We proposed and demonstrated a hitless wavelength channel selective switch (hitless tunable Add/Drop filter) using Thermo-Optic (TO) effect of double series coupled dielectric microring resonator. Using a dielectric material as the core, the response time was reduced to 105 μs (rise time) and 15 μs (fall time), which are fifteen-fold and hundred-fold faster than that of polymer material, and the reproducibility by the heat cycle test was also improved to less than 0.01nm. The tuning range of wavelength selective switch was expanded to 13.3nm using the Vernier effect, and a large extinction ratio of more than 20dB was realized. In this review, the principle and recent progress of microring resonator wavelength selective switch will be introduced.

Channel drop filters with ring/disk resonators in a plasmon-polaritons metal are studied. It shows that light can be efficiently dropped. Results obtained by the finite difference time domain method are consistent with those from the coupled mode theory. It also shows, without considering the loss of the metal, that the quality factor for the channel drop system can be very high. The quality factor decreases significantly if we take into account the loss, which also leads to a weak drop efficiency.

Wavelength tuning and dynamic characteristics due to the vernier effect of a coupled-ring reflector (CRR) laser diode including active region within resonators are investigated using a split-step time-domain modelling approach. A CRR provides a strong mode-selection filter and could significantly extend the effective cavity length of a conventional Fabry-Perot laser. The simulation results for a particular design show that the tuning range as wide as 33 nm is possible with side mode suppression ratio exceeding 35dB. The modulation frequency could be 8 GHz and the frequency chirp could be in the range of 50 ~ 200 MHz/mA.

Based on the solution of the density matrix of motion equations we gave a theoretical simulation and experimental measurement of reduction of light propagation for a doped Er+ fiber by population oscillation. The effects of the concentration of doped Er ion density and fiber length were given in the paper. The measured data was agree with theoretical simulation well. Maximum delay is 10.75ms. The delay is tunable by changing modulation frequency and laser input power. The result showed the delay based on slow light in fiber can be proposed a novel optical delay line.

We designed and fabricated an arrayed waveguide grating demultiplexer of 70 × 75 μm² size, which consists of Si slab and wire waveguides on silicon-on-insulator substrate. By optimizing the connection between components and the layout of arrayed waveguides, the internal light scattering and the phase error were suppressed. As a result, the clear demultiplexing characteristics were observed with a channel spacing of 8 nm and a sidelobe level of -22 dB in the wavelength range around 1.55 μm.

In this paper, we report a fast and facile method for fabricating colloidal photonic crystals inside microchannels of radially symmetric microfluidic chips. As the suspension of monodisperse silica or polystyrene latex spheres was driven to flow through the channels under the centrifugal force, the colloidal spheres were quickly assembled into face centered cubic arrangement which had photonic stop bands. The optical reflectance spectrum was modulated by the refractive-index mismatch between the colloidal particles and the solvent filled in the interstices between the particles. Therefore, the present microfluidic chips with built-in colloidal photonic crystals can be used as in-situ optofluidic microsensors for high throughput screening, light filters and biosensors in integrated adaptive optical devices.

Using MOCVD butt-joint regrowth technique, we fabricated a μ-laser monolithically integrated with a passive waveguide in GaInAsP photonic crystal slab, for the first time. We observed the lasing operation and light extraction from the waveguide with an external quantum efficiency of 8%/facet. It will be the key technology for advanced photonic integration based on photonic crystal slab.

A polarization beam splitter based on a two-dimensional photonic crystal slab is presented. In the designed photonic crystal, the TE polarization is negatively refracted, while the TM polarization is positively refracted. The photonic crystal is composed of silicon pillars in a triangular lattice. The whole device is very compact (20×20μm²), and is compatible with the planar integration. The measurement results show that the extinction ratio is ~15dB from 1530 to 1610 nm.

The emission of erbium ions in silicon at 1550nm is enhanced by the presence of the two-dimensional (2D) plasmonic crystal. The nonradiative surface plasmon polaritons (SPPs) are generated through the relaxation of the excited erbium ions. The SPPs can be highly confined at the edge of the band of the 2D plasmonic crystal. By adding or reducing a multiple of external wave vector, the large amount of SPPs will be coupled out. The emission efficiency of erbium ions with help of reradiated SPPs can increase nearly 40 times comparing to that of one-dimensional (1D) metal grating. The emission angles of enhanced spectrum are discussed.

An InGaAlAs short-cavity DBR laser enabling 1.3-μm, uncooled, 10-Gbit/s operation at lower drive currents is demonstrated. This laser consists of a short InGaAlAs-MQW active region butt-jointed to an InGaAsP-DBR region. This structure provides moderate chip power, low threshold current, and a large relaxation oscillation frequency simultaneously, because it has an optimum cavity length in the range between 10 and 100 μm, at which both VCSELs and conventional edge-emitters cannot be formed because of their difficulty of manufacture. The fabricated 75-μm short-cavity laser demonstrated 100°C, 10-Gb/s operation at a record low drive current of 14 mA_p-p. Furthermore, it achieved side-mode suppression ratio of more than 37 dB at a high yield of 95%, because of the naturally high single-mode stability of its short-cavity DBR structure.

A laser and high speed modulator are integrated using a common shallow etched ridge waveguide for the first time. Performances of 10Gbps electro-absorption modulation transmission and the reliability of EM lasers are demonstrated. Low dispersion penalty for transmission over 90km of standard single mode fiber (SMF) is achieved. The devices have been fabricated and qualified for TDM and DWDM operation.

A highly integrated 10 Gb/s transmitter optical sub-assembly was fabricated and characterized for XFP transceiver. As a light source, uncooled 1.3 μm high-speed distributed feedback laser diode (DFB-LD) was fabricated and assembled on AlN sub-mount with a monitoring PD, a matching-resistor, and a bias-Tee with spiral-inductor. A glass sealed metallic low-loss TO-stem with in-line leads was newly presented. We developed a small-signal equivalent circuit model based on measured S-parameters in order to verify RF characteristics of LD and passive components. The eye-diagram of 10 Gb/s NRZ patterns with a PRBS 2³¹ -1 was opened clearly without mask violation. At 85°C, -3-dB bandwidth was measured as high as 11 GHz and 75-km transmission was successfully achieved with very low penalty.

Pulsation characteristics of multi-section DFB lasers with amplifying optical feedback have been investigated for various coupling coefficients. Wide current ranges showing stable pulsations could be obtained in the multi-section DFB lasers with anti-phase gratings compared to those with in-phase gratings and index gratings.

Lasing dynamics of photonic-crystal single-cell cavity is studied by Lorentz-dispersive Gain FDTD method. From hexapole mode of a photonic-crystal single-cell cavity, the generation of laser modes and the relaxation oscillation are observed.

The basic properties including physical properties and absorption spectrum were measured firstly. Then the chromatic dispersion of air-guiding photonic band-gap fibers (PBGF) was measured with phase-shift method and pulse delay method, from experimental results we can see that their test results are accordant. The differential group delay (DGD) of the PBGF was also measured with a method of Sagnac interferometer. We found that the DGD is in direct proportion to its chromatic dispersion.

Wideband wavelength tunable lasers are now beginning to play important roles in dense wavelength-division multiplexing (DWDM) optical transmission systems, because these lasers can reduce the required number of inventory lasers and inventory costs. They are also key components in the evolution towards future photonic network systems, particularly in reconfigurable optical add/drop multiplexing (ROADM) systems and optical cross-connect (OXC) systems. However, before the tunable laser can be a viable alternative for the conventional fixed wavelength laser (e.g. DFB laser) it should have the same characteristics over its entire tuning range. Additionally, the device should only be marginally more expensive than its fixed counterpart. This paper reviews our recent activities on the development of high performance full-band wavelength tunable lasers. Our approach utilizes an external cavity configuration, which makes use of a liquid crystal (LC) tunable mirror. We demonstrate high performance of an external cavity wavelength tunable laser (ECTL) with an intracavity etalon. This ECTL module shows a tuning range of 45 nm, fiber-coupled output power of higher than 14 dBm with SMSR of better than 59 dB, wavelength accuracy of +/-0.6 GHz, relative intensity noise (RIN) of better than - 150 dB/Hz, and linewidth of narrower than 1 MHz.

The butt-coupled (BT) sampled grating distributed Bragg reflector laser diode (SGDBR-LD) was designed and fabricated using planar buried heterostructure (PBH), enabling a low threshold current and a stable fundamental transverse mode. The but-coupled SGDBR-LD's with target tuning ranges of 44.4nm was fabricated, and the tuning ranges were experimentally measured to be 44.4nm. The measured peak periods of the fabricated SGDBR-LD's were well matched with theoretical values and output power is close to calculated values. The side mode suppression ratio of more than 35dB was obtained in the whole tuning range. The output power variation was less than 5dB, which is 4dB smaller than that of RWG structure.

We presented the high temperature operation of 1200-nm band highly strained GaInAs/GaAs ridge-waveguide lasers. Active layer consists in three quantum wells with highly strained GaInAs. The In composition is 32%. The maximum operating temperature reaches at over 200°C and temperature characteristic T₀ is 222K at 30-80°C with continuous wave operations, showing excellent temperature characteristics of highly strained QWs. We obtained a relaxation oscillation frequency of 2 GHz at 170°C.

We demonstrate 1.25 Gbps 1300 nm InGaAsP/InP DFB lasers employing asymmetric MQWs structures operable over a wide temperature range of -40°C to 100°C, which has an optical power of 10 mW and a side mode suppression ratio (SMSR)and a Fabry-Perot suppression ratio (FPSR) of >30dB.

Ultra-compact silicon-photonic-crystal-waveguide-based thermo-optic and electro-optical Mach-Zehnder interferometers have been proposed and fabricated. Thermal and electrical simulations and optical characterizations have been performed. Experimental results were in good agreement with the theoretical predictions.

Woodpile structure which is a simple three-dimensional photonic crystal was fabricated by a two-photon polymerization. Resins with high two-photon absorption cross-section, transparent at near-IR spectral region but highly absorbing around 400 nm, undergo a two-photon absorption and polymerization upon irradiation with 780nm from a Ti:Sapphire femtosecond laser. The mechanical nano-positioning system was adopted to identify the position of the focal point inside the polymer and to fabricate the high resolution nano woodpile structures. The resultant structures have a face-centered-cubic symmetry.

The influence of etching slope on cavity Q-factors in two-dimensional (2D) photonic crystal (PhC) slab is studied. Through FDTD simulation, it is confirmed that the Q-factor decreases with etching slope. The main loss comes from the horizontal coupling into propagating TM-modes. We designed three-lattice-long modified linear cavities having high Q-factors. However, the measured Q-factor was about 250. This small Q-factor is attributed to the non-vertical (13°) side wall.

Spontaneous parametric down-conversion in two-dimensional photonic crystal made of semiconductor material with large quadratic nonlinear susceptibility is proposed to generate polarization entangled photon pairs. On one hand, the large quadratic nonlinear susceptibility of AlGaAs crystal insures the high nonlinear conversion efficiency; on the other hand, the abnormal dispersion property of two-dimensional photonic crystal causes the satisfaction of the phase-matching condition. In particular, the dispersion property sensitively depends on the structure of photonic crystal. Then the walk-off between the down-converted photons with orthogonal polarizations can be minimized through appropriate structure parameter design; hence compensation measures to mitigate labeling effect on polarization entangled photon pairs can be eliminated.

We demonstrate the light transmission characteristics in a photonic crystal (PC) slab with liquid crystal (LC) as an anisotropic medium. In PC slabs, the guided resonance occurs for normal incidence of light. In this study, we first fabricated a SiN PC slab without LC and observed this resonance. Also, we added an ellipticity to holes of the PC slab and observed a polarization filtering effect arising from the structural birefringence. Then, we filled the holes with LC. By making a phase transition in the LC, the resonant wavelength was tuned by maximally 33 nm. This will be applicable to surface-normal-type devices which control the transmission and polarization characteristics of light.

The ability to dynamically control the properties of a photonic crystal by basing such a crystal on a media exhibiting electromagnetically induced transparency (EIT) is discussed. As an example of this, simulation on a tunable photonic crystal slab exhibiting negative refraction is presented.

The planewave based transfer matrix method has been developed with rational function interpolation to efficiently simulate photonic crystal devices. Cavities embedded in three-dimensional layer-by-layer photonic crystal are systematically studied as an example to show the power of transfer matrix method with the relation between resonant frequencies and the cavity size obtained.

A 1300nm, high power, vertically emitting Fabry Perot laser is presented with a monolithically integrated photodiode. The lasers use ridge waveguide technology with a 45° etched facet to create 30mW of vertically emitted light. Two types of monolithically integrated back facet monitor diodes are discussed togther with their merits for adequate collection efficiency and tracking error. HCSELs with integrated MPDs have passed over 3000 hours of reliability testing.

This paper suggests a passive aligned optical subassembly (OSA) using 54.7° mirrors of a silicon optical bench (SiOB) and a 57° angled fiber array for a vertical cavity surface emitting laser (VCSEL). This OSA is very cost-effective because the OSA was fabricated by only one-axis alignment along the V-groove's direction and flip-chip-bonding the VCSEL. In addition, this paper describes a 2.5-Gbps x 12-channels parallel optical transmitter module fabricated with the OSA.

Vertical-cavity surface emitting lasers (VCSELs) can always perform single longitudinal mode operation. However, in VCSELs, when the emission aperture becomes large, single transverse mode operation will be challenged. In this paper, we have discussed the VCSEL designs to achieve single transverse mode operation. Scalar transfer matrix method analysis has been used to analysis of antiresonant reflecting optical waveguides (ARROW) VCSEL.

This work presents complete 2D electro-opto-thermal simulation of the intracavity contacted oxide confined vertical cavity surface emitting lasers (ICOC VCSEL). The analysis represents the influence of geometrical parameters on power and modulation properties on such devices. The optimized values with maximum of modulation bandwidth are presented.

Focused ion beam (FIB) etching technology is a highly efficient post-processing technique with the functionality to perform sputter etching and deposition of metals or insulators by means of a computer-generated mask. The high resolution and the ability to remove material directly from the sample in-situ make FIB etching the ideal candidate for device prototyping of novel micro-size photonic component design. Furthermore, the fact that arbitrary profile can be etched directly onto a sample without the need to prepare conventional mask and photolithography process makes novel device research with rapid feedback from characterisation to design activities possible. In this paper, we present a concise summary of the research work in Cambridge based on FIB technology. We demonstrate the applicability of focussed ion beam post processing technology to active photonic devices research. Applications include the integration of advanced waveguide architectures onto active photonic components. We documents details on the integration of lens structure on tapered lasers, photonic crystals on active SOA-integrated waveguides and surface profiling of low-cost gain-guided vertical-cavity surface-emitting lasers. Furthermore, we discuss additional functions of FIB in the measurement of buried waveguide structures or the integration of total-internal-reflection (TIR) mirror in optical interconnect structures.

We demonstrated the clear light focusing in a photonic crystal superlens. We fabricated it on SOI substarate with optimum interfaces for suppressing reflection and diffraction losses. We observed near field patterns of light propagation in the superlens, which showed the light focusing in a wide frequency range inside and outside of the light cone. We also confirmed that the focal length, NA, and the spot width, which ageed well with theory. The smallest spot width was 1.8 μm at ω= 1.39 μm. In addition, we demonstrated the real image formation using multiple light sources.

We propose efficient unidirectional light emitters, which are enabled by the use of large Purcell effect, defect engineering, and the bottom Bragg reflector. Enhanced spontaneous emission rate enables us to achieve very efficient light sources, in which most of the emitted photons can be funneled into a specific resonant mode of interest. The far-field radiation properties of a photonic crystal resonant mode are modified by tuning the cavity geometry and by placing a reflector below the cavity. As a result, > 80% of the photons generated inside a photonic crystal resonator can be collected from the top, within a small divergence angle of ±30°.

We present an investigation of free-carrier screening in coupled asymmetric GaN quantum discs with embedded AlGaN barriers using time-integrated and time-resolved micro-photoluminescence measurements, supported by three-dimensional multi-band k.p computational modeling. We observe that with increasing optical excitation the carrier lifetime decreases and emission energy blue-shifts. This originates from the screening of built-in piezo- and pyroelectric fields in the quantum discs by photo-generated free-carriers. Due to non-resonant tunneling of carriers from the smaller disc to the larger disc, free carrier screening is enhanced in the larger disc. The non-resonant tunneling was found to have a significant role in samples with a thin barrier, as the screening decreased with barrier thickness (i.e. decreased tunneling). Computational modeling was in good agreement with the experimental results.

We investigate the influence of oxide aperture size on the performance of intracavity contacted oxide-aperture vertical-cavity surface-emitting lasers with asymmetric current injection. Several counteracting mechanisms are shown to result in size dependent behavior, which limits the performance of very small cavities. Reducing the oxide aperture is shown to improve the threshold current and the 3dB bandwidth. However, significant increase of optical losses is observed that is attributed to increase the threshold current density and to decrease the maximum output power. From the far-field measurement, we have shown that the smaller aperture VCSELs have large FWHM. Also, we have achieved the small signal modulation bandwidth of 10.3GHz with 4.5μm oxide aperture diameter at 9mA bias current.

Low loss, single mode rib waveguides, based on PECVD deposited multi-layer amorphous silicon are fabricated. These waveguide are refractive index and mode-matched to III/V laser waveguides. Methods for monolithic integration of these passive amorphous silicon waveguides with InGaAsP/InP gain sections are demonstrated. Results of a multi-wavelength laser based on an amorphous silicon arrayed waveguide grating integrated on a single chip with InGaAsP gain sections are presented.

Integrated optical demultiplexing and receiving device based on one-mirror-inclined three-mirror cavity (OMITMiC) structure, or OMITMiC wavelength-selective photodetector, is a kind of novel integrated multifunction optoelectronic device which was proposed in 1996 and first realized with GaAs-based materials for short wavelength (less than 1μm) operation in 2001. Recently, after great efforts on developing controllable self-retreating dynamic mask (CSRDM) wet etching method for InP-based epitaxial layer and low temperature InP/GaAs wafer bonding technique, such a device operating at long wavelength region (1550nm) had also been successfully demonstrated and the measurement results shown that it features high-speed (12GHz with a mesa area of 40×36 μm²), high quantum efficiency (66%~78.4%), ultra-narrow spectral linewidth (0.6 nm) and wide range tuning (more than 10 nm ) simultaneously. In addition, a long wavelength monolithic OMITMiC photodetector with GaInNAs absorption layer has also been demonstrated. These achievements could have a significant impact on wavelength-division-multiplexed (WDM) optical fiber transmission systems and networks.

Enhanced slow light propagation is predicted in a coupled resonator optical waveguide structure possessing highly dispersive elements using the finite-difference time-domain method. The group velocity is shown to be below 0.01c_o.

A general structure of Ge-on-Si Resonant cavity enhanced (RCE) PIN photodetectors with modified top and bottom mirrors operation at 1.55μm is presented. Resonating mirrors with higher reflectance are obtained. Different cases on the relationship of quantum efficiency and wavelength are discussed in detail. The number of Si-SiO₂ layers in top and bottom mirrors for the highest quantum efficiency is analyzed. The frequency response considering both the transit time and the capacitance with different device areas is also simulated. These results offer a complete guideline for designing and fabricating a high speed, high quantum efficiency Ge-on-Si RCE PIN device.

We discuss a new simple InGaAs/InAlAs avalanche photodiode (APD) with a planar buried multiplication region. Some of the advantages compared to standard APDs are: 1. The thickness of the avalanche and the charge control regions are accurately controlled by molecular beam epitaxy (MBE) growth in contrast to the standard diffusion process; 2. InAlAs is the multiplication material (avalanching faster electrons) instead of InP (avalanching slower holes); 3. InAlAs avalanche gain has a lower noise figure than that for InP; 4. A guard ring is not required; 5. Fabrication is as simple as that for a p-i-n detector; 6. The APD has high wafer uniformity, and high reproducibility; 7. The InAlAs breakdown voltage is lower than InP, and its variation with temperature is three times lower than that for InP; 8. Excellent aging and reliability including Telcordia GR-468 qualification for die and modules; 9. High gain-bandwidth product as high as 150GHz; and 10.High long range (LR-2) bit error rate (BER) 10^-12 receiver sensitivity of -29.0dBm at 10Gb/s, -28.1dBm at 10.7Gb/s and -27.1dBm at 12.5Gbs.

We demonstrate a wavelength-selective photodetector that combines a Fabry-Perot filtering-cavity (FPC) with a taper absorption-cavity (TAC). The taper cavity shows non-resonant effect but exhibits absorption enhancement effect, so that high-speed, high quantum efficiency, wide tuning range and ultra-narrow spectral linewidth can be achieved simultaneously. Device performance was theoretically investigated by including key factors such as taper angle, finite-size diffracting-beam input, and lateral walk-off in the taper cavity. The device was fabricated by bonding a GaAs-based FPC, which can be tuned via thermal-optic effect, with an InP-based TAC. The experiment results of the devices were reported in another paper.

Some recent advances in photonic integrated circuits are presented, including an InP-based monolithically integrated optical channel monitor, an arrayed waveguide grating (AWG) triplexer and a novel design of polarization insensitive AWGs. The optical channel monitor comprises a flat-field echelle grating and a slab photodetector array. The channel passband is flattened and widened without loss penalty. The shape of the slab waveguide detectors and the layer structure are optimized to obtain low crosstalk and low polarization dependence. It also has a smaller size due to the elimination of output waveguides. For triplexers in fiber access networks, a cross-order AWG design is proposed to overcome the device layout difficulty due to the wide spectral range of the wavelength channels. The spectral periodicity of the grating is utilized to reduce the free spectral range requirement. Consequently, the AWG can operate at a higher diffraction order with a smaller overall size. Finally, a novel design of a polarization insensitive arrayed waveguide grating is presented. Unlike conventional AWGs where the optical path length difference is obtained only in the arrayed channel waveguides, we design the star coupler regions according to Rowland circle construction with an oblique incidence/diffraction angle. As a result, the optical path length difference is produced in both the channel waveguides and the slab waveguide regions. By using the birefringence difference between the channel waveguide and the slab waveguide, a polarization dispersion compensated AWG is realized without any additional fabrication step.

A modulation efficiency enhancement scheme for an InGaAsP waveguide-coupled micro-ring cavity resonator utilizing the self-aligned total internal reflector mirrors was proposed and investigated numerically and experimentally. Unlike the conventional electroabsorption modulators, the micro-ring cavity resonator was found to exhibit a singular modulation characteristics depending on the coupling coefficient between the micro-ring cavity and the input/output waveguide, which can be exploited to enhance the optical modulation efficiency as well as the extinction ratio.

An active microlens device is demonstrated by using a stacked layer structure of UV curable polymer, liquid crystalline polymer (LCP) and a liquid crystal (LC). The incident linearly polarized light is focused after passing through the combined refractive type microlens array system of UV curable polymer and LCP. Because used LCP shows highly birefringent macroscopic property from the well-ordered molecular structure, the additional polarization state control layer was inserted to modulate the dynamic focusing characteristics of the device. From the additional twisted LC layer's electro-optic response, we obtained good focal switching characteristics of microlens array with a small operation voltage application. This enhanced dynamic focusing characteristic of device was originated from the separate operation of polymer lens structure's beam focusing and twisted LC layer's polarization control ability. The measured focal length was well matched to the calculated one. This proposed LC microlens array is expected to play a critical role in the various real photonic components such as highly reliable optical switch, beam modulator and key device for 3-D imaging system.

We report on the development of GaN-based violet laser diodes (LDs) for the high-capacity optical storage application and blue LDs for the laser projection display application. InGaN LDs with emission wavelength of ~405 nm are already being adopted for next-generation optical-storage systems. We present results on >400 mW single-mode output power under pulsed operation which can be employed in 100 Gbyte multi-layer BD systems. We designed LD layer structures to exhibit high level of catastrophic optical damage (COD) and small beam divergence. In addition, GaN-based blue LDs with emission wavelength of ~450 nm have also been developed for the application to the blue light sources of laser display systems. We demonstrate single-mode blue InGaN LDs with >100 mW CW output power. Interestingly, we observed anomalous temperature characteristics from the blue InGaN LDs, which has shown highly-stable temperature dependence of output power or even negative characteristic temperature (T₀) in a certain operation temperature range. This unusual temperature characteristic is attributed to originate from unique carrier transport properties of InGaN QWs with high In composition, which is deduced from the simulation of carrier density and optical gain.

We investigated the dependency of waveguide structures on ripples of far-field patterns in 405nm GaN-based laser diodes theoretically and experimentally. As the n-type cladding layer thickness decreases, the passive waveguide modes strongly interact with an active layer mode. This suggests that the thicknesses of n-AlGaN/GaN superlattice clad and n-GaN waveguide layers have significant influences on FFP ripples. We successfully obtained very smooth far-field patterns perpendicular to the junction plane by optimizing both n-AlGaN/GaN clad layer thickness and n-GaN waveguide layer thickness.

Intracavity-contacted resonant cavity enhanced photodetectors (IC RCEPDs) have been fabricated for monolithic integration with IC VCSELs. A low parasitic capacitance of 0.39 pF and an extrinsic 3-dB bandwidth of 9 GHz are demonstrated by using coplanar metal contacts. Optimization issues on device and epi designs are discussed. The largest frequency saturation photocurrent below which the extrinsic 3-dB bandwidth exceeds 6.5 GHz is 4.2 μA.

A novel transmitter product from Bookham with high performance has been reported. This sophisticated telecommunications optical source consists of a high power DFB laser with a wavelength locker and a Bookham's third generation InP-based Mach-Zehnder Modulator. This device possesses nearly zero-chirp characteristics, small variation of dispersion penalty values within 1.0dB over both positive and negative 800 ps/nm transmission links at 10.70 Gbit/s.

The 1.3/1.55 μm bi-directional small form pluggable (SFP) optical transceiver with an integrated wavelength division multiplexing (WDM) subassembly using accurate ceramic blocks has been developed. The WDM subassembly on which a laser diode (LD), a receiver photodiode (r-PD), a WDM filter, and two micro lenses are integrated is only 2.0 x 2.1 x 0.6 mm³ in size and inserted in a TO-CAN package. The SFP transceiver coupled with single mode fiber has been operated at a 622Mbps data rate. The transmitted optical output power is -2.8 dBm and the measured value of sensitivity is -32 dBm at 10^-10 Bit Error Rate (BER).

In this paper, a novel scheme of all optical logic gates by using an injection-locked semiconductor laser is proposed. We use a model to describe the dynamics of the injection-locked laser. The simulation results show that NOR and XOR gates can be achieved with properly designed parameters.

The structure of the optical path of a novel VOA integrated receiver is presented. The method to enhance the attenuation performance of the Receiver is described in detail. The standard coplanar package module exhibits a fluent attenuation curve and can achieve more than -20dB attenuation at ~ 6.5V drive voltage. S21, S22 performance and specifications of the module are explained in the paper. All these features provide customers considerable benefits, including high quality, low power consumption and cost, board real estate flexibility and ease of use.

The responding time delay characteristic of the detector of PIN photodiode is caused by the different intensity of the signal received by the detector. The responding time delay usually is less several tens nano-seconds. Aaccording to the principle of the responding time delay, the testing system is designed to measure the responding time delay. The test of responding time delay with different laser powers is carried on, and the result of the responding time delays is given.

An electrically driven light emission from silicon is a long-standing problem in silicon photonics. Recently, significant progress has been made using silicon quantum dots embedded in silicon nitride thin films, transparent doping layers and electrodes, and surface modified structures. This paper provides an overview of progress in the device physics and fabrications of the nanocrystal silicon light emitting diodes including new device structures to improve the light extraction efficiency as well as highlights in growth of silicon quantum dots and their quantum confinement effects.

We developed a nano-photodiode that confines and absorbs the sub-wavelength-size optical near field in small-scale silicon. A surface plasmon resonance antenna is used to enhance the near field in silicon. The response time of the nanophotodiodes is shorter than that of conventional photodiodes because the separation between anode and cathode and the size of the electrodes can be as small as one thousandth of that for conventional photodiodes. The full-width at halfmaximum of the impulse response of the silicon nano-photodiode was as fast as ~20 ps even when the bias voltage was less than 1 V. This nano-photodiode technology can be applied to other semiconductor materials such as germanium and ternary compound semiconductors.

We present detailed investigations about the elastic strain field distributions of the lens shaped InAs/GaAs selforganization quantum dots array, incorporating the effect of the longitudinal and transverse period distributions. The results show that the period distributions especially for the longitude period distribution have obvious influences to the strain field. The influence of the longitudinal and transverse periods on the strain field are just opposite in effect, especially for the line-scan path along the quantum dot's center-axis. Under suitable conditions, the effects of both periods on the strain field distribution can be partly eliminated. The results also demonstrate that one must take the quantum dot's period distribution into account when the effect of the strain field on the electronic structure is calculated. We conclude that using the isolated quantum dot model is not appropriate to simulate the strain field, or evaluate the influence of it on electronic structure.

Optically pumped terahertz emission has been observed in a wide range of semiconductors. We show that InAs quantum dots on GaAs can be used to significantly enhance terahertz emission compared with a bare GaAs surface.

Recently, ultrafast nonlinear optical responses of single-wall carbon nanotubes (SWCNTs) in suspensions and in films have been investigated intensively. Transient photobleaching has been observed with femtosecond laser pulses at photon energies of 0.8 ~ 1.1 eV (wavelengths = 1100 ~ 1550 nm), resonant with the lowest interband transitions of semiconducting SWCNTs. Here, we report both absorptive and refractive nonlinearities in a film of multiwalled carbon nanotubes (MWCNTs) grown mainly along the direction perpendicular to the surface of quartz substrate. Such MWCNT films are prepared by a method of plasma enhanced chemical vapor deposition. By employing Z-scans with 180-fs laser pulses at wavelengths ranging from 720 to 1550 nm, we have observed that both absorptive and refractive nonlinearities are of negative. More importantly, the degenerate pump-probe measurement reveals an ultrafast recovery time of ~ 1 ps. In addition, we also present a demonstration that the ultrafast nonlinear optical properties can be manipulated by a hybrid system in which MWCNTs are coated with ZnO nanoparticles. At wavelengths of interest, it is known that ZnO possesses three-photon (or four-photon) absorption, which is of positive sign and can be used to balance off the negative nonlinearity of MWCNTs.

The progress in Si-based optoelectronic devices and their integration for optical communications are summarized. Integrations on different material platforms are described, but emphasis is given to new micro/nano scale emitters, detectors, and light beam controlling devices. The perspective of micro/nano scale monolithic integration of optical devices and electronic devices on a single chip by standard CMOS technologies is presented. The possibility of using these devices for computer data communications, board-to-board and chip-to-chip, is also discussed.

Ytterbium doped sodium phosphotellurite glasses are made with different Yb³⁺ concentrations. Physical properties of the new glasses are reported. The glasses show high absorption and emission cross-sections and higher lifetime of [equation] transition. Laser performance parameter σ_e × τ_f, with a value as high as 2.14×10^-20 cm²-ms and the laser threshold intensity as low as 1.79 KW/cm² are reported. The laser parameters of these Yb³⁺ doped glasses show that they are potential to fabricate high power laser and broadband optical amplifiers.

We have explored a new doping method, which could disperse densely dyes into polymers, and have developed the present method into preparation of polymeric devices, such as storage media, luminescent media, switching devices, and waveguides. Preparation of the waveguide was based on dispersal of organofluorine (OF) compounds with a low refractive index in a polymer plate functioning as a cladding layer. The layer dispersed with the OF compound showed uniform concentration of the OF compound independent of the dispersal depth indicating that dispersion behavior was not governed by Fick's law. This means that the waveguide in the present study is step index. As a further development, we expanded the present method to preparation of a polymeric optical amplifier (POA) because reduction in intensity of a signal beam is clear due to driving of splitters in a short distance network using polymer optical fibers with large cores. A laser dye was dispersed densely within the core by the present method, followed by the dispersal of the OF compound to form the cladding layer resulting in the fabrication of POA with the large core. Amplification of the signal beam at 650 nm was confirmed in the waveguide upon irradiation of a pulsed laser.

Photopolymer based microholograms are gaining much importance in the field of security imaging, product authentication and prevention of document forgery. Security holograms, mass produced through soft or hard embossing, from electroformed metal master holograms are not amenable to store and retrieve variable data. On the other side, rapid developments in optical and digital technologies result in large scale counterfeiting of conventional security holograms and look-alike holograms of great exactitude is becoming a real threat to original manufacturers. In contrast to conventional recording materials, photopolymer holograms do not need wet processing. They are amenable to replication and, at the same time, can hold variable data. This, apart from security at various levels, facilitates machine readability, automation, easy tracking and effective maintenance of inventory. This paper presents design, development and performance evaluation of a photopolymer based holographic variable data storage system for security applications. A liquid crystal spatial light modulator (SLM) is used to create a modulated optical data beam that varies from hologram to hologram. Photopolymer films in tape form are applied for continuous recording of micro-holograms, synchronous with the variable data content. This is a novel, but simple data storage system and can be used to give added security, in conjunction with conventional holograms. Easy and on site verification by applying special reading devices and dedicated software is the other charm of the proposed system. Moreover, for added protection, variable key based data encryption can be applied effectively. System parameters like diffraction efficiency, recording speed, preprocessing requirements etc. are analyzed and the response of the photopolymer material is also evaluated.

A practical numerical model for liquid crystal cell is set up based on the geometry structure of liquid crystal optical phased arrays (LCOPA). Model parameters include width and space of electrodes, thickness of liquid crystal layer, alignment layers, electrodes and glass substrates, pre-tilted angles, dielectric constants, elastic constants, the refractive indexes and so on. Especially, the thickness of alignment layer is first considered to the best of our knowledge. According to electrostatic field theory and Frank-Oseen elastic continuum theory, two dimension (2D) electric potential distribution and 2D director distribution are calculated by means of the finite difference method on non-uniform grids. And the phase delay of LCOPA is derived from crystal optics. The influence of cell sizes on phase delay distribution is analyzed. The evaluation function of fringing field effect is provided. And the methods to decrease fringing field effect between electrodes are also discussed.

The optical and structural properties of binary and ternary III-V nanowires including GaAs, InP, In(Ga)As, Al(Ga)As, and GaAs(Sb) nanowires by metal-organic chemical vapour deposition are investigated. Au colloidal nanoparticles are employed to catalyze nanowire growth. Zinc blende or wurtzite crystal structures with some stacking faults are observed for these nanowires by high resolution transmission electron microscope. In addition, the properties of heterostructure nanowires including GaAs-AlGaAs core-shell nanowires, GaAs-InAs nanowires, and GaAs-GaSb nanowires are reported. Single nanowire luminescence properties from optically bright InP nanowires are reported. Interesting phenomena such as two-temperature procedure, nanowire height enhancement of isolated ternary InGaAs nanowires, kinking effect of InAs-GaAs heterostructure nanowires, and unusual growth property of GaAs-GaSb heterostructure nanowires are investigated. These nanowires will play an essential role in future optoelectronic devices.

In this paper, the transmission characteristic in nano-size metallic photonic crystals (MPCs) is studied with the finite difference time domain (FDTD) method. It's show that the size and the layout of the metal decide the transfer characteristic of the metallic photonic crystals.

Band structures of two-dimensional photonic crystal composed of metallic cylinders are numerically studied with FDTD method. Four kinds of metal: Cu, Ag, Au and Al are considered. Bandgaps varing with lattice constant (a) and filling factor (f) are drawn and analyzed.

We investigate the resonant transmission of light through a sub-wavelength metallic nanoslit surrounded by various medium. The relationship between the resonant wavelength through a sub-wavelength metallic nanoslit and refractive index of surrounding medium is obtained and the distributions of magnetic field intensity at the resonant peaks are numerical simulated by boundary integral method. The results show that there exists a linear relationship between the resonant transmission and the refractive index of surrounding medium. This analysis will provide useful information in nano-sensor technology.

Semiconductor modeling and characterization of ternary materials have been performed using simulation. Ternary semiconductors such as Al_xGa_1-xAs, Ga_xIn_1-xAs and Al_xIn_1-xAs have been studied using Monte Carlo Simulation. Standard parameters like device dimension, donor concentration, temperature, electric field, etc have been considered. Drift velocity with respect to applied electric field in the range of 10kv/cm - 60kv/cm has been observed for three ternary semiconductors. It has been found that up to 30kv/cm, velocities for different compositions show major variations. But after 30kv/cm, they show less variation from each other. It is observed that, with an increase in applied field, the G valley electron population decreases while L and X valley population increases. The scattering methods that have been considered during the study of transport properties are Polar Optical Phonon Scattering (absorption and emission), Intervalley scattering, Ionized Impurity Scattering.

We developed a technique of bonding magneto-optic garnets to III/V compound semiconductors for integrating an optical isolator with a semiconductor optical device. Some approaches to realize an isolator integrated with a laser diode will be presented.

1.5μm wavelength region is the window of optical communication. Absolute frequency references that satisfy ITU-T standards in 1.5 μm region for frequency calibration are widely needed in DWDM system. In this paper, an approach of obtaining precise frequency points which ITU-T has produced has been experimented, using an acoustic frequency shift, by calibrating Fabry-Perot etalon using double acetylene absorption lines. In the experiment, mode 1955 of F-P etalon is offset frequency-locked to acetylene absorption line P14, 195500.4023GHz, with frequency stability 108 for a 30-minute measurement. After calibrating, a series of spaced 100GHz artificial absolute frequency references in 1.5μm region are obtained.

We demonstrated the room temperature lasing of GaAs-based 1.3 μm quantum-dot laser diode (QDLD) grown by atomic layer epitaxy (ALE). The active region of a QDLD consists of 3-stacked InAs quantum-dots (QDs) in an In_0.15Ga_0.85As quantum well (dots-in-a-well: DWELL), which was grown by molecular beam epitaxy (MBE). For advanced performances of QDLD, the high-growth-temperature spacer layer and p-type modulation doping were applied to QDLD active region. We fabricated ridge waveguide structure LDs which had 10 ~ 50 μm ridge width with several cavity lengths and applied a high reflection (HR) coating on one-sided mirror facet. The threshold current density was 95 A/cm² under a pulsed operation and 247 A/cm² under a CW operation, respectively. The lasing wavelength was 1.31 μm under a pulsed operation condition and 1.32 μm under a CW operation at room temperature. The QDLD showed a simultaneous lasing and a state switching to the higher-order state. The lasing wavelength switching from the ground state to the excited state depends on the cavity length, the injection current and operating temperature.

In this paper, based on self-reproduction theory, harmonic mode-locked (HML) and rational harmonic mode-locked (RHML) fiber ring lasers consisting of two semiconductor optical amplifiers (SOAs) was numerically researched, respectively. Harmonic mode locking makes a target of obtaining ultra-short pulse, but, in rational harmonic mode locking, it urgently needs to be solved that pulse amplitude becomes uneven with the increase of the order of rational harmonic, which results in the different work conditions of both. After obtaining the optimal work condition, the system parameters effects on the characteristic of HML pulse and the quality of pulse-amplitude equalization in rational harmonic mode locking have been investigated, respectively.

We investigate the Goos-Hanchen (GH) shifts from an asymmetric configuration with single-negative materials by means of the stationary phase theory. The transmission and reflection coefficients for both TE- and TM-polarized incident beams are obtained using the transfer matrix method. A large GH shift was observed in the asymmetric configuration with single-negative materials when the surface polariton is properly excited for the TM polarization. The GH shift of the reflected beam is not equal to that of the transmitted beam. Furthermore, it is found that there is an optimum thickness and an optimum incident angle for the maximum GH shift time. The GH shift of the reflected beam can be detectable due to its large value and high reflectivity.

Accelerometers are one of the important sensors for a number of applications. In this paper, an accelerometer utilizing optical means for sensing the acceleration is introduced. The proposed accelerometer combines micro optical, electrical and mechanical means (MOEMS) to achieve high sensitivity, wide measurable range, as well as minimal losses in optical part. The optical subsystem is a highly integrated system where passive and active components are all monolithically integrated together. Photo-elastic effect and electro-optical modulation are used when sensing and processing the acceleration. The simulation models of the parts of optical subsystem and its parts and the results are also analyzed.

This paper presents a polarimeter for simultaneous measurement of variation in magnitude phase retardation, azimuth and principal axis angle of the incident light. The configuration is based on dual-liquid-crystal-modulator (Dual-LCM) which is used to enhance the measurement's S/N and resolution. The changes of these three parameters are measured by a very simple digital signal processing system. Compared to the conventional configurations, the proposed polarimeter offers an easier setup, high system stability and repeatability. The results show that our system has the advantages of wide measurable domain for the whole-polarization and high measurement precision. The results indicate that the maximum absolute errors of the phase retardation, azimuth and principal axis angle are 0.24°, 0.14° and 0.31°, respectively. Finally, the dynamic ranges of the principal-axis angle measurement and the phase-retardation measurement extend as far as 360° and the measurement precision is below 0.31°.

In this paper, we provide a detailed performance analysis of an nc-Si EDWA for the real application. Optical gain (small signal / saturation regime), noise figure and required pump density has been assessed in terms of the device structure. Results show a high feasibility of achieving 10dBm output power with 0dBm of input signal, using an array of commercially available high-power blue-green LEDs as the top pump. In numerical model section, we suggest simplified coupled rate equation and 2-D propagation equation constructed to investigate amplifier performance. In performance analysis section, firstly, we compared population inversion characteristics for Er with/without nc-Si condition to confirm the widening of high inversion region by introducing nc-Si sensitizer, which means less pump intensity requirement for same input signal power. In addition, to test the feasibility of the NC-EDWA for the metro-network applications, we simulated the amplifier performance for varying the width of amplifier region. With 50x7μm² active core and bottom mirror, only 15.8 W/cm² (8.8 W/cm² for 100μm width) of pump intensity was sufficient to meet the target operating condition. We also compare the inversion distributions of NC-Si EDWA, for the 4 types of EDWA structures under investigation (straight without/with mirror, adiabatic without/with mirror). As another key performance factor, we also calculated noise figures for different NC-Si EDWA structures.

We propose and demonstrate a highly linear-polarized hybrid integrated external cavity laser (ECL) on a planar lightwave circuit (PLC). An in-PLC polarizer is designed to improve polarization extinction ratio (PER). The in-PLC polarizer based on a 45°-tilted grating was simply formed in the laser cavity composed of a Fabry-Perot laser diode and a Bragg grating. The proposed ECL exhibited singled longitudinal mode oscillation with polarization extinction ratio of 30dB.

A grating cavity laser and a Mach-Zehnder interferometer are monolithically integrated for a tunable wavelength converter, consisting of semiconductor optical amplifiers, a deeply etched diffraction grating, a beam deflector, and passive waveguides with alternating structures. The interferometers for wavelength conversion have demonstrated at 2.5 Gb/s using the probe source from the device with an eight channel grating cavity laser which can provide wavelength tuning range of 43 nm with an SMSR > 40 dB.

The x-ray imaging technology has been improved greatly since Roentgen discovered x-ray in 1895 and it is very important in some fields such as medical treatment, nondestructive test and industrial detection. At present, there are many kinds of x-ray imaging technologies based on the different image detectors. To meet the requirement of real time observation, digital image processing and long-distance communication, digital radiography (DR) and direct digital radiography (DDR) are advantageous. The image intensifier is often used by large number of hospitals due to its affordable cost. As the key device of the x-ray imaging system, the conventional x-ray detector is mainly imported from Japan, or France because there is no domestic manufacture, which increases the cost of the whole x-ray imaging system, moreover the imported x-ray image intensifier has 9 inch visual field, thus limiting its application in some fields further. By combing the x-ray imaging technology and the low light level (L³) imaging technology, our research group designed the novel combined x-ray image intensifier, which means it is composed of the x-ray intensifying screen, zoom lens and the L³ image intensifier. In this paper, the principles on how to match the components of the x-ray image detector and how to design the structure of the x-ray image detector are introduced. At the end of the paper, the imaging performance of the x-ray image detector is given by showing the produced images and the MTF curve of the whole x-ray imaging system, which proved that the novel combined x-ray image intensifier is a good choice for many users and can be applied in many fields, such as the medical treatment, nondestructive test and industrial detection and even takes the place of the conventional x-ray image intensifier in some application fields.

In this paper, stabilization of an unstable state of a pattern forming system or selecting of arbitrary pattern formation is presented using a spatial perturbation method in the far-field configuration. Selection and tracking of unstable rolls, squares, and hexagons are demonstrated by numerical simulations through semiconductor microresonators in passive and far-field configurations.

A compact configuration of all-optical adders implemented with a single semiconductor optical amplifier (SOA) and optical bandpass filter (OBF) is presented in this paper. A comprehensive SOA model is put forward to investigate the output performance of all-optical adders. The numerical simulation results demonstrate the influence of these key parameters including input pulse peak power, pulsewidth, repetition rate, and filter characteristics. Moreover some design rules are extracted for the proper selection of these parameters so as to ensure optimum performance. The obtained results confirm the feasibility of our configuration.

A technique using a holographic optical element to split one incident laser beam into hundreds is proposed. The holographic optical element is fabricated with hexagonal packed lattice structure using 4-beam interference method. When the element is illuminated by a single laser beam with normal incidence, hundreds of beams are generated by diffraction. The element has the potential to be used as the device for interconnection and clock distribution in optical and electronic systems.

A kind of GaAs-based F-P(Fabry-Perot) cavity filter was presented. Its structure and tunability were analysed theoretically. Numerical simulation shows transmitted centre wavelength of the filter is 1.55μm, FWHM is 1.8nm and 0.6nm with 17 and 23 pairs of DBR respectively, a red shift of 7.2nm with temperature change of 100K, transmissivity almost keeps unchanged during tuning, and tuning wavelength is linear to temperature change. Further, long-wavelength-absorbed integrated photodetectors based on this filter structure was fabricated. Experimental results show a tuning wavelenghth of 10.2nm with power change of 200mW applied to the device, FWHM of about 0.6nm, and 4% quantum efficiency fluctuation during tuning. Good agreement with the simulation has been achieved.

In order to resolve the trade-off between quantum efficiency and response speed in resonant cavity enhanced (RCE) photodiode, the scheme of the transparent and unique pattern ohmic contact (TUPOC) microstructure was proposed. This scheme can be used for improving device's response speed by reducing the diode capacitance without influencing quantum efficiency. Three kinds of the TUPOC microstructures were proposed and simulated, the first one was realized. The response speed of the finished device with mesa area of 50×50μm² is remarkably increased, 3dB bandwidth of 18GHz has been demonstrated, but the device without the TUPOC microstructure only have 3dB bandwidth of 9.47GHz.

Generally, in the section of photonic crystal fiber, all air holes are arranged to a triangular regulation, when the size of air holes, the pitch of between neighboring air holes, and refractive index of background material are mapped optimally, one missing air hole in the central of section can localize optical field and form a single mode fiber. Here, each air hole is replaced by twin air holes with fixed distance and axis direction. Accordingly, we can think the central of section where twin air holes missing is the core of fiber, and optical field is guided in here. In the novel photonic crystal fiber, all twin air holes arranged according to identical axis direction in the cladding of PCF bring an asymmetry structure of section, and birefringence can come into being in this novel PCFs. After some parameters are selected optimally, the effective refractive index difference between two orthogonal directions Δn_eff can reach the magnitude of 10^-4. From the result of numerical calculation, we also can see that the birefringence parameter Δn_eff can increase slightly when the distance between twin air holes is shortened a little but keeping each air holes size and the pitch of neighboring cell composed by twin air holes.

Experimental results are reported on temporal instability of phase conjugate beam in a self-pumped Ce: BaTiO₃ phase conjugator at 532nm. The transition from stable output to unstable one is studied for various input powers, beam diameters and incident angles. Novel results that the phase conjugate output will be unsteady when control parameter P_S is within a range of 2220~3897 mW/cm² are presented. A qualitative analysis to temporal instability that instability behaviors result from competitions among the backscattering centers is given.

Single mode fiber (SMF) has been used widely in local and access network since the early 1980's. Fiber loss reduction will accelerate the construction of various transmission systems with longer repeater spacing. It has been reported that the transmission loss in SMF increases due to residual stress, which is caused by viscosity mismatch between the core and cladding material. The idea of viscosity matching is to match the viscosity of core and cladding doping to minimize the viscosity difference on the cross-section of the fiber or preform. The dopant concentrations can be chosen so that the viscosity of core and the cladding are equal in ideal step index fiber. However, all the reported viscosity-matching design is based on single dopant, for example, only GeO2 is doped in the central core and only F in the cladding. In this paper, optical loss reduction through viscosity-matched design for the SMF with GeO2-F codoped silica core and cladding is described. The impact of viscosity-matching on optical loss of silica-based single-mode fiber has been investigated in detail theoretically and experimentally based on PCVD fiber. For PCVD optical fiber, F is introduced in core is to reduce the water peak. Single mode optical fiber with low attenuation fabricated by PCVD process can be gotten through viscosity-matching design between core and cladding. Viscosity-matching can reduce the sensitivity of attenuation to drawing tension. The model for estimating the viscosity matching has been deduced, which can not only be used for conventional SMF but also for the fiber with arbitrary index profile.

A multi-information measurement system is used to activate the GaAs photocathode. During the experiment, the curves that show the change of vacuum pressure and photocurrrent are recorded also. The cathode used in the experiment is heavily p-type GaAs (100). The doping concentration is 1×10¹⁹cm^-3. The cathode is grown by molecular beam epitaxy (MBE) and the thickness is 1.6μm. GaAs cathode is degreased before being sent into ultra-high vacuum system to be heat cleaned. The activation technique is "high-low temperature" two-step activation. High temperature of heating is 600° and low temperature of heating is 410°. During the high temperature activation the integrated sensitivity is 1380μA/lm, the surface escape probability is 0.3 and the electron diffusion length is 3.1μm. During the low temperature activation the integrated sensitivity is 2140μA/lm, the surface escape probability is 0.6 and the electron diffusion length is 3.8μm.

High-performance reflection-mode GaAs photocathode (named cathode 1 for short) with the integral sensitivity of 2140μA/lm is prepared by adopting "high-low temperature" two-step activation and using heavily p-type Be-doped GaAs materials, which is grown by molecular beam epitaxy (MBE) technique. Moreover, spectral response characteristic and cathodes performance parameters of two cathodes are obtained by spectral response database we compiled, one is the reflection-mode photocathode (named cathode 2 for short) with the integral sensitivity of 1800μA/lm reported by G. H. Olsen in the 70s; the other is the transmission-mode photocathode (named cathode 3 for short) with the integral sensitivity 3070μA/lm reported by O. H. W. Siegmund in 2003. A transmission-mode cathode (named cathode 4 for short) is acquired by computer simulation on the basis of cathode 1, and its integral sensitivity is 1907μA/lm, then we compare the reflection-mode cathodes (cathode 1 and cathode 2) and the transmission-mode cathodes (cathode 3 and cathode 4), respectively, and analyze the cause for performance difference among these cathodes, the results show that the surface escape probability of cathode 1 reach to 0.62, which is lower slightly that of cathode 2, so preparation technique of cathode 1 has gotten higher the surface escape probability, but the electron diffusion length of cathode 1 and the back interface recombination velocity of cathode 4 is not better compared to cathode 2 or cathode 3. Which shows preparation technique of cathode 1 obtains better surface barrier, it need to be optimized all the same for achieving higher performance GaAs photocathodes.

The photocurrent curves and spectral response curves of GaAs photocathodes are measured by the multi-information measurement system, and the photocurrent variation has been investigated as a function of Cs/O current ratios. The identical Zn doped (1×10¹⁹cm^-3) p-type GaAs (100) wafers, identical methods of chemical cleaning and heat cleaning of wafers are used in the performed three experiments. From the experimental results, we find the envelopes of three photocurrent curves approximately satisfy parabola after the exposure to oxygen, while the detailed variation process and the ultimate photocurrent of them are different. The photocathode activated with the smallest Cs/O current ratio has the least consumed time and the largest photocurrent during the first exposure to cesium, and the most alteration times. The photocathode activated with the moderate ratio has the most rapid increase of photocurrent during the first exposure to oxygen, and has the highest quantum efficiency and stability after activation. The photocathode activated with the largest ratio has the fewest alteration times and the lowest quantum efficiency. These phenomena have a close relationship with the coadsorption mechanism of cesium and oxygen on GaAs, and in which the oxygen plays an important role. Due to the exposure to oxygen, the cesium atoms adsorbed on the surface becomes Cs⁺, their radius decrease to 1.67Å from 2.71Å, and form the dipoles with O^-2, this is the main reason of above phenomena appeared.

In this paper we review simply the surface models. These models have several technical problems not solved appropriately except for having deficiency themselves. So we present a new negative electron affinity (NEA) photocathode photoelectric emission model: [GaAs (Zn): Cs]: O - Cs. After discussing photocathodes activation technique on the model, we design a activation technique, which increases the Cs current to decrease the first peak in appropriate degree after using smaller Cs current to achieve the first peak of photoemission (GaAs (Zn)-Cs dipole layer), then set out Cs-O alternation and do not end the technique until gaining maximal photoemission (Cs-O-Cs dipole layer), in the photocathodes with GaAs (Zn) (100)2×4 reconstruction surface. In the present material configuration and level of technique, it is difficult that the integral sensitivity of cathode excess 3500 μA/lm. However, it is likely to excess 4000 μA/lm by varied doping As-rich GaAs (Zn) (100)2×4 reconstruction surface.

Photonic crystals have been widely studied in the fields of physics, material science and optical information technology. In general, the standard rectangular FDTD method is used to predict the performances of photonic crystals even if it is very time consuming and inefficient for the structures with non-orthogonal structures or inhomogeneous media. The current authors developed a software called GCFE, which is based on non-orthogonal FDTD method .The upgraded version of GCFE software can be used to calculate the photonic band structures, states density, transmission and reflection coefficients of one dimensional to three dimensional photonic crystals. It has the characteristic of efficient calculation and simple manipulation. In the present paper, the system structure of GCFE software is presented and the implementation of the algorithm module and the result display module are described in detail. Finally the band structures, transmission and reflection coefficients and photonic states density for the photonic crystal fibers with cube structures are calculated by our GCFE software and the numerical application results are also shown and discussed.

Some PCF fabrication techniques are discussed. Effects of draw conditions on the capillaries and the geometry of the final photonic crystal fiber (PCF) are investigated experimentally. The cross-sectional hole structure can be adjusted to a certain extent by controlling the parameters such as the temperature, the feed and draw speed or combinations of these. Since the improvement is limited and hardly can a satisfactory fiber be obtained, the inert gas pressurization method is introduced. It is testified feasible and effective in tuning the geometry of the final PCF and PCFs of good uniformity are fabricated experimentally.

Uncooled microbolometer infrared detectors are being developed for a wide range of thermal imaging applications. To design and manufacture high-performance microbolometer infrared detectors, numerical calculation and simulation is necessary. In this work, finite element methods are performed to simulate the transient temperature field of thermistor films of microbolometer infrared detectors. The varisized supporting legs' impacts on the performance of detectors are discussed. At the same time, variation of the bias voltage and the substrate temperature's impacts on total noise, noise equivalent to temperature difference (NETD) and detectivity (D*) are also discussed in details. These performance analyses are helpful for optimum design of microbolometer infrared detectors' structure and rational choice of working temperature of infrared focal plane arrays.

The guided optical modes in the asymmetric slab waveguide with a core of the normal dielectric surrounded by two single-negative (SNG) materials are investigated. The condition for occurring surface waves in the SNG material is analyzed. It is found that the epsilon-negative (ENG) waveguide supports both oscillating and surface guided modes, which is a new feature that the conventional waveguide does not possess. The oscillating guided modes of TM polariton are absent of fundamental mode, while the TE polariton can support the fundamental mode. For higher frequency, larger slab thickness and constitutive parameters, the ENG waveguide accommodates more oscillating guided modes. Furthermore, it is found that the ENG waveguide only supports TM surface guided modes, while the mu-negative (MNG) waveguide only supports TE surface guided modes. The existence of various stable solutions to the surface guided modes depends on the combination of the different constitutive parameters and the structures of waveguide. Finally, the transverse profiles of the surface guided modes in three different regions are obtained.

In this paper, we investigate the solitary wave propagation through the nonlinear periodic structure that consists of alternating layers of both positive and negative Kerr nonlinear coefficients along the propagation direction, in which the pulse dynamics is governed by the nonlinear coupled mode (NLCM) equations. Using the multiple scale analysis, the NLCM equations are reduced into the perturbed nonlinear Schroedinger (PNLS) type equation, which incorporates both the higher order dispersive effects and self-steepening effect. From the PNLS equation, dark solitary solutions have been constructed by an extended Tanh-function expansion method. The effects of the physical parameters for nonlinear periodic structure on soliton propagation are discussed.

In this paper, on the base of simple introduction of inner structure of 320×240 pixels UFPA in electronics and calorifics, the relationship of NETD (noise equivalent temperature difference) and bias voltage are researched and presented through the formulas about noise and NETD. The relation between NETD and four kinds of temperatures is presented. Moreover the two bias voltages are adjusted to observe the changing of NETD. Some experiments on power consumption and image quality of thermal imaging system is done, the result data is given. On the basis of the theory and experiments, how to enhance the NETD performance of UFPA (Focal Plane Array) at much lower or higher than room temperature is researched by analyzing experiment data. At last, the conclusion is summarized: in order to get the best image and the lest power consumption, we should adjust these parameters to find the optimized configuration at different application conditions.

The spectral response curves of reflection-mode GaAs (100) photocathodes are measured in activation chamber by multi-information measurement system at RT, and by applying quantum efficiency formula, the variation of spectral response curves have been studied. Reflection-mode GaAs photocathodes materials are grown over GaAs wafer (100) by MBE with p-type beryllium doping, doping concentration is 1×10¹⁹ cm^-3 and the active layer thickness is 1.6μm. During the high-temperature activation process, the spectral response curves varied with activation time are measured. After the low-temperature activation, the photocathode is illuminated by a white light source, and the spectral response curves varied with illumination time are measured every other hour. Experimental results of both high-temperature and low-temperature activations show that the spectral response curve shape of photocathodes is a function of time. We use traditional quantum efficiency formulas of photocathodes, in which only the Γ photoemission is considered, to fit experimental spectral response curves, and find the theoretical curves are not in agreement with the experimental curves, the reason is other valley and hot-electron yields are necessary to be included in yields of reflection-mode photocathodes. Based on the two-minima diffusion model and the fit of escape probability, we modified the quantum efficiency formula of reflection-mode photocathodes, the modified formula can be used to explain the variation of yield curves of reflection-mode photocathodes very well.

A novel low temperature direct wafer bonding technology employing vacuum-cavity pre-bonding is proposed and applied in bonding of InGaAs/Si couple wafers under 300°C and InP/GaAs couple wafers under 350°C. Aligning accuracy of 0.5μm is achieved. During wafer bonding process the pressure on the couple wafers is 10MPa. The interface energy is sufficiently high to allow thinning of the wafers down from 350um to about 100um. And the tensile strength test indicates the bonding energy of bonded samples is about equal to the bonded samples at 550°C.

The effect of fabrication errors on surface plasmons excitation of a silver rectangular cylinder is analyzed by boundary integral method. The scattering cross section of silver rectangular cylinders with perfectly rectangular shape and different fabrication defects are numerical investigated. The results show that the resonance wavelength is shifted towards high frequency components as the fabrication flaw is increased. It indicates that the performance of surface plasmons excitation of such structure will be degraded for various fabrication errors.

Using plane wave expansion (PWE) method and finite-difference time-domain (FDTD) method, we study the photonic band gap (PBG) and transmission properties of two-dimensional photonic crystals (PCs). A holographic fabrication design is used to construct the two-dimensional PCs in our work. To produce the lattice holographically, we use an equation to demonstrate a light intensity distribution. In this equation, there are several variant to control the lattice constant, the shape of dielectric columns, etc. The PBG calculation results using PWE method and FDTD method agree with each other very well.

This paper analyzes the thermal characteristics of VCSELs. The thermal model is based on rate-equations and has been implemented in ADS by using SDDs (Symbolically defined devices). The effects of the temperature on the light-current (LI) characteristics, small signal response and the transient response of a VCSEL are predicted and discussed.

A multi-giga bits circuit-level model of a thin film metal-semiconductor-metal photodetector (MSM PD) for high speed optical receiver (Rx) was obtained using the RF measurement technique and widely used optimization routines in simulation tool such as ADS and SPICE. On-wafer measurement-based modeling technique was employed to exactly characterize DUTs in this paper. On-wafer calibration standard structures such as NiCr 50 Ω of load, short, and open were also fabricated and optimized using laser trimming for exact measurements. The obtained circuit-level model shows good agreement with measured s-parameters and wide eye open up to 20Gbps.

We study cross-talk noise in volume holographic memory with fractional Fourier transform. For the volume holographic medium with finite dynamic range for the linearity fractional Fourier transform has an advantage over the conventional Fourier transform because it yields a spectral distribution with no high peak.

We proposed the vertical mode coupling structure (VMCS) for monolithic integration of optoelectronic devices. The electroabsorption duplexer (EAD) chip was fabricated by monolithically integrating both a waveguide photodiode (PD) and an electroabsorption modulator (EAM) in association with traveling wave electrodes. Using an EAD we presented a transceiver (TR_x) module for dual functions of both electrical-to-optical (E/O) and optical-to-electrical (O/E) conversions at 60GHz band. The responsivity and the extinction ration of the EAD were 0.72 A/W and 20 dB at -4 V_dc, respectively. The coupling loss between the optical fiber and the device facet was as small as 1.96 dB. The small signal 3 dB bandwidth of E/O and O/E response was 25 GHz and 8 GHz, respectively. We also investigated the issues of RF packaging in which the optoelectronic and electronic amplifier devices were co-packaged in a single housing.

We fabricated SiO₂/TiO₂ one-dimensional photonic crystals by a sol-gel method. A picosecond pump-probe nonlinear optical measurement was performed in the one-dimensional photonic crystal, with the pump wavelength fixed at 355nm and the probe wavelength fixed at 532nm falling on the bandgap edges. The third order nonlinear optical response in an anatase TiO₂ film composing the one-dimensional photonic crystal is found to be responsible for the nonlinear optical transmission changes at both bandgap edges.

Using the cross-gain modulation (XGM) characteristics of semiconductor optical amplifiers (SOAs), multi-functional all-optical logic gates including XOR, AND, and OR gates are successfully demonstrated at 10 Gbps by using VPI component makerTM simulation tool. Multi-quantum well (MQW) SOA is used for the simulation of all-optical logic system. Our suggested system is composed of four MQW SOAs, SOA-1 and SOA-2 for XOR logic operation and SOA- 3 and SOA-4 for AND logic operation. By the addition of two output signals XOR and AND, all-optical OR logic can be obtained.

Two-dimensional (2D) photonic quasicrystals (PQCs) were fabricated by a holographic nanolithography technique. Using two laser beams with different angles incident on the sample, microcavities with 2D internal nanostructures are patterned with a few micrometer periods.