Detection of Microplastics Based on a Liquid-Solid Triboelectric Nanogenerator and a Deep Learning Method

Rapid detection of microfibres in environmental samples using open-source visual recognition models

Microplastics, particularly microfibres (< 5 mm), are a significant environmental pollutant. Detecting and quantifying them in complex matrices is challenging and time-consuming. This study presents two open-source visual recognition models, YOLOv7 and Mask R-CNN, trained on extensive datasets for efficient microfibre identification in environmental samples. The YOLOv7 model is a new introduction to the microplastic quantification research, while Mask R-CNN has been previously used in similar studies. YOLOv7, with 71.4 % accuracy, and Mask R-CNN, with 49.9 % accuracy, demonstrate effective detection capabilities. Tested on aquatic samples from Seyðisfjörður, Iceland, YOLOv7 rapidly identifies microfibres, outperforming manual methods in speed. These models are user-friendly and widely accessible, making them valuable tools for microplastic contamination assessment. Their rapid processing offers results in seconds, enhancing research efficiency in microplastic pollution studies. By providing these models openly, we aim to support and advance microplastic quantification research. The integration of these advanced technologies with environmental science represents a significant step forward in addressing the global issue of microplastic pollution and its ecological and health impacts.

Using artificial intelligence to rapidly identify microplastics pollution and predict microplastics environmental behaviors

With the massive release of microplastics (MPs) into the environment, research related to MPs is advancing rapidly. Effective research methods are necessary to identify the chemical composition, shape, distribution, and environmental impacts of MPs. In recent years, artificial intelligence (AI)-driven machine learning methods have demonstrated excellent performance in analyzing MPs in soil and water. This review provides a comprehensive overview of machine learning methods for the prediction of MPs for various tasks, and discusses in detail the data source, data preprocessing, algorithm principle, and algorithm limitation of applied machine learning. In addition, this review discusses the limitation of current machine learning methods for various task analysis in MPs along with future prospect. Finally, this review finds research potential in future work in building large generalized MPs datasets, designing high-performance but low-computational-complexity algorithms, and evaluating model interpretability.

ZIF-8 functionalized S-tapered fiber-optic sensor for polystyrene nanoplastics detection by electrostatic adsorption

Microplastic (MP) residues in marine have become an increasingly serious environmental pollution issue, and in recent years the detection of MPs in marine started to attract worldwide research interests. Optical-fiber-based environmental sensors have been extensively employed for their several merits such as high sensitivity, pressure resistance, compactness and ease of constructing communication networks. However, fiber-optic refractive index sensors are not specifically developed for distinguishing MPs from other inorganic particles suspended in water. In this paper, an metal-organic framework (MOF) ZIF-8 functionalized S-tapered fiber (STF) sensor is proposed for specific detection of polystyrene nanoplastics (PSNPs) in aqueous environment. ZIF-8 coordination nanoporous polymers with different film thickness were immobilized over the surface of the fabricated STF structure based on self-growth technique and yielding a large surface area over the sensor surface. High sensitivity detection can be achieved by converting the concentration perturbation of PSNPs into evanescent waves over the ZIF-8 functionalized STF surface through the strong electrostatic adsorption effect and π-π stacking, while the fabricated sensor is insensitive to gravels with silica as the primary component in water. It is found that the proposed detector with 18 film layers achieves a sensitivity up to 114.1353nm/%(w/v) for the PSNPs concentration range of 0.01 %(w/v) to 0.08 %(w/v).

Advancing microplastic surveillance through photoacoustic imaging and deep learning techniques

Microplastic contamination presents a significant global environmental threat, yet scientific understanding of its morphological distribution within ecosystems remains limited. This study introduces a pioneering method for comprehensive microplastic assessment and environmental monitoring, integrating photoacoustic imaging and advanced deep learning techniques. Rigorous curation of diverse microplastic datasets enhances model training, yielding a high-resolution imaging dataset focused on shape-based discrimination. The introduction of the Vector-Quantized Variational Auto Encoder (VQVAE2) deep learning model signifies a substantial advancement, demonstrating exceptional proficiency in image dimensionality reduction and clustering. Furthermore, the utilization of Vector Quantization Microplastic Photoacoustic imaging (VQMPA) with a proxy task before decoding enhances feature extraction, enabling simultaneous microplastic analysis and discrimination. Despite inherent limitations, this study lays a robust foundation for future research, suggesting avenues for enhancing microplastic identification precision through expanded sample sizes and complementary methodologies like spectroscopy. In conclusion, this innovative approach not only advances microplastic monitoring but also provides valuable insights for future environmental investigations, highlighting the potential of photoacoustic imaging and deep learning in bolstering sustainable environmental monitoring efforts.

Maximizing Triboelectric Nanogenerators by Physics-Informed AI Inverse Design

Triboelectric nanogenerators offer an environmentally friendly approach to harvesting energy from mechanical excitations. This capability has made them widely sought-after as an efficient, renewable, and sustainable energy source, with the potential to decrease reliance on traditional fossil fuels. However, developing triboelectric nanogenerators with specific output remains a challenge mainly due to the uncertainties associated with their complex designs for real-life applications. Artificial intelligence-enabled inverse design is a powerful tool to realize performance-oriented triboelectric nanogenerators. This is an emerging scientific direction that can address the concerns about the design and optimization of triboelectric nanogenerators leading to a next generation nanogenerator systems. This perspective paper aims at reviewing the principal analysis of triboelectricity, summarizing the current challenges of designing and optimizing triboelectric nanogenerators, and highlighting the physics-informed inverse design strategies to develop triboelectric nanogenerators. Strategic inverse design is particularly discussed in the contexts of expanding the four-mode analytical models by physics-informed artificial intelligence, discovering new conductive and dielectric materials, and optimizing contact interfaces. Various potential development levels of artificial intelligence-enhanced triboelectric nanogenerators are delineated. Finally, the potential of physics-informed artificial intelligence inverse design to propel triboelectric nanogenerators from prototypes to multifunctional intelligent systems for real-life applications is discussed.

Synergizing Machine Learning Algorithm with Triboelectric Nanogenerators for Advanced Self-Powered Sensing Systems

The advancement of the Internet of Things (IoT) has increased the demand for large-scale intelligent sensing systems. The periodic replacement of power sources for ubiquitous sensing systems leads to significant resource waste and environmental pollution. Human staffing costs associated with replacement also increase the economic burden. The triboelectric nanogenerators (TENGs) provide both an energy harvesting scheme and the possibility of self-powered sensing. Based on contact electrification from different materials, TENGs provide a rich material selection to collect complex and diverse data. As the data collected by TENGs become increasingly numerous and complex, different approaches to machine learning (ML) and deep learning (DL) algorithms have been proposed to efficiently process output signals. In this paper, the latest advances in ML algorithms assisting solid-solid TENG and liquid-solid TENG sensors are reviewed based on the sample size and complexity of the data. The pros and cons of various algorithms are analyzed and application scenarios of various TENG sensing systems are presented. The prospects of synergizing hardware (TENG sensors) with software (ML algorithms) in a complex environment and their main challenges for future developments are discussed.