Steven Hartman1,Ekaterina Dolgopolova1,Jennifer Hollingsworth1,Ghanshyam Pilania1
Los Alamos National Laboratory1
Steven Hartman1,Ekaterina Dolgopolova1,Jennifer Hollingsworth1,Ghanshyam Pilania1
Los Alamos National Laboratory1
To improve the efficiency of optical communications devices, it is necessary to generate strong light-matter interactions at length scales much smaller than those of conventional optics, and one of the most promising ways to do this is by harnessing the localized surface plasmonic response (LSPR) of free charge carriers. In metal nanoparticles, the collective excitation of surface electrons can greatly amplify the electromagnetic field of incoming photons, leading to strong absorption even if the nanoparticles are much smaller than the light’s wavelength. The frequency of plasmon response is determined by the concentration of free carriers in the metal, which is fixed at approximately 10<sup>23</sup>/cm<sup>3</sup> and leads to a visible-spectrum LSPR. Accessing the telecommunications frequencies in the infrared requires doped semiconductors with tunable carrier concentrations of 10<sup>18</sup>-10<sup>21</sup>/cm<sup>3</sup>, but these materials must also maintain high carrier mobility to ensure a narrow linewidth of the LSPR. The spinel family of A<sub>2</sub>BO<sub>4</sub> metal oxides offer great potential to meet these requirements, since many spinels can easily form A<sub>B</sub> or B<sub>A</sub> antisite defects, and they are also much cheaper and less prone to oxidation than currently used metal plasmonics.<br/>In this study we examine the dopability of about 100 spinel oxides with density functional theory, using the formation energy of neutral defects as an inexpensive proxy metric to down-select a limited set of promising materials for detailed charged defect calculations. We identify a family of chemically related spinel oxides which can be easily doped either n- or p-type, because one cation antisite defect is much easier to form than the other. We also determine the relevant chemical potential conditions which will result in the highest doping concentration. Further, by calculating the electronic band structure, we are able to confirm that some of the candidate plasmonic materials have light electron masses and high <i>n</i>-type carrier mobility. Finally, we show preliminary nanocrystal synthesis results which demonstrate an LSPR for the first time in these materials validating our predictions.