The procedure for demonstrating the reliability of laser diodes used in telecommunications equipment is documented in Telcordia document GR-468-CORE, Issue 1, December 1998. This procedure was intended for low-power single-mode diodes used in fiber-optic transmitters. A small section was added to the procedure to address high-power, single-mode diodes used as optical pumps for fiber amplifiers. Due to the invention of the double-clad fiber optic cavity, the next generation of fiber amplifiers will likely utilize high-power, multimode diode lasers as optical pumps. In this study, we report on the reliability assurance of multi-spatial mode laser diodes emitting 1 Watt from a 100 micron aperture at 915 nm. While additional study is required to verify the results, the present data indicate a total failure rate, combining both sudden and random failures, of less than 1000 FITs for an operational lifetime of 20 years.

We describe a prototype night vision system for automotive applications that uses a high power near-infrared (NIR) diode laser, compact optics, and a CCD camera. Because NIR radiation is invisible to the human eye, a high-beam illumination pattern can be formed permitting a clear view of objects on both sides of the roadway, even in the presence of oncoming traffic. A narrow band-pass filter in front of the camera passes only the laser wavelength and prevents blooming of the image due to the headlights of other vehicles. This system permits drivers to see objects at night (such as debris or pedestrians) that are in close proximity to oncoming vehicles. The diode laser operates at 810 nm and emits 5 - 10 W. The illuminator distributes the laser light using a combination of refractive, reflective, and holographic optics in a manner that meets the standards for Maximum Permissible Exposure. We discuss the performance of our prototype system as a function of laser power and camera field-of-view and sensitivity, and we provide comparisons with a commercially available automotive night-vision system that uses a thermal-imaging camera.

Although light-emitting diodes exhibit much higher efficiencies and greatly reduced power consumption compared to incandescent light sources, the use of LEDs in lighting applications is limited by their smaller size and subsequently lower light output. However, it has been found that these parameters can be increased significantly by cooling the diodes to cryogenic temperatures. This may make their use feasible for several applications requiring more efficient and brighter illumination for much less cost. In this paper, we compare the temperature-dependent behavior of five commercially available LEDs of different wavelengths down to liquid nitrogen temperatures. It was found that three AlInGaP diodes (red, yellow, and green) demonstrated significant operating improvements. The performance of InGaN-based blue LEDs declined at low temperatures, and because most white LEDs are simply blue LEDs coated with YAG, these exhibited similar behavior. However, the three AlInGaP LEDs demonstrated at least an order of magnitude improvement in illuminance, absolute intensity, and maximum operating current. The green LEDs showed the largest improvement factors, while the yellow LEDs produced the brightest illumination at low temperatures. The emissions of all five LEDs shifted to shorter wavelengths at low temperatures. This is significant in terms of lighting applications since the low-temperature AlInGaP diodes emitted more visible spectra.

The pulsed time-of-flight laser range finding techniques based on laser diode transmitter enables one to realize a mm-level accuracy to non-cooperative targets in a distance measurement range of several tens of meters in industrial applications such as the measurement of level heights in silos, positioning of tools and vehicles, velocity measurement, anti-collision radars, proximity sensors, etc. In this work the basic pulsed time-of-flight laser radar functions, the receiver channel and the time interval measurement unit, have been realized in the form of high-performance full-custom integrated circuits, which should pave the way for realizing a laser radar eventually as a component-like micro system thus increasing the number of possible applications for the developed techniques. The integrated BiCMOS receiver channel produces a logic level output pulse for the time interval measurement from the received laser echo. Two versions were realized, one detecting the leading edge of the received pulse and achieving an accuracy of +/- 35 mm in a dynamic range of 1:4000. In the other version gain control and constant fraction type of timing detection techniques are used to enhance the accuracy. With this circuit an input dynamic range of 1:650 (SNR > 10) can be achieved with a timing accuracy of about +/- 3 mm. The developed fully integrated CMOS time-to-digital converter realization is based on a counter and a novel parallel two- step interpolation method. The single-shot precision and measurement range of the unit are 50 ps and 2 microseconds, respectively. In averaging mode the linearity is better than +/- 5 ps. Even better single-shot precision can be realized with analog interpolation techniques, however, at the expense of reduced stability.

We propose for the first time the in-situ frequency discriminating and continuous phase-tuning of the microwave signal or pulse-train generated from directly modulated laser diode with respect to a free-running microwave clock by integrating a DC-voltage controlled optoelectronic phase shifter (OEPS) with the laser source. This technology facilitates the combination of the frequency synchronization and the phase shifting functions in one circuit. The transferred function of the phase shift versus the controlling voltage is linear with a maximum phase-tuning range and a tuning slope of up to 3.6(pi) (640 degrees) and 90 degrees/volt, respectively. The fluctuation and drift in phase of the controlled signal is about 0.05 degrees and 0.003 degrees/min. The tuning resolution of 0.2 degrees at a 3-mV increment of the controlling voltage is achieved by using a high-precision voltage regulator. Relative timing jitter of the controlled optical microwave clock is less than 5 ps. By using the delay-time tunable pulsed laser, we demonstrate for the first time the delay-line-free electro-optic sampling of the waveforms, which are RF sinusoidal and pulse signals with repetition rate of 500 MHz. The maximum sampling range and highest sampling resolution are about 1.9 periods and 0.2 ps/mV, respectively.

One of the main reasons for result the pantograph-catenary system fault in electrified railway is the contact wire zigzag value exceeding quota. In this paper, an infrared LED based photoelectric detecting system which is directly installed in the electric locomotives is presented and the method of dynamic measuring of the contact wire bias is realized. The system designed which consists of IR-LED and corresponding photo-detectors, encoder, signal transmission channel insulating from high to low voltage side, decoder and data processing device is described in details. It is shown that the zigzag value examining sensors are mounted on the head of pantograph, each neighbor unit get interval distance of 10 mm and there are 120 units altogether. The signals are output when the overhead line slipping over the slide's surface of the pantograph. The experimental results of the zigzag value obtained from the field test are reported and it has been practically proved that measuring error < 0.5%. The results are of important significance for the dynamic performance analysis as well as fault diagnosis for the pantograph- catenary system.

The presence of surface states, which pins the Fermi level within the bandgap, contributes to the degradation in the performance of light-emitting diodes (LEDs) and p-i-n photodiodes due to an increase in the non-radiative recombination rate. Chemical modification on the facet or device surface can greatly enhance the output power of LED's and photocurrent of p-i-n photodiodes. Adsorbing molecules can change either the density or energy distribution of surface states. This effect leads to changes in surface recombination which result in systematic variations in light output, and thus the effect can be used for detection of analytes. This mechanism was used to realize a compact chemical sensor based on III/V LED/Detector structures. Initially, we fabricated InGaAlP/InGaP/InGaAlP double heterostructure (DH) LEDs (400 X 1000 micrometer²) with three different active region thicknesses: 50, 250 and 500 nm. In constant current mode, the DH LED exhibits electroluminescence (EL) at approximately 670 nm for an InGaP active region. The EL intensity changes of the LED in various gaseous ambients (NH₃, NH₂(CH₃), NH(CH₃)₂, N(CH₃)₃, and SO₂) are measured. The data show reproducible trends: DH LED structures with thicker active regions result in larger emission intensity changes due to analyte adsorption. Our findings are consistent with active-layer surface area dependence. Thicker active layer devices have larger carrier losses due to nonradiative surface recombination, and thus show a stronger sensitivity to the surface chemistry. Furthermore, we used this DH LED design to build a highly versatile compact sensor. The MOCVD-grown LED wafer is patterned and chemically etched to fabricate integrated GaAs/AlGaAs edge-emitting LEDs and p-i-n photodiode units. The light emitted from the edge-emitting LEDs is absorbed at the sidewall of the adjacent photodiode, and the resulting photocurrent is measured. The device design concept is based on increasing the ratio of analyte-accessible facet area to the volume of the active region. This integrated LED- photodiode device can serve as an on-line chemical sensor for a variety of analytes.

Recent developments in the use of diode lasers as ultra sensitive sensors are presented. Particular emphasis is placed on higher harmonic detection using the wavelength modulation technique. Experimental results are presented illustrating that this method, which uses a compact, portable apparatus, yields measurements with very high precision. Theoretical analyses of these experimental results are also provided. Novel ways of extracting detailed information of the probed sample are described. These include measurements of fine features, as well as simultaneous measurements of multiple overlapping lines. Such fine features allow one to accurately probe the collision dynamics of gaseous samples yielding values of temperature, density, and pressure. The method described here utilizes several novel features of higher harmonic detection. These features include a set of self-consistent measurements, by employing coherent detection up to the eighth harmonic order. The highest harmonic detection order achievable with this technique is limited by the bandwidth of the synchronous phase-sensitive demodulation apparatus and the noise figure for the detection system. Applications in basic science, in pollution monitoring and detection, and in industrial environments are briefly discussed.

We have stabilized frequency of 35mW 680nm visible laser diode (LD) by using simple frequency stabilization system based on optical-electrical double feedback method. For the long-term frequency stabilization, frequency of the LD was stabilized to a reference confocal Fabry-Perot cavity (CFP cavity) by negative electrical feedback to the injection current of the LD based on Pound-Drever technique. By employing optical feedback from another tilted CFP cavity along with electrical feedback, the residual frequency noise has been efficiently suppressed. The achieved PSD of frequency fluctuation under optical-electrical double feedback condition was less thani x 1 05[Hz2 1Hz] within the Fourier frequency <10 MHz. The reduced linewidth under the double feedback condition is estimated to be narrower than at least 400 kFIz, which seems to be the resolution limit of our measurement. The minimum square root of the Allan variance is 3 .9 x 10' at the integration time of 0.1 msec under the double feedback condition.

Optical triangulation displacement sensors are widely used for their non-contact measurement characteristics, sub-micron order resolution, simple structure, and long operation range. However, errors originating from surface inclination, speckle effect, light source fluctuation, and detector noise limit the wider use. In order to minimize these errors, the structure for optical triangulation displacement sensors, which is composed of an incoherent source and a linear CCD, has been proposed. But using a linear CCD causes several problems in signal processing. In this paper, we propose an adequate signal processing system for the proposed structure. With the help of the proposed algorithm, the limited resolution problem of CCD can be solved.

Optical triangulation displacement sensors detect linear displacements of objects without mechanical contact. They have simple structure, good resolution, and long operating range. However, there are several errors generated from speckle effects, environmental effects, and electronic noises, etc. To reduce errors from the electronic noises, the easiest way is to average the measurement outputs. Because the electronic noises are random in nature, their variance can be reduced with the averaging operation. However, this method is inherently time consuming process. To decrease the averaging time, several sensors or better signal processing hardwares are needed. So it increases the size of the measurement system and is not costeffective. In this paper, we propose a simple and cost-effective system structure for optical triangulation displacement sensors, which simplifies the averaging by inserting a transmission-type diffraction grating. When an incident ray enters to the diffraction grating, the grating separates the incident ray into several rays by the diffraction effect. The diffraction grating helps us to attain several signals simultaneously. Theoretical analysis is given and the feasibility of the proposed system is verified through experiments.

Negative luminescence (NL) operation at ∼4 µm for p-InAsSbP/n-InAs and 4÷5 µm for p-InAsSb/n-InAsSb(P) diodes with FWHM∼0.1⋅λmax is reported for a reverse bias operation mode. NL output at 180oC is as high as 3÷5 µW and negative apparent temperature contrast is as strong as ∆T= - (6÷10°C) which show advantages of InAs and InAsSb based NL devices for high temperature applications. The remarkable feature is the uniformity of spatial NL output distribution, which is a confirmation of the existence of a potential barrier in narrow band p-n junction at elevated temperatures.

The coded IR proximity detector is able to generate an electrical signal when the distance between the detector and a certain object becomes shorter than a preset value. The detector may be used as an automotive indicator and control. The detector consists of four parts : a controller, a transmitter, a receiver and the optical system, used both for transmission and reception. The transmitter contains a high-power laser diode driven by a recurrent sequence of position coded pulses. The receiver consists of a PIN photodiode amplifier module, a decoder and an output stage. The outgoing beam is separated from the incident (reflected) beam using a beam-splitter. A CMOS microcontroller comands the generation of the coded pulses, the reception and the decoding of the received pulses. The device features a very low power consumption due both to the pulsed regime and to the use of CMOS technology of the controller. A personalization through code for each detector is possible due to the use of the microcontroller, avoiding unauthorized or parasite commands from other similar devices in the neighborhood. The device’s detection range is 10-15 m, enabling its use in military and robotics applications.

A fiber optics explosion primer is proposed to remove the major drawback of the electrical cable links by transmiting of the ignition primer signals through an optical fiber cable, insensitive to the electromagnetic induction generated by the atmospheric electric discharges. The primer consists of a transmitter, an optical cable and a receiver. The system has some features that assures a completely safety operation. The first feature is the signal encoding and the microcontroller supervision. Both transmitter and receiver are microcontroller assisted and all comands for the receiver operation and priming are coded and password-protected. The other feature is the absence of the power source (battery) on the receiver, preventing from accidental priming. The receiver power supply is replaced by a capacitor that is charged by the energy of the light beam coming through the optical cable to a PiN photodiode operating in photovoltaic regime. The primer operates in two steps. In the first step the capacitor is charged and all devices of the receiver are powered, while in the second step all the optical signals coming through the fiber are considered as commands. Wrong or unauthorized (nonmatching passwords) commands reset the receiver and cut off the power supply by discharging the capacitor.

The component, principles as well as practical applications for photoelectric encoding transmission system are described in detail. The infrared LED and corresponding JR receiver were used in the system to realize E-O and O-E transformation. The inspected data are transmitted by JR-LED after being converted and encoded to serial signals. The infrared modulated light which is obtained by JR receiver are decoded and converted to the original electric signals. Thereby, air-isolated photoelectric data transmission comes true. Signals are picked up with 8098 single-chip microcomputer, then transmitted to the computer via R5232 for further processing. Test results are reported which were obtained from the dynamic parameter measurement of catenary and pantograph system in electrified railway.

Novel single-mode VCSEL designs based on antiguided structures are discussed. Modal characteristics of single-element, 8-12m-dia, ARROW-type VCSELs and two-dimensional phase-locked arrays (4x4 elements) of antiguided VCSELs are studied. ARROW-type VCSELs exhibit low loss for the fundamental mode and strong high-order mode suppression as a result of edge radiation leakage. Preliminary results are 1mW single-mode (thermally limited) from 12im-dia devices. Two-dimensional arrays of antiguided VCSELs are shown to exhibit resonant coupling if the interelement spacing corresponds to an integer number of half-waves of the lateral radiation leakage. Single-mode emission with all array elements locked "in-phase" is demonstrated for the first time from a monolithic 2-D array of VCSELs. Large phase-locked arrays have potential for <100mW of single-mode output power with further optimization and efficient heatsinking.

Ultra high power 14XX (1400-1520) nm lasers are required for erbium doped fiber amplifiers (EDFAs) and Raman amplifiers in dense wavelength division multiplexing (DWDM) systems. To realize both high-power and high-reliability performances, we introduced a novel concept in laser diode (LD) chip design in terms of energy conversion efficiency. We also investigated the dependency of laser cavity length on LD characteristics to minimize the total power consumption of both the LD-chip and the LD-module with thermal electric cooler. Using this concept, we successfully reduced the total power consumption of the LD-module by 2Watts, comparing with an old-generation chip design. We have developed the next-generation chips for ultra high power LD-modules with over 300mW fiber-coupled-power in the range of 1400nm-1520nm lasing wavelength, which are suitable pumping sources for Raman amplifiers. The stable operation over 4000hrs in the reliability test under 60 degree C was obtained at 90% of maximum rollover power, indicating that the next-generation chip is robust enough to commercial use. In addition, the cavity length dependency on the reliability performances has been studied. We found that activation energy was 0.62eV, which is consistent with any cavity lengths, and that no degradation of reliability is observed as the facet optical power increases.

High power InGaAs multi-mode broad area semiconductor lasers emitting between 190 nm and 980 nm are required as optical pumps for Er+ and Yb+ doped double clad fiber lasers and amplifiers. In this paper, we present performance and reliability of two generations of 100 micrometer aperture broad area devices emitting at 920 nm and 970 nm. The first generation devices have been deployed in the field with up to 2.5 W ex-facet optical power. More than 500,000 device-hrs of actual multi-cell lifetest data, and nearly 100 million accelerated device-hrs have been accumulated with 91FIT at 1.2W and 25 degrees Celsius or 1.9 million hrs MTBF at 2W and 25 degrees Celsius. A next-generation design further reduces thermal resistance, optical loss, and far-field divergence resulting in up to 4W ex-facet CW output power with superb reliability. Multi-mode fiber coupled modules demonstrate high coupling efficiency due to the reduced divergence angles of the new design. Lifetest of the new generation devices demonstrate the reliability of 167 FIT at 2W and 25 degrees Celsius or 499,000 hrs MTBF at 4W and 25 degrees Celsius.

High-power diode lasers serve as small and highly efficient laser-beam sources. Their output power has been increased steadily over the recent years. Nowadays, the output power of one diode laser bar is sufficient for many different applications such as pumping of solid state lasers, direct material processing, medical applications or printing. The successful use of high-power diode lasers depends on their high reliability in combination with a long lifetime. For a further increase in the quality of high-power diode lasers the properties of semiconductor lasers have to be improved as well as the cooling and mounting techniques for these bars onto specially designed heat sinks.

Tunable lasers are becoming critical in DWDM systems for reasons of increased system functionality, system adaptability and costs. Key issues that arise are wavelength tuning range, characterization of the devices, wavelength control, mode stabilization, wavelength switching times, output power and long term stability of operation. We have developed tunable EML components and transmitters that address all these issues. Transmission at 2.5 Gb/sec over 640 km of fiber has been demonstrated using an EA-DBR capable of being tuned to any one of 20 wavelengths spaced at 50 GHz. These lasers are integrated into a single package with wavelength stabilization elements and can be stabilized such that both the desired wavelength and mode are maintained during operation. These integrated modules are also incorporated into a small form factor transmitter capable of operating in DWDM systems.

We report on a novel flip-chip packaging technology capable of interconnecting devices packed at very high density. The process utilizes micro-machined cantilevers for establishing electrical contact, where package assembly is performed at room temperature without solder. The cantilevers, called micro-springs, are fabricated by sputtering, masking, and releasing a stress-engineered conductive thin film on a quartz substrate. The film is patterned into electrical routing wires whose ends are released from the substrate. Upon release, the film stress forces the ends to curl up into compliant springs. Packages are formed by pressing the micro-springs against a set of device contact pads, much like probing pads using tungsten needle probes. The connections between springs and contact pads are anchored by an encapsulating acrylic adhesive. We utilize this packaging technology to interconnect 200-element arrays of independently addressable VCSELs with 4 micrometer-wide pads on 6 micrometer pitch to silicon CMOS driver chips with equally dense output lines. Tests show the technology produces good contacts with excellent robustness.

Optical Time Division Multiplexing (OTDM) is one of the important techniques by which data transmission capacity (aggregate bit rates) can be increased enormously. In this work, we demonstrate an 80 Gb/s time division multiplexing and demultiplexing experiment. By gain switching a DFB laser with 10 GHz RF signal we produced approximately 7 ps FWHM pulses. By using TDM technique we obtained 80 GHz signal. The demultiplexing of this TDM signal is carried out using two cascaded LiNbO₃ Mach-Zehnder modulators.

MSM photodiodes attracted attention due to their high-speed performance and ease of integration, but this interest has waned recently. This paper endeavors to explore why this occurred and tries to address these issues. MSM photodiodes have a much lower capacitance per unit area than p-i-n photodiodes, and are often transit time limited. MSM photodiodes are comprised of back-to-back Schottky diodes using an interdigitated electrode configuration on top of an active light collection region. The transit time is related to the spacing between these interdigitated electrodes. MSM photodiodes are more easily integrated with pre-amplifier circuitry than p-i-n photodiodes. One reason is that MSM photodiodes do not require doping which eliminates any parasitic capacitive coupling between the photodiode and doped regions within the active transistors. Another reason is that the Schottky electrodes of the MSM photodiodes are essentially identical to the gate metallization of field effect transistors (FET), which might eliminate one photolithography step. But, MSM photodiodes suffer from very low external quantum efficiency (EQE) and high leakage currents. MSM photodiodes exhibit low EQE because the metallization for the electrodes shadows the active light-collecting region. Shadowing can limit the incident light from reaching the active region of the MSM detector and prevent an ideal MSM from achieving high EQE. Transparent conductors have been shown to nearly double responsivity. Leakage currents are determined primarily by the Schottky barrier heights. These can be unreliable. However, thin wide bandgap cap layers can be inserted below the Schottky and different metals used for the anode and cathode to break symmetry and to circumnavigate these concerns.

Light output degradation of nonhermetic packaged AlGaAs LEDs has been characterized, following application of accelerated stress testing of the LED component. Our data have been applied to a previously reported 'Black Box' acceleration model, to determine the model coefficients for temperature, ambient humidity and current biase of the LED device. The stress testmparameters included 3 levels of moisture, between 55% and 100%, 3 elevated temperature levels between 35 degrees Celsius and 121 degrees Celsius, and 3 current bias conditions between 0mA and 40mA. Where device failure has not occurred during the test period, the time to failure (TTF) for an individual part has been projected to the time at which 50% reduction in the light output power will occur. The TTF data was then fit to a lognormal distribution to obtain the mean time to failure (MTTF) for each stress level of the study. The dependence of MTTF on temperature and relative humidity is assumed to be, C₀(DOT)exp[Ea/KT](DOT)exp[-Arh(RH)²]. The activation energy (Ea), and relative humidity coefficient (Arh) were determined from our accelerated test data. Our results show different behavior when testing with no in-test current bias, indicating a different physical mechanism for degradation under these conditions. Finally, physical analysis of aged devices point to a moisture drive corrosion mechanism, leading to eventual failure of the LED.

The effects of gamma radiation on rare-earth doped optical fibers have been investigated over the range of 0.01 to 145 kilorads(Si) per hour to total doses exceeding 100 kilorads(Si). The effects of 3 to 4 MeV protons have also been investigated. The level of radiation induced damage has been found to be strongly dependent on radiation dose rate and fiber composition. The existence of dose rate dependence is explained by the process of simultaneous creation of color center defects and annihilation via thermal annealing. Through the use of a suitable kinetic analysis technique, we have determined appropriate rate constants for this process and have developed an empirical model based on our experimental results. In addition to these results, we present the effects of gamma radiation on the performance of operational high power erbium-ytterbium optical amplifiers.

We report on the generation of a 100 GHz train of pulses from an Er-doped fiber laser, by using rational harmonic active mode-locking. The pulse width is approximately 4 ps. The supermode noise in the spectrum is reduced significantly with a semiconductor optical amplifier in the cavity.

Within silicon silver is an impurity with fast diffusivity and deep levels. It forms effective recombination centers in silicon acting as either acceptor or donor levels. That has been confirmed by a depth profile analysis with the SIMS. The silver atoms do exist near the barrier region of a solar cell with Ti/Pd/Ag electrodes heated at 245 degrees Celsius for 308 hours. The open circuit voltage at low injection decreases as recombination actions increase in the barrier region. According to these phenomena, an estimation for the lifetime of solar cells is given by using acceleration tress tests based on Arrhenius equation.

Rotating polarizer and rotating retarder plate polarimeters are widely used in high-resolution polarimetry, for example in remote sensing, fiber optic measurements and biomedics; as a consequence the analysis of the performances of these devices is very important from the instrumental point of view. To compare the two methods, we developed a synchronous polarimeter based on a mechanically rotating stage, where a rotating Glan-Thompson linear polarizer or a wave retarder can be easily mounted. A specific design allows to acquire synchronously the intensity signals digitally process the data to extract the polarization Stokes parameters. We investigate the two cases along with their impact on measurement techniques. Performance curves are shown for various polarization input parameters and light levels. Specifically, we address issues concerning the accuracy and the systematic and statistical measuring errors. Moreover, computer simulations and measurement results are presented and discussed.