Neural Inverse Design of Nanostructures (NIDN).

In the recent decade, computational tools have become central in material design, allowing rapid development cycles at reduced costs. Machine learning tools are especially on the rise in photonics. However, the inversion of the Maxwell equations needed for the design is particularly challenging from an optimization standpoint, requiring sophisticated software. We present an innovative, open-source software tool called Neural Inverse Design of Nanostructures (NIDN) that allows designing complex, stacked material nanostructures using a physics-based deep learning approach. Instead of a derivative-free or data-driven optimization or learning method, we perform a gradient-based neural network training where we directly optimize the material and its structure based on its spectral characteristics. NIDN supports two different solvers, rigorous coupled-wave analysis and a finite-difference time-domain method. The utility and validity of NIDN are demonstrated on several synthetic examples as well as the design of a 1550 nm filter and anti-reflection coating. Results match experimental baselines, other simulation tools, and the desired spectral characteristics. Given its full modularity in regard to network architectures and Maxwell solvers as well as open-source, permissive availability, NIDN will be able to support computational material design processes in a broad range of applications.

Perfect Photon Indistinguishability from a Set of Dissipative Quantum Emitters.

Single photon sources (SPS) based on semiconductor quantum dot (QD) platforms are restricted to low temperature (T) operation due to the presence of strong dephasing processes. Although the integration of QD in optical cavities provides an enhancement of its emission properties, the technical requirements for maintaining high indistinguishability (I) at high T are still beyond the state of the art. Recently, new theoretical approaches have shown promising results by implementing two-dipole-coupled-emitter systems. Here, we propose a platform based on an optimized five-dipole-coupled-emitter system coupled to a cavity which enables perfect I at high T. Within our scheme the realization of perfect I single photon emission with dissipative QDs is possible using well established photonic platforms. For the optimization procedure we have developed a novel machine-learning approach which provides a significant computational-time reduction for high demanding optimization algorithms. Our strategy opens up interesting possibilities for the optimization of different photonic structures for quantum information applications, such as the reduction of quantum decoherence in clusters of coupled two-level quantum systems.

Light absorption enhancement and radiation hardening for triple junction solar cell through bioinspired nanostructures.

Multi-junction solar cells constitute the main source of power for space applications. However, exposure of solar cells to the space radiation environment significantly degrades their performance across the mission lifetime. Here, we seek to improve the radiation hardness of the triple junction solar cell, GaInP/Ga(In)As/Ge, by decreasing the thickness of the more sensitive middle junction. Thin junctions facilitate the collection of minority carriers and show slower degradation due to defects. However, thinning the junction decreases the absorption, and consequently, the expected photocurrent. To compensate for this loss, we examined two bioinspired surface patterns that exhibit anti-reflective and light-trapping properties: (a) the moth-eye structure which enables vision in poorly illuminated environments and (b) the patterns of the hard cell of a unicellular photosynthetic micro-alga, the diatoms. We parametrize and optimize the biomimetic structures, aiming to maximize the absorbed light by the solar cell while achieving significant reduction in the middle junction thickness. The density of the radiation-induced defects is independent of the junction thickness, as we demonstrate using Monte Carlo simulations, allowing the direct comparison of different combinations of middle junction thicknesses and light trapping structures. We incorporate the radiation effects into the solar cell model as a decrease in minority carrier lifetime and an increase in surface recombination velocity, and we quantify the gain in efficiency for different combinations of junction thickness and the light-trapping structure at equal radiation damage. Solar cells with thin junctions compensated by the light-trapping structures offer a promising approach to improve solar cell radiation hardness and robustness, with up to 2% higher end-of-life efficiency than the commonly used configuration at high radiation exposure.

Location-Specific Spectral and Thermal Effects in Tracking and Fixed Tilt Photovoltaic Systems.

The efficiency of photovoltaic modules in the field is generally lower than the efficiency under standard testing conditions due to temperature and spectral effects. Using the latest spectral dataset available from the National Solar Radiation Database, we report spectral correction factors ranging from -2% to 1.3% of the produced energy for silicon modules depending on location and collector geometry. We find that spectral effects favor trackers if silicon modules are used, but favor a fixed tilt instead if perovskites or CdTe are used. In high-irradiance locations, the energy yield advantage of silicon-based trackers is underestimated by 0.4% if spectral sensitivity effects are neglected. As the photovoltaic market grows to a multi-terawatt size, these seemingly small effects are expected to have an economic impact equivalent to tens of billions of dollars in the next few decades, far outweighting the cost of the required research effort.

GaAs nanoscale membranes: prospects for seamless integration of III-Vs on silicon.

The growth of compound semiconductors on silicon has been widely sought after for decades, but reliable methods for defect-free combination of these materials have remained elusive. Recently, interconnected GaAs nanoscale membranes have been used as templates for the scalable integration of nanowire networks on III-V substrates. Here, we demonstrate how GaAs nanoscale membranes can be seamlessly integrated on silicon by controlling the density of nuclei in the initial stages of growth. We also correlate the absence or presence of defects with the existence of a single or multiple nucleation regime for the single membranes. Certain defects exhibit well-differentiated spectroscopic features that we identify with cathodoluminescence and micro-photoluminescence techniques. Overall, this work presents a new approach for the seamless integration of compound semiconductors on silicon.

Numerical Study on Mie Resonances in Single GaAs Nanomembranes.

GaAs nanomembranes grown by selective area epitaxy are novel structures. The high refractive index of GaAs makes them good candidates for nanoantennas. We numerically studied the optical modal structure of the resonator. The nanomembrane geometry introduces a strong light-polarization dependence. The scattering is dominated by an electric dipole contribution for polarization along the nanomembrane long dimension and by a magnetic dipole contribution in the orthogonal direction. The dependence on the geometry of the resonances close to the GaAs band gap was modeled by a single coefficient. It describes the resonance shifts against up-to 40% changes in length, height, and width. We showed that the nanomembranes exhibited field enhancement, far-field directionality, and tunability with the GaAs band gap. All these elements confirm their great potential as nanoantennas.

Fabrication of Silicon Nanobelts and Nanopillars by Soft Lithography for Hydrophobic and Hydrophilic Photonic Surfaces.

Soft lithography allows for the simple and low-cost fabrication of nanopatterns with different shapes and sizes over large areas. However, the resolution and the aspect ratio of the nanostructures fabricated by soft lithography are limited by the depth and the physical properties of the stamp. In this work, silicon nanobelts and nanostructures were achieved by combining soft nanolithography patterning with optimized reactive ion etching (RIE) in silicon. Using polymethylmethacrylate (PMMA) nanopatterned layers with thicknesses ranging between 14 and 50 nm, we obtained silicon nanobelts in areas of square centimeters with aspect ratios up to ~1.6 and linewidths of 225 nm. The soft lithographic process was assisted by a thin film of SiOx (less than 15 nm) used as a hard mask and RIE. This simple patterning method was also used to fabricate 2D nanostructures (nanopillars) with aspect ratios of ~2.7 and diameters of ~200 nm. We demonstrate that large areas patterned with silicon nanobelts exhibit a high reflectivity peak in the ultraviolet C (UVC) spectral region (280 nm) where some aminoacids and peptides have a strong absorption. We also demonstrated how to tailor the aspect ratio and the wettability of these photonic surfaces (contact angles ranging from 8.1 to 96.2°) by changing the RIE power applied during the fabrication process.

Cloaking of solar cell contacts at the onset of Rayleigh scattering.

Electrical contacts on the top surface of solar cells and light emitting diodes cause shadow losses. The phenomenon of extraordinary optical transmission through arrays of subwavelength holes suggests the possibility of engineering such contacts to reduce the shadow using plasmonics, but resonance effects occur only at specific wavelengths. Here we describe instead a broadband effect of enhanced light transmission through arrays of subwavelength metallic wires, due to the fact that, in the absence of resonances, metal wires asymptotically tend to invisibility in the small size limit regardless of the fraction of the device area taken up by the contacts. The effect occurs for wires more than an order of magnitude thicker than the transparency limit for metal thin films. Finite difference in time domain calculations predict that it is possible to have high cloaking efficiencies in a broadband wavelength range, and we experimentally demonstrate contact shadow losses less than half of the geometric shadow.

Light-trapping in photon enhanced thermionic emitters.

A series of photonic crystal structures are optimized for a photon enhanced thermionic emitter. With realistic parameter values to describe a p-type GaAs device we find an efficiency above 10%. The light-trapping structures increases the performance by 2% over an optimal bilayer anti-reflective coating. We find a device efficiency very close to the case of a Lambertian absorber, but below its maximum performance. To prevent an efficiency below 10% the vacuum gap must be dimensioned according to the concentration factor of the solar irradiance.

Silicon nanowires: where mechanics and optics meet at the nanoscale.

Mechanical transducers based on nanowires promise revolutionary advances in biological sensing and force microscopy/spectroscopy. A crucial step is the development of simple and non-invasive techniques able to detect displacements with subpicometer sensitivity per unit bandwidth. Here, we design suspended tapered silicon nanowires supporting a range of optical resonances that confine and efficiently scatter light in the visible range. Then, we develop an optical method for efficiently coupling the evanescent field to the regular interference pattern generated by an incoming laser beam and the reflected beam from the substrate underneath the nanowire. This optomechanical coupling is here applied to measure the displacement of 50 nm wide nanowires with sensitivity on the verge of 1 fm/Hz(1/2) at room temperature with a simple laser interferometry set-up. This method opens the door to the measurement of the Brownian motion of ultrashort nanowires for the detection of single biomolecular recognition events in liquids, and single molecule spectroscopy in vacuum.

Optomechanics with silicon nanowires by harnessing confined electromagnetic modes.

The optomechanical coupling that emerges in an optical cavity in which one of the mirrors is a mechanical resonator has allowed sub-Kelvin cooling with the prospect of observing quantum phenomena and self-sustained oscillators with very high spectral purity. Both applications clearly benefit from the use of the smallest possible mechanical resonator. Unfortunately, the optomechanical coupling largely decays when the size of the mechanical system is below the light wavelength. Here, we propose to exploit the optical resonances associated to the light confinement in subwavelength structures to circumvent this limitation, efficiently extending optomechanics to nanoscale objects. We demonstrate this mechanism with suspended silicon nanowires. We are able to optically cool the mechanical vibration of the nanowires from room temperature to 30-40 K or to obtain regenerative mechanical oscillation with a frequency stability of about one part per million. The reported optomechanical phenomena can be exploited for developing cost-optimized mass sensors with sensitivities in the zeptogram range.