltage between two conductive layers to cause a tunnel current to flow through the dielectric layer 22, or by applying heat to the dielectric layer 22 by passing a current through the conductive material.

In general, the light detection sensitivity of a photodetector and the emission intensity of a light emitting element behave in common from the perspective of optical resonance due to the principle of ray retrograde motion. That is, the conditions that allow effective resonance to occur in the photodetector element also enable effective resonance to occur in the light emitting element. Hereinafter, the photodetecting element will be mainly explained, and the specific details of the light emitting element will be explained individually as appropriate.

The device 101 of the present invention has major characteristics in the shape, size, structure, and arrangement of the photosensitive piece 12 and the wire piece. In addition, the dielectric layer 22 also has features to further improve performance.

The photosensitive piece 12 has a cavity structure consisting of three layers: a first conductor layer 21, a dielectric layer 22, and a second conductor layer 23, and the thickness T of the dielectric layer 22 is greater than 0. λ/(2n) or less, and the area of the photosensitive piece 12 is {3λ/(8n)} ² or more and {5λ/(8n)} ² or less.
It has been found that by adopting this structure, half-wavelength resonance of the TM gap plasmon mode occurs efficiently while noise is suppressed, and infrared light is efficiently absorbed. In addition, in this structure, since the dielectric layer 22 is thin, the electric field generated by resonance is large, resulting in a large sensitivity enhancement, and the volume is small, resulting in a reduction in dark current. Since dark current can be suppressed, the device can be used even in a low-temperature environment such as a liquid nitrogen environment.

Further, when the photosensitive piece 12 is a square with a size L, n・(1+2δ/T) ¹ where δ is the average skin depth of the first conductor layer and the second conductor layer. When the effective refractive index of the lowest-order in-plane propagation mode polarized in _the second direction _of the photosensitive piece ¹² expressed as _eff ) or less.
By adopting this structure, it was optimized for a square photosensitive piece, and it was discovered that the half-wavelength resonance of the TM gap plasmon mode occurred efficiently and infrared light was efficiently absorbed while suppressing noise. .
Here, λ is the peak wavelength of infrared light that this element attempts to detect or emit, and n is the refractive index of the dielectric layer 22 with respect to the wavelength λ. Note that infrared light in the present invention refers to light with a wavelength of 0.76 μm or more and 1000 μm or less. Further, the wavelength range of infrared light that is particularly effective in the present invention is 2.5 μm or more and 25 μm or less.

In the present invention, the photosensitive pieces 12 having a planar shape of main size L are arranged two-dimensionally with a period P, and the adjacent photosensitive pieces 12 are arranged with wire pieces 14 two-dimensionally arranged with a period P. , by electrically connecting the group of photosensitive pieces 12 to a resonant state and absorbing infrared light with high efficiency. Here, the two-dimensional arrangement is typically a matrix arrangement, and includes a square lattice, a rectangular lattice, an orthorhombic lattice, and a triangular lattice. Among these, the square lattice shape is the easiest to design and can be preferably used.
The planar shape of the photosensitive piece 12 is preferably highly symmetrical, and specific examples include a regular triangle, a square, a regular hexagon, a regular octagon, and a perfect circle. Among these, square shapes are particularly preferred because they are easy to manufacture. It is preferable that the planar shape of the photosensitive piece 12 has high symmetry because the dependence on the direction and polarization of incident light is reduced, and the sensitivity is stabilized at a high level in a state where it is less susceptible to these disturbances.

Here, in order to increase the degree of integration of the photosensitive pieces 12 by narrowing the interval between adjacent photosensitive pieces 12 and to make the element 101 more compact, the wire piece 14 has a maximum width of more than 0 and less than or equal to L/5. It is assumed that the conductor layer has a folded shape having a Z-shaped or S-shaped bent part with a non-linear part. An example is shown in FIG. FIG. 2 is a plan view of a part of the main part of the element 101 seen from above, (a) is when only the photosensitive piece is arranged, and (b) is the photosensitive piece 12 and the straight wire piece 14a. (c) is the case when the photosensitive piece 12 and the Z-shaped wire piece 14b are arranged, and (d) is the case when the photosensitive piece 12 and the S-shaped wire piece 14c are arranged. This is the case. Here, in order to narrow the area and create an efficient folded plane shape, it is preferable to include a right-angled portion.
Here, the main size L is the length of one side if the photosensitive piece 12 is a square, the length of one side if it is an equilateral triangle, and the length of one side if the photosensitive piece 12 is a regular n-gon (n is an even number of 6 or more) such as a regular hexagon or a regular octagon. In the case of circles, it is the diameter; in other cases, it is the diameter of circles with equal area.

As a result of detailed studies, we found that in order to match the resonance state of the group of photosensitive strips 12 to the wavelength λ, it is necessary to satisfy the following conditions, taking into account the shape of the bent portion and the photonic band gap. Ta.
The element 101 causes an electric field to propagate through the photosensitive pieces 12 and the wire pieces 14 in a first direction, which is the direction in which the photosensitive pieces 12 and wire pieces 14 are arranged, and in a second direction, which is a direction perpendicular to the plane on which the photosensitive pieces 12 are placed. For the infrared light having a polarized in-plane propagation mode and incident from the second direction,
|k _a ×L+k _w ×(PL)−2×m×π|≦π/2 (m is an appropriate positive integer)...(1)
The following relationship exists when the positive propagation constants of the photosensitive piece 12 portion and the wire piece 14 portion of the propagation mode traveling in the first direction are k _a and k _w , respectively.

In the element 101, the incident light (detection light) is subjected to a resonance (two resonances) in which two optical resonances, the individual optical resonance of the photosensitive piece 12 and the overall optical resonance via the wire piece 14, are combined. heavy resonance). In this case, the resonance due to light incident perpendicularly to the element 101 is a very strong resonance, but the resonance characteristics suddenly deviate from the resonance condition for light incident obliquely to the element 101.
In a photodetector for infrared light, background noise is mostly caused by thermal radiation from the surroundings. In the element 101, in order to suppress heat radiation from the surroundings easily, at low cost, and to make it compact and to make the element with less noise, the above-mentioned double resonance is used to increase directivity, in other words, FOV (Field Of View) narrowed to reduce sensitivity to thermal radiation (infrared rays) from the surroundings. As a result, in the case of an infrared light detection element where heat radiation from the surroundings becomes a problem, the element 101 becomes an element with high detection sensitivity and excellent S/N. In particular, in the case of infrared light detection elements that utilize intersubband transitions in quantum wells, the quantum wells themselves have resonant sensitivity (electronic resonance) at specific wavelengths. This matches double optical resonance, resulting in triple resonance, resulting in an element that is particularly strong and has a high S/N ratio that is unaffected by unnecessary thermal radiation in terms of wavelength and orientation. Alternatively, in an infrared light emitting device where unnecessary heat radiation to the surroundings is a problem, the device can achieve high emission intensity and also have excellent directivity.

Further, in the element 101, the planar shape of the functional section consisting of the photosensitive section 11 and the wire section 13 is 3-fold or more symmetrical, preferably 4-fold symmetrical. That is, the photosensitive pieces 12 and the wire pieces 14 are arranged at a period P so as to have a three-fold symmetry or more, preferably a four-fold symmetry. This substantially eliminates the polarization dependence (angle dependence of polarization) of the incident light, contributing to improved sensitivity.

The shapes of the photosensitive parts and wire parts formed in the first conductor layer 21, dielectric layer 22, and second conductor layer 23 are not limited to those shown in the bird's-eye view of FIG. The wire portion 13 may or may not sandwich the dielectric layer 22, and even if the first conductor layer 21 and the dielectric layer 22 are patterned, they are flat plates with a wide area on which a plurality of photosensitive pieces and wire pieces are placed. It may be in the form of

FIG. 3 is a plan view of FIG. 1 divided into the first conductor layer 21, dielectric layer 22, and second conductor layer 23. FIG. 3(a) shows the second conductive layer 23, and FIG. 3(b) shows the dielectric layer 22. The first conductor layer 21 is shown in FIG. 3(c). In the case of FIG. 3, the planar shape of the second conductor layer 23 is a shape having the photosensitive part 11 and the wire part 13, and the planar shape of the dielectric layer 22 is the same, and The planar shape 21 is a wide planar shape on which a plurality of photosensitive pieces and wire pieces can be placed.
FIG. 4 shows the planar shape of each layer when the dielectric layer 22 includes only the photosensitive part 11 and does not include the wire part, and FIG. In FIG. 6, the first conductive layer 21 and the second conductive layer 23 have the same planar shape including the photosensitive part 11 and the wire part 13 and are dielectric. The layer 22 has a planar shape with only the photosensitive part 11 and no wire part, and FIG. and the dielectric layer 22 has a wide planar shape on which a plurality of photosensitive pieces and wire pieces are placed, and in FIG. In this case, the dielectric layer 22 has a wide planar shape with only a photosensitive portion and a plurality of photosensitive pieces and wire pieces are placed thereon. The element 101 of the present invention may have any of the above planar shapes.

Note that when a wide planar shape on which a plurality of photosensitive pieces and wire pieces are placed is used, the formation of the layer does not involve the formation of a fine pattern, making manufacturing easier. Further, in the case shown in FIGS. 5 and 6, since the wire pieces serve as springs and connect the respective photosensitive pieces, when formed on a flexible substrate, flexibility and expansion/contraction durability are improved, which is preferable.

The dielectric layer 22 is made of a layer including a semiconductor layer and/or an insulator layer.
The semiconductor layer is a group consisting of Si, Ge, AlAs, GaAs, InAs, AlSb, GaSb, InSb, AlP, GaP, InP, AlN, GaN, InN, PbS, PbSe, ZnTe, CdTe, HgTe, ZnSe, CdSe, MgSe. One or more of them can be mentioned, and is particularly suitable for obtaining high detection sensitivity and high S/N when the element 101 is used as a photodetecting element.

When the semiconductor layer includes a quantum well, it becomes suitable for detecting infrared light in a predetermined narrow wavelength range determined by the design of the quantum well structure with extremely high sensitivity and S/N. That is, when a semiconductor layer including quantum wells is used, it becomes a highly sensitive detection element in a narrow wavelength range, while thermal radiation, which is a noise source, generally has a wide wavelength range, so a high S/N can be obtained.
Here, the quantum wells include GaAs and AlGaAs, InGaAs and InAlAs, InGaAs and InP, InGaAs and AlAsSb, InGaAsP and InP, GaAs and GaInP, GaAs and AlInP, GaN and AlGaN, GaN and AlN, InAs and AlAsSb, ZnCdSe and Examples include one or more selected from the group consisting of ZnCdMgSe, InAs and GaSb, InAs and GaInSb, InAs and InAsSb, InGaAs and GaAsSb, and InAs and AlGaSb.

If a quantum well has a defect, the photodetection sensitivity and S/N will decrease, so it is preferable that the number of quantum wells be small. Furthermore, in general, light absorption by a quantum well in the absence of absorption enhancement by a cavity is approximately proportional to the number of quantum wells, but the device of the present invention has a cavity structure using half-wavelength resonance of the TM gap plasmon mode. Therefore, even if the number of quantum wells is small, the absorption is sufficiently large. In addition, photoconductive gain (the ease with which electrons move through a group of quantum wells), which is another factor that determines sensitivity in addition to absorption, is approximately inversely proportional to the number of quantum wells. Responsiveness is better when the number is small. As a result of detailed study, it was found that the number of quantum wells is preferably 1 or more and 5 or less, more preferably 1 or more and 3 or less, and even more preferably 1.

Moreover, it is preferable that both the first conductor layer 21 and the dielectric layer 22 and the dielectric layer 22 and the second conductor layer 23 are in ohmic contact. The ohmic contact smoothes the flow of electrons at the interface between the first conductor layer 21 and the dielectric layer 22 and the interface between the dielectric layer 22 and the second conductor layer 23, allowing efficient photocurrent flow. It becomes possible to take it out. If a Schottky contact is used instead of an ohmic contact, electrons accumulated in the dielectric layer 22 will remain, making it difficult to extract a sufficient photocurrent.

FIG. 9 shows an explanatory diagram of a conduction band structure when a semiconductor layer with one quantum well is used as the dielectric layer 22. This explains the conduction band structure of the dielectric layer used in the device of Example 1.
The vertical axis in FIG. 9 represents the energy of electrons, and the shaded area represents that the area is filled with electrons. Since the quantum well is thin, electronic levels exist discretely.
In addition to the ground level, electrons in the quantum well can exist in continuous excited levels that are continuously distributed in a predetermined energy width. In this state, when a photon of infrared light with energy exactly equal to the energy difference between the two levels enters, the electron at the ground level is excited to the excited level. When a bias voltage is applied between the first conductor layer and the second conductor layer, the excited electrons move in the direction according to the bias voltage due to the electric field, and pass through the contact layer to the conductor layer. photocurrent flows.
Although FIG. 9 shows the case where a Schottky barrier exists at the boundary between the contact layer and the conductor layer, an ohmic contact is preferable as described above. However, even with a Schottky barrier, if the contact layer is doped with sufficient electrons, this barrier becomes extremely thin as shown in Figure 9, and electrons tunnel and flow freely between the contact layer and the conductor layer. This makes it possible to use it.

When used as a light emitting element, a structure in which the dielectric layer 22 includes an insulating layer can be preferably used. There are two methods for emitting light: thermal excitation and tunneling current.
In the method using thermal excitation, a current is passed through one or both of the first conductor layer 21 and the second conductor layer 23 to apply Joule heat to the dielectric layer 22, causing it to emit light. Another method for applying heat is light absorption. In this case, the thickness of the insulating layer is preferably 1 nm or more and 500 nm or less, more preferably 10 nm or more and 300 nm or less.
In the method using tunnel current, a bias voltage is applied between the first conductor layer 21 and the second conductor layer 23 to cause a tunnel current to flow through the insulator layer, causing light emission. In this case, the thickness of the insulator layer must be set to allow a tunnel current to flow. Specifically, the thickness of the insulator layer is preferably 1 atomic layer or more and 10 nm or less, more preferably 3 atomic layers or more and 3 nm or less.

Examples of the insulating layer include diamond, Si, Ge, MgO, Al ₂ O ₃ , SiO ₂ , BeO, CaO, TiO ₂ , Ti ₂ O ₃ , TiO, V ₂ O ₅ , Cr ₂ O ₃ , MnO, _Cu2O , CuO _{, ZnO, Sc2O3, Y2O3, ZrO2, Ga2O3, Nb2O5 , SnO2 , CeO2 , HfO2 , Ta2O5 , LiF , NaF} , MgF ₂ , _AlF3 , _CaF2 , _{SrF2 , YF3} , CsF, _BaF2 , LaF3, CeF3, _NdF3 , _GdF3 , _YbF3 , _{PbF2 , ThF4} , ZnS, ZnSe, ZnTe, CdSe, CdTe, AgGaS ₂ , AgGaSe ₂ , Zn ₃ P ₂ , ZnGeP ₂ , As ₂ S ₃ , As ₂ Se ₃ , AlAs, AlSb, GaP, GaAs, GaSb, GaSe, GaTe, InP, SiC, BN, AlN, GaN, NaCl, List one or more selected from the group consisting of KCl, AgCl, CsCl, KBr, RbBr, AgBr, CsBr, KI, RbI, AgI, CsI, TlI, TlBr, CaCo ₃ , SrTiO ₃ , LiTaO ₃ , ZrSiO ₄ , KNbO ₃ be able to.

The first conductor layer 21 and the second conductor layer 23 are not particularly limited as long as they are conductive materials, but examples include Ti, Cr, Ni, Au, Pt, Ag, Pd, W, Cu, A group consisting of Yb, Sm, Y, Tb, Ho, Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, Al One or more metals selected from Ti, Cr, Ni, Au, Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er, Nd, Sc, Zr, Hf, An alloy containing one or more metals selected from the group consisting of Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, Al, Ti, Cr, Ni, Au, Pt, Ag, Pd , W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In , Al, and any one of ITO, AZO, GZO, IZO, IGZO, ATO, FTO, FZO, and TiN.

As described above, the element 101 of Embodiment 1 provides a compact detection element and light emitting element for infrared light having a high S/N ratio and high photoelectric conversion efficiency.

<Element manufacturing method>
The element 101 of Embodiment 1 can be manufactured by the following steps.

The first method is particularly suitable when the dielectric layer is an insulating film.
A substrate is prepared and a first conductor layer is formed thereon. Here, the first conductor layer may also serve as the substrate. Substrates include GaAs substrates, Si substrates, InP substrates, sapphire substrates, glass substrates such as synthetic quartz glass, borosilicate glass, and soda lime glass, ceramic substrates such as alumina and silicon nitride, acrylic, polystyrene (PS), and polypropylene ( Organic material substrates such as PP), polyethylene terephthalate (PET), and polycarbonate (PC), and metal substrates such as aluminum, iron, stainless steel, and copper can be preferably used. When the substrate is not insulating, it is also preferable to form an oxide film, a nitride film, and/or an oxynitride film on the surface of the substrate. Here, the substrate may be a material with sufficient rigidity such as Si or synthetic quartz, or a material with high flexibility such as PET.
Next, a first conductor layer is formed on the substrate. Examples of methods for forming the first conductive layer include sputtering, thermal evaporation, electron beam evaporation, CVD (Chemical Vapor Deposition), MOCVD (Metal Organic Vapor Deposition), and atomic layer deposition.
A dielectric layer is then formed on the first conductor layer. Examples of methods for forming the dielectric layer include sputtering method, thermal evaporation method, electron beam evaporation method, CVD method, MOCVD method, sol-gel method, spin coating method, dip coating method, bonding method, and atomic layer deposition method. be able to.
After that, a second conductive layer is formed on the dielectric layer. Examples of methods for forming the second conductive layer include sputtering, thermal evaporation, electron beam evaporation, CVD, MOCVD, and atomic layer deposition.

Next, lithography and etching are performed to form a photosensitive part pattern having a plurality of photosensitive pieces and a wire part pattern having a plurality of wire pieces on the second conductor layer. Then, the dielectric layer is also etched using these patterns as a mask.
Here, as the lithography, electron beam lithography, EUV lithography, and nanoimprinting, which have excellent microfabrication properties, are preferable, but KrF lithography, ArF lithography, and ArF immersion lithography may also be used. For etching, dry etching such as reactive ion etching, which has excellent microfabrication properties, can be preferably used.
Further, in forming the pattern of the second conductor layer, a lift-off method may be used instead of etching. That is, a lift-off resist pattern is formed on the dielectric layer by lithography, a second conductive layer is deposited by sputtering or vapor deposition, and then the resist is removed to form a second conductive layer having a pattern of photosensitive parts and wire parts. Two conductor layers may be formed. Since pattern formation of the dielectric layer is performed in self-alignment using the pattern of the second conductive layer as a mask, problems such as misalignment do not occur. Since the wire piece is thin, this self-aligned formation allows it to be formed at a high yield.
Through the above steps, the element 101 having the structure shown in FIG. 1 is manufactured.

The second method is a method in which a dielectric layer is formed first, and is a suitable manufacturing method when the dielectric layer includes a semiconductor layer having a crystal structure. In the second method, a high-quality semiconductor layer can be easily formed, so that the performance of the element 101 can be easily improved.

First, a first base having sufficient rigidity is prepared, and a buffer layer and a sacrificial layer responsible for surface flattening are sequentially formed thereon.

Any material can be used as the first substrate as long as it has sufficient rigidity and allows a semiconductor layer to be epitaxially formed thereon.
The buffer layer may be made of any material as long as its surface can be made sufficiently flat and smooth.
Examples of methods for manufacturing the buffer layer include MBE (Molecular Beam Epitaxy), MOCVD, MOPVE (Metal Organic Vapor Phase Epitaxy), HVPE (Hydride Vapor Phase Epitaxy), and LPE ( liquid phase epitaxy), but these It is not limited to.
The thickness of the buffer layer is not particularly limited, but may be, for example, 50 nm or more and 1000 nm or less.

The sacrificial layer may be any film that can have an etching rate higher than the etching rate of the material of the semiconductor layer. For example, when using GaAs as the semiconductor layer, AlGaAs, particularly AlGaAs with an Al composition ratio of 50% or more and 100% or less, can be preferably used. This is because AlGaAs having this Al composition ratio can be easily removed by wet etching with a hydrofluoric acid aqueous solution.
Examples of methods for manufacturing the sacrificial layer include, but are not limited to, MBE, MOCVD, MOPVE, HVPE, and LPE.
The thickness of the sacrificial layer is not particularly limited, but may be, for example, 500 nm or more and 2000 nm or less.

Next, a first interfacial layer is formed on the sacrificial layer.
Examples of the method for forming the first interface layer include, but are not limited to, MBE, MOCVD, MOPVE, HVPE, and LPE.

A semiconductor layer is then formed over the first interface layer.
The semiconductor layer is preferably formed epitaxially by a method such as MBE, MOCVD, MOPVE, HVPE, or LPE.
Although a laminated film can be preferably used as the semiconductor layer, a single layer film can also be used. When a single layer film is used, a pn junction is formed in the semiconductor layer by creating an impurity distribution.
When the semiconductor layer is a laminated film, for example, a plurality of combinations of a GaAs layer doped with Si, an undoped GaAs layer, and an AlGaAs layer are laminated by the MBE method.

A second interfacial layer is then formed over the semiconductor layer.
Examples of the method for forming the second interface layer include, but are not limited to, MBE, MOCVD, MOPVE, HVPE, and LPE.

A first conductor layer is then formed over the second interfacial layer.
Methods for forming the first conductive layer include sputtering, thermal evaporation, electron beam evaporation, MOCVD, etc., but are not limited to these methods. Any formation method that provides excellent surface flatness can be used.

Thereafter, the sample is turned upside down and bonded onto the second substrate so that the first conductive layer is in contact with the sample. Note that the second base becomes the substrate of the element 101.
Examples of this bonding method include Au--Au diffusion bonding.
In this method, for example, Ti with a thickness of 10 nm and Au with a thickness of 500 nm are stacked on the second substrate. The first conductor layer also has at least its surface side made of Au, and the two are brought into contact under pressure and heat. The conditions include, for example, a pressure of 5 to 10 MPa and a temperature of 250 to 330° C. for 1 hour.
Here, in order to reduce the stress generated at this time, it is also preferable to subsequently apply heat treatment under the same conditions without applying pressure.
As the second bonding method, an epoxy bonding method can be used.
Other bonding methods include eutectic bonding (soldering, silver soldering), anodic bonding, and surface activated bonding (cleaning the surface with Ar ⁺ ions under ultra-high vacuum and bonding at room temperature). , diffusion bonding using Au fine particles, etc. can also be mentioned.

The second substrate can be used as long as its surface has flatness and smoothness suitable for bonding. For example, as the second substrate, a glass substrate such as a GaAs substrate, a Si substrate, an InP substrate, a sapphire substrate, a synthetic quartz glass, a borosilicate glass, a soda lime glass, a ceramic substrate such as alumina or silicon nitride, an acrylic substrate, a polystyrene (PS) substrate, etc. ), organic material substrates such as polypropylene (PP), polyethylene terephthalate (PET), and polycarbonate (PC), and metal substrates such as aluminum, iron, stainless steel, and copper.

Thereafter, the first substrate is etched away, and subsequently the buffer layer is also etched away. These etchings may be mechanical polishing, wet etching, or dry etching.
After that, the sacrificial layer is removed by etching. This etching may be wet etching or dry etching.

Thereafter, a second conductor layer is formed on the exposed first interface layer.
Next, lithography and etching are performed to form a photosensitive part pattern having a plurality of photosensitive pieces and a wire part pattern having a plurality of wire pieces on the second conductor layer. Then, the dielectric layer is also etched using these patterns as a mask.
Here, as the lithography, electron beam lithography, EUV lithography, and nanoimprinting, which have excellent microfabrication properties, are preferable, but KrF lithography, ArF lithography, and ArF immersion lithography may also be used. For etching, dry etching such as reactive ion etching, which has excellent microfabrication properties, can be preferably used.
Further, in forming the pattern of the second conductor layer, a lift-off method may be used instead of etching. That is, a lift-off resist pattern is formed on the dielectric layer by lithography, a second conductive layer is deposited by sputtering or vapor deposition, and then the resist is removed to form a second conductive layer having a pattern of photosensitive parts and wire parts. Two conductor layers may be formed. Since pattern formation of the dielectric layer is performed in self-alignment using the pattern of the second conductive layer as a mask, problems such as misalignment do not occur. Since the wire piece is thin, this self-aligned formation allows it to be formed at a high yield.
Through the above steps, the element 101 having the structure shown in FIG. 1 is manufactured.

(Embodiment 2)
In Embodiment 2, an element (an infrared light detection element and a light emitting element) with improved functionality such as directivity and light gathering ability will be described.
As shown in FIG. 10(a), the element 106 of Embodiment 2 has the element shown in Embodiment 1 arranged in the center as a core region 51, and a peripheral region 52 shown below arranged outside the core region 51. The configuration is as follows.
In the peripheral area 52, second photosensitive pieces and second wire pieces, which will be described below, are arranged in a two-dimensional manner. The second photosensitive piece has a structure in which a dielectric layer and a second conductive layer described in Embodiment 1 are sequentially laminated on a surface made of the first conductive layer described in Embodiment 1. have The second wire piece has a maximum width of more than 0 and has a conductivity of L/5 or less, and connects at least the second conductor layer of the adjacent second photosensitive piece with a non-linear part. It is a two-dimensional array of wires. Here, the wire length of the second wire piece changes depending on the distance of one-dimensional, two-dimensional, or polar coordinate radius from the center of gravity of the array of photosensitive pieces in the core region 51 as the origin. It has become something like that. Here, the second wire pieces may be arranged with a period P. By fixing the period of the second wire piece to P, there is an effect that the wire length of the second wire piece can be designed easily.
Here, the planar shape of the element 106 is not limited to a rectangle, although both the core region 51 and the peripheral region 52 have a rectangular outer shape. It may be circular (element 107) as shown in FIG. 10(b), triangular, hexagonal, or octagonal.

As described in Embodiment 1, the core region 51 having the same device structure as Embodiment 1 exhibits a narrow FOV, high S/N ratio, and high photoelectric conversion efficiency. On the other hand, in the peripheral region 52, by adjusting the wire length of the second wire piece and controlling the phase, it is possible to provide the element 106 with directivity, light convergence, and divergence as a kind of phased array. can. Here, in order to improve the light convergence to one point (focal point) or the light divergence from one point (focal point), it is preferable that the wire length of the second wire piece changes monotonically.
The elements 106 and 107 of the second embodiment make it possible to provide an element that has a high S/N ratio, high photoelectric conversion efficiency, and has desired directivity, light convergence, and divergence.

Furthermore, in the elements 106 and 107 of the second embodiment, the substrate (not shown) is made of a flexible material such as PET, polycarbonate, or rubber, or a membrane made of SiN, etc., and the substrate is bent to give a variable curvature. By changing the shape of the wire piece within the plane by expanding and contracting it, it is also possible to create a variable focus element. In this case as well, the core region 51 is formed on a highly rigid substrate as a fixed region to ensure a narrow FOV, high S/N ratio and high photoelectric conversion efficiency, and the peripheral region 52 is formed on a flexible substrate. Preferably, the deformation is performed to provide variability. For example, the core region 51 is formed on a relatively thick substrate, and the peripheral region 52 is formed on a thin film substrate, and is deformed by applying pressurization or negative pressure to the substrate like a diaphragm, or by expanding and contracting the periphery of the substrate. can be done.
The propagation coefficient k _w of a wire piece in a folded shape such as an S-shape changes due to expansion and contraction, and generally, the interval between the second light-sensitive pieces also changes, so an in-plane phase distribution occurs in the peripheral region 52 and the light focusing property and It becomes possible to vary the divergence.
In this case, for at least the second wire piece, it is preferable to use phosphor bronze, beryllium copper, carbon-containing iron, etc., which are often used as spring materials, because of their high expansion and contraction durability. Further, the structure of the second wire piece is preferably an aerial wire-like structure in which the dielectric layer 22 is not formed on the second wire piece as shown in FIG. 6, since it has high elasticity.

It should be noted that the present invention is not limited to the above-mentioned embodiments, and the above-mentioned embodiments are merely examples for explaining the present invention, and are substantially the same as the technical idea described in the claims of the present invention. Anything that has the same configuration and produces similar effects is included within the technical scope of the present invention.

EXAMPLES Hereinafter, the present invention will be explained in more detail using examples, but the present invention is not necessarily limited to the following examples.

(Example 1)
In Example 1, an element according to Embodiment 1 was manufactured and its characteristics were evaluated.
First, prior to the experiment, the resonance characteristics of a conventional cavity structure (patch antenna structure) and the resonance characteristics of the structure of the present invention were simulated. The results are shown in FIG. Here, the left side of FIG. 11 shows the structure of the element, and the right side shows the light absorption resonance characteristics (a diagram of the relationship between wavelength and light absorption).

FIG. 11A shows an example of a conventional structure, which is an element 102 in which square cavities each having a width L on one side, which are photosensitive pieces, are arranged in a matrix at a pitch P. Here, the refractive index of the dielectric layer 22 was assumed to be 3.05+0.03i, and the thickness was 200 nm. It is assumed that both the first conductor layer 21 and the second conductor layer 23 are made of Au and have a thickness of 100 nm. L is 1.08 μm, and P is 2.0 μm. This size and arrangement are such that vertically incident light with a wavelength of 6.7 μm causes resonance, and a peak light absorption is actually observed at a wavelength of 6.7 μm. Note that in this structure, the photosensitive pieces are not electrically connected to each other, so it is difficult to extract the photoelectrically converted electrical signal.

FIG. 11(b) shows an element 103 obtained by connecting the photosensitive pieces shown in FIG. 11(a) with a straight wire piece having a width W of 0.1 μm. The wire piece has a layered structure in which a wire-shaped dielectric layer 22 and a second conductor layer 23 are laminated on a first conductor layer 21 . It can be seen that by connecting the photosensitive pieces with wire pieces, the wavelength that causes resonance shifts to the shorter wavelength side. In order to cause resonance at a wavelength of 6.7 μm using a straight wire piece, it is necessary to widen the period P, which increases the size of the element 103.

FIG. 11(c) shows an element 104 obtained by connecting the photosensitive pieces shown in FIG. 11(a) with a Z-shaped wire piece having a width W of 0.1 μm. The layer structure of the wire piece was the same as that of element 103, and S shown in the figure was 0.45 μm. At a wavelength of 6.7 μm, the propagation constant of this photosensitive piece is k _a =2.83 μm ⁻¹ , and the propagation constant of the Z-shaped wire piece with S=0.45 μm is k _w =3.14 μm ⁻ It is ¹ . The structure of this wire piece is k _a × L + k _w × (PL) - 2 × m × π = -0.34 for a positive integer m = 1, so the above equation (1) The relationship |k _a ×L+k _w ×(PL)−2×m×π|≦π/2 is satisfied. As a result, resonance was observed at a wavelength of 6.7 μm, the peak level of light absorption was higher than that of element 102, the half-width was narrow, and the resonance had a high Q value.
In this structure, since the photosensitive pieces are electrically connected to each other, it is possible to extract the photoelectrically converted electrical signal. Further, in this structure, the photosensitive pieces are arranged in a period P, and the photosensitive pieces are electrically connected to each other while maintaining the period, so that the element 104 is compact.

FIG. 11(d) shows an element 105 obtained by connecting the photosensitive pieces shown in FIG. 11(a) with an S-shaped wire piece having a width W of 0.1 μm. The layer structure of the wire piece was the same as that of element 103, and S shown in the figure was 0.38 μm. The structure of this wire piece is that the _{photosensitive} piece part is the same ^as that in FIG. Since the propagation constant is k _w = 3.65 μm ^-1 , for a positive integer m = 1, k _a ×L + k _w × (PL) - 2 × m × π = +0.13, and the above The relationship of formula (1) is satisfied. As a result, resonance was observed at a wavelength of 6.7 μm, the peak level of light absorption was higher than that of element 102, the half-width was narrow, and the resonance had a high Q value.
In this structure, since the photosensitive pieces are electrically connected to each other, it is possible to extract the photoelectrically converted electrical signal. Further, in this structure, the photosensitive pieces are arranged in a period P, and the photosensitive pieces are electrically connected to each other while maintaining the period, so that the element 105 is compact.

Note that the propagation constant k _a of the photosensitive piece portion and the propagation constant k _w of the wire piece portion of various shapes are determined by precise numerical calculations. As research progresses in the future, it is expected that it will be possible to obtain reliable values at a practical level using appropriate equivalent circuit depictions. It is also possible to obtain these values experimentally if a special sample is prepared for this purpose, but at present there is no way to know these values other than relying on numerical calculations. Here, we will explain the meanings of ka _{and kw} and how to find them in this study.
The propagation constant k represents how many radians the phase of a mode of interest in a certain waveguide structure rotates per unit length. More specifically, if λ _P is the propagation wavelength of the mode within the waveguide structure, k=2π/λ _P is defined. The propagation wavelength is generally different from the wavelength λ in air or vacuum, and in the field of microwave waveguides, λ _P is sometimes called the guide wavelength. Here, both the photosensitive piece and the wire piece are MIM waveguides, and the propagation wavelength λ _P and propagation constant (also called wave number) k of the TM gap plasmon mode propagating there are discussed in Non-Patent Document 1 etc. .

The propagation constant _{k a} of the photosensitive piece is determined by considering an infinitely long MIM waveguide with a width L = 1.08 μm, and an electric field polarized in the second direction at the center of the end face of the width L, An electric dipole that vibrates at a frequency that would emit infrared light with a wavelength of 6.7 μm in a vacuum was placed, and the propagation constant of a propagation mode that propagated infinitely in the longitudinal direction of the MIM waveguide was determined.
The propagation constant of a Z-shaped or S-shaped wire piece is shown in Figures 11(c) and (d) when only a wire piece with width W and length PL extending in the x direction is infinitely extended in the x direction. Consider a MIM waveguide made only of repeatedly connected wire pieces, and an electric field is polarized in the second direction at the center of the end face of width W, and in a vacuum, it emits infrared light with a wavelength of 6.7 μm. An electric dipole that vibrates at a frequency of
Various numerical calculation methods that can solve Maxwell's equations can be used for this calculation, but in this study we used the finite element method.

Next, an element having the structure shown in FIG. 11 was manufactured and its characteristics were evaluated. Here, the first conductor layer 21 is a Ti metal layer with a thickness of 3 nm and an Au metal layer with a thickness of 150 nm, and the dielectric layer 22 is a GaAs semiconductor layer in which a single quantum well having a band structure shown in FIG. 9 is formed. and a contact layer with a total thickness of 200 nm, and the second conductive layer 23 is a 3 nm thick Ti and 100 nm thick Au metal layer.
For reference, FIG. 12 shows an enlarged SEM photo of the fabricated device, and FIG. 13 shows an SEM photo of the entire device.

This device was manufactured by the method shown below.
First, the method for manufacturing the quantum well layer will be described. After preparing a GaAs (100) substrate as the first substrate and removing the oxide film on its surface by heating at 580°C, use a GaAs buffer for surface flattening using an MBE device (COMPACT21T, manufactured by RIBER). The layer was grown to a thickness of 300 nm. Here, the subsequent films containing Ga and As were formed using the same MBE apparatus.
Thereafter, an AlGaAs sacrificial layer with an Al composition of 90% was formed to a thickness of 900 nm, and an AlGaAs sacrificial layer with an Al composition of 55% was formed to a thickness of 100 nm. The formation temperature at this time is 580°C.
Subsequently, a total of seven layers of films containing Ga and As were formed as a first contact layer. The first contact layer is made of Si-doped GaAs (Si: 5×10 ¹⁸ /cm ^{3 ) with a total thickness of 28 nm, which is made by laminating seven layers of Si-doped GaAs (Si: 5×10 18 /cm 3} ) with a thickness of 4 nm. It was doped with δ a total of 7 times at 3×10 ¹² /cm ² . The formation temperature for this layer and subsequent layers is 530°C. Thereafter, GaAs with a thickness of 15 nm and a Si volume content of 3×10 ¹⁸ /cm ³ and undoped GaAs with a thickness of 5 nm were formed. After the sacrificial layer, the 48 nm thick portion up to this point corresponds to the contact layer on the left side of FIG.

After that, in order to form a quantum well structure that plays a central role in the device, an Al _0.3 Ga _0.7 As barrier layer with a thickness of 50 nm and a Si volume content of 3 × 10 ¹⁸ with a thickness of 4 nm are formed. A GaAs quantum well layer (the last 0.85 nm is undoped) with a thickness of /cm ³ and an Al _0.3 Ga _0.7 As barrier layer with a thickness of 50 nm were formed.
Finally, as the second contact layer on the right side of FIG. 9, after forming GaAs with a thickness of 15 nm and a Si volume content of 3×10 ¹⁸ /cm ³ , a 4 nm thick Si-doped GaAs (Si: 5×10 ¹⁸ /cm ³ ) were laminated, and just before forming each layer, δ-doping was performed with Si at 3×10 ¹² /cm ² a total of 7 times.

Next, the formation of the first conductor layer and the process of inverting the quantum well structure using it will be described. A first conductive layer of Ti with a thickness of 3 nm and Au with a thickness of 150 nm was formed by sputtering on the second contact layer located on the top surface of the quantum well structure. In addition, Ti to a thickness of 10 nm and Au to a thickness of 500 nm are formed by sputtering on another GaAs (100) substrate serving as a second base. The substrate on which quantum wells were formed was turned upside down to bring the two Au surfaces into contact with each other, and while applying a load of 5 MPa, the substrate was heated at 250° C. for 60 minutes in a nitrogen gas atmosphere. As a result, the first and second base bodies were integrated. Next, the first GaAs substrate was mechanically polished leaving about 100 μm, and the remaining GaAs substrate and buffer layer were mixed with a 1 g/ml citric acid aqueous solution and a hydrogen peroxide solution in a ratio of 10:1. The AlGaAs sacrificial layer was selectively removed using a solution prepared at 38° C. to expose the AlGaAs sacrificial layer. Finally, the AlGaAs sacrificial layer was selectively removed using a hydrofluoric acid aqueous solution.

The structure shown in FIG. 12 was fabricated by electron beam lithography, lift-off, and dry etching. After applying an electron beam resist and drawing a pattern in an area of 100 μm square (each of which becomes a detector) using an electron beam, a pattern corresponding to the second conductor layer is drawn by development. 1 contact layer was exposed. On top of that, Ti with a thickness of 3 nm and Au with a thickness of 150 nm were formed as a second conductor layer. By dissolving the resist (lift-off method), a second conductor layer having the expected pattern was formed. Finally, using this second conductor layer as a mask, the dielectric layer was vertically removed by inductively coupled plasma etching using a mixed gas of chlorine and nitrogen.
Separately, electrode pads connected to each of the individual 100 μm square detectors were formed by photolithography and brought into contact with the second conductive layer of each detector. The electrode pad is approximately 400 μm square and made of 3 nm thick Ti and 150 nm thick Au formed on a 100 nm thick SiO ₂ insulating layer.
Finally, the necessary parts including the detector and electrode pads on the second substrate were cut out by cleavage, mounted on an 8-pin ceramic package with conductive epoxy, and each electrode pad was bonded with Ag paste to the pins of the package using Au wire. .

Next, the characteristics of the manufactured device will be explained.
FIG. 14 shows the wavelength dependence of the light absorption rate and sensitivity of the fabricated infrared light detection element to vertically incident light. Sensitivity was measured using a Fourier transform infrared spectrophotometer (FT/IR-6200, manufactured by JASCO Corporation). Since it is difficult to directly measure the light absorption rate using vertically incident light, it was measured as follows. First, infrared light was incident at an angle of 26° with respect to normal incidence, and the reflected light was monitored with a microscope Fourier transform infrared spectrophotometer, and the absorption rate was determined from the intensity of the reflected light. After confirming that the values matched the simulation values and that the simulation results were reliable, a simulation was performed for vertically incident light to calculate the absorption rate. Here, the length L of one side of the actually produced photosensitive piece is 1.17 μm, the period P is 2.0 μm, the width W of the wire piece is 0.1 μm, and the wire shape is Z-shaped (Fig. 11 ( c), and was plotted using S as a parameter.

As explained in connection with FIG. 11, the resonant wavelength can be tuned by changing the shape of the wire. In FIG. 14, S=0 μm corresponds to the straight wire in FIG. 11(a), and as seen there, at this time, the resonance wavelength is the shortest. As S increases, the wire becomes longer and the resonant wavelength accordingly becomes longer. In FIG. 14, when S=0.29 μm, the resonance wavelength exactly coincided with the optical absorption wavelength (electronic resonance) unique to the quantum well of 6.7 μm, and the maximum sensitivity was obtained. At this time, in addition to the optical resonance of the individual photosensitive pieces, double optical resonance is caused by electrically connecting the photosensitive pieces with a wire piece having a non-linear portion, and the wavelength of 6.7 μm is generated. It can be seen that triple resonance is achieved at , resulting in extremely high light absorption and extremely high sensitivity. The maximum sensitivity observed individually is 3.3 A/W, which is 61% when expressed in terms of quantum efficiency. This value is comparable to that of the HgCdTe detector, which is currently used exclusively as a high-sensitivity detector in these wavelength ranges. However, since the detector according to the present invention does not use toxic elements such as Hg or Cd, the present invention is also valuable in that it allows conventional toxicity detectors to be replaced with low toxicity materials.

FIG. 15 shows the dependence of detection sensitivity on the angle of incidence of infrared light. The incident angle when the normalized sensitivity is 0.5 represents FOV/2.
Here, in the structure of the present invention, L of the photosensitive piece is 1.17 μm, P is 2.0 μm, the width of the wire piece is 0.1 μm, the shape is Z-shaped, and S is 0.45 μm.
The prior art is an element 102 in which only square photosensitive pieces with one side L are arranged in a matrix with a period P as shown in FIG. 11(a). Here, L is 1.17 μm and P is 2.0 μm. It is actually impossible to extract a current signal from just a square photosensitive piece that is not connected by wires. The results shown here are hypothetical results obtained through simulation to clarify the effect that the presence of a wire piece has on the incident angle dependence.
While the prior art element 102 has a wide FOV of 117°, the FOV of the present invention is 66°, which is about half as narrow. Because of the narrow FOV, the device of the present invention is less likely to pick up noise due to background light (thermal radiation from the surroundings).

FIG. 16 shows the polarization dependence of the detector sensitivity.
Here, in the structure of the present invention, L of the photosensitive piece is 1.17 μm, P is 2.0 μm, the width of the wire piece is 0.1 μm, the shape is Z-shaped, and S is 0.45 μm.
The prior art in this figure is a Brewster angle incidence detector using the same quantum wells but without any cavity absorption enhancement. A conventional infrared light detection element using a quantum well employs an oblique incidence structure in order to obtain incident light whose electric field is polarized in the second direction (defined as a polarization angle of 0°). One form of this is Brewster's angle incidence, which efficiently captures light at a special angle that allows the incident light to efficiently penetrate into the quantum well without reflection at the surface. However, polarization dependence has always been a problem because quantum wells have no sensitivity to incident light polarized in orthogonal directions (90° polarization angle). FIG. 16 shows the situation based on actual measurements.
In the conventional element, the sensitivity decreases monotonically as the polarization angle increases from 0° to 90° and becomes almost zero at 90°, whereas in the case of the present invention, the sensitivity hardly changes from 0° to 90°. By adopting a 4-fold symmetrical structure, polarization dependence is eliminated and maximum sensitivity can be exhibited for any incident light.

FIG. 17 shows the environmental temperature dependence of detection sensitivity. In this figure, the shapes of the characteristic curves and the positions of the peak wavelengths are normalized and plotted for comparison. It can be seen that this element has sensitivity not only at 78K but also at 293K (20°C). This is because the quantum well is incorporated into a patch antenna structure (a cavity structure that utilizes half-wavelength resonance of the TM gap plasmon mode) to enhance sensitivity.

(Example 2)
In Example 2, the characteristics of the element according to Embodiment 2 were evaluated.
Here, we show that a structure in which the wire gradually becomes longer from the center can create an infrared light detection element that is particularly sensitive to spherical waves with a specific curvature. The structure shown in FIG. 11(d) is the basic structure. Wire pieces whose S value is _S0 in FIG. 11(d) are connected to the central photosensitive piece on four sides. A wire piece is connected to the outer region so that the value of S gradually increases depending on the distance from the center in the horizontal and vertical directions. This makes it possible to resonate strongly with wavefronts that have locally different phases.

L is 1.08 μm, P is 2.0 μm, and W is 0.1 μm. The refractive index of the dielectric layer 22 was assumed to be 3.05+0.03i, and the thickness was 200 nm. It is assumed that both the first conductor layer 21 and the second conductor layer 23 are made of Au and have a thickness of 100 nm. Although FIG. 18 is a schematic diagram, the structure actually studied is a complicated one in which 19×19 photosensitive pieces 12 are arranged. Let the central photosensitive piece be the 0th photosensitive piece, and let S _i be the value of S of the wire piece that connects the i-th photosensitive piece and the (i+1)th photosensitive piece in each of the X and Y directions. Specifically, S ₀ =0.25 μm, S ₁ =0.30 μm, S ₂ =0.35 μm, S ₃ =0.38 μm, S ₄ =0.41 μm, S ₅ =0.43 μm, S ₆ = They were selected to increase monotonically: 0.44 μm, S ₇ =0.45 μm, S ₈ =0.46 μm, and S ₉ =0.47 μm. The wire pieces located at the same position in the X direction were all chosen to have the same S value, and the wire pieces located at the same position in the Y direction were all chosen to have the same S value.
Here, the central photosensitive piece 12 is connected in the X direction and the Y direction with four wire pieces whose S value is S ₀ , and the central photosensitive piece 12 is connected to the four photosensitive pieces 12 arranged at a period P. The region surrounded by the five photosensitive pieces 12 satisfies the condition of formula (1).

FIG. 19 shows the relationship between the light source position and the detection element sensitivity when the point light source that emits infrared light with a wavelength of 6.1 μm is at a distance Zs from the surface of the first conductive layer. The maximum sensitivity was obtained at Zs=10 μm, which means that this detection element has a strong sensitivity specifically to a spherical wave with a radius of curvature of 10 μm.
When this structure is used for an infrared light emitting element, it is possible to strongly condense infrared light of a specific wavelength or diverging light from an ultrathin light emitting element to a specific focal position without using a lens. can be emitted.

The present invention provides a compact detection element and light emitting element for infrared light having a high S/N ratio and high photoelectric conversion efficiency.
Infrared light detection elements are in strong demand in fields such as human sensors, surveillance sensors, temperature sensors, and gas sensors, where they are required to be compact and have both a high S/N ratio and high detection sensitivity.
Furthermore, infrared light emitting elements are in strong demand in fields such as gas sensors, where they are required to have both high directivity and strong emission intensity.
The detection element and light emitting element according to the present invention satisfy the above performance and are considered to be industrially useful.

11 Photosensitive section 12 Photosensitive piece 13 Wire section 14, 14a, 14b, 14c Wire piece 21 First conductor layer 22 Dielectric layer 23 Second conductor layer 51 Core region 52 Peripheral region 53 Core region 54 Peripheral region 101, 102, 103, 104, 105, 106, 107 elements

Claims (20)

An element that detects and emits infrared light with a peak wavelength of λ,
The element has at least a photosensitive part that is sensitive to the infrared light and a wire part that makes an electrical connection,
The photosensitive section is composed of a plurality of photosensitive pieces each having a square shape with a side length L or a circular shape with a diameter L, arranged two-dimensionally with a period P,
The photosensitive piece has a layered structure in which a dielectric layer and a second conductor layer are sequentially laminated on a surface made of a first conductor layer,
The thickness T of the dielectric layer is greater than 0 and less than λ/(2n), where n is the refractive index of the dielectric layer with respect to λ,
The area of the photosensitive piece is {3λ/(8n)} ² or more and {5λ/(8n)} ² or less,
The wire portion is a wire piece having a maximum width of more than 0 and having conductivity of L/5 or less, which connects at least the second conductor layers of the adjacent photosensitive pieces with a non-linear part. Consisting of
The element has an in-plane propagation mode in which the electric field propagates in a first direction, which is the arrangement direction of the photosensitive pieces and the wire pieces, and is polarized in a second direction, which is a direction perpendicular to the plane. has
With respect to the infrared light incident from the second direction,
|k _a ×L+k _w ×(PL)−2×m×π|≦π/2 (m is an appropriate positive integer)
When the positive propagation constants of the photosensitive piece part and the wire piece part of the propagation mode traveling in the first direction are respectively k _a and k _w ,
The device, wherein the planar shape of the functional section consisting of the photosensitive section and the wire section has three-fold symmetry or more. 2. The device of claim 1, wherein the dielectric layer comprises a semiconductor layer. The semiconductor layer is made of Si, Ge, AlAs, GaAs, InAs, AlSb, GaSb, InSb, AlP, GaP, InP, AlN, GaN, InN, PbS, PbSe, ZnTe, CdTe, HgTe, ZnSe, CdSe, MgSe. The element according to claim 2, comprising one or more selected from the group. 3. The device of claim 2, wherein the semiconductor layer includes a quantum well. 5. The device according to claim 4, wherein the number of quantum wells is 1 or more and 5 or less. 5. A device according to claim 4, wherein the number of quantum wells is one. The quantum wells include GaAs and AlGaAs, InGaAs and InAlAs, InGaAs and InP, InGaAs and AlAsSb, InGaAsP and InP, GaAs and GaInP, GaAs and AlInP, GaN and AlGaN, GaN and AlN, InAs and AlAsSb, and Z. nCdSe, ZnCdMgSe, InAs 7. The element according to claim 4, comprising one or more selected from the group consisting of: and GaSb, InAs and GaInSb, InAs and InAsSb, InGaAs and GaAsSb, and InAs and AlGaSb. 8. The device according to claim 1, wherein the first conductive layer and the dielectric layer, and the dielectric layer and the second conductive layer are both in ohmic contact. 2. The device of claim 1, wherein the dielectric layer is made of an insulator. The dielectric layer is made of diamond, Si, Ge, MgO, Al ₂ O ₃ , SiO ₂ , BeO, CaO, TiO ₂ , Ti ₂ O ₃ , TiO, V ₂ O ₅ , Cr ₂ O ₃ , MnO, Cu ₂ O, CuO, ZnO, _Sc2O3 , _Y2O3 , _{ZrO2 , Ga2O3} , _{Nb2O5 ,} SnO2 _{, CeO2} , _{HfO2 , Ta2O5} , LiF, NaF _{, MgF2 ,} AlF ₃ , CaF ₂ , SrF ₂ , YF ₃ , CsF, BaF ₂ , LaF ₃ , CeF 3 , NdF ₃ , GdF ₃ , YbF ₃ , PbF _{2 ,} ThF ₄ , ZnS, ZnSe, ZnTe, CdSe, CdTe, AgGaS ₂ , AgGaSe ₂ , Zn ₃ P ₂ , ZnGeP ₂ , As ₂ S ₃ , As ₂ Se ₃ , AlAs, AlSb, GaP, GaAs, GaSb, GaSe, GaTe, InP, SiC, BN, AlN, GaN, NaCl, KCl, A claim consisting of one or more selected from the group consisting of AgCl, CsCl, KBr, RbBr, AgBr, CsBr, KI, RbI, AgI, CsI, TlI, TlBr, CaCo ₃ , SrTiO ₃ , LiTaO ₃ , ZrSiO ₄ , KNbO ₃ Item 1. Element according to item 1. The first conductor layer and the second conductor layer are made of Ti, Cr, Ni , Au, Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er. , Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, Al, Ti, Cr, Ni , Au , Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh , Re, Ir, In, an alloy containing one or more metals selected from the group consisting of Al, Ti, Cr, Ni , Au, Pt, Ag, Pd, W, Cu, Yb, Sm, Y, Tb, Ho, A compound containing one or more metals selected from the group consisting of Tm, Gd, Er, Nd, Sc, Zr, Hf, Ta, Nb, V, Fe, Mo, Ru, Co, Rh, Re, Ir, In, and Al. , and any one of ITO, AZO, GZO, IZO, IGZO, ATO, FTO, FZO, and TiN. The planar shape of the photosensitive piece is a square with a size of one piece L,
L is 3λ/(8n _eff ) or more and 5λ/(8n _eff ) or less, where n _eff is the effective refractive index of the lowest order in-plane propagation mode polarized in the second direction of the photosensitive piece. and
1 . The effective refractive index n _eff is n·(1+2δ/T) ^1/2 , where δ is the average skin depth of the first conductor layer and the second conductor layer. The device according to any one of 11 to 11. 13. The element according to claim 1, wherein the planar shape of the wire piece includes a right-angled section. 14. A device according to any one of claims 1 to 13, wherein the symmetry is 4-fold symmetry. 15. A device according to any one of claims 1 to 14, wherein the arrangement is matrix-like. 16. The device according to claim 1, wherein the second conductor layer of at least a plurality of the photosensitive pieces is formed above one of the first conductor layers. 17. The device according to claim 1, wherein the second conductive layer of at least a plurality of the photosensitive pieces is formed on one of the dielectric layers. On the outside of the element according to any one of claims 1 to 17,
a second photosensitive piece in which the dielectric layer and the second conductor layer are sequentially laminated on a surface made of the first conductor layer;
A second wire having a maximum width of more than 0 and having a conductivity of L/5 or less, which connects at least the second conductor layers of the adjacent second photosensitive pieces with a non-linear part. The pieces are arranged in a two-dimensional manner,
The wire length of the second wire piece changes according to the distance in one dimension, two dimensions, or a radius vector of polar coordinates with the origin of the array gravity point of the array of photosensitive pieces. The device according to any one of Items 1 to 17 . 19. The element of claim 18, wherein the second wire pieces are arranged with the period P. The element according to claim 18 or 19, wherein the change in wire length is such that the wire length of the wire piece increases as the distance from the center of gravity of the array of the photosensitive pieces becomes the origin. .

that's all