Numerical Study of a Solar Thermophotovoltaic Energy Converter with High Performance 2d Photonic Crystals

Solar thermophotovoltaic (STPV) systems convert solar energy into electricity via thermally radiated photons at tailored wavelength to increase energy conversion efficiency. In this work we report the design and analysis of a STPV using a highfidelity 2D axisymmetric thermal-electrical hybrid model that includes thermal coupling between the absorber/emitter/PV cell and accounts for non-idealities such as temperature gradients and parasitic thermal losses. The radiative spectra of the absorber and emitter are engineered by using two-dimensional periodic square array of cylindrical holes on a tantalum (Ta) substrate. The optimal solar concentration and resulting temperature are determined by considering the energy losses associated with re-emission at the absorber, low energy (below band gap) emission at the emitter, and carrier thermalization /recombination in the PV cell. The modeling results suggest that the overall efficiency of a realistic planar STPV consisting of Ta PhCs and existing InGaAsSb PV cells with a filter can be as high as ~8%. The use of high performance PhCs allows us to simplify the system layout and operate STPVs at a significantly lower optical concentration level and operating temperature compared with STPVs using metallic cavity receivers. This work shows the importance of photon engineering for the development of high efficiency STPVs and offers design guidelines for both the PhC absorber/emitter and the overall system. INTRODUCTION Solar thermophotovoltaic (STPV) systems use an intermediate module that absorbs the solar radiation, and reradiates photons at high temperatures and tailored wavelengths toward a photovoltaic (PV) cell (Fig.1). By converting the solar radiation to a narrow-banded thermal emission matching the spectral response of the PV cell, STPVs have the potential to overcome the Schockley-Queisser limit for PVs (<~40%) [1, 2]. STPVs are also highly scalable for a wide range of power capacity, have no moving parts, and allow solar energy storage and the use of an alternative fuel to generate electricity. Despite the significant potential, very few experimental results have reported the overall efficiency. Of the studies reported, the demonstrated values were quite low due to the poor performance of the emitter/absorber/cell and the lack of understanding of the highly coupled energy transport processes among the components. A previous study using an eutectic emitter reported an extremely low (~0.025%) efficiency [3] and a recent experiment with a cylindrical tungsten thermal cavity and Ge cells demonstrated ~0.7% overall efficiency [4]. With a similar cylindrical layout, ~1% efficiency was achieved with a tantalum absorber/emitter and GaSb cells [5]. Recently, the use of photonic crystals (PhCs) with 1D periodic metal/dielectric layers or 2D array of cylindrical holes have been suggested to achieve a narrow-banded thermal emission with a tailored cut-off wavelength [6-8]. Long wavelength reflection filters also have been developed to reflect low energy emission back to the emitter [9, 10]. These previous studies, however, have focused on the component level performance rather than the system, which made it difficult to properly estimate overall performance of STPVs with these components. In STPVs, the spectral performance of absorber/emitter and the parasitic system-level thermal losses are strongly affected by operating temperature determined from the complex energy transport among the components (see Fig.1). In this work, we developed a high-fidelity 2D axisymmetric thermal-electrical hybrid system-level model for STPVs with 2D Ta PhC absorbers/emitters. Our model includes radiative and conductive thermal coupling between the absorber/emitter/PV cell, and precisely accounts for nonidealities such as temperature gradients and parasitic thermal losses via the side wall (t) and gap (g) (Fig.1). The desired spectra are achieved by adjusting the cavity resonant frequency through changes in the micro-cavity dimensions and by matching the quality factor. Unlike previous studies [6, 8] the Proceedings of the ASME 2012 Summer Heat Transfer Conference HT2012 July 8-12, 2012, Rio Grande, Puerto Rico 1 Copyright © 2012 by ASME HT2012-58222 Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 10/07/2013 Terms of Use: http://asme.org/terms angular dependence of PhCs are included in the system-level analysis and the PhCs are designed through the optimization process. By incorporating Ta PhC absorber/emitter, we show that ~8% overall efficiencies (without including a collector loss) can be achieved with the developed PhCs and existing cells/filters at a relatively low solar concentration (~100 Suns). The predicted efficiency is substantially higher than previously demonstrated values and our design has a much simpler layout than the previous STPVs with a cylindrical metal cavity. MODEL FORMULATION The simplified schematic of a planar STPV is shown in Figure 1. The concentrated solar energy is converted into heat at the absorber and reemitted through the emitter that is thermally coupled to the absorber. The thermally radiated high energy photons create electron-hole pairs and generate electricity while low energy photons are wasted as heat. The photons reflected on the PV cell surface or emitted from the cell are reabsorbed on the emitter. Figure 1: (a) Schematic of a planar STPV that converts wide spectrum solar radiation (b) into narrow-banded thermal emission (c). R and t: radius and thickness of a circular absorber/emitter module, g: gap between the emitter and cell. We developed a hybrid model using the finite element method and an equivalent circuit model. The radiative heat transfer is coupled with conduction and convection heat transfer on each infinitesimal boundary element defined in a 2D or 3D framework (Eq.1): ( ) dA on T T h q T k n conv rad ) ( inf + = Ñ × r . (1) On the absorber side (Eq.2), the radiative heat flux applied to each infinitesimal element (qrad in Eq.1) is determined by incoming solar radiation and re-emission loss from the absorber: { } l l l l e l l l a d T E T E d I C q b a b a solar a opt rad ) , ( ) , ( ) ( ) ( ) ( inf 0 0 × = ò ò ¥ ¥ . (2) The standard solar spectrum for concentrated solar applications (AM1.5D) is used for the entire calculations. On the emitter side (Eq.3), the emission loss, re-absorption via multiple reflections between the emitter and the PV cell, and emission from the PV cell determine the radiative heat flux on each element (qrad in Eq.1): l l r l r l l e l e l l r l r l r l l e l l l e d F F T E d F F T E d T E q