Nuno Azevedo Silva

Abstract
Optical tweezers are an interesting tool to enable single cell analysis, especially when coupled with optical sensing and advanced computational methods. Nevertheless, such approaches are still hindered by system operation variability, and reduced amount of data, resulting in performance degradation when addressing new data sets. In this manuscript, we describe the deployment of an automatic and intelligent optical tweezers setup, capable of trapping, manipulating, and analyzing the physical properties of individual microscopic particles in an automatic and autonomous manner, at a rate of 4 particle per min, without user intervention. Reproducibility of particle identification with the help of machine learning algorithms is tested both for manual and automatic operation. The forward scattered signal of the trapped PMMA and PS particles was acquired over two days and used to train and test models based on the random forest classifier. With manual operation the system could initially distinguish between PMMA and PS with 90% accuracy. However, when using test datasets acquired on a different day it suffered a loss of accuracy around 24%. On the other hand, the automatic system could classify four types of particles with 79% accuracy maintaining performance (around 1% variation) even when tested with different datasets. Overall, the automated system shows an increased reproducibility and stability of the acquired signals allowing for the confirmation of the proportionality relationship expected between the particle size and its friction coefficient. These results demonstrate that this approach may support the development of future systems with increased throughput and reliability, for biosciences applications.

Abstract
Laser-induced breakdown spectroscopy is a spectroscopic technique that allows for fast elemental mapping of heterogeneous samples. Yet, detailed maps need high-resolution sampling grids, which can turn the task into a time-consuming process and can increase sample damage. In this work, we present the implementation of an imaged-based intelligent mesh algorithm that makes use of superpixel segmentation to optimize elemental mapping processes. Our results show that the approach can increase the elemental mapping resolution and decrease acquisition times, fostering opportunities for applications that benefit from minimal sample damage such as heritage analysis, or timely analysis such as industrial applications. © 2022 The Author(s)

Abstract
Imaging the spatial distribution of chemical elements at a sample surface is a common application of laserinduced breakdown spectroscopy with vast scientific and technological applications. Yet, typical imaging solutions only explore the creation of two-dimensional maps, which can limit the interpretability of the results and further diagnostics in three-dimensional settings. Within this context, this work explores the combination of spectral imaging techniques and photogrammetry to deploy a versatile solution for the creation of threedimensional spectral imaging models. First, by making use of a numerical algorithm that is able to match features in the spectral image with those of the three-dimensional model, we show how to match the mesh from distinct sensor modalities. Then, we describe a possible visualization workflow, making use of dedicated photogrammetry and visualization software to easily deploy interactive models. Overall, the results demonstrate the versatility of our approach and pave for the development of novel spectral imaging diagnostic strategies that are able to deliver better qualitative analysis and insight in the three-dimensional space.

Abstract
Extreme Learning Machines (ELMs) are a versatile Machine Learning (ML) algorithm that features as the main advantage the possibility of a seamless implementation with physical systems. Yet, despite the success of the physical implementations of ELMs, there is still a lack of fundamental understanding in regard to their optical implementations. In this context, this work makes use of an optical complex media and wavefront shaping techniques to implement a versatile optical ELM playground to gain a deeper insight into these machines. In particular, we present experimental evidences on the correlation between the effective dimensionality of the hidden space and its generalization capability, thus bringing the inner workings of optical ELMs under a new light and opening paths toward future technological implementations of similar principles.

Abstract
Anode-less batteries are a promising innovation in energy storage technology, eliminating the need for traditional anodes and offering potential improvements in efficiency and capacity. Here, we have fabricated and tested two types of anode-less pouch cells, the first using solely a copper negative current collector and the other the same current collector but coated with a nucleation seed ZnO layer. Both types of cells used the same all-solid-state electrolyte, Li2.99Ba0.005ClO composite, in a cellulose matrix and a LiFePO4 cathode. Direct and indirect methods confirmed Li metal anode plating after charging the cells. The direct methods are X-ray photoelectron spectroscopy (XPS) and laser-induced breakdown spectroscopy (LIBS), a technique not divulged in the battery world but friendly to study the surface of the negative current collector, as it detects lithium. The indirect methods used were electrochemical cycling and impedance and scanning electron microscopy (SEM). It became evident the presence of plated Li on the surface of the current collector in contact with the electrolyte upon charging, both directly and indirectly. A maximum average lithium plating thickness of 2.9 mu m was charged, and 0.13 mu m was discharged. The discharge initiates from a maximum potential of 3.2 V, solely possible if an anode-like high chemical potential phase, such as Li, would form while plating. Although the ratings and energy densities are minor in this study, it was concluded that a layer of ZnO, even at 25 degrees C, allows for higher discharge power for more hours than plain Cu. It was observed that where Li plates on ZnO, Zn is not detected or barely detected by XPS. The present anode-less cells discharge quickly initially at higher potentials but may hold a discharge potential for many hours, likely due to the ferroelectric character of the electrolyte.