Abstract
Harnessing photoexcited “hot” carriers in metallic nanostructures could define a new phase of non-equilibrium optoelectronics for photodetection and photocatalysis. Surface plasmons are considered pivotal for enabling efficient operation of hot carrier devices. Clarifying the fundamental role of plasmon excitation is therefore critical for exploiting their full potential. Here, we measure the internal quantum efficiency in photoexcited gold (Au)–gallium nitride (GaN) Schottky diodes to elucidate and quantify the distinct roles of surface plasmon excitation, hot carrier transport, and carrier injection in device performance. We show that plasmon excitation does not influence the electronic processes occurring within the hot carrier device. Instead, the metal band structure and carrier transport processes dictate the observed hot carrier photocurrent distribution. The excellent agreement with parameter-free calculations indicates that photoexcited electrons generated in ultra-thin Au nanostructures impinge ballistically on the Au–GaN interface, suggesting the possibility for hot carrier collection without substantial energy losses via thermalization.
Highlights
Harnessing photoexcited “hot” carriers in metallic nanostructures could define a new phase of non-equilibrium optoelectronics for photodetection and photocatalysis
Numerous experimental studies based on internal photoemission[14,15,16] (IPE) of hot electrons in metal–semiconductor photodiodes have shown a close correlation between the plasmonic resonance of the nanoantenna and the device responsivity[12,17,18,19,20,21,22,23]
While surface plasmons are well known to enhance light absorption[24] (Fig. 1b), deeper insight into their fundamental role in the physics of hot carrier devices requires a careful analysis of the internal quantum efficiency (IQE, Fig. 1c), which deconvolutes absorption and transport
Summary
Harnessing photoexcited “hot” carriers in metallic nanostructures could define a new phase of non-equilibrium optoelectronics for photodetection and photocatalysis. Previous work has relied on a semi-classical Fowler theory for interpreting the experimental IQE spectra[17,21,22,23,26,27,28,29] Failures of this approximation in the visible regime[18,23,30], where interband absorption in metals may be dominant, have required making ad hoc assumptions regarding the effect of plasmon excitation in electronic transport processes[26,31], in contrast with results of recent ab initio calculations[32]. Our analysis reveals that the measured photocurrents arise from ballistically injected hot electrons at photon energies below the threshold for interband transitions (~2 eV)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have