This work presents a machine-learning-based methodology for optimizing perovskite solar cells performance using a freely accessible synthetic dataset generated from numerical simulations of devices in SCAPS-1D. The dataset includes systematic variations of the device’s geometric, optical, and defect-related parameters, enabling an extensive exploration of the design space without the need for additional simulations. A Support Vector Regression (SVR) model with a radial basis function (RBF) kernel was trained as a surrogate model to predict power conversion efficiency (PCE) from the geometrical and optical variables of the device. For the optimization stage, a multi-objective framework based on Pareto-front analysis was implemented to balance the trade-off between maximizing PCE and minimizing prediction errors. As result, a cell architecture with a maximum PCE of 29.82% was identified, with predictive performance characterized by R² = 0.9124, MAE = 1.8470, and RMSE = 2.8773, supporting the use of SVR-RBF as a surrogate model on open synthetic data. To enhance physical interpretability, the SHAP framework was applied to quantify the influence of each optical parameter to the PCE, revealing the dominant factors associated with layer thicknesses, optical properties, and defect densities. The proposed approach demonstrates the potential of interpretable machine learning as a physics-informed design tool for high-performance perovskite solar cells, providing clear guidelines for device optimization in optoelectronic and photonic applications.

Reliable measurement of electric current is essential in many technological and industrial contexts, particularly in environments where conventional electrical sensors are limited by electromagnetic interference, electrical contact requirements, or accessibility constraints. Optical sensing strategies based on luminescent materials offer an attractive alternative due to their noninvasive nature, immunity to magnetic fields, and potential for remote operation. In this work, an optical approach for current monitoring is presented using thermally sensitive upconversion luminescence. The sensing system is based on NaYF₄:Yb³⁺,Er³⁺ particles dispersed in a photocurable resin and deposited as a coating on a resistive element. Under near-infrared excitation, the material exhibits temperature-dependent upconversion emissions arising from thermally coupled energy levels of Er³⁺. A calibration procedure was first carried out by correlating the luminescence intensity ratio with temperature over the range 298–328 K. Subsequently, electrical current was applied to a ceramic resistor, and the resulting self-heating was monitored optically through changes in the luminescent response. A clear relationship between the applied current and the luminescence intensity ratio was observed, reflecting the temperature increase induced by Joule heating. The sensor was evaluated for direct currents between 0 and 2.2 A, a range defined by the electrical and thermal characteristics of the resistor. The measurements demonstrated good repeatability and an estimated current resolution of approximately 0.01 A. The calibration was shown to remain stable across different ambient temperatures within the studied range. The proposed luminescence-based strategy enables accurate, remote measurement of electric current without direct electrical contact. Its flexibility, precision, and adaptability to different resistive components make it a promising tool for current sensing in environments where traditional techniques are impractical or invasive.

Frequency conversion plays a key role in controlling relaxation dynamics in rare-earth-doped materials and underlies numerous photonic applications, particularly in photovoltaics. In this context, spectral conversion aims to optimize solar radiation absorption by shifting poorly absorbed wavelengths toward ~980 nm, corresponding to the bandgap region of crystalline silicon. Glass–ceramic systems have been shown to be especially effective for tailoring energy transfer mechanisms between rare earth ions [1]. Among them, silica–hafnia binary matrices have emerged as promising hosts for incorporating rare earth ions into hafnia nanocrystals, resulting in enhanced luminescence efficiency and expanded application potential.
In this work, we investigate the Tb³⁺/Yb³⁺-mediated down-conversion process in silica–hafnia glass–ceramic thin films as a strategy to improve the spectral response of silicon-based solar cells. The films were prepared via a sol–gel spin-coating method followed by an appropriate thermal treatment, enabling precise control over composition, morphology, and optical properties [2].
Photoluminescence measurements performed under 476 nm excitation reveal a strong emission from Yb³⁺ ions associated with the 2F5/2→2F7/2 transition after Tb³⁺ excitation, demonstrating an efficient energy transfer from Tb³⁺ to Yb³⁺. These results confirm the efficiency of the down-conversion mechanism in the silica–hafnia glass–ceramic system and highlight its relevance for photovoltaic spectral conversion applications.
In summary, this study demonstrates that Tb³⁺/Yb³⁺-codoped silica–hafnia glass–ceramic thin films constitute a promising and tunable system for efficient spectral down-conversion in silicon-based photovoltaic technologies.

[1] Salima El Amrani , Lamyae Oulmaati , Giancarlo C. Righini , Maurizio Ferrari, Mohammed Reda Britel, Adel Bouajaj. Ceramics International (2025) 51(12) 16645-16649. https://doi.org/10.1016/j.ceramint.2024.10.448.

[2] Salima El Amrani, Michael Sun, Sirona Valdueza-Felip, Fernando B. Naranjo, Mohammed Reda Britel, Maurizio Ferrari, Adel Bouajaj, Effect of temperature and excitation power on down-conversion process in Tb3+/Yb3+-activated silica-hafnia glass-ceramic films,Ceramics International,2024,ISSN 0272-8842 .https://doi.org/10.1016/j.ceramint.2024.06.392.

Optical beam profiling is essential for characterizing beam quality and diffraction for the optimization of laser-based systems. However, in wavelength ranges such as near-infrared, commercially available beam profilers are often costly, limiting their accessibility. There is therefore a clear need for low-cost and flexible beam profiling approaches that can operate across a broad wavelength range while maintaining acceptable accuracy and robustness. We present a beam-profiling system based on a rotating photodiode that reconstructs the 1D beam intensity profile via angular scanning. A single photodiode on a motorized stage rotates around the beam's axis with an angular resolution less than 0.1 degrees, corresponding to a spatial resolution of 0.1mm, while a microcontroller with a fast-sampling analog-to-digital converter (ADC) performs real-time data acquisition of the detected signal. We validated the system using a 2 µm laser source. At a propagation distance of 140 mm, the rotating diode profiler measured a beam width of 7.3 mm, compared to 6.5 mm from a commercial Thorlabs profiler. Given the photodiode width of 0.3mm, the step resolution, variations in the fit results from beam asymmetries, and variations in the beam itself, the difference is within the expected tolerances. Reconstructed profiles showed good agreement with a Gaussian distribution, and the measured beam diameters were highly repeatable. Our rotating diode profiler is a cost-effective and scalable alternative to scanning slit and camera-based systems, especially for near-infrared wavelength regimes, making it suitable for research, development and instructional applications. Commercial infrared beam profilers cost 6,000 USD and up, while our rotating photodiode system can be implemented for under 400 USD—representing a greater than 10-fold cost reduction.

Based on the Integral Equation Method (IEM), the influence of target curvature on electromagnetic scattering depolarization characteristics is investigated. The studied target is a blunt-nosed cone, whose nose geometry is modeled using three typical surfaces—spherical, ellipsoidal, and parabolic caps—in order to analyze the effect of different curvature distributions on polarization scattering mechanisms. Surface roughness parameters are incorporated into the model, and the depolarization characteristics of various nose geometries are systematically examined under different roughness conditions and incidence angles. Within the IEM framework, both single-scattering and multiple-scattering contributions are taken into account to compute the co-polarized and cross-polarized scattering components, from which the depolarization ratio is obtained.

Numerical simulation results indicate that nose curvature has a significant impact on depolarization characteristics, and this influence exhibits distinct dependence on the incidence angle. Compared with the spherical nose, the ellipsoidal and parabolic noses, due to their continuously varying local curvature, show stronger polarization coupling effects at moderate and large incidence angles. As surface roughness increases, multiple scattering becomes more pronounced, leading to enhanced polarization mixing and a noticeable increase in the depolarization level. These results provide theoretical insight into polarization scattering modeling and depolarization mechanism analysis for complex curved targets.

Fiber-optic Mach–Zehnder interferometers are fundamental building blocks in modern photonic and quantum information systems, enabling precise phase manipulation for applications ranging from coherent communications to quantum state preparation and measurement. Nevertheless, their performance is inherently limited by phase noise induced by environmental perturbations, such as temperature fluctuations and mechanical vibrations. This challenge becomes increasingly severe as the dimensionality and complexity of the interferometric architecture grow, particularly in multi-path and high-dimensional fiber-optic networks. In this talk, we present an adaptive control approach for the stabilization of phase fluctuations in complex fiber-optic interferometers. The proposed method dynamically adjusts the control action based on the system response, allowing it to operate effectively in nonlinear regimes without requiring an explicit system model. Numerical studies performed on high-dimensional N×NN \times NN×N interferometric architectures illustrate the scalability and robustness of the approach under realistic noise conditions. Experimental validation is carried out using correlated photon pairs, demonstrating stable and sustained phase locking in fiber-based interferometers relevant to quantum information applications. The results show that the adaptive strategy provides strong resilience to phase noise while maintaining long-term stability. Furthermore, the control scheme is well suited for real-time implementation on FPGA platforms, supporting low-latency operation and practical deployment. These features make the proposed approach attractive for both classical photonic systems and emerging quantum technologies based on fiber-optic interferometry.

Polarization is a basic electromagnetic unit of information. It can easily be controlled and adjusted for plane waves propagating in the bulk media. However, it is still a challenge to manipulate the polarization of the localized light, i.e. in the near-field region.

In this work, we discover the near-field polarization degree of freedom offering two configurations for the practical implementation based on (i) all-dielectric metasurfaces in visible and near-infrared ranges, and (ii) self-complementary metasurfaces in microwaves. The first approach assumes a general design strategy for all-dielectric nanostructures that supports TE-TM degenerate guided modes. It is broadly applicable across a wide spectral range, including the visible, infrared, terahertz, and microwave regimes. In what follows, we consider a metasurface consisting of identical disk-shaped high-index dielectric resonators. Each disk exhibits the Mie resonances, and we adjust the collective Mie-like response. The second approach is based on the Babinet’s duality principle implemented in a self-complementary metasurface. It has been recently investigated that surface waves supported by the self-complementary metasurface exhibit degenerate TE-TM dispersion in certain directions, leading to a near-field polarization degree of freedom.

Finally, we demonstrate the excitation of surface waves with on-demand polarization and a planar compact near-field waveguide polarizer. Both devices are subwavelength and have been verified experimentally for microwaves. The results obtained form a new direction for near-field polarization photonics, leading to a plethora of compact planar polarization devices for manipulating localized light.

The analysis of the efficiency of laser radiation transmission through the atmosphere from the point of view of an alternative to existing radio communication methods of information transfer is based on determining the influence of negative effects such as scattering, the absorption of radiation, and the turbulence of air flows on the optical beam. The influence of scattering and absorption in the atmosphere is usually minimized by choosing an appropriate radiation wavelength, primarily the in near-IR spectrum. Atmospheric turbulence occurs due to uneven heating of the underlying earth's surface. Such heating causes local fluctuations in the refractive index of air, which leads to the appearance of an optical path difference of propagating optical rays. As the radiation propagates along the transmission path, the number of fluctuations increases; consequently, the wavefront of the beam arriving at the receiving focusing system has a shape that is far from flat, which manifests itself as a focal spot with a shape different from the diffraction-limited one. Various modifications of adaptive optics systems are typically used to reduce turbulence effects. As a result, an optical circuit was created that simulates a wireless communication channel, with two laser beams simultaneously directed along the atmospheric path, with wavelengths of 1550 nm and 830 nm. The first beam transmits the primary radiation, while the second beam analyzes the characteristics of the introduced turbulence onto a Shack–Hartmann wavefront sensor, as the sensor's sensitivity lies in the frequency range of 400–1100 nm. Artificial turbulence was generated using a fan heater, and the resulting phase fluctuations were corrected using the bimorph wavefront corrector.

This work focuses on the acquisition and analysis of the multispectral behavior of mammary tissue representing glandular tissue, adipose tissue, and/or a tumor, incorporating a perfusion system, in order to characterize the constructed structure using image-processing algorithms applied to the acquired data. A synthetic model was fabricated from organic materials in relative proportions that simulate the two typical breast tissues—glandular and adipose—together with a material that presumably represents the tumor. The specimens consist of an outer ring (glandular or adipose tissue) containing the tumor at its center, and the entire system is traversed by two tubes carrying a fluid that mimics blood circulation. Two lamps irradiate the structure to thermally stimulate the surface, thereby enabling the simultaneous acquisition of thermographic images and spectral images in the visible region, which were subsequently processed using Python. An experimental and computational procedure was developed to acquire and process multispectral images of the simulated tissues. The algorithm implemented for thermographic acquisition enabled the identification of two differentiated regions during thermal wave propagation by employing the K-means algorithm. In addition, the temporal evolution during the cooling process allowed the cooling constant to be identified as a characteristic parameter of the process. Regarding the acquisition of multispectral images in the visible range (MSI), it was possible to differentiate diffuse reflectance among the different simulated tissues and the vascularization system. The applied methodology enabled a comprehensive analysis that facilitated the understanding of thermal responses as a function of time in the implemented experimental model. The algorithm allowed the segmentation of regions of interest, delineating regions associated with distinctive spectral contrasts. The results are consistent with the mathematical models studied, enabling an adequate differentiation among the tissues.

Chirality is used to describe the asymmetry of a physical system that cannot be superimposed on its mirror images. Despite being chemically identical, chiral molecules can differ drastically in their biological activity, exhibiting therapeutic or toxic effects depending on their handedness. The primary challenge during the measurement of chiral substances is that most conventional methods are often limited by weak chiroptical signals at low concentrations of molecules.

To address this gap, we investigated a system consisting of a 50 nm gold film coated with a chiral analyte on a quartz substrate. By employing the Kretschmann prism-coupling configuration, we excited surface plasmon polaritons with a TM-polarized plane wave to analyze the resulting reflection spectra. It is observed that the presence of a chiral medium leads to a small non-zero response in orthogonal (transverse electric) polarization due to magneto-electric coupling. As a result, an angular difference emerges between the spectral resonances in terms of right- and left-handed circularly polarized waves. The magnitude of this angular shift determines the sensitivity of the sensor—a larger shift corresponds to higher sensitivity.

To overcome the film's limitations, we replaced it with a gold subwavelength grating. This approach amplified the chiral sensitivity by nearly two orders of magnitude. The reason for this enhancement is explained by the Rayleigh anomaly, which occurs when a diffracted wave propagates tangentially to the grating surface, transitioning into an evanescent mode. Consequently, the coupling between the evanescent wave strongly localized at the edges of the grating and the chiral substance leads to the significant amplification of the chiroptical response.

We demonstrated that the structural transition from thin films to subwavelength gratings significantly improves the limits of chiral detection. These findings offers a promising platform for detecting small concentrations of chiral biomolecules in real time for pharmaceutical and biomedical applications.