p-i-n and p-n betavoltaic cells are modeled combining Monte Carlo and drift-diffusion simulations to assess device performance under a nickel-63 radioactive source. Semiconductor layer thicknesses and doping were optimized using a particle swarm optimization algorithm to maximize the power output. Simulations show that beta particles can penetrate deeply in semiconductor materials, creating electron-hole pairs up to 10 µm deep. A current gain multiplier of 6500 and 5100, compared to the beta particle flux, were demonstrated, respectively, for the optimal p-i-n devices and p-n devices. We show that the p-i-n heterostructures present an enhanced carrier collection efficiency over their p-n counterparts due to their wider depletion region, and that introducing an intrinsically doped region enables an increase in betavoltaic device power output by 9 %.

Photonic power converters (PPCs) are photovoltaic cells that convert monochromatic light into electric power. The impact of luminescent coupling (LC) on InGaAs-based PPCs is studied. Multi-junction PPCs are simulated using an experimentally validated drift-diffusion model, and the contribution of LC is quantified. Up to 85% of the photons emitted across the InGaAs layers are re-absorbed in the dual-junction device considered. This number increases to 96% when a back reflector is included due to improved light management. Interference effects produced by multiple reflections are examined as a function of the emission angle.

We have developed a machine learning empowered computational framework to facilitate design space exploration for optoelectronic devices. In this work, we apply dimensionality reduction and clustering machine learning algorithms to identify optimal ten-junction C-band photonic power converter (PPC) designs. We outline our framework, design optimization procedure, calibrated optoelectronic model, and experimental calibration devices. We report on top performing device designs for on-substrate and flat back-reflector architectures. We comment on the design sensitivity for these PPCs and on the applicability of dimensionality reduction and clustering algorithms to assist in optoelectronic device design.

Photovoltaic devices containing InGaAs absorbers, lattice matched to InP, have shown excellent performance in many applications. We explore this material’ potential in photonic power converter (PPC) applications. We have developed a machine learning empowered computational framework to explore design space for optoelectronic devices. We apply dimensionality reduction and clustering machine learning algorithms to identify optimal ten-junction C-band PPC designs. We introduce a photo current figure of merit to find optimal designs with low computational cost. We compare modeled performance to experimental devices for 1, 2 and 10 junctions. We discuss the role of luminescent coupling in devices under test, and show increased effect when devices have a back reflector. We compare results obtained using dimensionality reduction with traditional Beer-Lambert calculations.

Wavefront distortion of free-space optical beams propagating through atmospheric turbulence leads to signal fading in communications applications. The wavefront distortion can be compensated by adaptive optics, but conventional methods based on wavefront sensing may fail to ensure adequate signal coupling in typical communications scenarios. Alternative methods rely on focal plane intensity measurements. In this work, we apply computational imaging using a digital micromirror device to reconstruct the focal plane intensity image of a beam distorted by simulated atmospheric turbulence generated by a phase-only spatial light modulator. The image fidelity of the experimental focal plane images is compared to calculated point spread functions as a function of the compression ratio for undersampled images.

Under extreme solar concentrations, series resistance is a primary solar power loss mechanism for concentrator photovoltaic modules. In this paper, a detailed balance based approach demonstrates the mitigation of these losses by the application of subcell segmentation, a current matching architecture which divides each subcell into multiple pn junctions. By increasing the number of series-connected junctions, power extraction shifts from high current to high voltage, thus shifting theoretical efficiency peaks from hundreds to thousands of suns. Simultaneously, the high-efficiency bandgap design space expands across all concentrations by leveraging the current matching potential of subcell segmentation.

The ternary alloy silicon germanium tin is a versatile candidate to extend the industry standard lattice matched InGaP/InGaAs/Ge multijunction solar cell to four junctions. Here, the SiGeSn composition space is discussed and its bandgap trend is visualized. Then, InGaP/InGaAs/SiGeSn/Ge solar cells are simulated using drift-diffusion modelling to ascertain SiGeSn quality limits, and the design challenges in the four-junction material system. Power conversion efficiencies of 42.6% and 41.6% at 1000 suns AM1.5D are determined for designs implementing surface recombination velocities of 103 cm/s and 5×104 cm/s, respectively, at important interfaces in the device. These signify absolute efficiency gains of 1.3% and 0.4% with respect to like-modelled InGaP/InGaAs/Ge designs. The obtained power conversion efficiencies assume a Shockley-Read-Hall recombination lifetime of 1 µs in the SiGeSn material, however, lifetimes of 100 ns drop the efficiency by only ~1 % (absolute). The external quantum efficiency of the four-junction devices is near 90% across most of the solar spectrum. A plot of the fraction of incident light lost to various physical mechanisms in the solar cell is given and has been used to optimize the surface field layers to reduce minority charge carrier loss currents. Designs have been optimized for output power by thinning the top three subcells to ensure that the germanium does not limit the device’ 11.25 A/cm 2 operating current at its maximum power point. This result indicates that current-matching limited by inefficient absorption and Auger recombination in the Ge subcell is one of the main design challenges of this material system and suggests avenues for possible improvements to the design, such as improved light trapping, refined bandgap engineering, and subcell segmentation.

Detailed balance calculation of subcell segmentation in the extension of the InGaP/InGaAs/Ge system to 4-subcell designs shows performance above the level of the standard optimal bandgap is achievable in conjunction with other practical advantages.