Design of Plasmonic Thin-Film Solar Cells with Broadband Absorption Enhancements By Ragip A. Pala, Justin White, Edward Barnard, John Liu, and Mark L. Brongersma* Basic design rules are developed for the use of metallic nanostructures to realize broadband absorption enhancements in thin-film solar cells. They are applied to a relevant and physically intuitive model system consisting of a two-dimensional, periodic array of Ag strips on a silica-coated Si film supported by a silica substrate. We illustrate how one can simultaneously take advantage of 1) the high near-fields surrounding the nanos- tructures close to their surface plasmon resonance frequency and 2) the effective coupling to waveguide modes supported by the thin Si film through an optimization of the array properties. Following this approach, we can attain a 43% enhancement in the short circuit current as compared to a cell without metallic structures. It is suggested that 3-dimensional nanoparticle arrays with even larger boosts in short circuit current can also be generated using the presented framework. Photovoltaic (PV) cells can provide virtually unlimited amounts of energy by effectively converting sunlight into clean electrical power. Silicon has been the material of choice for PV cells due to low cost, earth abundance, non-toxicity, and the availability of a very mature processing technology. The cost of current PV modules still needs to be significantly reduced and efficiency substantially increased to enable large scale implementation. Thin-film, second-generation Si solar cells may provide a viable pathway towards this goal because of their low materials and processing costs. [1] Unfortunately the materials quality and resulting energy conversion efficiencies of such cells are still substantially lower than crystalline, wafer-based cells. This is a direct result of the large mismatch between electronic and photonic length scales in these devices; the absorption depth of light in Si is significantly longer than the electronic (minority carrier) diffusion length in deposited thin-film materials for photon energies close to the band-gap. As a result, charge extraction from optically thick cells is challenging due to carrier recombination in the bulk of the semiconductor. If light absorption could be improved in ultra-thin layers of active material it would lead directly to lower recombination currents, higher open circuit voltages, and higher conversion efficiencies. Conventional, planar anti-reflection (AR) coatings do not provide high transmission efficiencies over the entire solar spectrum and do not enable effective light trapping to increase absorption. Light trapping schemes using diffusely scattering surface textures were first suggested in the 1980s and are by now fairly-well understood. [2,3] Texturing surfaces of thin film cells is not ideal as it leads to enhanced surface recombination. For this reason, some interesting alternative trapping configurations have been proposed that utilize structuring at length-scales orders of magnitude larger than the cell thickness. [4] More than a decade ago, it was first proposed to use the unique optical properties of metallic (i.e., plasmonic) structures to boost the efficiency of PV cells; [5,6] those metallic nanostructures exhibit easily accessible collective electron oscillations known as surface plasmons. Surface plasmon excitations enable unparalleled light concentra- tion and trapping. Since these pioneering efforts, plasmonics has also been used to enable new photodetector designs that exploit lateral and in-depth light concentration to increase their signal-to-noise ratio and speed in the visible, near-IR, and mid-IR wavelength ranges. [7–9] Recently, the use of metallic nanostructures for PV has received renewed attention with the availability of new nanofabrication tools and the growing understanding of their optical properties provided by the burgeoning field of plasmonics. [10,11] In different cell designs both near-field light concentration close to the individual particle resonance and effective light trapping by nanometallics have been explored. [12–20] Experimentally, high peak enhancements in the tens of percent range at specific wavelengths [19] and overall efficiency enhancements of 40%, 8.3%, and 8% have been achieved with the use of plasmonic structures for cells employing organics, [21] a-Si, [16] and GaAs, [22] respectively. Separate efforts have focused on increasing the more omni-directional absorption characteristics for solar tracking and operation in diffuse sunlight. [23] These results are very promising, although no detailed comparisons have yet been made to cells employing alternative light trapping technologies. Moreover, there is a clear need for effective optimization strategies that lead to broadband absorption enhancements over the entire solar spectrum. This type of optimization for nanostructured solar cells is now within the realm of possibilities; the recent advances in full-field electromagnetic simulations and computer hardware have resulted in the development of extremely accurate and robust optimization tools that are now commonly used by the PV community. [24–26] In this paper, we illustrate a straightforward and physically intuitive procedure to optimize the net overall absorption of a thin-film Si solar cell over the entire solar spectrum; this simultaneously takes advantage of 1) the high near-fields surrounding the nanostructures close to their surface plasmon COMMUNICATION www.advmat.de [*] Prof. M. L. Brongersma, R. A. Pala, J. White, E. Barnard, J. Liu Geballe Laboratory for Advanced Materials Stanford University Stanford, CA 94305 (USA) E-mail: brongersma@stanford.edu DOI: 10.1002/adma.200900331 3504 ß 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2009, 21, 3504–3509