Adaptive Learning for Soil Classification in Laser-Induced Breakdown Spectroscopy Streaming

LIBS measurements inside the FTU vacuum vessel by using a robotic arm

Time-resolved data acquisition is crucial for compositional analysis using Laser-Induced Breakdown Spectroscopy (LIBS). It can be managed by adjusting the delay time and gate width of the spectrometer. This study describes the compositional analysis of molybdenum (Mo) ore utilizing charge coupled device (CCD) and intensified charge-coupled device (ICCD) based LIBS systems. LIBS was successfully used for both quantitative and qualitative analysis of molybdenum ore. Spectra were recorded with CCD and ICCD spectrometers. Plasma parameters were estimated from LIBS spectra to diagnose the plasma thoroughly. A performance comparison was conducted among LIBS measurements using ICCD, CCD, and x-ray fluorescence (XRF). The quantitative analysis of molybdenum ore was carried out using calibration-free LIBS. Both major and trace elements were identified in the spectra. Optical emission spectra showed varying element concentrations of Mo, silicon, aluminum, iron, calcium, potassium, tungsten, and sodium. The XRF was used to compare the performance of LIBS techniques. Results from the compositional analysis of molybdenum ore using ICCD, CCD-based LIBS, and XRF showed good agreement. It is demonstrated that CCD can produce nearly the same results as ICCD within acceptable error margins. The CCD-based LIBS system is a cost-effective analytical method with additional benefits for field applications because it does not require a temperature-controlled environment for operation.

A feasibility of real-time quantitative multielemental analysis during parts growth by additive manufacturing technology has been demonstrating for the first time using laser induced breakdown spectroscopy (LIBS). A low weight and compact LIBS probe was developed to equip the laser cladding head installed at six axis industrial robot. LIBS quantitative analysis of key components (carbon and tungsten) has been demonstrated in real-time conditions during synthesis of high wear resistant coatings of nickel alloy reinforced with tungsten carbide particles. Hot solidified clad as well as a melt pool surface was sampled by LIBS probe but better analytical results were achieved for the latter case due to low partial sinking of heavy tungsten carbide particles in nickel matrix. LIBS system was calibrated for quantitative measurements by offline analysis with energy dispersive X-ray spectrometry (EDX). We didn’t observe any impact of laser ablation (LIBS sampling) on clad dimensions even when LIBS probe ablated melt pool. LIBS system was capable to quantify concentrations for both light (carbon, silicon) and heavy (nickel, tungsten, chromium, etc.) elements at frequency of 10 Hz. A good correspondence was achieved between online LIBS measurements and offline analysis opens wide perspectives for LIBS technique as a powerful tool for controlling process of parts growth by additive manufacturing

Front-end signal processing for metal scrap classification using online measurements based on laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy (LIBS) is one of the most promising methods for quantitative in-situ determination of fuel retention in plasma-facing components (PFCs) of magnetically confined fusion devices like ITER and JET. In this article, the current state of understanding in LIBS development for fusion applications will be presented, based on a complete review of existing results and complemented with newly obtained data. The work has been performed as part of a research programme, set up in the EUROfusion Consortium, to address the main requirements for ITER: (a) quantification of fuel from relevant surfaces with high sensitivity, (b) the technical demonstration to perform LIBS with a remote handling system and (c) accurate detection of fuel at ambient pressures relevant for ITER. For the first goal, the elemental composition of ITER-like deposits and proxies to them, including deuterium (D) or helium (He) containing W–Be, W, W–Al and Be–O–C coatings, was successfully determined with a typical depth resolution ranging from 50 up to 250 nm per laser pulse. Deuterium was used as a substitute for tritium (T) and in the LIBS experiments deuterium surface densities below 1016 D/cm2 could be measured with an accuracy of ∼30%, confirming the required high sensitivity for fuel-retention investigations. The performance of different LIBS configurations was explored, comprising LIBS systems based on single pulse (pulse durations: ps–ns) and double pulse lasers with different pulse durations. For the second goal, a remote handling application was demonstrated inside the Frascati-Tokamak-Upgrade (FTU), where a compact, remotely controlled LIBS system was mounted on a multipurpose deployer providing an in-vessel retention monitor system. During a shutdown phase, LIBS was performed at atmospheric pressure, for measuring the composition and fuel content of different area of the stainless-steel FTU first wall, and the titanium zirconium molybdenum alloy tiles of the toroidal limiter. These achievements underline the capability of a LIBS-based retention monitor, which complies with the requirements for JET and ITER operating in DT with a beryllium wall and a tungsten divertor. Concerning the capabilities of LIBS at pressure conditions relevant for ITER, quantitative determination of the composition of PFC materials at ambient pressures up to 100 mbar of N2, the D content could be determined with an accuracy of 25%, while for atmospheric pressure conditions, an accuracy of about 50% was found when using single-pulse lasers. To improve the LIBS performance in atmospheric pressure conditions, a novel approach is proposed for quantitative determination of the retained T and the D/T ratio. This scenario is based on measuring the LIBS plume emission at two different time delays after each laser pulse. On virtue of application of a double pulse LIBS system, for LIBS application at N2 atmospheric pressure the distinguishability of the spectra from H isotopes could be significantly improved, but further systematic research is required.

Spectrochemical analysis of powdered biological samples using transversely excited atmospheric carbon dioxide laser plasma excitation

Depth profiling of surface degradation of silicone rubber composite insulators by remote laser-induced breakdown spectroscopy

LIBS (Laser Induced Breakdown Spectroscopy) systems are capable of real-time chemical analysis with little or no sample preparation. A Q-switched laser is configured such that laser induced plasma is produced on targeted material. Chemical element line spectra are created, collected and analyzed by a fiber spectrometer. Line spectra emission data is instantly viewed on a head mounted display. "Eye-safe" Class I erbium glass lasers provide for insitu LIBS applications without the need for eye-protection goggles. This is due to the fact that Megawatt peak power Q-switched lasers operating in the narrow spectral window between 1.5um and 1.6um are approximately 8000 times more "eye-safe" than other laser devices operating in the UV, visible and near infrared. In this work we construct and demonstrate a LIBS system that includes a hand held eye-safe laser gun. The laser gun is fitted with a micro-integrating sphere in-situ target interface and is designed to facilitate chemical analysis in remote locations. The laser power supply, battery pack, computer controller and spectrophotometer components are packaged into a utility belt. A head mounted display is employed for "hands free" viewing of the emitted line spectra. The system demonstrates that instant qualitative and semi-quantitative chemical analyses may be performed in remote locations utilizing lightweight commercially available system components ergonomically fitted to the operator.

Machine Learning (ML) models have been applied to solve problems in various fields, which necessarily involves proper evaluation of models to ensure performance. Once deployed, ML models are subject to performance issues, such as those related to changes in data (drift). This type of issue has prompted efforts in model analysis and maintenance, as well as in continual learning, which seeks the ability to continuously learn from a (continuous) stream of data. Therefore, it's important to understand and develop methodologies that can be used to evaluate ML models, making their use in real-world environments feasible. Amongst current areas of application for ML, one that stands out, in particular, is Machine Learning for Healthcare, especially in conjunction with Software for Decision Support of Medical Applications, which presents specific challenges for the evaluation and monitoring of models, particularly given that incorrect prediction or classification can lead to life-threatening situations. This paper presents a systematic literature review that aims at identifying state-of-the-art techniques for evaluating and maintaining ML models for healthcare in effective use in the real world.

Response surface methodology for optimizing LIBS testing parameters: A case to conduct the elemental contents analysis in soil

Aim: The majority of machine learning (ML) models in healthcare are built on retrospective data, much of which is collected without explicit patient consent for use in artificial intelligence (AI) and ML applications. The primary aim of this study was to evaluate whether clinicians and scientific researchers themselves consent to provide their own data for the training of ML models. Materials and Methods: This survey was conducted through an anonymous online survey, utilizing platforms such as Telegram, LinkedIn, and Viber. The target audience comprised specific international groups, primarily Russian, German, and English-speaking, of clinicians and scientific researchers. These participants ranged in their levels of expertise and experience, from beginners to veterans. The survey centered on a singular, pivotal question: “Do You Consent to the Use of Your Biological and Private Data for Training Machine Learning and AI Models?” Respondents had the option to choose from three responses: “Yes” and “No”. Results: The survey was conducted in January 2024. A total of 119 unique and verified individuals participated in the survey. The results revealed that only 50% of respondents (63 persons) expressed consent to provide their own data for the training of ML and AI models. Conclusion: In the development of ML and AI models, particularly open-source ones, it is crucial to ascertain whether participants are willing to provide their private data. While ML algorithms can transform the nature of data, it is important to remember that the primary owner of this data is the individual. Our findings show that in 50% of the cases, even participants from scientific research and clinical backgrounds – individuals typically accountable for ensuring data quality in AI and ML model development – do not consent to the use of their data in AI and ML settings. This highlights the need for more stringent consent processes and ethical considerations in the utilization of personal data in AI and ML research.

In situ radiogenic isotope measurements to obtain the absolute age of geologic events on planets are of great scientific value. In particular, K-Ar isochrons are useful because of their relatively high technical readiness and high accuracy. Because this isochron method involves spot-by-spot K measurements using laser-induced breakdown spectroscopy (LIBS) and simultaneous Ar measurements with mass spectrometry, LIBS measurements are conducted under a high vacuum condition in which emission intensity decreases significantly. Furthermore, using a laser power used in previous planetary missions is preferable to examine the technical feasibility of this approach. However, there have been few LIBS measurements for K under such conditions. In this study, we measured K contents in rock samples using 30 mJ and 15 mJ energy lasers under a vacuum condition (10-3 Pa) to assess the feasibility of in situ K-Ar dating with lasers comparable to those used in NASA's Curiosity and Mars 2020 missions. We obtained various calibration curves for K using internal normalization with the oxygen line at 777 nm and continuum emission from the laser-induced plasma. Experimental results indicate that when K2O < 1.1 wt%, a calibration curve using the intensity of the K emission line at 769 nm normalized with that of the oxygen line yields the best results for the 30 mJ laser energy, with a detection limit of 88 ppm and 20% of error at 2400 ppm of K2O. Futhermore, the calibration curve based on the K 769 nm line intensity normalized with continuum emission yielded the best result for the 15 mJ laser, giving a detection limit of 140 ppm and 20% error at 3400 ppm K2O. Error assessments using obtained calibration models indicate that a 4 Ga rock with 3000 ppm K2O would be measured with 8% (30 mJ) and 10% (15 mJ) of precision in age when combined with mass spectrometry of 40Ar with 10% of uncertainty. These results strongly suggest that high precision in situ isochron K-Ar dating is feasible with a laser used in previous and upcoming Mars rover missions.

Machine learning-based water quality prediction using octennial in-situ Daphnia magna biological early warning system data

Sensitive detection of chlorine in iron oxide by single pulse and dual pulse laser-induced breakdown spectroscopy

Machine Learning Can Accurately Predict Overnight Stay, Readmission, and 30-Day Complications Following Anterior Cruciate Ligament Reconstruction