Optical device
A photonic crystal with a nanowire lattice and optical confinement section supports high optical confinement of TM-polarized light, addressing the confinement challenge in AlGaN-based devices and enabling efficient TM-polarized laser oscillation and related applications.
Patent Information
- Application Number
- PCT/JP2024/005518
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-02-16
- Publication Date
- 2025-08-21
AI Technical Summary
Conventional photonic crystals struggle to confine TM-polarized light, which is perpendicular to the electric field, making it difficult to form a resonator structure with high optical confinement and a high Q factor for TM-polarized light in AlGaN-based optical devices.
A photonic crystal structure composed of nanowires arranged in a lattice pattern on a substrate, with a photonic band gap for TM-polarized light, and an optical confinement section without nanowires, incorporating a columnar section with an active layer made of a compound semiconductor, allowing for high optical confinement of TM-polarized light.
The structure enables effective confinement and laser oscillation of TM-polarized light, achieving a Q factor of 1000 or more, facilitating applications such as filters, modulators, and second-harmonic generation devices.
Smart Images

Figure JP2024005518_21082025_PF_FP_ABST
Abstract
Description
Optical Devices
[0001] The present invention relates to optical devices.
[0002] High-performance optical devices using photonic crystal structures have been reported (Non-Patent Documents 1, 2, and 3). For example, light-emitting diodes (LEDs) and surface-emitting lasers using photonic crystal structures can increase the light extraction efficiency from their surfaces. Furthermore, by using photonic crystal structures, optical devices can be scaled down compared to existing optical devices, achieving high efficiency and low power consumption. In particular, in the optical communication wavelength band, not only small transmitting and receiving devices but also single-photon and squeezed light sources for quantum information communication applications have been widely researched.
[0003] The photonic crystals used in these optical devices have columnar holes in their bases that are periodically arranged at intervals equal to or less than the wavelength of the target light. For example, in the case of Si photonic crystals, high optical confinement is achieved by processing an SOI substrate to form periodically arranged holes in the surface silicon layer.
[0004] On the other hand, in the case of light emission involving heavy holes (HH) in the valence band of a semiconductor in a quantum well structure, the electric field is horizontally oriented, resulting in TE polarization, while in the case of light holes (LH) or crystal field splitting holes (CH), the TM polarization becomes stronger. In particular, TM emission has been reported in light-emitting devices using AlGaN with a high Al composition. Wurtzite Al x Ga 1-x N is a direct transition type up to the Al composition x=1.
[0005] To date, research into AlGaN-based optical devices has focused on wurtzite-type films grown in the c-axis direction. Polarization is caused by optical transitions between the conduction band and the valence band. AlN has a smaller c / a ratio (c is the lattice constant in the c-axis direction, and a is the lattice constant in the a-axis direction) than GaN, and the relative coordinates of anions and cations are also different. As a result, the crystal field separation energy is positive in GaN but negative in AlN.
[0006] In the wurtzite structure, the upper edge of the valence band has three band edges due to the crystal field splitting and spin-orbit coupling. 6 9v (HH), Γ 6 7v (LH), Γ 1 7v The valence band is composed of the order of Γ (CH) and Γ (CH). 1 7v (CH), Γ 6 9v (HH), Γ 6 7v The order is (LH). The wave function of the HH band has the characteristic of |X±iY>, so it is TE polarized. The wave functions of the LH and CH bands have the characteristic of |z>, so it is TM polarized.
[0007] In AlGaN grown along the c-axis, the probability of spontaneous emission of TE-polarized light decreases and the TM-polarized component increases as the Al composition increases. In unstrained AlGaN, TM-polarized light becomes dominant at an Al composition of approximately 0.2. Furthermore, in compressively strained AlGaN grown coherently on an AlN substrate, TM-polarized light becomes dominant at an Al composition of approximately 0.65.
[0008] Deep-ultraviolet light-emitting devices using AlGaN-based nitride semiconductors are expected to play an important role in a wide range of fields in the future, including sterilization, virus inactivation, material modification, processing, sensing, communications, lithography, etc. Because AlGaN-based materials have a small refractive index, a photonic crystal structure is required to increase light confinement.
[0009] RK Gangwar et al., "Recent Progress in Photonic Crystal Devices and Their Applications: A Review", Photonics, vol. 10, no. 11, 1199, 2023.CH Lin et al., "A GaN photonic crystal membrane laser", Nanotechnology, vol. 22, no. 2, 025201, 2011.S. Matsuo et al., "Electrically-pumped photonic crystal lasers for optical communications", 2012 38th European Conference and Exhibition on Optical Communication, DOI: 10.1364 / ECEOC.2012.Th.1.E.2, 2012.
[0010] However, with the above-mentioned conventional photonic crystals, it is difficult to confine TM-polarized light, in which the electric field is perpendicular. With the above-mentioned conventional photonic crystals, it is not possible to form a photonic band gap for TM-polarized light, making it difficult to form a resonator structure with high optical confinement for TM-polarized light, i.e., a high Q factor. Until now, there has been no photonic crystal structure capable of high optical confinement for an AlGaN active layer, which emits significant TM-polarized light. Thus, the prior art has had the problem of being unable to confine TM-polarized light in photonic crystals.
[0011] The present invention has been made to solve the above problems, and has as its object to make it possible to confine TM polarized light in a photonic crystal.
[0012] The optical device of the present invention comprises a photonic crystal composed of nanowires arranged in a lattice pattern at multiple locations on a substrate and formed perpendicular to the surface of the substrate, the photonic crystal having a lattice constant such that a photonic band gap is formed for TM polarized light; an optical confinement section composed of an area where no nanowires exist; and a columnar section disposed in the optical confinement section and formed perpendicular to the surface of the substrate, the columnar section comprising an active layer made of a compound semiconductor.
[0013] As described above, according to the present invention, a photonic crystal is constructed from nanowires arranged in a lattice pattern at a plurality of locations on a substrate, and therefore TM polarized light can be confined in the photonic crystal.
[0014] FIG. 1 is a perspective view showing the configuration of an optical device according to an embodiment of the present invention. FIG. 2A is a photonic band diagram for TM-polarized light of a conventional photonic crystal using columnar holes as lattice elements. FIG. 2B is a photonic band diagram for TE-polarized light of a conventional photonic crystal using columnar holes as lattice elements. FIG. 3A is a photonic band diagram for TM-polarized light of a photonic crystal according to an embodiment using a nanowire array. FIG. 3B is a photonic band diagram for TE-polarized light of a photonic crystal according to an embodiment using a nanowire array. FIG. 4A is a diagram showing the electric field intensity distribution of a photonic crystal according to an embodiment obtained by a three-dimensional simulation using the finite element method. FIG. 4B is a diagram showing the electric field intensity distribution of a photonic crystal according to an embodiment obtained by a three-dimensional simulation using the finite element method. FIG. 4C is a diagram showing the electric field intensity distribution of a photonic crystal according to an embodiment obtained by a three-dimensional simulation using the finite element method. FIG. 5 is a characteristic diagram showing the changes in the wavelength and Q value of an eigenmode when the Al composition of active layer 142 is changed. FIG. 6 is a photograph showing an electron microscope image of actually fabricated nanowires and columns. FIG. 7 is a photograph showing the light emission state of an optical device that was actually fabricated.
[0015] An optical device according to an embodiment of the present invention will now be described with reference to Fig. 1. This optical device comprises a photonic crystal 110 made of nanowires 102 arranged at multiple locations, and a columnar portion 104 disposed in an optical confinement portion 103 of the photonic crystal 110.
[0016] The nanowires 102 are arranged in a lattice pattern at multiple locations on the substrate 101. This arrangement can be a triangular lattice pattern. The nanowires 102 are formed perpendicular to the surface of the substrate 101. The nanowires 102 are minute columnar structures. The photonic crystal 110 has a lattice constant that forms a photonic band gap for TM polarized light. The lattice constant is the spacing between the nanowires 102. The photonic crystal 110 uses the nanowires 102 as lattice elements that have a refractive index different from that of the surroundings.
[0017] According to the embodiment, the photonic crystal 110 is configured by a nanowire array of nanowires 102 arranged at a plurality of locations. Therefore, by appropriately setting the lattice constant, the photonic crystal 110 can be made to have a photonic band gap for TM polarized light.
[0018] The optical confinement section 103 is a region where no nanowires 102 exist. The columnar section 104 is formed perpendicular to the surface of the substrate 101 and includes an active layer 142 made of a compound semiconductor, as shown in FIG. 1B. The active layer 142 can be a bulk compound semiconductor layer or can have a multiple quantum well structure. Note that in FIG. 1B, the nanowires 102 are not shown, and only the columnar section 104 is shown on the substrate 101. Note that an insulating layer made of silicon oxide (SiO2) or the like can be filled between adjacent nanowires 102 and between the columnar section 104 and the nanowire 102 adjacent to the columnar section 104.
[0019] The columnar section 104 may include a first semiconductor 141 formed below the active layer 142 and a second semiconductor layer 143 formed above the active layer 142. The first semiconductor 141 is formed below and in contact with the active layer 142, and the second semiconductor layer 143 is formed above and in contact with the active layer 142. The first semiconductor layer 141 may be made of a first conductivity type compound semiconductor having a refractive index smaller than that of the active layer 142. The second semiconductor layer 143 may be made of a second conductivity type compound semiconductor having a refractive index smaller than that of the active layer 142. The first semiconductor 141 and the second semiconductor layer 143 function as cladding for the active layer 142, which may serve as a core.
[0020] Furthermore, a contact layer 144 made of a second conductivity type compound semiconductor with a higher impurity concentration is formed on the second semiconductor layer 143, and an electrode 145 is formed on the contact layer 144. For example, the substrate 101 can be made of a first conductivity type compound semiconductor. The contact layer 144, the electrode 145, the substrate 101, and an electrode formed in a region of the substrate 101 (not shown) can form a current injection structure that injects current between the first semiconductor layer 141 and the second semiconductor layer 143. For example, the first conductivity type can be n-type, and the second conductivity type can be p-type. The contact layer 144 may be omitted.
[0021] The first semiconductor layer 141, the active layer 142, and the second semiconductor layer 143 form a so-called pin structure, and the active layer 142 can emit light by injecting a current into the active layer 142 via the first semiconductor layer 141 and the second semiconductor layer 143. For example, by forming the above-mentioned quantum well structure using AlGaN with a high Al composition, the active layer 142 can emit TM-polarized light. According to the embodiment, a photonic band gap is formed in the photonic crystal 110 for TM-polarized light, and therefore, light emitted from the active layer 142 is confined in the light confinement section 103.
[0022] Figures 2A and 2B show photonic band diagrams of a conventional photonic crystal using columnar holes as lattice elements, while Figures 3A and 3B show photonic band diagrams of a photonic crystal according to an embodiment using a nanowire array. Figures 2A and 3A show photonic band diagrams for TM-polarized light, and Figures 2B and 3B show photonic band diagrams for TE-polarized light. In the wave vectors on the horizontal axis, Γ represents the center of the hexagonal region (Brillouin zone) in the reciprocal lattice of a triangular lattice, K represents the vertex of the hexagon in the reciprocal lattice of the triangular lattice, and M represents the midpoint of the side of the hexagon in the reciprocal lattice of the triangular lattice. Here, the refractive index difference between the lattice element and its surroundings is set to 0.77, the period (lattice constant) of the triangular lattice is set to 84 nm, and the columnar holes and nanowires are cylindrical with a radius of 25.2 nm. Two-dimensional calculations were performed using the finite element method.
[0023] In the calculations, a Floquet periodic boundary condition was used around the unit cell. As shown in Figures 2A and 2B, a photonic crystal with columnar holes has no band gap for TM-polarized light when the electric field is out of plane. On the other hand, as shown in Figures 3A and 3B, a photonic crystal with a nanowire array has a band gap for TM-polarized light. Thus, a photonic crystal with a nanowire array exhibits high reflectivity for TM-polarized light within the band gap in both two-dimensional directions. Therefore, by introducing a defect structure where no nanowires are formed in the center of a photonic crystal, a resonator with a high Q factor can be formed, with this region serving as an optical confinement region.
[0024] Next, the electric field intensity distributions obtained by three-dimensional simulation using the finite element method are shown in Figures 4A, 4B, and 4C. For an optical device designed to obtain a high Q value near a wavelength of 280 nm, the electric field intensity distributions are shown for an eigenmode obtained by three-dimensional simulation using the finite element method, with a Q value of 1153 (λ = 279.66 nm). Figure 4A shows the electric field intensity distribution in the xy plane passing through the center of the active layer 142 of the optical device, Figure 4B shows the electric field intensity distribution in the yz plane of the optical device, and Figure 4C shows the electric field intensity distribution in the xz plane of the optical device. Note that the plane of the substrate 101 is parallel to the xy plane.
[0025] In the simulation, first, the substrate 101 is made of Al 0.64 Ga 0.36 The photonic crystal 110 was constructed from AlGaN, and nanowires 102 were arranged on top of it in a regular triangular lattice with a spacing (lattice constant) of 96 nm so that light with a wavelength of 280 nm was within the photonic band gap. The nanowires 102 were also constructed from AlGaN and were cylindrical with a radius of 24 nm. A light confinement section 103 without nanowires 102 was provided in a long, hexagonal region in the center of the photonic crystal 110 in a planar view.
[0026] Furthermore, the optical confinement section 103 thus configured was provided with a rectangular columnar section 104 having a width of 208 nm and a length of 800 nm in plan view, with triangles added above and below the rectangular columnar section to match the lattice. The columnar section 104 was made of AlGaN.
[0027] The height of the nanowire 102 and the columnar portion 104 was set to 1000 nm. The active layer 142 was made of bulk Al. 0.5 Ga 0.5 The first semiconductor layer 141 and the second semiconductor layer 143 are made of Al. 0.64 Ga 0.36 The nanowire 102 and the columnar portion 104 were assumed to be surrounded by SiO2. This simulation was performed assuming that the nanowire 102 also had the same layer structure as the columnar portion 104.
[0028] In the simulation, to reduce the load on the simulator, the optical device configured as described above was divided into four parts at the center of the columnar portion 104 in a plan view, and one quarter of this area was extracted. Calculations were performed using perfect magnetic conductor (PMC) boundary conditions on the surface of the extracted part. Therefore, the center of the columnar portion 104 was exposed on one of the adjacent side surfaces of the rectangular parallelepiped used for the simulation. An x-y-z coordinate system was set, with the side between the two side surfaces where the center of the columnar portion 104 is exposed as the z-axis.
[0029] As shown in Figure 4A, it can be seen that the electric field is strong from the active layer 142 portion of the columnar section 104. Furthermore, from the distribution in the yz plane at the position x = 0 shown in Figure 4B and the distribution in the xz plane at the position y = 0 in Figure 4C, it can be seen that light is concentrated in the active layer 142, which has a high refractive index. The light in this mode is a TM wave, and a sufficient resonator with a Q value of 1000 or more is obtained, making it possible to generate TM-polarized laser oscillation.
[0030] Here, the calculations are performed with the Al composition of the active layer 142 set to 0.5, but the design can be performed in the same way for structures with an even higher Al composition of 0.6 or more. Figure 5 shows how the wavelength and Q value of a mode similar to the above-mentioned configuration change when the Al composition of the active layer 142 is changed.
[0031] When the Al composition of the active layer 142 is 0.6, the Al composition of the substrate 101, the first semiconductor layer 141, and the second semiconductor layer 143 is 0.72, and the Al composition of the contact layer 144 is 0.048.
[0032] When the Al composition of the active layer 142 is 0.7, the Al composition of the substrate 101, the first semiconductor layer 141, and the second semiconductor layer 143 is 0.79, and the Al composition of the contact layer 144 is 0.15.
[0033] When the Al composition of the active layer 142 is 0.8, the Al composition of the substrate 101, the first semiconductor 141, and the second semiconductor layer 143 is 0.88, and the Al composition of the contact layer 144 is 0.298.
[0034] When the Al composition of the active layer 142 is 0.9, the Al composition of the substrate 101, the first semiconductor layer 141, and the second semiconductor layer 143 is 1.0, and the Al composition of the contact layer 144 is 0.465.
[0035] As the Al composition of the active layer 142 increases, the emission wavelength becomes shorter. Therefore, the emission wavelength was set to 272 nm when the Al composition of the active layer 142 was 0.6, 256 nm when it was 0.7, 241 nm when it was 0.8, and 228 nm when it was 0.9. The eigenvalues near these wavelengths were calculated. In actual optical devices, the emission wavelength can be controlled by the composition and thickness of the quantum wells in the active layer 142, which has a multiple quantum well structure. As the Al composition increases, the refractive index decreases, and the difference in refractive index with the surrounding SiO2 decreases, narrowing the stop band. Therefore, as shown in Figure 5, the Q factor also tends to decrease. Nevertheless, since the Q factor is 450 or higher, laser oscillation is possible as a light-emitting device, for example. An active layer 142 with a high Al composition can confine light originating from the CH band, enabling TM-polarized laser oscillation.
[0036] In addition, by integrating an optical waveguide on the same substrate and injecting TM light into this optical device from the outside, it is possible to apply it to filters that transmit only the resonant mode, modulators that change the resonant frequency by changing the refractive index using current or voltage, and second-harmonic generation that utilizes the relatively large optical SHG coefficient in the C-axis direction.
[0037] Next, we will explain the optical device that was actually fabricated. First, a 2 μm-thick buffer layer made of GaN, an n-type AlN layer, and a sapphire substrate whose main surface is the (0001) plane were formed on the substrate. x Ga 1-x A first semiconductor layer of 3.4 μm thick made of N (x=0.01-0.04) is formed. 0.1 Ga 0.9 A 9-layer multi-quantum well layer is formed, consisting of a 4.7 nm thick well layer made of N and a 8.6 nm thick barrier layer made of GaN. x Ga 1-x A second semiconductor layer made of N (x=0.1-0.4) and having a thickness of 100 nm is formed.
[0038] Next, a 150 nm thick SiO2 insulating layer is formed on the second semiconductor layer, and then a 10 nm thick Ni metal pattern is formed using electron beam lithography. This metal pattern serves as a mask pattern for forming the nanowires and columnar portions. For example, after depositing a nickel layer, the metal pattern can be formed using a lift-off method.
[0039] After forming the metal pattern, the insulating layer is etched by reactive ion etching (RIE) using a fluorine compound, and then etched down to a thickness of about 450 nm down to the first semiconductor layer by inductively coupled plasma (ICP) etching using a chlorine compound. Using this metal pattern as a mask, the second semiconductor layer, the multiple quantum well layer, and the first semiconductor layer are etched, forming nanowires at multiple locations on the buffer layer and forming columnar portions.
[0040] The nanowires and pillars formed by this etching have a tapered shape with diameters that become thinner as they go upward. Therefore, by immersing them in a 22% aqueous solution of TMAH (tetramethyl ammonium hydroxide) at 70°C for 2 minutes, the sidewalls of the nanowires and pillars are made perpendicular to the plane of the substrate.
[0041] Electron microscope images of the nanowires and pillars that were actually fabricated are shown in Figure 6. A good pattern was formed, and it was clear that vertical nanowires had been formed even when observed from an oblique direction.
[0042] Next, the metal pattern is removed by etching with an aqueous solution of H2SO4 / H2O2 / H2O=30 / 10 / 10 for 2 minutes, and then the insulating layer is removed by etching with an aqueous solution of HF / H2O=1 / 10 for 1 minute 30 seconds.
[0043] Next, an alumina layer was formed to a thickness of 5 nm by atomic layer deposition (ALD), and SiO2 was deposited to a thickness of approximately 290 nm by sputtering to fill the gaps between the nanowires. Next, the columnar portion was exposed using known photolithography techniques, and while the periphery was protected with a resist layer, etching was performed by RIE down to the top of the nanowires. Further, Pd / Pt / Au = 30 nm / 30 nm / 50 nm was deposited and lifted off to form a p-electrode that makes an ohmic contact with the second semiconductor layer.
[0044] Next, to form an n-type electrode on the exposed portion of the first semiconductor layer, patterning was performed by photolithography, Ti / Pt / Au = 10 nm / 30 nm / 250 nm was evaporated and lifted off to form an electrode portion for ohmic contact with the first semiconductor layer, and annealing was performed for 5 minutes at 500° C. Finally, wiring electrodes connected to each electrode were formed by photolithography and evaporation of Ti / Au = 10 nm / 250 nm and lift off.
[0045] When current was injected between the n-electrode and p-electrode via the wiring electrodes, light emission from the device was confirmed, as shown in Figure 7. It is believed that light from the active layer is emitted by reflection from the nanowire array. Figure 7(a) is a micrograph, and Figure 7(b) is an EL image.
[0046] In the above description, both the nanowire and the columnar portion are fabricated by etching from a stacked structure of a pin diode structure, but this is not limited to this. For example, first, only the columnar portion is fabricated as a diode structure having a light-emitting layer. Then, the nanowire portion can be fabricated by regrowth or sputter deposition of insulating, wide-bandgap AlN, BAlN, AlNO, or SiON.
[0047] In the above example, AlGaN was examined, but TM light laser oscillation is also possible with AlGaAs or InGaAsP-based materials by matching the resonant frequency of the resonator structure to light emission involving the LH band, the SO (spin-orbit spin split-off) band in the case of zinc blende, or the CH band in the case of wurtzite.Furthermore, by introducing TM light from an external source, it is possible to operate the device as a filter, a modulator, or a second harmonic generator utilizing a relatively large optical SHG coefficient in the C-axis direction.
[0048] As described above, according to the present invention, a photonic crystal is constructed from nanowires arranged in a lattice pattern at multiple locations on a substrate, making it possible to confine TM-polarized light in the photonic crystal. The present invention makes it possible to provide, for example, microscopic coherent optical devices in the deep ultraviolet region, which is expected to lead to further technological developments in a wide range of fields, including chemistry, medicine, lithography, analysis, processing, communications, and sensors.
[0049] It should be noted that the present invention is not limited to the embodiments described above, and it is clear that many modifications and combinations can be made by a person having ordinary knowledge in the art within the technical concept of the present invention.
[0050] 101...substrate, 102...nanowire, 103...light confinement portion, 104...columnar portion, 110...photonic crystal, 141...first semiconductor, 142...active layer, 143...second semiconductor layer, 144...contact layer, 145...electrode.
Claims
1. An optical device comprising: a photonic crystal composed of nanowires arranged in a lattice pattern at multiple locations on a substrate and formed perpendicular to the surface of the substrate, the photonic crystal having a lattice constant that forms a photonic band gap for TM polarized light; an optical confinement section composed of an area where the nanowires are not present; and a columnar section disposed in the optical confinement section, formed perpendicular to the surface of the substrate, and having an active layer made of a compound semiconductor.
2. An optical device according to claim 1, wherein the columnar portion comprises: a first semiconductor layer formed below the active layer and made of a compound semiconductor of a first conductivity type having a refractive index smaller than that of the active layer; and a second semiconductor layer formed above the active layer and made of a compound semiconductor of a second conductivity type having a refractive index smaller than that of the active layer.
3. An optical device according to claim 2, further comprising a current injection structure for injecting a current between said first semiconductor layer and said second semiconductor layer.
4. An optical device according to any one of claims 1 to 3, wherein the nanowires are arranged in a triangular lattice pattern at a plurality of locations on the substrate.
Citation Information
Patent Citations
Semiconductor laser device
JP2001308452A
Semiconductor light-emitting element and production method
WO2015008776A1