arXiv subjects

Jacob LaMountain

Publications and source records attributed to Jacob LaMountain.

3 recordsLinked to original sources

arXiv2025

Subdiffractive confinement of ultrashort mid-IR pulses with photonic funnels

The ability to control the spatial distribution of light, particularly in deep sub-wavelength areas, is important for a range of materials science, microscopy, and communications applications. Separately, materials science and communications rely on the ability to temporally shape the evolution of electromagnetic pulses. In this work we investigate theoretically the propagation of ultrafast pulses inside hyperbolic metamaterials-based photonic funnels, which have been recently used to achieve deep subwavelength (wavelength/30) concentration of monochromatic mid-infrared light. By analyzing the complex spatio-temporal dynamics of the pulse-funnel interaction, we show that photonic funnels, in general, broaden bandwidth-limited ultrafast Gaussian pulses. We demonstrate that this broadening can be mitigated by pre-chirping the incoming light, realizing simultaneous intensity enhancement and spatio-temporal compression of mid-wave IR light in the all-semiconductor "designer metal" funnel platform. Our analysis suggests that, in combination with linear chirp, designer-metal-based photonic funnels can be utilized with 100 fs bandwidth- and diffraction-limited pulses to produce wavelength/30-scale signals of ~200 fs duration, with intensity enhancement on the order of 5. Lowering material absorption can further enhance the peak intensity. The results presented can be used to assess the perspectives of ultrafast sub-diffraction light manipulation in other portions of the electromagnetic spectrum by adjusting the (meta)material composition of the funnels.

physics.optics↗

arXiv2025

Physics-guided hierarchical neural networks for Maxwell's equations in plasmonic metamaterials

While machine learning (ML) has found multiple applications in photonics, traditional "black box" ML models typically require prohibitively large training data sets. Generation of such data, as well as the training processes themselves, consume significant resources, often limiting practical applications of ML. Here we demonstrate that embedding Maxwell's equations into ML design and training significantly reduces the required amount of data and improves the physics-consistency and generalizability of ML models, opening the road to practical ML tools that do not need extremely large training sets. The proposed physics-guided machine learning (PGML) approach is illustrated on the example of predicting complex field distributions within hyperbolic metamaterial photonic funnels, based on multilayered plasmonic-dielectric composites. The hierarchical network design used in this study enables knowledge transfer and points to the emergence of effective medium theories within neural networks.

physics.optics↗

arXiv2024

Nano-Focusing of Vortex Beams with Hyperbolic Metamaterials

The synergy of judiciously engineered nanostructures and complex topology of light creates unprecedented opportunities for tailoring light-matter interactions on the nanoscale. Electromagnetic waves can carry multiple units of angular momentum per photon, stemming from both spin and orbital angular momentum contributions, offering a potential route for modifying the optical transition selection rules. However, the size difference between a vortex beam and quantum objects limits the interaction strength and the angular momentum exchange. Here, we demonstrate the sub-diffraction-limited focusing of a vortex beam using the high in-plane wave number modes present in hyperbolic metamaterials. The spin-orbit interaction within the hyperbolic structure gives rise to the formation of an optical skyrmion with a deep subwavelength structure, which may enable the exploration of new light-matter interaction phenomena.

physics.optics↗