ABSTRACT
A test method that measures spectrally resolved irradiance distribution for a concentrator photovoltaic (CPV) optical system is presented. In conjunction with electrical I-V curves, it is a means to visualize and characterize the effects of chromatic aberration and nonuniform flux profiles under controllable testing conditions. The indoor characterization test bench, METHOD (Measurement of Electrical, Thermal and Optical Devices), decouples the temperatures of the primary optical element (POE) and the cell allowing their respective effects on optical and electrical performance to be analysed. In varying the temperature of the POE, the effects on electrical efficiency, focal distance, spectral sensitivity, acceptance angle and multi-junction current matching profiles can be quantified. This work presents the calibration procedures to accurately image the spectral irradiance distribution of a CPV system and a study of system behavior over lens temperature.
ABSTRACT
Lithographic fabrication may be used to define laterally coupled gratings of high refractive index contrast on waveguide ridges, eliminating the need for regrowth steps in such distributed feedback lasers. These may be made more amenable to fabrication by employing higher-order gratings. Reliable exploration of the laser design space requires that the radiating partial waves be accurately incorporated in numerical simulations. We modify the coupled-mode approach to fully consider the two-dimensional cross section, analyzing rectangular, sinusoidal, triangular and trapezoidal grating shapes. Effective coupling coefficients are determined for grating orders from first to third. We show that, by tailoring the grating microstructure, effective coupling coefficients up to double that of a 0.5 duty cycle rectangular grating can be achieved. The actual grating microstructure of an as-fabricated grating was analyzed and its effective coupling coefficient predicted as [Formula: see text]. This was found to be in excellent agreement with the value extracted from the amplified spontaneous emission spectrum, [Formula: see text].
ABSTRACT
We observed for the first time a strong angle-independent modification of spontaneous emission spectra from laser dyes in photonic crystals, made of inverse opals in titania. Comparison with spectra from such crystals with much smaller lattice spacing, for which emission is in the long wavelength limit, reveals inhibition of emission up to a factor approximately 5 over a large bandwidth of 13% of the first order Bragg resonance frequency. The center frequency and bandwidth of the inhibition agree with calculated total density of states, while the measured inhibition of vacuum fluctuations is much larger. Because of the specific location of the dye molecules, we likely probe the strongly modulated local photonic density of states.
ABSTRACT
The character of wave transport through a strongly scattering medium, excited by a pulsed plane-wave source, is investigated as a function of sample thickness over the range from about one to 13 mean free paths. To examine the behavior theoretically, we perform a first-principles calculation of both the frequency correlation function of the transmitted field and the time-domain profile of the transmitted intensity. These quantities are investigated experimentally using an ultrasonic technique, which allows us to separate the ballistic and scattered components of the total transmitted field, and hence to measure the scattered component unambiguously in thin samples. For sample thicknesses greater than about four mean free paths, we find good agreement between our theory, the diffusion approximation, and our experimental data for both the frequency correlation function and the intensity time profile. In thinner samples, there are systematic differences between theory and experiment. To characterize the transition from ballistic to diffusive behavior in thin samples, we focus on the arrival time of the peak in the scattered component of the transmitted intensity; unexpectedly we find that the scattered peak arrival time exhibits an abrupt crossover between ballistic and diffusive behavior when the ratio of sample thickness to mean free path, L/l, is approximately equal to 3. Excellent agreement is obtained between our theory and experiment for this crossover behavior over the entire range of sample thicknesses investigated.
