Performance and economic limits of passively cooled hybrid thermoelectric generator-concentrator photovoltaic modules

Abstract Concentrator photovoltaic technology has doubled the efficiency of conventional non-concentrating photovoltaic systems. However, still ≈60% of the incident energy is dissipated as heat, and costs of concentrator photovoltaics need to be lowered for these systems to be competitive. One promising way for taking advantage of the heat generated in the solar cells is the hybridization with thermoelectric generators. The feasibility of hybrid concentrator photovoltaic-thermoelectric modules has been shown in recently published work for enhancing the efficiency and lowering the cost, with special emphasis on actively cooled designs. However, if feasible, the use of simple and reliable passive cooling would accelerate the development of new hybrid prototypes. The performance and cost reduction limits achievable with passively cooled designs have not yet been studied in detail. In this paper, an electric/thermal/economic model of concentrator photovoltaic-thermoelectric module is developed. As the main novelty, the model allows the thermoelectric generator area to be adjusted. Some optimisation problems are formulated and solved to evaluate the efficiency improvement and cost reduction of the hybrid system in comparison to a typical 800x light concentration factor and 36.4% efficiency concentrator photovoltaic-only module. The analysis showed that optimising the thermoelectric generator area is essential for the cell operating temperature limits not to be exceeded, and that conventional passive cooling is enough for achieving efficiency gains and cost reduction considering a state-of-the-art highly efficient 3 mm × 3 mm triple junction solar cell. With advanced thermoelectric materials and cell temperature of 100 °C, a maximum efficiency of 39.2% for the hybrid system can be achieved at 800x concentration factor by using low thermal resistance heat sinks, and a maximum cost reduction of 46.0% can be achieved at the maximum analysed concentration factor (1900x) by using moderate thermal resistance heat sinks. A trade-off between enhancing the efficiency and lowering the cost was observed. The sensitivity analysis on the thermoelectric parameters showed that parameters other than thermoelectric generator area and ZT figure-of-merit barely influence the results. The existing experimental concentrator photovoltaic-thermoelectric prototypes are far from the efficiency and cost benefits predicted in this paper. Thus, the study can help in the development of future prototypes.