Abstract
The fabrication of optical waveguides with femtosecond laser direct writing is reported in two materials, Suprasil1 and Eagle2000. The influence of typical fabrication parameters, such as pulse energy and scan velocity, on the waveguide's spectral characteristics is explored from 500 to 1700 nm. Tests conducted in Suprasil1 evidence a strong presence of Rayleigh scattering, hindering the production of low-loss waveguides at short wavelengths. On the other hand, optical waveguides fabricated in Eagle2000 exhibited lower insertion losses at short wavelengths, enabling the fabrication of low-loss broadband optical waveguides with a two order of magnitude higher scan velocity when compared with Suprasil1.

Abstract
A Fabry-Perot interferometer was fabricated inside a fused silica substrate through femtosecond laser micromachining. The influence of the waveguide's writing parameters on the measured signal's quality was studied for an interferometer with a 27-mu m wide cavity. Optimal signal-to-noise ratio and fringe visibility were obtained for waveguides written at 75 nJ and 50 mu m/s. The same device was characterized with different refractive index liquids, and a maximum sensitivity of 1181.4 +/- 23.6 nm/RIU was obtained in the index range of 1.2962 to 1.3828 (at 1550 nm) for the spectral order m = 46.

Abstract
Optical waveguides were fabricated in alkaline earth boro-aluminosilicate glass, by femtosecond laser direct writing, with varying pulse energy and scan velocity. A spectral characterization, from 500 nm to 1700 nm, was made in order to determine their losses and understand its dependence on the processing parameters. Three major loss mechanisms were identified. At longer wavelengths, loss is mainly due to weak coupling. On the other hand, the behavior at shorter wavelengths is governed by propagation loss due to Rayleigh scattering, which was shown to be practically eliminated (& x003C; 0.05 dB $\cdot$ cm $<^>{-1} {\cdot }\,\,\mu \text{m}<^>{4}$ ) at higher scan velocities. Bulk absorption was also found to have an influence in the propagation losses at higher wavelengths. The combination of intermediate pulse energies (between 125-250 nJ) and high scan velocities (above 6 cm/s) allowed the fabrication of optical waveguides offering low losses across the entire range of wavelengths tested, facilitating applications that require larger wavelength working bands. Furthermore, since optimal fabrication conditions are achieved at higher scanning velocities, mass production with reduced fabrication times can be achieved.

Abstract
This article presents a review of the numerical techniques employed in simulating plasmonic optical sensors based on metal-dielectric nanostructures, including examples, ranging from conventional D-type fiber sensors, to those based on photonic crystal D-type fibers and incorporating metamaterials, nanowires, among other new materials and components, results and applications. We start from the fundamental physical processes, such as optical and plasmonic mode coupling, and discuss the implementation of the numerical model, optical response customization and their impact in sensor performance. Finally, we examine future perspectives.

Abstract
We consider broadband radiation interacting with a gas of self-gravitating dust grains. We show that photon-bubble formation can occur, due to a modified Jeans instability, which will imply the formation of two different kinds of dust density perturbations. This could be useful for understanding the B-mode signal observed in the CMB polarization survey, and other astrophysical processes, such as the formation of protoplanets and voids in dust clouds.

Abstract
The interaction of light with matter in near-to-resonant conditions opens a path for the exploration of nontrivial optical response that can play an important role in future photonics-driven technology. But as the attention shifts towards many-level atomic systems and involving multi-dimensional experimental scenarios, the complexity of the physical systems makes the analytical approach to the semiclassical model of the MaxwellBloch equations impossible without any strongly-limiting approximations. In this context, robust and highperformance computational tools are mandatory. In this work, we describe the development and implementation of a cross-platform Maxwell-Bloch numerical solver that is capable to exploit the different hardware available to tackle efficiently the problems under consideration. Moreover, it is demonstrated that this simulation tool can address a vast class of problems with considerable reduction of simulation time, featuring speedups up to 30 when running in massive parallel GPUs compared with the same codes running on a CPU, showing its potential towards addressing a large class of modern problems in photonics.