The engineering of broadband absorbers to harvest white light in thin-film semiconductors is a major challenge in developing renewable materials for energy harvesting. Many solution-processed materials with high manufacturability and low cost, such as semiconductor quantum dots, require the use of film structures with thicknesses on the order of 1 μm to absorb incoming photons completely. The electron transport lengths in these media, however, are 1 order of magnitude smaller than this length, hampering further progress with this platform. Herein, we show that, by engineering suitably disordered nanoplasmonic structures, we have created a new class of dispersionless epsilon-near-zero composite materials that efficiently harness white light. Our nanostructures localize light in the dielectric region outside the epsilon-near-zero material with characteristic lengths of 10-100 nm, resulting in an efficient system for harvesting broadband light when a thin absorptive film is deposited on top of the structure. By using a combination of theory and experiments, we demonstrate that ultrathin layers down to 50 nm of colloidal quantum dots deposited atop the epsilon-near-zero material show an increase in broadband absorption ranging from 200% to 500% compared to a planar structure of the same colloidal quantum-dot-absorber average thickness. When the epsilon-near-zero nanostructures were used in an energy-harvesting module, we observed a spectrally averaged 170% broadband increase in the external quantum efficiency of the device, measured at wavelengths between 400 and 1200 nm. Atomic force microscopy and photoluminescence excitation measurements demonstrate that the properties of these epsilon-near-zero structures apply to general metals and could be used to enhance the near-field absorption of semiconductor structures more widely. We have developed an inexpensive electrochemical deposition process that enables scaled-up production of this nanomaterial for large-scale energy-harvesting applications.
KAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): CRG-1-2012-FRA-005
Acknowledgements: For computer time, we used the resources of the KAUST Supercomputing Laboratory and the Redragon cluster of the Primalight group. A.F. acknowledges funding from KAUST (Award CRG-1-2012-FRA-005). E.H.S. acknowledges funding from the Ontario Research Fund.