"Since micro- and nanotechnology already allow to design devices which only require minuscule amounts of power, the ability to convert ""free"" energy ""harvested"" from the environment into electric energy has tremendous potential, especially with reference to implantable devices for medical applications and wireless sensors networks. However, existing approaches for energy harvesting are not sufficiently efficient, cheap, compact, and versatile; in particular, the output power of conventional integrable piezoelectric devices is rather low. Recently, Prof. Zhong Lin Wang has used arrays of ZnO nanowires as piezoelectric nanotransducers for harvesting mechanical energy; though several outstanding advantages of piezoelectric nanogenerators have already been demonstrated, existing nanogenerators are largely sub-optimal: first, external acoustic energy is not optimally transduced into deflection of nanowires; second, there can be significant differences among different nanowires in the array, thus reducing efficiency; third, despite its compatibility with practically all MEMS processing steps (in particular, growth temperatures below 100 °C are possible), the aqueous chemical growth technique does not currently allow to grow sufficiently long nanowires.The goal of this proposal is to deal with all these dominant issues of existing piezoelectric nanogenerators. In practice we will use a modified (low-temperature) aqueous chemical growth process with localized heating, in order to improve the nanowires homogeneity as well as to increase the nanowires length; such an approach will permit the co-integration of sufficiently long ZnO nanowires with suspended microstructures (diaphragms and bridges), which will allow the optimal transduction of incoming acoustic energy into deflections of nanowires. We expect that co-integration of suspended microstructure and sufficiently long piezoelectric nanowires will allow to harvest output power density up to about 1mW/cm2."