In A. Shahsafi et al, Nano Letters 20, 8483 (2020), we describe a new type of mid-infrared polarizer based on polarization-selective coupling to surface-plasmon polaritons (SPPs). The deceptively simple structure comprises porous anodic aluminum oxide on aluminum, which enables incident light to couple to SPPs at the interface between the aluminum and aluminum oxide -- a process that can only occur for one linear polarization due to the nature of surface plasmons.

This work builds our previous observation that the refractive index (n) of aluminum oxide (and many other polar dielectrics) can drop below 1 close to vibrational resonances in the material. We used this observation to demonstrate a variety of unique optical effects such as external reflection at infrared frequencies (in analogy to total internal reflection) as well as direct coupling to surface plasmons from free space without the need for gratings, prisms, or other structures conventionally required for surface plasmon coupling [A. Shahsafi et al, Physical Review Applied 10, 034019 (2018)].

Our surface-plasmon-based reflective polarizers have several performance metrics (extinction ratio, efficiency) that are competitive with the much-more-expensive commercial wire-grid polarizers, but are narrowband, limiting applications to lasers and other narrowband light sources.

This work was supported by the Office of Naval Research.

(2020) Adjoint-optimized nanoscale light extractor for nitrogen-vacancy centers in diamond

In R. Wambold et al, Nanophotonics 10, 1 (2020), we tackled a challenging problem in the growing research field of nitrogen-vacancy (NV) centers in diamond. NV centers are optical emitters (fluorescing when illuminated by green light), whose level structure is highly sensitive to external perturbations, which makes them excellent localized sensors of highly localized electric and magnetic fields, temperature, and strain. NV centers are also excellent solid-state systems for quantum computing and communication and the study of quantum phenomena such as quantum entanglement and superposition.

However, efficiently extracting NV fluorescence is often challenging due to the high index of refraction in diamond (∼2.4), which results in high reflectance at the diamond–air interfaces and total internal reflection for emission angles larger than the critical angle. Only ~3% of light emitted by an NV center can ultimately emerge from a polished diamond sample. Over the last decade, several research groups have tackled this problem by etching the diamond surface to make nanostructures (nanowires, gratings, metasurfaces, etc.), but diamond etching is challenging and can degrade the quantum properties of NV centers.

Our new approach is based on a nanoscale light extractor (NLE) based on silicon that can be fabricated directly on a flat diamond surface without etching it, and pulls light out from the NV center -- simultaneously increasing the outcoupling efficiency and enhancing the spontaneous emission rate via the Purcell effect. The structure is quite unintuitive: we optimized it using the adjoint-optimization technique, which enables free-form optimization of nanostructures given a particular figure of merit (FoM) -- here, the extraction of light from a near-surface NV center into the air, in the direction directly normal to the diamond surface. Our NLE can enhance the optical output of near-surface NV emitters by more than 35× compared with the unpatterned case, directing the light into a narrow cone that can be easily collected with simple optical systems.

This work was carried out in collaboration with the groups of Robert Hamers, Mark Eriksson, Jennifer Choy, and Shimon Kolkowitz, all a UW-Madison. Our group's contributions were supported by the National Science Foundation.

(2019) Nanosecond mid-infrared pulse generation via modulated thermal emissivity

An object emitting thermal radiation is typically slow to switch off and on again, because of the time it takes to heat up and cool down. In Y. Xiao et al, Light: Science and Applications 8, 51 (2019), we demonstrate the generation of nanosecond-scale pulses of thermal radiation by rapidly changing an object's emissivity, which is the propensity to emit thermal radiation.

The experiment is conceptually simple: an undoped semiconductor (silicon or gallium arsenide), which has very low emissivity, is briefly illuminated by an ultrafast visible light pulse, on the order of 200 femtoseconds in duration. Because the energy of the incident photons is higher than the band gap of the semiconductor, many electrons and holes are generated, which momentarily increases the emissivity of the semiconductor, which then falls as the electrons and holes recombine. This process results in the emission of an infrared thermal pulse, with duration down to about 1 nanosecond. To our knowledge, the nanosecond modulation of thermal emissivity in this work is three orders of magnitude faster than what has been previously demonstrated.

This work was carried out in collaboration with Alberto Pique's group at the Naval Research Laboratory, and was funded on the UW-Madison end by the Office of Naval Research.

(2019) Temperature-independent thermal radiation

Thermal radiation is the process by which all objects at nonzero temperatures emit light. For most solids, the thermally radiated power increases with increasing temperature in a one-to-one relationship that enables applications such as infrared ("thermal") imaging and noncontact thermometry (e.g., a forehead thermometer). In A. Shahsafi, PNAS 116, 26402 (2019), we demonstrated coatings that violate this one-to-one relationship via the use of samarium nickel oxide (SmNiO3), a strongly correlated quantum material that undergoes a fully reversible, temperature-driven solid-state phase transition.

As our SmNiO3-based coating is heated, it becomes more metallic, and hence less thermally emissive. This reduction in the emissivity cancels out the expected increase in radiation due to increasing temperature, and the overall thermally radiated power remains constant. In our initial demonstration, this effect persists over a temperature range of ∼30 °C, centered around ∼120 °C. The ability to decouple temperature and thermal emission opens a gateway for controlling the visibility of objects to infrared cameras and, more broadly, opportunities for quantum materials in controlling heat transfer.

This work was performed with collaborators in Shriram Ramanathan's group at Purdue, Riccardo Comin's group at MIT, and Jerzy Sadowski at Brookhaven. The UW-Madison contributions were funded by the Office of Naval Research and the National Science Foundation.