Quantum cascade laser

Abstract

A QCL has a semiconductor substrate and an active layer provided on the semiconductor substrate. The active layer has a cascade structure that is formed by stacking a plurality of unit laminates including a light-emitting layer that generates light and an injection layer that transports electrons from the light-emitting layer. The light-emitting layer and the injection layer each have a quantum well structure in which a quantum well layer and a barrier layer are alternately stacked. A separation layer is provided between the light-emitting layer and the injection layer in the unit laminate, and the separation layer includes a separation quantum well layer having a layer thickness that is smaller than an average layer thickness of the quantum well layer included in the light-emitting layer and smaller than an average layer thickness of the quantum well layer included in the injection layer.

Description

Technical Field

[0001] This disclosure relates to a quantum cascade laser. Background Technology

[0002] A quantum cascade laser (hereinafter referred to as "QCL") is known to generate a terahertz wave of the frequency difference between a first frequency ω1 and a second frequency ω2, i.e., a third frequency ω3 (=|ω1-ω2|), by difference-frequency generation (DFG). (For example, see Patent Document 1 (Japanese Patent Publication No. 2010-521815)). Summary of the Invention

[0003] In the QCL described above, two pump light components (ω1, ω2) of mid-infrared light are generated, and terahertz waves are produced through nonlinear optical effects (NL: nonlinear mixing) and difference frequency generation (DFG) within the QCL. This technology utilizes high mid-infrared to terahertz conversion efficiency (high second-order nonlinear sensitivity χ). (2) For example, a known design involves a combined dual-upper-state (DAU) structure that fully injects charge carriers (electrons) into both luminescent upper levels. However, conventional DAU structures suffer from the problem of difficult continuous operation at room temperature due to the high threshold current density required for laser oscillation.

[0004] Therefore, one aspect of this disclosure is to provide a quantum cascade laser capable of reducing threshold current density.

[0005] One aspect of this disclosure provides a quantum cascade laser comprising: a substrate; an active layer disposed on the substrate, the active layer having a cascade structure, the cascade structure being a multi-level unit stack comprising a light-emitting layer for generating light and an injection layer for transporting electrons from the light-emitting layer, the light-emitting layer and the injection layer respectively having a quantum well structure with alternating quantum well layers and barrier layers, a separation layer disposed between the light-emitting layer and the injection layer in the unit stack, the separation layer comprising a separated quantum well layer having a thickness smaller than the average thickness of the quantum well layers contained in the light-emitting layer and smaller than the average thickness of the quantum well layers contained in the injection layer.

[0006] In the quantum cascade laser, a separation layer is provided between the emitting layer and the injection layer in each unit stack constituting the active layer. The separation layer includes a separated quantum well layer, which has a thickness smaller than the average thickness of the quantum well layers contained in the emitting layer and the average thickness of the quantum well layers contained in the injection layer. Based on this separated quantum well layer, nonlinear energy levels that contribute to nonlinear optical effects can be formed in the subband energy level structure of the quantum well structure in the unit stack. Furthermore, because the amount of carriers (electrons) injected from the injection layer of the preceding unit stack into the nonlinear energy level is relatively small, the carrier number of the nonlinear energy level is suppressed to a low level. As a result, compared with conventional DAU structures that actively inject electrons into both emitting upper energy levels, the threshold current density required for laser oscillation in the active layer can be reduced.

[0007] Alternatively, the thickness of the separated quantum well layer can be smaller than the thickness of the first quantum well layer adjacent to the separated quantum well layer in the light-emitting layer, and smaller than the thickness of the second quantum well layer adjacent to the separated quantum well layer in the injection layer. Alternatively, the thickness of the separated quantum well layer can be less than half the thickness of the first quantum well layer, and less than half the thickness of the second quantum well layer. According to this structure, the efficiency of electron transport from the light-emitting layer to the injection layer via the separated layer can be improved. As a result, the efficiency of laser oscillation can be improved.

[0008] Alternatively, the unit stack in the subband energy level structure based on the quantum well structure may have a luminescent upper energy level, a luminescent lower energy level, and a nonlinear energy level caused by the base energy level of the separated quantum well layer. According to this structure, by forming a nonlinear energy level separate from the luminescent upper and lower energy levels due to the base energy level of the separated quantum well layer, it is possible to reduce the threshold current density and simultaneously achieve a second-order nonlinear responsibility χ. (2) The improvement.

[0009] Alternatively, the energy gap between the luminescent upper energy level and the nonlinear energy level can be set to the energy E of the longitudinal optical phonon. LO Small. Based on the described structure, because a nonlinear energy level can be stably set for the luminescent upper energy level, good device characteristics can be obtained.

[0010] Alternatively, the anti-crossing gap between the lowest energy level (low energy level) in the first unit stack (which is a unit stack) and the luminescent upper energy level in the second unit stack (which is a unit stack that follows the first unit stack) can be set to be larger than the anti-crossing gap between the low energy level and the nonlinear energy level in the second unit stack. According to this structure, the current (i.e., the amount of electron injection) flowing from the first unit stack to the luminescent upper energy level in the second unit stack can be made sufficiently large relative to the current flowing from the first unit stack to the nonlinear energy level in the second unit stack. This effectively suppresses the injection of charge carriers (electrons) into the nonlinear energy level.

[0011] Alternatively, in a unit stack, the separated quantum well layer can be any one of the 4th to 6th quantum well layers counting from the foremost quantum well layer. According to this structure, it is possible to suitably achieve a structure where the wavefunction of the nonlinear energy level induced by the base energy level of the separated quantum well layer does not reach the foremost barrier layer (injection barrier) in the emitting layer. Thus, the injection of electrons from the injection layer of the preceding unit stack to the nonlinear energy level of the subsequent unit stack can be effectively suppressed.

[0012] Alternatively, the unit stack can be configured to generate light at a first frequency ω1 and a second frequency ω2 as mid-infrared light through a dual-resonance process involving the resonance of the upper luminescent level, the lower luminescent level, and the nonlinear level, as well as the difference frequency ω between the first frequency ω1 and the second frequency ω2. THz Terahertz waves. According to the structure described above, by forming the aforementioned nonlinear energy level, it is possible to reduce the threshold current density while maintaining a high mid-infrared to terahertz conversion efficiency (second-order nonlinear sensitivity χ). (2) This generates terahertz waves.

[0013] The separation layer may also include separation barrier layers, which are disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack and serve as barrier layers. The thickness of the separation barrier layer may also be smaller than the average thickness of the barrier layers contained in the light-emitting layer and smaller than the average thickness of the barrier layers contained in the injection layer. According to the structure described above, the electron transport efficiency from the light-emitting layer to the injection layer via the separation layer can be further improved.

[0014] The separation layer may also include separation barrier layers, which serve as barrier layers, disposed on both sides of the separation quantum well layer along the stacking direction of the unit stack. The thickness of the separation barrier layer may also be smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer. According to this structure, the electron transport efficiency from the light-emitting layer to the injection layer via the separation layer can be further improved.

[0015] According to one aspect of this disclosure, a quantum cascade laser capable of reducing threshold current density can be provided. Attached Figure Description

[0016] Figure 1 This is a perspective view of a laser module including one embodiment of a quantum cascade laser.

[0017] Figure 2 This is a cross-sectional view of a quantum cascade laser.

[0018] Figure 3 It is along Figure 2 A cross-sectional view of a quantum cascade laser along line III-III.

[0019] Figure 4 This is a diagram illustrating an example of the subband energy level structure in the active layer of a quantum cascade laser.

[0020] Figure 5 This is a diagram illustrating an example of the structure of a unit layer that constitutes the active layer.

[0021] Figure 6 This is a diagram illustrating an example of the structure of a unit stack with one cycle in the active layer.

[0022] Figure 7 This is a graph representing the current-optical output characteristics of a quantum cascade laser. Detailed Implementation

[0023] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Furthermore, in the description of the drawings, the same symbols are used to denote the same elements, and repeated descriptions are omitted. Also, the scale of the drawings may not necessarily be consistent with the description in the specification.

[0024] like Figure 1 As shown, the laser module 1 includes a quantum cascade laser (hereinafter referred to as "QCL") 10 and an auxiliary support 20 on which the QCL 10 is mounted. The QCL 10 is a monopolar laser element that generates light by electron migration between subbands in a semiconductor quantum well structure. The auxiliary support 20 is, for example, a ceramic substrate containing aluminum nitride (AlN).

[0025] [Structure of a quantum cascade laser]

[0026] The QCL10, for example, is configured as a terahertz light source capable of outputting terahertz waves at room temperature. Figure 2As shown, the QCL10 has a first end face 10a and a second end face 10b that are opposite each other along a direction (located on opposite sides of each other). The QCL10 is configured to emit light in the mid-infrared region (e.g., light of a first frequency ω1 and light of a second frequency ω2 (<ω1)) from the first end face 10a and the second end face 10b (specifically, the end faces 31a and 31b of the active layer 31 described later). In this specification, for convenience, the direction in which the first end face 10a and the second end face 10b are opposite each other (i.e., the direction along the light emission direction) is referred to as the X-axis direction, the thickness direction of the QCL10 (i.e., the stacking direction of the semiconductor layer 3 described later) is referred to as the Z-axis direction, and the width direction of the QCL10 (i.e., the direction orthogonal to the X-axis direction and the Z-axis direction, respectively) is referred to as the Y-axis direction.

[0027] like Figures 1-3 As shown, the QCL10 includes a semiconductor substrate 2, a semiconductor layer 3, an insulating film 4, an upper electrode 5, and a lower electrode 6. Furthermore, Figure 2 This is a cross-sectional view along the XZ plane in the central portion of the QCL10 in the width direction. The semiconductor substrate 2 is, for example, a rectangular plate-shaped InP single-crystal substrate (semi-insulating substrate: a high-resistivity semiconductor substrate doped with Fe). The semiconductor substrate 2 has a main surface 2a on which the semiconductor layer 3 is disposed and a back surface 2b opposite to the main surface 2a. The main surface 2a and the back surface 2b are flat surfaces perpendicular to the Z-axis direction, and the back surface 2b is mounted on an auxiliary support 20. The length of the semiconductor substrate 2 (length along the X-axis direction) is, for example, about several hundred μm to several millimeters. The width of the semiconductor substrate 2 (length along the Y-axis direction) is, for example, about several hundred μm to several millimeters. The thickness of the semiconductor substrate 2 (length along the Z-axis direction) is, for example, about several hundred μm.

[0028] Semiconductor substrate 2, as part of the first end face 10a of QCL10, has a side surface 2c connecting the main surface 2a and the back surface 2b. Side surface 2c has a first surface 2d and a second surface 2e. The first surface 2d is a flat surface connected to the back surface 2b and extending from the back surface 2b toward the main surface 2a. The first surface 2d is inclined relative to the main surface 2a and the back surface 2b in such a way that it moves further away from the second end face 10b as it moves toward the main surface 2a from the back surface 2b. The angle θ1 formed by the first surface 2d and the main surface 2a is, for example, about 30° to 80°. The first surface 2d is, for example, a polished surface formed by polishing the corner of a semiconductor substrate that is originally formed as a rectangular plate. The second surface 2e is a flat surface connecting the end of the first surface 2d on the main surface 2a side to the main surface 2a. The second surface 2e is inclined relative to the first surface 2d. The second surface 2e is approximately orthogonal to the main surface 2a and the back surface 2b.

[0029] Near the corner 2f formed between the first surface 2d and the second surface 2e, the components of the terahertz wave propagating from the active layer 31 toward the interface of QCL10 and the components of the terahertz wave reflected at the interface reinforce each other. Therefore, by, for example, arranging the corner 2f close to the lens surface of a lens (not shown), the terahertz wave can propagate efficiently inside the lens, thereby improving the extraction efficiency of the terahertz wave.

[0030] Semiconductor layer 3 is disposed on the main surface 2a of semiconductor substrate 2. The thickness of semiconductor layer 3 is approximately 10 μm to 20 μm. Semiconductor layer 3 has an active layer 31, an upper guiding layer 32, a lower guiding layer 33, an upper cladding layer 34, a lower cladding layer 35, an upper contact layer 36, a lower contact layer 37, and a support layer 38. The lower contact layer 37, the lower cladding layer 35, the lower guiding layer 33, the active layer 31, the upper guiding layer 32, the upper cladding layer 34, and the upper contact layer 36 are sequentially stacked from the main surface 2a side of semiconductor substrate 2. The support layer 38 is disposed on both sides (both sides in the Y-axis direction) of the ridge-shaped active layer 31, the upper guiding layer 32, and the lower guiding layer 33 between the lower cladding layer 35 and the upper cladding layer 34. In addition, the lower contact layer 37 has a portion extending outward (outward in the Y-axis direction) from the lower covering layer 35 that is stacked on the lower contact layer 37.

[0031] The lower contact layer 37 is, for example, a high-concentration Si-doped InGaAs layer with a thickness of about 200 nm (Si: 1.0 × 10⁻⁶). 18 / cm 3 The lower cladding layer 35 is, for example, a Si-doped InP layer with a thickness of about 5 μm (Si: 1.5 × 10⁻⁶). 16 / cm 3 The lower guiding layer 33 is, for example, a Si-doped InGaAs layer with a thickness of about 250 nm (Si: 1.5 × 10⁻⁶). 16 / cm 3 The active layer 31 is a layer in which a quantum cascade structure is formed. Details of the layer structure of the active layer 31 will be described later.

[0032] The upper guiding layer 32 is, for example, a Si-doped InGaAs layer with a thickness of about 350 nm (Si: 1.5 × 10⁻⁶). 16 / cm 3A diffraction grating structure, functioning as a distributed feedback (DFB) structure, is formed in the upper guide layer 32 along the resonance direction A0 of the first and second pump lights (details to follow). The upper guide layer 32 includes diffraction grating layers 32a and 32b arranged side-by-side along the resonance direction A0 as the diffraction grating structure. Diffraction grating layer 32a causes the first pump light to oscillate in a single mode. Diffraction grating layer 32b causes the second pump light to oscillate in a single mode. Furthermore, in this embodiment, the resonance direction A0 is parallel to the X-axis direction.

[0033] The upper cladding layer 34 is, for example, a Si-doped InP layer with a thickness of 5 μm (Si: 1.5 × 10⁻⁶). 16 / cm 3 The upper contact layer 36 is, for example, a high-concentration Si-doped InP layer with a thickness of about 15 nm (Si: 1.5 × 10⁻⁶). 18 / cm 3 The support layer 38 is, for example, an Fe-doped InP layer.

[0034] like Figure 3 As shown, the insulating film 4 is formed to cover the upper surface 36a of the upper contact layer 36 and the side surface 3a of the semiconductor layer 3 intersecting the Y-axis direction. A contact hole 4a is formed on the insulating film 4 to expose a portion of the upper surface 36a of the upper contact layer 36. The contact hole 4a extends along the X-axis direction such that the central portion of the upper surface 36a in the Y-axis direction is exposed. The insulating film 4 is formed, for example, of SiN. An upper electrode 5 is formed on the upper surface 36a of the upper contact layer 36. The upper electrode 5 is electrically connected to a portion of the upper surface 36a of the upper contact layer 36 via the contact hole 4a. A lower electrode 6 is formed on a portion of the lower contact layer 37 extending outward in the Y-axis direction from the lower covering layer 35, and is electrically connected to the lower contact layer 37. Thus, by allowing current to flow from the lower electrode 6 to the upper electrode 5, the QCL10 can be driven.

[0035] [Structure of the active layer]

[0036] The active layer 31 is configured to generate a first pump light with a first frequency ω1 and a second pump light with a second frequency ω2 through inter-subband luminescence migration of electrons, and to generate the difference frequency ω between the first frequency ω1 and the second frequency ω2 through difference-frequency generation (DFG) induced by Cherenkov phase matching. THz Terahertz waves of (=|ω1-ω2|).

[0037] The active layer 31 has an end face 31a and an end face 31b opposite to the end face 31a. End faces 31a and 31b are surfaces that intersect the X-axis direction, respectively. End face 31a is part of the first end face 10a of the QCL10, and end face 31b is part of the second end face 10b of the QCL10. End faces 31a and 31b constitute a resonator that causes the first pump light and the second pump light to oscillate.

[0038] like Figure 4 As shown, the active layer 31 has a cascaded structure formed by stacking multiple levels of unit stacks 16. Each unit stack 16 includes a light-emitting layer 17 that generates light and an injection layer 18 that transfers electrons from the light-emitting layer 17. The light-emitting layer 17 is the part that primarily functions to generate light. The injection layer 18 is the part that primarily functions to transfer electrons from the light-emitting layer 17 to the upper energy level of the light-emitting layer 17 in the subsequent unit stack 16. The light-emitting layer 17 and the injection layer 18 each have a quantum well structure with alternating quantum well layers and barrier layers. Thus, an energy level structure, i.e., a sub-band energy level structure, based on the quantum well structure is formed in each unit stack 16.

[0039] The unit stack 16 has a luminescent upper level (second level) L in its subband energy level structure. up =L2 and the nonlinear energy level (third level) with higher energy than the luminescent upper energy level. NL =L3. Additionally, the unit stack 16 possesses a lower luminescent energy level (first level) L in its subband energy level structure. low =L1 and moderate energy level L r The aforementioned lower energy level (first energy level) L that emits light low =L1 has a higher energy level than the luminescent upper level L up Low energy, the above-mentioned moderate energy level L r Having a lower energy level L that emits light low Low energy.

[0040] An injection barrier layer is provided between the light-emitting layer 17 and the injection layer 18a of the preceding unit stack to prevent electrons from being input into the light-emitting layer 17 from the injection layer 18a. A thin separation layer 19, sufficient to allow sufficient wave function overflow, is provided between the light-emitting layer 17 and the injection layer 18. That is, the light-emitting layer 17 and the injection layer 18 are separated by the separation layer 19. The separation layer 19 has a pair of barrier layers (separation barrier layers) sandwiching a quantum well layer (separation quantum well layer) and the quantum well layer. Thus, instead of an exit barrier layer (one barrier layer) that can be provided in the conventional QCL active layer structure, the separation layer 19 is provided in the active layer 31.

[0041] The spacing structure of each level in the subband energy level structure of the unit stack 16 is as follows. That is, the luminescent upper energy level L up and the lower energy level L of light emission low Energy interval ΔE 21 The energy E2 of the pump light at the second frequency ω2 is approximately the same. Nonlinear energy level L NL and the lower energy level L of light emission low Energy interval ΔE 31 The energy E1 of the pump light at the first frequency ω1 is approximately the same. The upper energy level L of the emitted light... up and nonlinear energy level L NL Energy interval ΔE 32 The difference frequency ω between the first frequency ω1 and the second frequency ω2 THz The energy E (=E1-E2) of the terahertz waves is roughly the same.

[0042] In the aforementioned subband energy level structure, electrons enter from the easing energy level L of the injection layer 18a of the preceding stage. r The luminescent layer 17 is injected via an injection barrier. This, along with the moderated energy level L... r The luminescent upper energy level L of the junction up They are strongly excited. Here, in the conventional DAU structure, it is set that carriers (electrons) are actively input to both of the two luminescent upper energy levels. On the other hand, in the sub-band energy level structure of active layer 31, it is set that: for the two energy levels on the upper side (luminescent upper energy level L...)... up and nonlinear energy level L NL The luminescent upper energy level L in ) up Actively injecting electrons, in contrast, into the nonlinear energy level L NL The amount of electrons injected is relatively small. That is, in the conventional DAU structure, sufficient electrons are injected into each of the two luminescent upper energy levels. In contrast, the subband energy level structure of the active layer 31 is configured to inject sufficient electrons only into the luminescent upper energy level L. up That is, in the subband energy level structure of active layer 31, the nonlinear energy level L NL It is set to primarily contribute to nonlinear optical effects.

[0043] In the active layer 31, at, for example, the difference frequency ω THz At frequencies above approximately 2 THz, through what is called the luminescent upper energy level L up , lower energy level L low and nonlinear energy level L NL The double resonant process uses a nonlinear technique to generate the difference frequency ω. THz Terahertz waves. For example, in the case of the luminescent upper energy level L... up and the lower energy level L of light emission lowThe second pump light, corresponding to the second frequency ω2, which corresponds to the electron migration between them, is incident on the active layer 31 and interacts with the nonlinear energy level L. NL and the lower energy level L of light emission low When the first pump light, corresponding to the first frequency ω1 of the electron migration between them, is incident on the active layer 31, it interacts with the upper energy level L of the emission layer. up and nonlinear energy level L NL The difference frequency ω corresponding to the migration of electrons between them THz Terahertz waves are generated and emitted.

[0044] As shown in equations (1) and (2) below, the power W(ω) of the difference frequency ω light generated by the above non-resonant process, the power W(ω1) and W(ω2) of the mid-infrared pump light of the first frequency ω1 and the second frequency ω2, and the coherence length l coh It is proportional to the square of the value; in addition, the nonlinear resistivity χ (2) With the migration dipole moment z nm Proportional. Here, in the above formula, e represents electric charge, h (the horizontal bar symbol for h) represents the Dirac constant, ΔN represents the inversion distribution number, and Γ nm This represents the emission half-width of each migration. As shown in equation (2) above, the second-order nonlinear resistivity χ between the three levels... (2) It is represented by the product of the dipole moments in the corresponding migration.

[0045]

[0046]

[0047] Migration to the luminescent lower energy level L low Electrons are moderated at energy level L r This is mitigated. Thus, through the emission from the lower energy level L... low Electrons are extracted, and at the luminescent upper energy level L up Nonlinear energy level L NL , lower energy level L low An inverted distribution is formed between these points to achieve laser oscillation. The moderated energy level L... r The moderately injected electrons are cascaded into the luminescent upper energy level L of the subsequent luminescent layer 17b via injection layer 18. up Furthermore, the moderate energy level L r It is not limited to being composed of only one energy level; it can be composed of multiple energy levels, or it can be composed of mini bands.

[0048] As described above, electron injection, electron luminescence migration, and electron relaxation are repeated in the multiple unit stacks 16 constituting the active layer 31, cascaded light generation occurs in the active layer 31. When electrons cascade through the multiple unit stacks 16, a first pump light with a first frequency ω1 and a second pump light with a second frequency ω2 are generated through inter-subband luminescence migration of electrons in each unit stack 16. Furthermore, the difference frequency ω between the first frequency ω1 and the second frequency ω2 is generated through difference frequency generation caused by Cherenkov phase matching. THz Terahertz waves of (=|ω1-ω2|).

[0049] The structure of the active layer 31 will be further explained. Figure 5 and Figure 6 The subband energy level structure shown is the composition of Figure 4 An example of an active layer structure in a subband energy level structure is shown. Furthermore, in Figure 5 The diagram shows a portion of the repeating structure formed by the emitting layer 17 and the injection layer 18, illustrating the quantum well structure and subband energy level structure in its operating electric field. In this structural example, the designed value of the first frequency ω1 is 930 cm⁻¹. －1 The design value for the second frequency ω2 is 1000 cm⁻¹. －1 Additionally, the luminescent upper energy level L up and nonlinear energy level L NL Energy interval ΔE 32 The design value is 16 meV, and the luminescent upper energy level L up and the lower energy level L of light emission low Energy interval ΔE 21 The design value is 126 meV, and the center wavelength of the gain in active layer 31 is set to approximately 10 μm. Nonlinear energy level L NL and the lower energy level L of light emission low Energy interval ΔE 31 The design value is 142 meV. Furthermore, the active layer 31 is formed by, for example, stacking 50 cycles of unit layers 16.

[0050] In the active layer 31, in order to generate terahertz waves caused by difference frequency generation, it is necessary to be able to generate pump light components of two wavelengths and to have a high second-order nonlinear resistivity χ relative to the two pump light components. (2) In this structural example, by setting two diffraction grating layers 32a and 32b in the upper guide layer 32, the generation of the first pump light with a first frequency ω1 and the second pump light with a second frequency ω2, as well as the difference frequency ω, are achieved with a single active layer design. THz The generation of terahertz waves.

[0051] In this structural example, the carrier concentrations of the lower luminescent level L1, the upper luminescent level L2, and the nonlinear level L3 are represented as n1 to n3, respectively. If we assume that "n1 = n3", then the carrier density in each sub-band level is "n2 - n1 = 1.0 × 10⁻⁶". 15 / cm 3 "and "n3-n1=0.3×10 15 / cm 3 This carrier density is derived analytically using a non-equilibrium Green's function. Based on this derived carrier density, the design frequency (i.e., the frequency ω of the terahertz wave) is then determined. THz When the frequency is set to 2THz, the second-order nonlinear resistivity χ generated by this structural example will be... (2) The absolute value of the sum is calculated as "|χ (2) |=15nm / V。

[0052] Design frequency ω THz The first frequency ω1 and the second frequency ω2 are determined by the DFB structure (in this embodiment, a diffraction grating structure formed on the upper guide layer 32). The frequency of the finally obtained terahertz wave is determined by "ω". THz =ω1-ω2” is determined. In this structural example, the design frequency ω is... THz The frequency is set to 2THz. In addition, in the active layer 31, by using a two-cycle DFB structure, the first frequency ω1 and the second frequency ω2 are made to perform single-mode operation together, so that the terahertz wave performs single-mode operation.

[0053] like Figure 5 and Figure 6 As shown, a unit stack 16 of one period is constructed into a quantum well structure by alternating layers of 13 quantum well layers Q1 to Q13 and 13 barrier layers B1 to B13. Each quantum well layer Q1 to Q13 is, for example, an InGaAs layer, and each barrier layer B1 to B13 is, for example, an InAlAs layer.

[0054] In each unit stack 16, the portion consisting of five quantum well layers Q1 to Q5 stacked alternately from the side of the preceding unit stack 16 and five barrier layers B1 to B5 stacked alternately from the side of the preceding unit stack 16 constitutes the light-emitting layer 17, which mainly has the function of emitting light. Furthermore, the portion consisting of seven quantum well layers Q7 to Q13 stacked alternately from the side of the following unit stack 16 and six barrier layers B8 to B13 stacked alternately from the side of the following unit stack 16 constitutes the injection layer 18, which mainly has the function of electron transport. Additionally, the first barrier layer B1 in the light-emitting layer 17 functions as an injection barrier layer.

[0055] A separation layer 19 is provided between the light-emitting layer 17 and the injection layer 18 in each unit stack 16. The separation layer 19 is composed of a separation quantum well layer Q6, which serves as a quantum well layer, and two barrier layers B6 and B7, which are arranged on both sides of the separation quantum well layer Q6 in the stacking direction (i.e., the Z-axis direction) of the unit stack 16.

[0056] The thickness of the isolated quantum well layer Q6 is set to be sufficiently thin so as not to impede the transport of electrons from the light-emitting layer 17 to the injection layer 18. Specifically, the thickness of the isolated quantum well layer Q6 (1.7 nm in this structural example) is smaller than the average thickness of the quantum well layers Q1 to Q5 contained in the light-emitting layer 17 (approximately 4.7 nm in this structural example), and smaller than the average thickness of the quantum well layers Q7 to Q13 contained in the injection layer 18 (approximately 3.3 nm in this structural example).

[0057] Furthermore, in this structural example, the thickness of the separated quantum well layer Q6 (1.7 nm) is smaller than the thickness of the quantum well layer Q5 (the first quantum well layer) adjacent to the separated quantum well layer Q6 among the quantum well layers Q1 to Q5 contained in the light-emitting layer 17 (4.3 nm in this structural example), and smaller than the thickness of the quantum well layer Q7 (the second quantum well layer) adjacent to the separated quantum well layer Q6 among the quantum well layers Q7 to Q13 contained in the injection layer 18 (3.4 nm in this structural example). In this structural example, the thickness of the separated quantum well layer Q6 is set to be the thinnest among the quantum well layers Q1 to Q13 contained in the unit stack 16. Moreover, the thickness of the separated quantum well layer Q6 is less than half the thickness of the quantum well layer Q5, and less than half the thickness of the quantum well layer Q7. According to the above structure, the electron transport efficiency from the light-emitting layer 17 to the injection layer 18 via the separated layer 19 can be improved. That is, by making the thickness of the separated quantum well layer Q6 sufficiently thin, it is possible to prevent the transport of electrons from the light-emitting layer 17 to the injection layer 18 from being obstructed. As a result, the efficiency of laser oscillation can be improved.

[0058] Similarly, the thicknesses of the separation barrier layers B6 and B7 are set to be sufficiently thin so as not to impede the transport of electrons from the light-emitting layer 17 to the injection layer 18. Specifically, the thickness of the separation barrier layer B6 (0.7 nm in this structural example) and the thickness of the separation barrier layer B7 (0.9 nm in this structural example) are smaller than the average thickness of the barrier layers B1 to B5 contained in the light-emitting layer 17 (approximately 1.6 nm in this structural example), and smaller than the average thickness of the barrier layers B8 to B13 contained in the injection layer 18 (approximately 2.1 nm in this structural example). Furthermore, the thickness of the separation barrier layers B6 and B7 is smaller than the thickness of the barrier layer B5 (first barrier layer) adjacent to the separation barrier layer B6 among the barrier layers B1 to B5 contained in the light-emitting layer 17 (1.3 nm in this structural example) and the thickness of the barrier layer B8 (second barrier layer) adjacent to the separation barrier layer B7 among the barrier layers B8 to B13 contained in the implantation layer 18 (1.5 nm in this structural example). According to the above structure, by making the thickness of the separation barrier layers B6 and B7 sufficiently thin, it is possible to prevent the transport of electrons from the light-emitting layer 17 to the implantation layer 18 from being obstructed. As a result, the transport efficiency of electrons from the light-emitting layer 17 to the implantation layer 18 via the separation layer 19 can be further improved.

[0059] In each unit stack 16, the luminescent upper energy level L up and nonlinear energy level L NL Energy interval ΔE 32 The energy E of the longitudinal optical phonon is set as follows. LO Small. Specifically, in this structural example, the condition of equation (3) below is satisfied. If the condition of equation (3) is not satisfied, the luminescent upper energy level L... up and nonlinear energy level L NL The level located on the high-energy side (in this structural example, the nonlinear energy level L) NL The energy level rapidly decreases to a lower energy level due to longitudinal optical phonon scattering (in this structural example, the luminescent upper energy level L). up This results in a different stage structure from the example structure. As a result, it is difficult to obtain good device characteristics (i.e., the threshold current density and second-order nonlinear resusceptibility obtained through the example structure). On the other hand, under the condition of satisfying equation (3), because it is possible to control the upper energy level L of the light emission stage... up Stable setting of nonlinear energy level L NL Therefore, it is possible to obtain good equipment characteristics.

[0060] E 32 <E LO =34meV(In) 0.52 Ga 0.48 As)…(3)

[0061] In this structural example, the lowest energy level (base level) in the unit stack 16 (hereinafter referred to as the "first unit stack") on the front-stage side, i.e., the moderate energy level L, will be used. r (Lower level) and the luminescent upper energy level L in the unit stack 16 (hereinafter referred to as "second unit stack") arranged in the subsequent level of the first unit stack. up The anti-crossing energy gap (anti-crossing energy gap) between them is set to be lower than the moderate energy level L in the first unit stack. r Nonlinear energy level L in the second unit stack NL The anti-crossing energy gap between them is large. In this structural example, the moderate energy level L in the first unit stack is... r The luminescent upper energy level L in the second unit stack up The anti-cross gap is 6.8 meV, and the moderate energy level L in the first unit stack is... r Nonlinear energy level L in the second unit stack NL The anti-crossing gap between them is 5.0 meV. Here, the moderate energy level L in the first unit stack is... r The current flowing into each stage of the second unit stack is proportional to the square of the anti-crossing bandgap. Therefore, in this structural example, in each unit stack 16, the current relative to the luminescent upper level L... up The magnitude of the current flowing through it is relative to the nonlinear energy level L. NL The current flowing through is 1.84 times the value of the current. Thus, electrons are injected from the injection layer 18 (e.g., the easing energy level L) of the unit stack 16 of the preceding stage. r To nonlinear energy level L NL The injection amount is relatively small. That is, the nonlinear energy level L NL The electron density and number of electrons are significantly smaller than those of the luminescent upper energy level L. up The electron density and number of electrons. In other words, by setting the anti-cross gap as described above, it is possible to allow the electrons to flow from the injection layer 18 of the preceding unit stack 16 (first unit stack) to the luminescent upper energy level L of the subsequent unit stack 16 (second unit stack). up The current (i.e., the amount of electrons injected) relative to the nonlinear energy level L flowing from the injection layer 18 of the first unit stack to the second unit stack. NL The current is large enough. Therefore, it is possible to effectively suppress the influence of charge carriers (electrons) on the nonlinear energy level L. NL The injection.

[0062] As mentioned above, the nonlinear energy level L NL It is mainly formed by the separated quantum well layer Q6. Specifically, the nonlinear energy level L NL This is the energy level caused by the base level of the separated quantum well layer Q6. That is, the nonlinear energy level L. NLThe center of the wavefunction is located in the split quantum well layer Q6. Furthermore, the split quantum well layer Q6 is positioned somewhat away from the preceding unit stack 16. That is, the nonlinear energy level L... NL The wavefunction is set to be such that it does not actually reach the injection barrier layer B1 of the emitting layer 17. This is achieved by connecting it to the emitting upper energy level L. up and the lower energy level L of light emission low The nonlinear energy level L formed separately by the substrate energy level of the separated quantum well layer Q6 NL This enables a reduction in threshold current density while simultaneously achieving a second-order nonlinear resistivity χ. (2) The improvement is achieved by suppressing the amount of electron injection and forming a nonlinear energy level L that contributes to nonlinear optical effects (in this structural example, it contributes to difference frequency generation based on a dual-resonance process). NL Thus, the aforementioned effect was achieved.

[0063] Furthermore, in this structural example, in the unit stack 16, the separated quantum well layer Q6 is composed of any one of the 4th to 6th (6th in this structural example) quantum well layers counting from the foremost quantum well layer Q1. In other words, the separating layer 19 is positioned somewhat away from the injection layer 18 of the preceding unit stack 16 (e.g., approximately at the center of the unit stack 16). Therefore, it is easy to set the nonlinear energy level L... NL The wave function does not reach the injection barrier layer B1 of the emitting layer 17. As a result, the nonlinear energy level L of the unit stack 16 from the injection layer 18 of the preceding unit stack 16 can be effectively suppressed. NL The injection.

[0064] Next, the emission angle (Cherenkov emission angle) θ of the terahertz wave generated by the aforementioned active layer 31 will be determined. C An explanation will be provided. Here, we take the following case as an example: the first and second pump lights are mid-infrared light, the terahertz wave frequency range is 1THz to 6THz, and the semiconductor substrate 2 is an InP single-crystal substrate. In this case, Figure 2 As shown by arrow A1, the terahertz wave generated by difference frequency generation has a radiation angle θ as shown in the following equation (4). C The angle (relative to the resonance direction A0) propagates as a plane wave (i.e., in phase) within the semiconductor substrate 2. In equation (4), n MIR It is the refractive index of the InP single-crystal substrate (semiconductor substrate 2) relative to mid-infrared light (first pump light and second pump light), n THz It is the refractive index of the InP single crystal substrate (semiconductor substrate 2) relative to terahertz waves.

[0065] θ C =cos－1 (n MIR / n THz (4)

[0066] [Manufacturing Methods of Quantum Cascade Lasers]

[0067] An example of the manufacturing method of QCL10 will be described. First, on a semi-insulating InP substrate (semiconductor substrate 2), InGaAs with a high concentration of Si (Si: 1.0 × 10⁻⁶) is doped. 18 cm －3 A 200 nm layer was grown as the lower contact layer 37, and Si was doped with InP (Si: 1.5 × 10⁻⁶). 16 cm －3 A 5 μm thick lower cladding layer (35) is grown, and Si is doped with InGaAs (Si: 1.5 × 10⁻⁶). 16 cm －3 A 250 nm layer is grown as the lower guiding layer 33. Next, the above-mentioned unit stack 16 is stacked, for example, for 50 cycles to form the active layer 31. The active layer 31 is formed by sequentially epitaxially growing an InGaAs quantum well layer and an InAlAs barrier layer that match the InP lattice using methods such as molecular beam epitaxy (MBE) or metal-organic chemical vapor deposition (MOCVD). Then, a 350 nm layer of Si-doped InGaAs (Si: 1.5 × 10⁻⁶) is disposed on the active layer 31. 16 cm －3 As the upper guide layer 32, which also serves as a DFB diffraction grating layer, a diffraction grating (diffraction grating layers 32a and 32b) with a depth of, for example, 250 nm is formed by etching. After forming the diffraction grating, ridges with a width of, for example, 12 μm are formed along a direction perpendicular to the periodic structure of the diffraction grating (Y-axis direction). Furthermore, Fe-doped InP layers (support layers 38) are buried and regrown on both sides of the width direction (Y-axis direction) of the lower guide layer 33, the active layer 31, and the upper guide layer 32 by metal-organic chemical vapor deposition or the like. Next, Si-doped InP (Si: 1.5 × 10⁻⁶) is formed. 16 cm －3 A 5 μm upper coating layer 34 was formed by growing it, and by doping InP with a high concentration of Si (Si: 1.5 × 10⁻⁶). 18 cm －3The upper contact layer 36 is formed by growing an InGaAs layer. Next, after etching to expose the InGaAs contact layer (lower contact layer 37) on the semiconductor substrate 2 side, an insulating film 4 is formed from SiN or the like. Then, contact holes (contact holes 4a, etc.) are formed on the insulating film 4 according to the upper and lower InGaAs contact layers (upper contact layer 36 and lower contact layer 37), and a thick Au film (approximately 5 μm thick) is formed by vapor deposition and plating, thereby forming an electrode. Finally, the electrode is separated into upper and lower electrodes (upper electrode 5 and lower electrode 6) by Au etching. Based on the above, a QCL10, serving as a quantum cascade laser element, is obtained.

[0068] [Functions and Effects]

[0069] As described above, in QCL10, in each unit laminate 16 constituting the active layer 31, a separation layer 19 is provided between the light-emitting layer 17 and the injection layer 18. Figure 5 and Figure 6 As shown, the separation layer 19 includes a separated quantum well layer Q6 with a thickness smaller than the average layer thickness of the quantum well layers Q1 to Q5 contained in the light-emitting layer 17 and the average layer thickness of the quantum well layers Q7 to Q13 contained in the injection layer 18. Based on this separated quantum well layer Q6, nonlinear energy levels L that contribute to nonlinear optical effects can be formed in the subband energy level structure based on the quantum well structure of the unit stack 16. NL Furthermore, because charge carriers (electrons) move from the injection layer 18 of the unit stack 16 in the preceding stage to the nonlinear energy level L... NL The injection amount is relatively small, so the nonlinear energy level L NL The carrier number is suppressed to a low level. As a result, compared with the previous DAU structure that actively injects electrons into both of the two emitting upper energy levels, the threshold current density required for laser oscillation in the active layer 31 can be reduced.

[0070] The unit stack 16 is configured to emit light through the upper energy level L. up , lower energy level L low and nonlinear energy level L NL The resonant dual-resonance process generates light at a first frequency ω1 and a second frequency ω2 as mid-infrared light, as well as the difference frequency ω between the first frequency ω1 and the second frequency ω2. THz Terahertz waves. Based on the above structure, by forming the aforementioned nonlinear energy level L... NL It can reduce the threshold current density while maintaining a high mid-infrared to terahertz conversion efficiency (second-order nonlinear sensitivity χ). (2) This generates terahertz waves.

[0071] In the case of continuous oscillation, for example, by QCL10, the temperature of QCL10 rises because current is constantly applied to it. Therefore, the threshold current density used for laser oscillation also increases. Thus, in THz-QCLs performing continuous laser oscillation, it is particularly necessary to reduce the threshold current density. As described above, based on QCL10, a lower energy level L with a specific emission level is formed in the same manner as in conventional DAU structures. low The two high-energy levels (luminescent upper level L) up and nonlinear energy level L NL It can achieve a high second-order nonlinear responsibility χ. (2) Moreover, because it almost never rises from the preceding unit stack 16 to one of the two higher levels (nonlinear level L)... NL By injecting electrons, the threshold current density can be reduced compared to conventional DAU structures. Therefore, according to QCL10, it is possible to maintain a high mid-infrared-terahertz conversion efficiency (i.e., a high second-order nonlinear sensitivity χ). (2) At the same time, the threshold current density is reduced to enable continuous operation at room temperature.

[0072] [Example]

[0073] Figure 7 This is a graph showing the current-light output characteristics of mid-infrared (MIR) and terahertz (THz) waves measured with respect to the QCL10 of the embodiment. In this graph, the horizontal axis represents current (A) or current density (kA / cm²). 2 The vertical axis represents the peak power (W) of the MIR light or the peak power (μW) of the THz wave. Measurements were performed using an infrared indicator of a THz detector after the light generated from the QCL10 was focused using a parabolic mirror under nitrogen purging conditions at room temperature (25°C). The QCL10 was driven at a repetition frequency of 100 kHz and a pulse width of 200 ns, and its signal was detected by a lock-in amplifier.

[0074] exist Figure 7 In the diagram, Figure G1 shows the current dependence of the peak power of the MIR light with a first wavelength λ1 corresponding to the first frequency ω1. Figure G2 shows the current dependence of the peak power of the MIR light with a second wavelength λ2 corresponding to the second frequency ω2. Figure G3 shows the current dependence of the peak power of the terahertz wave generated by difference frequency generation. Figure 7 As shown, in the QCL10 of the embodiment, a threshold current density of 2.2 kA / cm² was confirmed to be relatively low compared to MIR light. 2Specifically, it was confirmed that the threshold current density could be reduced to, for example, the same level as that of the bound-to-continuum structure. Furthermore, the total peak output of the MIR light was confirmed to reach approximately 1W, achieving higher performance for the MIR-QCL than previous active layer structures. Additionally, a peak power of approximately 32μW was confirmed for the terahertz wave output. In other words, it was confirmed that mid-infrared to terahertz conversion efficiency comparable to previous DAU structures could be achieved, while simultaneously achieving a lower threshold current density than the DAU structure.

[0075] [Variation Example]

[0076] The above describes one embodiment of the present disclosure, but the present disclosure is not limited to the above embodiment.

[0077] For example, instead of the two diffraction grating layers 32a and 32b, one diffraction grating layer can be disposed on the semiconductor layer 3, or three or more diffraction grating layers can be disposed on the semiconductor layer 3. The diffraction grating layer, which functions as a distributed feedback structure, can cause at least one of the first pump light and the second pump light to oscillate in a single mode. Alternatively, instead of the diffraction grating layer, which functions as a distributed feedback structure, a structure can be constructed using the difference frequency generated in a structure where the oscillation spectrum width in the Fabry-Perot operation is extended to more than 1 THz. In such a structure, compared to the diffraction grating layer which functions as a distributed feedback structure, the output of terahertz light is reduced, but a wide THz spectrum can be obtained.

[0078] In addition, the active layer 31 is not limited to including only Figures 4-6 The structure of the unit laminate 16 shown may also include two or more active layer structures (unit laminates).

[0079] Furthermore, the semiconductor substrate 2 is not limited to a semi-insulating InP single crystal substrate, but can also be, for example, an undoped InP single crystal substrate (cc (carrier concentration): ~5×10⁻⁶). 15 / cm 3 ) or low-doped InP single-crystal substrate (Si: 5×10 15 ~1×10 17 / cm 3 In this case, a lower electrode can be formed on the back side of the semiconductor substrate. However, when the doping concentration of the semiconductor substrate is increased, the light absorption rate inside the substrate increases, thus reducing the extraction efficiency of the terahertz wave. Therefore, from the viewpoint of reducing the absorption loss of the output light of the semiconductor substrate, the semiconductor substrate is preferably an undoped semi-insulator substrate as described in the above embodiment. In addition, the semiconductor substrate 2 can be a substrate other than an InP substrate, for example, it can be a silicon single crystal substrate.

[0080] Furthermore, in the above embodiment, an active layer 31 with a lattice-matched structure for an InP single-crystal substrate was shown, but the active layer 31 could also use a structure incorporating strain compensation. Regarding the semiconductor material system of the active layer 31, it is not limited to InGaAs / InAlAs described above; various semiconductor material systems such as GaAs / AlGaAs, InAs / AlSb, GaN / AlGaN, and SiGe / Si can be used. Additionally, various methods can be applied to the crystal growth method of the semiconductor.

[0081] Furthermore, in the above embodiment, in the sub-band energy level structure of the unit stack 16, the nonlinear energy level L NL Formed from the higher energy level L of the light emission level up Relying on the high-energy side, but the nonlinear energy level L NL It can also be formed at the higher energy level L of the luminescent site. up On the low-energy side. That is, the relationship between the first frequency ω1 and the second frequency ω2 can also be the opposite of the above implementation (ω1 < ω2).

[0082] Furthermore, QCL10 is not limited to a terahertz light source that generates terahertz waves through difference frequency generation; it can also be configured as a mid-infrared light source that only generates mid-infrared (MIR) light. That is, the structure of the active layer 31 described above (the structure with the separation layer 19) can also be applied to a mid-infrared light source that only generates mid-infrared (MIR) light, i.e., a QCL.

[0083] Furthermore, in the above embodiment, the separation layer 19 is composed of a quantum well layer and two barrier layers located on both sides thereon, but the separation layer 19 may also be composed of two or more separate quantum well layers and three or more barrier layers. When the separation layer 19 includes multiple separate quantum well layers, each of the multiple separate quantum well layers contained in the separation layer 19 may contribute to the formation of nonlinear energy levels. In this case, the "layer thickness of the separate quantum well layer" described in the above embodiment can be read as the average layer thickness of the multiple separate quantum well layers contained in the separation layer 19.

[0084] [Symbol Explanation]

[0085] 1. Laser Module

[0086] 2. Semiconductor substrate (substrate)

[0087] 10. QCL (Quantum Cascade Laser)

[0088] 16 unit stacks

[0089] 17. Emissive Layer

[0090] 18 Injection Layer

[0091] 19 Separation Layer

[0092] B1 Injection into the barrier layer (barrier layer)

[0093] B6 and B7 Separation of the barrier layer (barrier layer)

[0094] B2~B5, B8~B13… Barrier Layers

[0095] L up Luminous upper energy level

[0096] L low Emitting lower energy level

[0097] L NL Nonlinear energy levels

[0098] L r Moderate energy level (low energy level)

[0099] Q6 Separating the quantum well layer (quantum well layer)

[0100] Q1~Q5, Q7~Q13 quantum well layers.

Claims

1. A quantum cascade laser, comprising: substrate; An active layer is disposed on the substrate. The active layer has a cascaded structure, which is a multi-level stack of unit layers comprising a light-emitting layer that generates light and an injection layer from which electrons are transported. The light-emitting layer and the injection layer each have a quantum well structure consisting of alternating quantum well layers and barrier layers. A separation layer is provided between the light-emitting layer and the injection layer in the unit stack. The separation layer includes a separated quantum well layer having a thickness smaller than the average thickness of the quantum well layers contained in the light-emitting layer and smaller than the average thickness of the quantum well layers contained in the injection layer, which serves as the quantum well layer. The unit stack has a light-emitting upper energy level, a light-emitting lower energy level, and a nonlinear energy level caused by the base energy level of the separated quantum well layers in the subband energy level structure according to the quantum well structure. The anti-crossing gap between the lowest energy level (low energy level) in the first unit stack, which is the unit stack itself, and the luminescent upper energy level in the second unit stack, which is the unit stack itself and is the next level in the first unit stack, is set to be larger than the anti-crossing gap between the low energy level and the nonlinear energy level in the second unit stack.

2. The quantum cascade laser according to claim 1, wherein, The thickness of the separated quantum well layer is smaller than the thickness of the first quantum well layer adjacent to the separated quantum well layer in the quantum well layer contained in the light-emitting layer, and smaller than the thickness of the second quantum well layer adjacent to the separated quantum well layer in the quantum well layer contained in the injection layer.

3. The quantum cascade laser according to claim 2, wherein, The thickness of the separated quantum well layer is less than half the thickness of the first quantum well layer and less than half the thickness of the second quantum well layer.

4. The quantum cascade laser according to claim 1, wherein, The energy interval between the luminescent upper energy level and the nonlinear energy level is set to the energy E of the longitudinal optical phonon. LO Small.

5. The quantum cascade laser according to claim 2, wherein, The energy interval between the luminescent upper energy level and the nonlinear energy level is set to the energy E of the longitudinal optical phonon. LO Small.

6. The quantum cascade laser according to claim 3, wherein, The energy interval between the luminescent upper energy level and the nonlinear energy level is set to the energy E of the longitudinal optical phonon. LO Small.

7. The quantum cascade laser according to any one of claims 1 to 6, wherein, In the unit stack, the separated quantum well layer is composed of any one of the 4th to 6th quantum well layers counting from the first quantum well layer.

8. The quantum cascade laser according to any one of claims 1 to 6, wherein, The unit stack is configured to generate light of a first frequency ω1 and a second frequency ω2 as mid-infrared light through a dual-resonance process involving the resonance of the upper luminescent energy level, the lower luminescent energy level, and the nonlinear energy level, as well as the difference frequency ω between the first frequency ω1 and the second frequency ω2. THz Terahertz waves.

9. The quantum cascade laser according to claim 7, wherein, The unit stack is configured to generate light of a first frequency ω1 and a second frequency ω2 as mid-infrared light through a dual-resonance process involving the resonance of the upper luminescent energy level, the lower luminescent energy level, and the nonlinear energy level, as well as the difference frequency ω between the first frequency ω1 and the second frequency ω2. THz Terahertz waves.

10. The quantum cascade laser according to any one of claims 1 to 6, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the average thickness of the barrier layers contained in the light-emitting layer and smaller than the average thickness of the barrier layers contained in the injection layer.

11. The quantum cascade laser according to claim 7, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the average thickness of the barrier layers contained in the light-emitting layer and smaller than the average thickness of the barrier layers contained in the injection layer.

12. The quantum cascade laser according to claim 8, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the average thickness of the barrier layers contained in the light-emitting layer and smaller than the average thickness of the barrier layers contained in the injection layer.

13. The quantum cascade laser according to claim 9, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the average thickness of the barrier layers contained in the light-emitting layer and smaller than the average thickness of the barrier layers contained in the injection layer.

14. The quantum cascade laser according to any one of claims 1 to 6, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

15. The quantum cascade laser according to claim 7, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

16. The quantum cascade laser according to claim 8, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

17. The quantum cascade laser according to claim 9, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

18. The quantum cascade laser according to claim 10, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

19. The quantum cascade laser according to claim 11, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

20. The quantum cascade laser according to claim 12, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

21. The quantum cascade laser according to claim 13, wherein, The separation layer includes separation barrier layers that serve as barrier layers, disposed on both sides of the separation quantum well layer in the stacking direction of the unit stack. The thickness of the separation barrier layer is smaller than the thickness of the first barrier layer adjacent to the separation barrier layer in the light-emitting layer and the thickness of the second barrier layer adjacent to the separation barrier layer in the injection layer.

Images