To improve the competitiveness of solar cells, cell efficiency must increase and use of materials must be minimized. Light trapping measures can achieve this by allowing cells to absorb a greater fraction of the incident light. Traditional methods like surface texturing can negatively impact the cell's electrical characteristics and are generally unsuited to thin cell types. A plasmonic light trapping structure that avoids such issues can be formed using self-assembled hexagonal arrays of dielectric nanospheres in a continuous metal layer, at the rear surface of a cell. This can be easily fabricated toward the end of cell production, making it suitable for implementation with various solar cell types. 3D finite-difference time-domain simulations were conducted to investigate the potential of such structures, varying parameters including feature size and spacing, metal and absorber material and thickness, and the impact of random variations in the array. Significant improvements were found for a variety of topographies, with a peak increase in photocurrent from 2 μm silicon of 4.02 mA/cm 2 or 24.4%, relative to the case of a standard rear mirror with a 100 nm SiO 2 spacer layer. This also compares favorably to arrays of rear metal nanoparticles that previously yielded promising experimental results. We also identified critical parameters to control when designing such structures. A particular advantage of this structure is that it can offer light trapping advantages similar to those provided by metal nanoparticle arrays while still being able to serve as the rear contact of the cell due to the continuity of the metal layer.

Significant photocurrent enhancement has been demonstrated using plasmonic light-trapping structures comprising nanostructured metallic features at the rear of the cell. These structures have conversely been identified as suffering heightened parasitic absorption into the metal at certain resonant wavelengths severely mitigating benefits of light trapping. In this study, we undertook simulations exploring the relationship between enhanced absorption into the solar cell, and parasitic losses in the metal. These simulations reveal that resonant wavelengths associated with high parasitic losses in the metal could also be associated with high absorption enhancement in the solar cell. We identify mechanisms linking these parasitic losses and absorption enhancements, but found that by ensuring correct design, the light trapping structures will have a positive impact on the overall solar cell performance. Our results clearly show that the large angle scattering provided by the plasmonic nanostructures is the reason for the enhanced absorption observed in the solar cells.

Metal nanoparticles (MNP) supporting surface plasmon excitations have demonstrated efficiency improvements in solar cells through large angle scattering and light trapping. However, tarnishing of silver can degrade the scattering properties of the MNP, thereby limiting the potential for light trapping. In this work, we study the impact of ageing on the performance of silver MNPs over time. Our experimental results reveal that the degradation in photocurrent due to ageing can be as high as 7%, when compared to the case of freshly deposited silver MNPs. Simulation results further confirm that the degradation is indeed due to surface tarnish, which behaves as an undesirable over-coating layer. Our results highlight the sensitivity of the MNPs light trapping performance to varying properties of the dielectric material that surrounds them. It is important to prevent exposure of the MNPs to air to avoid tarnish. The use of encapsulation, over-coating, or embedding materials that have the potential to chemically alter the MNP surface should also be avoided. Experimental results reveal a method to mitigate these negative effects.