A simple laterally tapered waveguide for low-loss coupling tosingle-mode fibers

Low-loss coupling between semiconductor photonic devices and single-mode fibers is achieved using a simple InP/InGaAsP tapered waveguide. The proposed simple structure has a small and nearly square guiding core at its output facet. In this structure, the output field has a non-Gaussian profile, but low-pass filter coupling can be achieved by optimizing the design of the guiding core sizes. The waveguide is composed of a laterally tapered InGaAsP guiding layer and an InP cladding region on an InP substrate, facilitating integration of the waveguide with active devices using conventional processes. The waveguide is shown to have a total insertion loss of 2.6 dB, including a coupling loss of 0.9 dB and large ±2.5-μm misalignment tolerance in lateral and vertical directions with single-mode filters     

Specific features of the liquid-phase epitaxial growth of SiC epilayers in vacuum

4H-SiC epilayers were grown by a liquid-phase epitaxy in vacuum. It was found that the seed layer with steps characteristic of liquid-phase epitaxy should be preliminary deposited on the substrate. It is demonstrated that growth in a vacuum leads to a decrease in the concentration of uncompensated carriers N _d–N _a to the level of 2×10¹⁶ cm^?3.     

Design of a single-mode tapered waveguide for low-losschip-to-fiber coupling

The novel waveguide structures described in this paper have nonlinearly tapered shapes that result in low radiation losses despite their relatively short lengths. The core at the waveguide endface connected with the fiber has a very small cross section and an expanded mode field with a non-Gaussian shape. The taper structures are analyzed by using an improved step-transition method. This method is a based on the theory of enclosing a waveguide within electrical walls and that can therefore treat the radiation modes in a tapered waveguide as discrete mode spectra. Analyzing the relationships between the lengths and shapes of the tapers and the radiation loss due to the tapers show that appropriately tapered semiconductor waveguides operating at an optical wavelength of 1.55 μm and having a taper length of less than 0.7 mm can have a radiation loss of only 0.1 dB and a coupling loss with a conventional single-mode fiber of less than 0.5 dB     

Mode coupling in tapered structures

The theory of strong mode coupling involving nonorthogonal modes is extended to tapered structures. Initially, an attempt is made to show that previous formulations of the problem of nonorthogonal-mode coupling in tapered structures are incomplete. An equation is derived which is the generalization to tapered structures of the coupled-mode formalism with nonorthogonal modes     

Ridge laser with spot-size converter in a single epitaxial step forhigh coupling efficiency to single-mode fibers

We report on a 1.55-μm InGaAsP MQW laser diode with an integrated spot-size converter fabricated in a single epitaxial step using conventional photolithography. The laser structure uses a conventional ridge guide for the active layers and a second larger ridge for the passive waveguide. Low-beam divergence of typically 9°×9° results in about 3-dB coupling losses, with a cleaved optical fiber     

Optimal coding with a single standard run length

The optimum single standard run lengths for a binary first-order Markov source are derived and extended to multilevel first-order Markov sources. Maximization of the compression ratio is used as the criterion of optimality. When the output symbols are block coded, the optimal single standard run lengthn_ifor each symbol is shown to satisfy an implicit equation of the form(n_i - 1)(-ln q_{ii}) = 1 - q_{ii}^ {n_i}, whereq_{ii}is a transition probability. An expression for the overall compression ratio is derived for the binary case, and a comparison is made with enumerative source encoding. Compression ratio maxima are found by computer search for the binary independent source when the output symbols are subsequently Huffman coded, and a comparison of this scheme with ordinary run-length and source-extension coding is given.     

Analysis of the coupling characteristics of a tapered coupledwaveguide system

Optical coupling characteristics between one uniform waveguide and one tapered waveguide are discussed. This coupling structure is analyzed by utilizing coupled-mode theory and system step approximation. Based on numerical investigation, various coupling characteristics are presented. It is shown that the proposed structure can be used as a waveguide-type optical power divider by selecting an appropriate taper slope for the tapered waveguide or by separating the coupled waveguides properly for a desired power-dividing ratio     

High-efficiency coupling of laser diodes in tapered proton-exchanged waveguides

Low-loss coupling between a laser diode and a tapered waveguide fabricated by selective area proton-exchange in LiNbO/sub 3/ is reported. The effectiveness of this waveguide is demonstrated by generating 12.3 mW of second harmonic blue light pulse by Cerenkov radiation.<>     

GaAs-AlGaAs QW diluted waveguide laser with low-loss, alignment-tolerant coupling to a single-mode fiber

We report on low-loss, alignment-tolerant coupling between a single-quantum-well (SQW) GaAs-AlGaAs laser and a cleaved single-mode fiber. The laser was fabricated using conventional growth and processing techniques, and did not involve any regrowth. The mode size in the transverse direction was expanded by using a diluted waveguide structure. A butt-coupling efficiency of 66% was achieved against a theoretically possible 78% (which does not correct for Fresnel losses). In addition, large alignment tolerances for a 1-dB excess loss of /spl plusmn/2.0 pm and /spl plusmn/1.0 /spl mu/m were measured in the lateral and the transverse directions, respectively.     

MOCVD epitaxial growth of single crystal GaN,AlN and AlxGa1-xN

Ga begins to deposit from a stream of trimethylgallium (TMG) in H₂ at a minimum temperature of 475‡C. Addition of sufficient amounts of NH₂ results in the growth of textured polycrystalline GaN on basal plane sapphire substrates above 500‡C. A minimum temperature of 800‡C is required for the epitaxial growth of GaN on the substrate. Under similar conditions, but with the TMG replaced with trimethylaluminum (TMA), polycrystalline A1N begins forming at 400‡C (in the absence of NH₃,, the TMA starts pyrolyzing at 300‡C), but single crystal growth of A1N requires a temperature of at least 1200‡C. Epitaxial single crystal layers of Al_x Ga_1-x N can be grown in the temperature range 800−1200‡C, tne minimum temperature being approximately proportional to x, but preferential deposition of A1N on the hot walls of the reactor (>400‡C) precludes precise control of the alloy composition. This predeposition of A1N can be retarded by keeping the walls below 400‡C by using a water-cooled jacket, by rapid flow-rates, or by injecting the TMA through a nozzle close to the surface of the substrate.     

Multiquantum-well lasers with tapered active stripe for direct coupling to single-mode fiber

We demonstrate narrow beam divergence in 1.3-/spl mu/m wavelength multiquantum-well (MQW) lasers with an active stripe horizontally tapered over the whole cavity, for direct coupling to single mode-fibers. The lasers have reduced output beam divergence in a simple structure which does not contain an additional spot-size transformer. The fabricated laser shows narrow beam divergence of /spl sim/12/spl deg/, while a low-threshold current of 6.9 mA and a high efficiency of 0.62 mW/mA are realized. Furthermore, a direct-coupling efficiency to a single-mode fiber is -4.0-dB and -3-dB alignment tolerance is /spl plusmn/2.5 /spl mu/m.     

Mitigation for single event coupling delay

With advances in CMOS technology, circuits are increasingly more sensitive to transient pulses caused by single event particles. It has been predicted that the majority of the observed radiation induced soft failures in technologies below 65 nm will be because of transients that will occur in combinational logic (CL) circuits. Researchers mostly consider single event transients as the main source for CL related radiation-induced soft errors. However, for high reliability applications such as avionics additional sources need to be included in reliability analysis. In this work, we report a new error mechanism named ‘single event crosstalk delay’, investigate the vulnerability of recent technologies to these delay effects and then propose hardening techniques for single event crosstalk delay. Results are demonstrated using HSpice simulations with interconnect and device parameters derived in 130, 90 and 65 nm technology.     

Nucleation of a 60° glide dislocation in two-dimensional or three-dimensional growth of epilayers

The nucleation of a 60° glide dislocation in epitaxial thin films is analyzed. Nucleation from either a free surface or from a threading dislocation is considered. In this work, the presence of either infinitely long steps in a two-dimensional film or finite size steps in a two-dimensional or a three-dimensional film is proposed to reduce the activation energy associated with dislocation nucleation. The activation energy is reduced below 40 kT, a value predicted by standard nucleation theory for favorable nucleation, if the step is partially eliminated upon nucleation of a dislocation. However, when the coherency stress is small due to smaller lattice mismatch parameter, the activation energy can not be reduced below 40 kT. In addition, the activation energy is found to be a function of the size of the step but not that of the epilayer. Bowing of a threading dislocation by nucleation of kinks and propagation of the kinks along the dislocation is also analyzed using small angle approximation of the bow out. The activation energy associated with the bow out is again found to be independent of the size of the epilayer thickness. The critical size of the epilayer can be defined if the coherency strain energy remaining in the film is released in the form of thermal energy that helps to overcome the activation energy for nucleation of a dislocation.     

Parallel langmuir processes for silicon epitaxial growth in a SiHCl3-SiHx-H2 system

Parallel surface reactions in a SiHCl₃-SiH_x-H₂ system were numerically evaluated based on the heat and gas flow in the epitaxial reactor for clarifying the increase in the silicon epitaxial growth rate. H₂ and SiH_x are assumed to individually react with the intermediate surface species, *SiCl₂, which was formed by the chemisorption of SiHCl₃. The measurement was reproduced by the calculation using the rate constant obtained in this study for the surface reaction between *SiCl₂ and SiH_x. The decrease in the surface coverage by *SiCl₂ and the increase in the silicon yield were shown to be caused by the SiH_x. Thus, the surface reaction of the *SiCl₂ with the SiH_x was theoretically shown to be effective for increasing the growth rate to higher than the saturated value in an ordinary SiHCl₃-H₂ system. The rate equation for the parallel Langmuir processes in a general form was also described.     

Strain relief in epitaxial HgCdTe by growth on a reticulated substrate

In nearly all cases when an epitaxial layer of HgCdTe is grown on a CdZnTe substrate, there will be a finite lattice mismatch due to the lack of precise control over the ZnTe mole fraction. This leads to strains in the layer, which can be manifested in one or more ways: (1) as misfit dislocations near the interface, (2) as threading dislocations, (3) as surface topographical textures, and (4) as cross-hatch lines seen by x-ray topography. We have found that much of the strain can be relieved by growing on a reticulated substrate. Specifically, when the substrate has been etched to form mesas prior to growth of the layer, the resulting layer on the tops of the mesas shows evidence of significantly reduced strain. CdZnTe substrates oriented (111)A were prepared with two sets of mesas on 125 μm centers and 60 μm centers, and with other planar areas remaining for comparison. From a Hg melt, a layer of LWIR HgCdTe was grown about 16 μm thick on each substrate. Nomarski microscopy showed that the layers on the mesa tops were extremely flat, showing no sign of curvature or surface texture. X-ray topography showed no cross hatch on the mesa tops, while the usual cross hatch appeared in the planar regions. The LPE layer extended laterally beyond the edges of the original mesa because of faster growth in non-(111) directions. Samples were cleaved and examined in cross section. The linear density of etch pits seen in the cross section near the substrate, which represent misfit dislocations, was three times lower in the layer on the mesas than in the layer in the unpatterned region, although both regions have the same layer/substrate lattice mismatch. When an epilayer is grown on an unpatterned wafer (the conventional approach), the growth in any small region is confined laterally by the growing layer in the neighboring regions. However, when growth occurs on a reticulated surface, the lateral confinement is removed, providing strain relief and fewer defects.     

Resonant third-harmonic generation in a low-loss medium

The linear and nonlinear susceptibilities for resonant third-harmonic generation in an Mg-Kr system using the 3s2-3s 3d two-photon resonance at 431 nm are presented. The Mg-Kr system is shown to be a suitable low-loss medium because of its small one-photon and two-photon absorption cross section. A detailed quantitative analysis up to a power conversion efficiency of 1.2 . 10^-4was carried out. The phase-matching curves and the intensity dependence are in excellent agreement with a nonstationary modification of the theory of resonant third-harmonic generation in high density, phase-matched systems recently presented by Puell et al. [4].     

Application of selective silicon epitaxial growth for CMOS technology

The application of selective silicon epitaxial growth for device isolation is described. An improved selective epitaxial isolation technology is presented in the fabrication of CMOS LSI. This advanced process technology results from a superior selectivity for selective silicon deposition. A CMOS ring oscillator with a twin-well structure is fabricated by using this selective epitaxial isolation technology. The feasibility of using an oversized contact, due to the nature of its steeper oxide-to-silicon isolation boundary, is demonstrated.     

Confined lateral selective epitaxial growth of silicon for devicefabrication

An epitaxy technique, confined lateral selective epitaxial growth (CLSEG), which produces wide, thin slabs of single-crystal silicon over insulator, using only conventional processing, is discussed. As-grown films of CLSEG 0.9 μm thick, 8.0 μm wide, and 500 μm long were produced at 1000°C at reduced pressure. Junction diodes fabricated in CLSEG material show ideality factors of 1.05 with reverse leakage currents comparable to those of diodes built in SEG homoepitaxial material. Metal-gate p-channel MOSFETs in CLSEG with channel dopings of 2×10¹⁶ cm^-3 exhibit average mobilities of 283 cm²/V-s and subthreshold slopes of 223 mV/decade     

Selective epitaxial growth for the fabrication of CMOS integrated circuits

A process is described for the fabrication of CMOS integrated circuits which combines the epitaxial lateral overgrowth (ELO) technique with the concept of selective epitaxy. The resulting epitaxial material is shown to have a low defect density. Transistors fabricated in the selective epitaxy are shown to have characteristics which are a function of the epitaxial deposition conditions, the substrate orientation and dopant concentration, and the epitaxial layer thickness. Minimum device leakage currents were 250 pA/µm of channel width for n-channel devices fabricated in a p-well and 1.0 pA/µm for devices fabricated on p-substrates. The higher leakage currents for devices fabricated in a well are believed to be a result of the narrow vertical spacing (0.3-0.5 µm) between the n⁺source-drain regions and the n⁺substrate.