High-power single-mode laser diodes around 795 nm are widely used in applications such as Rb atomic clocks and nuclear magnetic resonance imaging. We simulate a high-power single-mode semiconductor laser around 795 nm based on a supersymmetric structure. In the lateral direction, the mode stability characteristics are investigated by varying the three waveguides widths and the distances between the middle main waveguide and the two sub-waveguides. Since the left and right waveguides have different widths, the optimal distance from them to the main waveguide is also different. In order to ensure the single-mode operating of the laser, there is a pair of optimized distances from the left and right waveguides to the main waveguide. The distances from the left and right waveguides to the main waveguide are 1 μm and 1.2 μm, respectively, when the widths of the left waveguide, right waveguide and main waveguide are set as 2.3 μm, 3.5 μm and 6 μm, respectively. In the longitudinal direction, a laterally-coupled grating structure is used to achieve longitudinal mode selection. Such lasers are expected to be the next generation of high-power, narrow-linewidth, singlemode laser diodes.
Microdisks and micro-rings are commonly used micro-optical devices that greatly enhance the interaction between light and matter within a cavity due to their high-quality factor and small mode volume, making them widely used in microcavity optical sensing. By introducing parity-time (PT) symmetric structures into the microcavity, the coupling efficiency of the optical field inside the cavity can be improved, which is conducive to obtaining higher sensing sensitivity. We theoretically verify the feasibility of using a PT-symmetric micro-ring coupled microdisk composite cavity as an active sensor based on the characteristics of exceptional point (EP) enhanced sensing in PT-symmetric systems. Gain is introduced to the microdisk cavity by injecting current until the system undergoes PT symmetry breaking, i.e., when the sensor is at the EP, the transmission of light will exhibit a nonlinear enhancement effect due to the degeneracy of eigenvalues and corresponding eigenvectors of the system, making the signal more sensitive to changes in the sensing medium. The results obtained through the finite difference time domain method show that the intensity sensitivity of the PT-symmetric microcavity at the EP is improved by about 11.8 times compared with the conventional microcavity when the working wavelength is in the range of communication band and different concentrations of the same gas are injected into the air, which is expected to provide reference and insight for the further development of microcavities for refractive index sensing.
We have designed a fiber coupling method based on the spatial beam combining of the Photonic Crystal (PC) Laser Diode (LD). The PC LD with a small fast-axis divergence angle makes it possible to reduce the requirements for the numerical aperture and the processing precision of the optical elements, increase the alignment tolerance of components, reduce the difficulty of shaping, and improve the product yield. In the module, there is no need for the collimation of the fast-axis and slow-axis beams, which can be simultaneously focused into the optical fiber through an aspheric cylindrical lens. The simulated results, obtained by the ray tracing method, have shown a coupling efficiency of around 91.4% when the PC LD is coupled into a fiber with a core diameter of 105 μm and the numerical aperture of 0.22. Then, we have performed the experiments, and the coupling efficiency of 71.5% has been achieved. By analyzing the deviation of the simulated and experimental coupling efficiency, we have proposed several solutions. Finally, according to the strategy of this beam shaping, we also list several promising arrangements, which further prove that the beam shaping method possesses broad application prospects.
The application of high-performance VCSELs is extending from consumer electronics to automotive applications. Wet oxidation is an important technology in the fabrication of VCSELs. In this paper, we studied the wet oxidation process and mechanism in order to accurately control the oxidation aperture and improve the power and the conversion efficiency. Current density distributions of VCSELs with different oxide apertures are simulated based on COMSOL Multiphysics. In the experiment, the output power, conversion efficiency and threshold current of single junction and five-junction 940 nm VCSELs varying with oxide apertures are measured. Five-junction VCSELs exhibit maximum power conversion efficiencies are more than 60% and slope efficiency are more than 5.28W/A with oxide aperture from 9 to 15 μm under room temperature pulse condition (50 µs pulse width, 0.5% duty cycle). In addition, 385-element five-junction VCSEL array exhibits maximum power conversion efficiency of 53.45%. The five-junction VCSELs can be used as the basic laser source for the automotive applications.
Photonic crystal laser diodes are characterized by low divergence angle and high brightness, but thermal effects have become a major obstacle to further improvement of output power and efficiency. The thermal characteristics of high- power photonic crystal laser diodes are of great importance to improve the output power and increase the lifetime. In this paper, the physical heat dissipation model of a single photonic crystal laser diode with CS-mount package is established. Steady-state thermal characteristics simulations are performed using the Finite Element Method (FEM) and the influences of different parameters, such as solder, transition heat sink and heat sink on the thermal characteristics are analyzed. The simulation results show that the thickness and thermal conductivity of the heat sink materials are the main factors impacting the heat dissipation of the laser. The thermal resistance of the laser can be reduced effectively by using heat sink materials of high thermal conductivity. On the premise of ensuring wettability and reliability, the thickness of the solder layer should be decreased. A photonic crystal laser diode with a cavity length of 4 mm and a stripe width of 350µm based on an optimized heat dissipation structure is designed and fabricated. The CW output power of 41.9W, the vertical divergence angle of 18.48° and the thermal resistance of 1.54 K/W are obtained under the injection current of 50A at 20 ℃.
High power and high beam quality laser sources are required in numerous applications such as nonlinear frequency conversion, optical pumping of solid-state and fiber lasers, material processing, and others. Here, we theoretically study and demonstrate a tapered laser diode with integrated metalens, which can greatly reduce the lateral far-field divergence of the device. A 980 nm tapered laser diode adopted in this design consists of a power-amplified tapered section and a narrow-ridged section, in which the latter restricts the lateral mode number, and the former is utilized to amplify the output power. The wavefront is carefully reshaped by preparing a one-dimensional (1D) trench metalens near the front facet of the tapered cavity. By precisely designing the length and width of the low refractive index elements at different positions, the approximate spherical wave formed by diffraction in the tapered cavity is transformed into an output plane wave while ensuring high transmittance (>90%), which reduces the divergence of the lateral far-field. The simulation results show that the lateral far-field divergence of the fundamental mode decreases from 3.2° to 2.0° (FWHM) after the integration of the metalens with a 500 μm length of tapered cavity.
Bound states in the continuum (BICs) remain localized even though they coexist with a continuous spectrum of radiating waves that can carry energy away. These modes can be almost perfectly localized in the structure, making lasers working at BIC or quasi-BIC have an ultrahigh quality factor (Q) and hence low threshold. Lowcontrast gratings (LCGs) have better mode selectivity than high-contrast gratings and promise higher single-mode output power for LCG-based vertical-cavity surface-emitting lasers. A quasi-BIC (i.e. supercavity mode) with a Q factor of 9.2 × 105 is obtained in the LCG, and a simplified three-layer slot laser with a Q factor of 9.66 × 106 is constructed. Further, a law of using the period of a grating to control resonant wavelength and using etched depth and width to control Q factor can be used for designing a high-Q structure at a specified wavelength. The calculated Q factor is optimized systematically by changing various parameters, and the highest Q factor obtained reaches 2.81 × 107 . The results of all these analyses are instructive to the design of grating-based low threshold electrically injected surface-emitting lasers or other high-Q devices.
High spatial coherence can maintain the beam stability of the laser after a long distance. However, it limits the applications in laser display including projection and imaging system, because the high coherence of laser diodes cause the artifacts such as speckle. In this work, we design a novel 6xx nm chaotic cavity laser diode, which consists of a Dshaped section used to achieve a large number of independent spatial modes thus reduce coherence and a stripe area to improve power. The radius of the D-shaped cavity is 500 μm and the length of stripe is 1000 μm. The red laser based on GaAs substrate is fabricated by standard photolithography and reactive ion etching process. To obtain an enough optical confinement by effective refractive index step, the etching depth exceeds the active region. The high-power chaotic cavity low-spatial coherence electrically pumped semiconductor laser is first realized with the wavelength around 630 nm. The spectrum width of 15 nm at full width at half maximum (FWHM) and output power of 300 mW is obtained under pulse operation. The speckle contrast is measured to be 5%, showing great potential of reducing speckle from the source directly for laser display.
The semiconductor laser diode has the advantage of low cost, high efficiency, and compactness, but the beam divergence is too large to directly use. The phase-locked laser array is an efficient way to control the lateral lasing mode, which can help to achieve narrow farfield.. Though the lasing mode of phase-locked laser array can be an in-phase mode via Ywaveguide, integrated with phase shifter and external cavity, it still has a large side lobe in the farfiled. We demonstrated an on-chip phase and amplitude manipulation method to suppress the side-lobe in the farfield. The intensity of the sidelobes decrease from 0.307 to 0.109 and the integral energy of the main lobe increase from 52.5% to 60.5%
The hybridization of active and passive platforms are always the hot area of material science and experimental physics, which also attracts our attention. We demonstrate a device composes silicon photonic crystal structure and perovskite. Single mode lasing is observed at 577nm, with full width half maximum (FWHM) of 0.3nm. While a thin film of allinoganic lead-halide perovskite is spin-coated atop, under the same pump situation, there exists a sharp peak at 565nm, with FWHM of 0.4nm. At the same time, the single peak at 470nm gradually shifts towards to longer wavelength and then splits into two peaks in photoluminescence (PL) spectra. Photonic band structure is calculated by the plane-wave expansion method. We choose the bandedge modes at Γ point for laser action from the band structure. Then the device is simulated as a whole and optimized by finite element method. Our works demonstrate that the visible light can resonant in silicon material, which indicates that active optical material such as perovskite can be hybridized with integrated circuits in future.
High-efficiency, high-power and high-brightness, fiber-coupled modules based on semiconductor laser diodes have been important sources in many fields, such as fiber laser pumping, material processing and defense applications. The coupling efficiency of fiber-coupled module has been limited due to the large vertical divergent angle of conventional semiconductor laser diodes. We present a high coupling efficiency module by using photonic-band-crystal (PBC) laser diodes with narrow vertical divergent angle. Fourteen PBC single-emitter laser diodes are combined into a fiber with core diameter of 200 μm and numerical aperture (NA) of 0.22. A high and stability coupling efficiency of 88% and peak otuput power of 47W with the injection current of 5 A are obtained. A comparison with the coupling efficiency of conventional laser diodes module is also presented. And there is a 5% increase of fiber-coupled efficiency based on PBC laser diodes module compared to conventional semiconductor laser diodes module.
Ridge-waveguide (RW) lasers based on photonic crystal structure were fabricated and measured. We investigated the effect of residual layer thickness (corresponding to etching depth) and ridge width on electro-optical characteristics of RW lasers. For deep-etching RW lasers, although lateral beam quality factor M2 is better than that of shallow-etching RW lasers, the other characteristics such as output power are much less than that of shallow-etching RW lasers. The calculating results indicate that RW lasers with ridge width w ≥ 8 μm will operate in mixing mode. The experimentally results of various ridge width RW lasers show that RW laser with 7 μm ridge operated in single mode over the whole measurement range and RW laser with 8 μm ridge change from single-mode operation to mixing-mode operation with the increasing of driving current. The device with 7-μm-wide ridge and 3-mm-long cavity obtain 2 W single-transverse-mode optical power and 59% maximum power conversion efficiency. The lateral beam quality factors M2 values are less than 1.7 over the whole measuring range.
Quantum walks (QWs) have been proposed as powerful tools in a broad range of fields. Discrete-time QW (DTQW) is an extension of the classical random walk, where the walker goes back and forth along a line and the direction at each step depends on the result of a fair coin flip. Continuous-time QW (CTQW) can be shown as a limit of DTQW. Many objects, such as atoms, trapped ions etc., have been used to simulate QW. But photon, with the wave-particle duality, is easy to generate and can be easily manipulated in many platforms, such as space light circuits and integrated optical platforms. Silicon-on-insulator (SOI) integrated optics have been widely used, among these self-collimation photonic crystal shows a great potential. In this paper we propose the simulation of CTQWs and DTQWs with self-collimation photonic crystal chip fabricated on 830 nm thick top silicon SOI. Similarity between theory results and simulations are analyzed.
Light propagation in strip and slot waveguide arrays for sensing are proposed and analyzed with a new theory of quantum walk. The waveguide arrays are designed on silicon-on-insulator and can be fabricated with mature and cost-efficient complementary metal-oxide semiconductor technology. A new slot waveguide array modified by conventional strip waveguide array with electric field mainly confined in the cladding region is investigated. Quantum walks have an exact mapping to classical phenomena as verified by experiments using bright laser light, so that they are introduced in our work as theoretical foundation. We take the width of waveguide of 450 nm and the coupling distance of 200 nm for strip waveguide array, and 420 nm and 180 nm for slot waveguide array, but with a 100nm slot in the center of waveguide. At last the waveguide array covered by a thin layer of graphene is investigated, which brings higher sensing property as well as a much better biocompatibility. With the monochrome light injection the intensity distribution at the end of the arrays changes with the refractive index of the sensing area (cladding region) and it can be explained by quantum walks theory. The designed waveguide arrays can possess compact footprint and high refractive index resolution, reaching 1E-11 RIU theoretically.
Search on improved-glued-binary-trees is a representative example of quantum superiority, where exponential acceleration can be achieved using quantum walk with respect to any classical algorithms. Here we analyzed the evolution process of this quantum-walk-based algorithm. Several remarkable features of the process are revealed. Generation of the model by introducing tunable defect strength and double defects is also discussed and the effects of these generalization on evolution process, arrival probability and residual probability are discussed in details. Physical implementation with silicon ridge waveguide array is presented. The design of the array with FEM method are presented and light propagation simulation with FDTD method shows that this kind of structure is feasible for the task. Lastly, preliminary experimental demonstration with classical coherent light simulation are presented. Our results show that silicon photonic chips are suitable for such search problems and opening a route towards large-scale photonic quantum computation.
We report the first demonstration of blue shift of optical pumping photonic crystal (PhC) laser. A femtosecond laser was
used to pump the InGaAsP based two dimensional photonic crystal laser at room temperature. Linear dependence of the
resonance wavelength with respect to the pump power is observed: dλ/dP=-1.5×10-2 nm/μW . Blue shift of overall
1.1nm was obtained with the increase power of pump laser. These results are in agreement with theoretical expectation
while the carrier-induced index change is introduced into the PhC semiconductor laser. It shows a possibility that by
proper wafer design and careful optimization, we may obtain wavelength stable photonic crystal laser, which is
important in photonic integration.
We design photonic crystal (PC) array surface emitting lasers with large-area coherence. The structure has six-fold rotational symmetry. By finite-difference time-domain method, we investigate the far-field characteristics of the individual element and the array. We demonstrate theoretically that the coherent PC array has lower far-field divergence angles and higher power compared to those of individual elements. Our PC array exhibits strong leaky coupling which has high mode stability and high intermodal discrimination. Thus, the coherent PC array shows great potential for high power low divergence in-phase surface laser emitting.
The Bragg diffraction condition of surface-emitting lasing action is analyzed and Γ2-1 mode is chosen for lasing. Two
types of lateral cavity photonic crystal surface emitting lasers (LC-PCSELs) based on the PhC band edge mode lateral
resonance and vertical emission to achieve electrically driven surface emitting laser without distributed Bragg reflectors
in the long wavelength optical communication band are designed and fabricated. Deep etching techniques, which rely on
the active layer being or not etched through, are adopted to realize the LC-PCSELs on the commercial AlGaInAs/InP
multi-quantum-well (MQW) epitaxial wafer. 1553.8 nm with ultralow threshold of 667 A/cm2 and 1575 nm with large
power of 1.8 mW surface emitting lasing actions are observed at room temperature, providing potential values for mass
production with low cost of electrically driven PCSELs.
A Low-loss Fabry-Perot interferometer (FPI) constructed in a two-dimensional photonic crystal (2D PhC) is proposed
and investigated. The 2D PhC consists of a square-lattice of cylindrical silicon rods in air. It has flat equal frequency
contours (EFCs) in the frequency range of 0.187-0.201c/a for TM modes. Two same line defects with spacing of d =
21√2 a, which is the physical length of the FP resonant cavity, are introduced in the PhC to form the FPI. The two line
defects have high reflectivity and low transmission. Their transmission is between 20.77% and 40.65% for the selfcollimated
lights with frequencies from 0.187c/a to 0.201c/a and thus they form the two partial reflectors. Lights
propagate in the FPI utilizing self-collimation effect. The transmission spectrum of the FPI has been investigated with
the finite-difference time-domain (FDTD) method. The calculation results show that even slight increase of d can cause
peaks shift left to lower frequencies. Through changing the configuration of the reflectors which results in transmission
between 19.97% and 38.77%, the varieties of the sharpness of peaks and the degree of extinction of the frequencies
between the peaks are obviously observed. Free spectral range (FSR) and peaks frequencies of its transmission decrease
when d increases. By raising the reflectivity of the reflectors, the full width at half maximum (FWHM) is decreased and
quality (Q) factor of peaks is increased.
A Folded Mach-Zehnder interferometer (FMZI) in a two-dimensional photonic crystal is proposed. The FMZI consists
of one splitter and several mirrors. Light propagates between them employing self-collimation effect. Its two interfering
branches have different path lengths. The two complementary transmission spectra at two FMZI output ports are both in
the shape of sinusoidal curves and have a uniform peak spacing in the frequency range from 0.255c/a to 0.270c/a. The
peak spacing becomes smaller when the length difference between the two branches is increased. As self-collimation
light beams can cross each other without coupling, this FMZI is much smaller than non-folded interference-type filters in
photonic crystals. This FMZI may work as a wavelength division demultiplexer in high-density photonic integrated
circuits.
The self-collimation frequencies (SCFs) in two-dimensional photonic crystals (2-D PhC) have been investigated
systematically by the plane-wave expansion method. In the wave-vector space, the square-lattice 2-D PhCs have some
square-shaped equifrequency contours (EFCs) both for TE modes and for TM modes. Narrow-beam lights with these
frequencies can propagate in the directions normal to the flat borders of the EFCs without any significant broadening,
which is known as self-collimation effect. We consider the 2-D PhCs consisting of a square lattice of air cylinders in a
dielectric material and the 2-D PhCs consisting of a square lattice of dielectric cylinders in air respectively. Calculation
results show how SCFs of TM and TE modes change with the radius of cylinders and the refractive index of the material.
These results can be applied to designing the PhC devices based on self-collimation effect.
A theoretical model of wavelength interleaver, which is based on an asymmetric Mach-Zehnder interferometer (AMZI) constructed in a two-dimensional photonic crystal (2D PhC), is proposed and numerically demonstrated. The 2D PhC consists of a square lattice of dielectric cylindrical rods in air. The AMZI includes two mirrors and two splitters. Light propagates between them employing self-collimation effect. The two interferometer branches have different path lengths. By using the finite-difference time-domain method, the calculation results show that the transmission spectra at two AMZI output ports are in the shape of sinusoidal curves and have a uniform peak spacing in the frequency range from 0.191c/a to 0.200c/a. When the path length of the longer branch is increased and the shorter one is fixed, the peaks shift to the lower frequencies and the peak spacing decreases nonlinearly. Consequently, the transmission can be designed to meet various application demands by changing the length difference between the two branches. For the dimensions of the wavelength interleaver are about tens of central wavelengths, it may be applied in future photonic integrated circuits.
A Fabry-Perot (FP) etalon constructed in a two-dimensional photonic crystal (2D PhC) utilizing self-collimation effect is
proposed and investigated. The 2D PhC consists of a square lattice of air holes in silicon. It has square-shaped equal
frequency contours (EFCs) in the frequency range of 0.275-0.295c/a for TE modes. The FP proposed consists of two
PhC reflectors and one cavity between them. Light propagates in the photonic crystal employing self-collimation effect.
The two reflectors have reflectivities of around 97.5% in the frequency range 0.275-0.295c/a. The FDTD calculation
results show that the transmission spectrum of the FP etalon has a uniform peak spacing between 0.275c/a and 0.295c/a.
The transmission spectrum shifts to the lower frequency as the refractive index of a fluid filling in the air holes in the FP
cavity is increased. Therefore this etalon can work as an optical sensor for a gas and a liquid. The fluids whose refractive
index vary within 1.0-1.5 can be sensed and detected. Its dimensions are only about tens of microns when the central
operating wavelength is equal to 1550nm. So it can be applied as a micro-scale sensor.
A Fabry-Perot interferometer (FPI) constructed in a two-dimensional photonic crystal (2D PhC) has been proposed and
demonstrated theoretically. The perfect 2D PhC consists of square-lattice cylindric air holes in silicon. Two same line
defects with spacing of d = 16a, which is the physical length of the FP resonant cavity, are introduced in the PhC to form
the FPI. The two line defects have high reflectivity and low transmission. Their transmission is between 4.81% and
11.1% for the self-collimated lights with frequencies from 0.275c/ato 0.295c/a and thus they form the two partial
reflectors. Lights propagate in the FPI employing self-collimation effect. The transmission spectrum of the FPI has been
investigated with the finite-difference time-domain method. The calculation results show that peaks have nearly equal
frequency spacing 0.0078c/a. Even slight increases of d can cause peaks shift left to lower frequencies. As a result, the
peak spacing decreases nonlinearly from 0.0142c/a to 0.0041c/a when dis increased from 9a to 30a. Through changing
the configuration of the reflectors which results in transmission between 4.18% and 7.73%, the varieties of the sharpness
of peaks and the degree of extinction of the frequencies between the peaks are obviously observed.
A Michelson interferometer (MI) constructed in a two-dimensional photonic crystal (2D PhC) utilizing self-collimation
effect is proposed and investigated theoretically. The 2D PhC consists of a square lattice of air holes in silicon. It has
square-shaped equal frequency contours (EFCs) in the frequency range of 0.26-0.275c/a for TE modes. The MI proposed
consists of two PhC mirrors and one defect-row splitter. Light propagates between them employing self-collimation
effect. The two interferometer branches have different path lengths L1 and L2. The FDTD calculation results show that
the transmission spectrum from 0.26c/a to 0.275c/a at the MI output port is comb-shaped. The transmission peaks have a
uniform spacing. Moreover, the peaks shift to the lower frequencies and the peak spacing decreases when the difference
between L1 and L2 is increased. For the operating wavelength around 1550nm, the dimensions of this MI are only tens of
microns. So this PhC Michelson interferometer may be applied in future photonic integrated circuits.
A theoretical model of wavelength division demultiplexer (WDD), which is based on an asymmetric Mach-Zehnder
interferometer (AMZI) constructed in a two-dimensional photonic crystal (2D PhC), is proposed and numerically
demonstrated. The 2D PhC consists of a square lattice of cylindric air holes in silicon. The AMZI includes two mirrors
and two splitters. Lights propagate between them employing self-collimation effect. The two interferometer branches
have different path lengths. By using the finite-difference time-domain method, the calculation results show that the
transmission spectras at two AMZI output ports are in the shape of sinusoidal curves and have a uniform peak spacing in
the frequency range from 0.26c/a to 0.27c/a. When the path length of the longer branch is increased and the shorter one
is fixed, the peaks shift to the lower frequencies and the peak spacing decreases nonlinearly. Consequently, the
transmission can be designed to meet various application demands by changing the length difference between the two
branches. For the dimensions of the WDD are about tens of operating wavelengths, this PhC WDD may be applied in
future photonic integrated circuits.
A theoretical model of a tunable Mach-Zehnder interferometer (TMZI) constructed in a 2D photonic crystal is proposed.
The 2D PhC consists of a square lattice of cylindric air holes in silicon. The TMZI includes two mirrors and two
splitters. Lights propagate between them employing self-collimation effect. The two interferometer branches have
different path lengths. Parts of the longer branch are infiltrated with a kind of liquid crystal (LC) whose ordinary and
extraordinary refractive indices are 1.522 and 1.706, respectively. The transmission spectra at two MZI output ports are
in the shape of sinusoidal curves and have a uniform peak spacing 0.0017c/a in the frequency range from 0.26c/a to
0.27c/a. When the effective refractive index neff of the liquid crystal is increased from 1.522 to 1.706, the peaks shift to
the lower frequencies over 0.0017c/a while the peak spacing is almost kept unchanged. So this TMZI can work as a
tunble power splitter or an optical switch. For the central operating wavelength around 1550nm, its dimensions are only
about tens of microns. So this device may be applied to photonic integrated circuits.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.