Struct Multidisc Optim DOI 10.1007/s00158-014-1077-z RESEARCH PAPER Topology optimization for highly-efficient light-trapping structure in solar cells Shuangcheng Yu · Chen Wang · Cheng Sun · Wei Chen Received: 29 May 2013 / Revised: 5 February 2014 / Accepted: 22 February 2014 © Springer-Verlag Berlin Heidelberg 2014 Abstract The limitation associated with the low opti- cal absorption remains to be the main technical barrier that constrains the efficiency of thin–film solar cells in energy conversion. Effective design of light-trapping struc- ture is critical to increase light absorption, which is a highly complex phenomenon governed by several compet- ing physical processes, imposing a number of challenges to topology optimization. This paper presents a general, yet systematic approach exploiting topology optimization for designing highly efficient light-trapping structures. We first demonstrate the proposed approach using genetic algorithm (GA) based non-gradient topology optimization (NGTO), which is robust for achieving highly-efficient designs of slot-waveguide based cells with both low-permittivity and high-permittivity scattering material at single wavelength or over a broad spectrum. The optimized light-trapping structure achieves a broadband absorption efficiency of 48.1 % and more than 3-fold increase over the Yablonovitch limit. The fabrication feasibility of the optimized design is also demonstrated. Next, the gradient topology optimiza- tion (GTO) approach for designing light-trapping structure is explored based on the Solid Isotropic Material with Penalization (SIMP) method. Similar designs are obtained through both GA based NGTO and SIMP based GTO, which verifies the validity of both approaches. Insights into the application of both approaches for solving the nanopho- tonic design problem with optimization nonlinearity are provided. S. Yu · C. Wang · C. Sun · W. Chen () Department of Mechanical Engineering, Northwestern University, Evanston, IL, USA e-mail: weichen@northwestern.edu Keywords Topology optimization · Light-trapping structure · Nanophotonic design · Genetic algorithm · SIMP · Solar cell 1 Introduction Organic solar cells promise an economically viable sus- tainable energy source due to their potential in low-cost production (Shah et al. 1999). As the diffusion length of organic active material is typically 10–20 nm, thin active layer is favorable for organic solar cell (Shaw et al. 2008). Yet the photon absorption length of the active material is an order of magnitude larger than the diffu- sion length (Park et al. 2009). A significant reduction of thickness of the active layer will result into a low opti- cal absorption, hence a poor power conversion efficiency (Shaw et al. 2008). Light trapping technology was there- fore developed (Yablonovitch 1982; Polman and Atwater 2012) to extend the path-length for light interacting with the active layer, which can boost the absorption to offset the reduction of the active layer. Recent advances in nanopho- tonics have opened up a new gateway for light trapping (Yu et al. 2010; Callahan et al. 2012) with enhancement of absorption exceeding the classical Yablonovitch Limit (Yablonovitch 1982). Light trapping in thin-film solar cells is governed by several competing physical processes, including light refraction, deflection, and absorption. Pursuing the opti- mal structure requires a comprehensive consideration of all these competing processes. A wide range of structures have been investigated by using forward design methods, such as the triangular or pyramid grating (Dewan et al. 2009; Feng et al. 2007), nanowires (Garnett and Yang 2010), nanoholes (Raman et al. 2011), nanocone (Wang