ABSTRACT
Recent progress in low power electronics, wireless technologies and miniaturization of electromechanical devices using complementary metal oxide semiconductor (CMOS) and micro-electromechanical systems (MEMS) technologies suggests that many intelligent embedded devices will soon operate on very small power budgets, in the range of 1–100 μW [1]. Closely associated with these developments is the challenge of developing portable sources of electric power. Electrochemical batteries are widely used but often restrict the performance and autonomy of microdevices in many applications. Furthermore, for important emerging applications such as wireless sensor networks employing a large number of nodes, it is not practical to use batteries. In fact, the task of replacing batteries is a major obstacle towards the widespread deployment of sensor networks for health monitoring applications and the Internet of things [2]. Hence, there is great interest in developing miniaturized energy harvesters that can convert ambient energy – thermal fluctuations and gradients, solar radiation, mechanical vibrations, fluid flow and radio-frequency waves – into electric power. Over the past 20 years, numerous device concepts have been proposed ranging from miniaturized engines and fuel cells to thermoelectric, thermophotovoltaic and piezoelectric devices. Many crucial questions remain open in this field. What are the performance limits of different types of harvesters? Can we develop effective strategies for selecting harvesters for specific applications and environments? Can we develop rational design methodologies to optimize the performance, reliability and manufacturability?
