Articles 2024

Résumé: The study investigates the impact of a vertically layered bathymetry, consisting of submerged vertical plates, on a ship wake through theoretical analysis and experimental realization. For subwavelength distances between the plates, the analysis relies on a homogenized model that provides an effective, anisotropic, dispersion relation for the propagation of water waves. Our findings reveal that a highly asymmetric wake can be achieved, with the degree of asymmetry contingent upon the ship propagation direction in relation to the plate orientation. This anisotropy is characterized with respect to water depth and to ship length using the dimensionless depth and hull Froude numbers. Laboratory experiments align closely with theoretical predictions, confirming that the asymmetry of the wake can indeed be managed through manipulation of bathymetric conditions.

Résumé: We theoretically study the propagation of light in one-dimensional space- and time-dependent disorder. The disorder is described by a fluctuating permittivity ε(x, t) exhibiting short-range correlations in space and time, without cross correlation between them. Depending on the illumination conditions, we show that the intensity of the average field decays exponentially in space or in time, with characteristic length or time defining the scattering mean-free path ℓs and the scattering mean-free time τs. In the weak scattering regime, we provide explicit expressions for ℓs and τs, that are checked against rigorous numerical simulations.

Résumé: Localisation microscopy of individual molecules allows one to bypass the diffraction limit, revealing cellular organisation on a nanometric scale. This method, which relies on spatial analysis of the signal emitted by molecules, is often limited to the observation of biological objects at shallow depths, or with very few aberrations. The introduction of a temporal parameter into the localisation process through a time-modulated excitation was recently proposed to address these limitations. This method, called ModLoc, is demonstrated here with an alternative flexible strategy. In this implementation, to encode the time-modulated excitation a digital micromirror device (DMD) is used in combination with a fast demodulation approach, and provides a twofold enhancement in localisation precision. Layout: Nowadays, we can use an optical microscope to observe how proteins are organised in 3D within a cell at the nanoscale. By carefully controlling the emission of molecules in both space and time, we can overcome the limitations set by the diffraction limit. This allows us to pinpoint the exact location of molecules more precisely. However, the usual spatial analysis method limits observations to shallow depths or causing low distortion of optical waves. To overcome these restrictions, a recent approach introduces a temporal element to the localisation process. This involves changing the illumination over time to enhance the precision of localisation. This method, known as ModLoc, is showcased here using a flexible and alternative strategy. In this setup, a matrix of micrometric mirrors, working together with a fast demodulation optical module, is used to encode and decode the time-modulated information. This combination results in a twofold improvement in localisation precision.

Résumé: Upon impact, the free surface of a solid metal may eject a cloud of fast and fine particles. Photon Doppler Velocimetry (PDV) is one of the optical diagnostics used to characterize these ejecta. Although the technique provides a direct way to estimate the particle velocities in the single scattering regime, it has been shown that multiple scattering cannot be neglected in real ejecta. Here, we derive a model for PDV measurements starting from the first principles of wave scattering. We establish rigorously the relationship between the specific intensity and the measured signal, as well as the Radiative Transport Equation (RTE) that describes the evolution of the specific intensity upon scattering and absorption in dynamic ejecta, including the effects of inelastic scattering and inhomogeneities in the optical properties. We also establish rigorously the connection between the Monte Carlo scheme used for numerical simulations and the solution to the RTE. Using numerical simulations, we demonstrate the crucial contribution of multiple scattering to PDV spectrograms as well as the effect of statistical inhomogeneities in particle size distribution. These results could substantially impact the analysis of ejecta by PDV.

Résumé: To produce sounds, we adjust the tension of our vocal folds to shape their properties and control the pitch. This efficient mechanism offers inspiration for designing reconfigurable materials and adaptable soft robots. However, understanding how flexible structures respond to a significant static strain is not straightforward. This complexity also limits the precision of medical imaging when applied to tensioned organs like muscles, tendons, ligaments and blood vessels among others. In this article, we experimentally and theoretically explore the dynamics of a soft strip subject to a substantial static extension, up to 180%. Our observations reveal a few intriguing effects, such as the resilience of certain vibrational modes to a static deformation. These observations are supported by a model based on the incremental displacement theory. This has promising practical implications for characterizing soft materials but also for scenarios where external actions can be used to tune properties.

Résumé: Nowadays, automated analysis of the electrocardiogram (ECG) signal is a popular choice to facilitates easy and expert-independent detection of lethal cardiovascular diseases (CVDs). Although, a majority of the state-of-the-art algorithms are found to be lagging due to the use of complicated methodologies, limited dataset, high feature dimension, feature selection or intense classification techniques. In this research an original, easy-to-use ECG based methodology is proposed for completely automated identification of multiple types of critical CVDs. Primarily, after preprocessing the algorithm uses a simplified technique for exact detection of ECG fiducial points. Then, based on detected fiducial points, some well interpretable and prominent ECG time domain features are efficiently extracted and a binary feature matrix has been derived using those features from different leads. Finally, a distinctive score is evaluated from the binary feature matrix calculating the sum of weighted feature value and only by utilizing the score,discrimination between the various types of CVDs is highly detectable. Proficiency of the algorithm is widely evaluated on the 12-lead ECG signal data collected from Physikalisch-Technische-Bundesanstalt (PTB) and PTB-XL database. The algorithm presents promising outcome with average accuracy, sensitivity and specificity of 99.43%, 98.27% and 99.59%, respectively. Evidently, the algorithm is capable enough and efficient as well in comparison with other reported techniques till date. Moreover, the use of a unique score derived from the binary matrix ascertains the exact detection of multiple cardiac abnormalities and the superior classification accuracy makes the algorithm promising for personal computerized health monitoring applications.

Résumé: 3D ceramic architectures are captivating geometrical features with an immense demand in optics. In this work, an additive manufacturing (AM) approach for printing alkaline-earth perovskite 3D microarchitectures is developed. The approach enables custom-made photoresists suited for two-photon lithography, permitting the production of alkaline-earth perovskite (BaZrO3, CaZrO3, and SrZrO3) 3D structures shaped in the form of octet-truss lattices, gyroids, or inspired architectures like sodalite zeolite, and C60 buckyballs with micrometric and nanometric feature sizes. Alkaline-earth perovskite morphological, structural, and chemical characteristics are studied. The optical properties of such perovskite architectures are investigated using cathodoluminescence and wide-field photoluminescence emission to estimate the lifetime rate and defects in BaZrO3, CaZrO3, and SrZrO3. From a broad perspective, this AM methodology facilitates the production of 3D-structured mixed oxides. These findings are the first steps toward dimensionally refined high-refractive-index ceramics for micro-optics and other terrains like (photo/electro)catalysis.

Résumé: In vitro cellular models denote a crucial part of drug discovery programs as they aid in identifying successful drug candidates based on their initial efficacy and potency. While tremendous headway has been achieved in improving 2D and 3D culture techniques, there is still a need for physiologically relevant systems that can mimic or alter cellular responses without the addition of external biochemical stimuli. A way forward to alter cellular responses is using physical cues, like 3D topographical inorganic substrates, to differentiate macrophage-like cells. Herein, protein secretion and gene expression markers for various macrophage subsets cultivated on a 3D topographical substrate are investigated. The results show that macrophages differentiate into anti-inflammatory M2-type macrophages, secreting increased IL-10 levels compared to the controls. Remarkably, these macrophage cells are differentiated into the M2d subset, making up the main component of tumour-associated macrophages (TAMs), as measured by upregulated Il-10 and Vegf mRNA. M2d subset differentiation is attributed to the topographical substrates with 3D fractal-like geometries arrayed over the surface, else primarily achieved by tumour-associated factors in vivo. From a broad perspective, this work paves the way for implementing 3D topographical inorganic surfaces for drug discovery programs, harnessing the advantages of in vitro assays without external stimulation and allowing the rapid characterisation of therapeutic modalities in physiologically relevant environments.

Résumé: A contactless method based on acousto-elastic transmission matrix analysis is proposed to recover the modal properties of weakly damped mechanical structures. The matrix is acquired using eight loudspeakers and a laser vibrometer probing hundreds of points. The matrix analysis is particularly interesting in case of overlapping modes. The proposed measurement set-up and associated data processing using the Singular Value Decomposition are applied to two symmetric samples, a gear and two monobloc impellers. Further analysis are performed taking advantage of their particular modal behavior, common to many rotationally symmetric structures. The method also enables to clearly identify the effect of damages on the modal organization. Additionally, the setup can also be used to excite specific patterns on the elastic structures. Finally, the acousto-elastic results are compared to the ones obtained with a classical impact hammer and high resolution algorithms.