Conference Agenda

Microplastics (MPs) are now documented across aquatic and terrestrial environments, biota, food systems, and human tissues, yet quantitative understanding of their flows, transformations, and impacts remains fragmented. Their heterogeneity in size, morphology, polymer composition, and environmental behavior challenges conventional environmental modeling approaches. Material flow analysis (MFA) and life cycle assessment (LCA) have emerged as two influential but largely disconnected frameworks for evaluating MP releases, transport, and impacts. This systematic mapping review assesses the current state, technical robustness, and research gaps of MFA and LCA studies that explicitly incorporate MPs, with the goal of identifying high-impact opportunities for methodological integration and advancement.

A structured literature search was conducted using manual mining in Google Scholar combined with backward citation tracing to maximize coverage of interdisciplinary work. Thirty-seven peer-reviewed studies were identified that explicitly incorporated MPs within MFA or LCA frameworks. For each study, we extracted methodological characteristics (static vs. dynamic modeling, spatial and temporal resolution, treatment of uncertainty), system boundaries, MP typology (primary, secondary, nanoplastics), data sources, and treatment of fate, exposure, and impact pathways. To strengthen completeness and contextualize modeling gaps, an additional 99 complementary modeling studies (including fate and transport models and quantitative risk assessments) were analyzed to identify transferable methods and data streams relevant to MFA and LCA integration.

Results reveal substantial methodological limitations and fragmentation. Only ten MFA studies incorporated MPs, predominantly static and regionally constrained models with limited temporal dynamics and uncertainty analysis. Most focused exclusively on primary MPs; only two incorporated secondary MPs, and none explicitly modeled nanoplastics. Twenty-seven LCA studies were identified, with the majority integrating MPs at the impact assessment stage through development or modification of characterization factors. Few studies incorporated MPs within life cycle inventory modeling, limiting traceability of emission pathways and transformation processes. Across both MFA and LCA, differentiation between MP size classes, polymer types, and environmental compartments was inconsistent, constraining comparability and technical rigor.

Key research priorities with high potential impact include: development of harmonized data collection protocols; establishment of a centralized, open-access global MP database; integration of dynamic MFA with fate and exposure modeling; generation of standardized ecotoxicity and human toxicity data; explicit modeling of secondary MP formation and nanoplastics; differentiation of MP types and pathways within life cycle stages; and incorporation of complex mechanisms such as vector (“Trojan horse”) effects.

By critically evaluating methodological quality and completeness across modeling approaches, this review demonstrates that advancing MP science will require integrated, multi-model frameworks rather than siloed analyses. Such integration is essential for generating decision-relevant insights capable of informing policy, product design, and environmental management strategies addressing plastic pollution.

Several Life Cycle Assessments (LCAs) have been conducted to assess the potential environmental impacts of packaging materials, with notable improvements in data quality over the years through regular updates and development. However, several data gaps remain unaddressed for packaging LCAs, particularly around the availability and quality of inventory data. This study explores data gaps in packaging LCAs, focusing on packaging-related product losses and application-specific end-of-life (EoL) rates.

Although packaging cannot be considered independently from the products it contains, LCA studies have primarily focused on packaging materials, often neglecting the potential impacts of the packaged products. Packaging failures and breakage during manufacturing, transportation, and distribution—and the resulting product losses—are frequently overlooked or omitted. This oversight represents a key gap because the environmental impacts of product losses can greatly exceed those of packaging. Data and methods from recent research addressing these gaps will be showcased.

EoL stage modeling is another area impacted by data gaps. Packaging LCAs sometimes rely on outdated estimates and generalized material-specific EoL disposition rates (e.g., landfill, recycling, incineration), typically sourced from country-level datasets (e.g., the US EPA and Eurostat) for different materials. With increasing investments towards addressing EoL challenges and improving recycling rates, developing more representative and robust LCAs will necessitate high-resolution, frequently updated data on EoL disposition rates. Research methods and insights from our work on this topic will be discussed.

Multi-stakeholder collaboration and participation throughout the product supply chain, along with engagement by LCA practitioners, can help bridge these data gaps by developing high-quality, reliable, and transparent life-cycle inventory datasets for packaging systems. The datasets developed can be integrated into LCA databases to support more comprehensive life-cycle inventories, serving as a resource for LCA practitioners to conduct robust LCAs and to inform business and policy decisions on packaging.

Chemical recycling is increasingly explored to address limitations of mechanical recycling, particularly for plastic waste streams that are difficult to recover through conventional pathways. However, the sustainability of emerging chemical recycling technologies is highly sensitive to early-stage process design choices, including solvent selection, which can influence the energy use, environmental impacts, and chemical hazards. Despite this, solvent choices are often evaluated primarily on the basis of reaction yield or efficiency, with limited attention to sustainability trade-offs. This study examines how solvent selection affects the sustainability performance of poly(ethylene terephthalate) (PET) chemical recycling to p-phenylenediamine (PPD), a high-value monomer used in advanced polymer applications.

The system boundary encompasses post-consumer plastic waste collection, mechanical preprocessing to obtain clean PET flakes, chemical depolymerization, and monomer extraction and purification. Two depolymerization routes, based on methanol and ethylene glycol, are experimentally evaluated to quantify differences in material and energy requirements and their implications for life cycle inventory development. For the monomer extraction step, a set of candidate aprotic solvents is initially screened and subsequently narrowed based on experimental performance. Selected solvents are then evaluated using a sustainability matrix that integrates environmental impacts, energy demand for process and solvent recovery, chemical costs, and chemical hazard profiles from NPFA 704.

Ethylene glycol–based depolymerization exhibits lower environmental impacts than methanol-based depolymerization, largely due to lower energy demand and reduced solvent losses rather than differences in reaction yield. For monomer extraction, solvents with similar extraction efficiencies show substantially different environmental, cost, and hazard characteristics. Ethyl acetate performs favorably across these metrics, reducing global warming potential by 29.5% relative to the baseline solvent while also reducing the cost and chemical hazards.

Evaluating solvent options based on process performance alone can overlook important sustainability trade-offs. By linking experimental data with LCA results, cost, and hazard indicators, this work provides a structured approach for assessing emerging chemical recycling technologies and supports more informed decision-making at early stages of process development.

Building material stock assessment (BMSA) and material flow analysis (MFA) increasingly underpin circular-economy strategies by quantifying where demolition-driven material outflows and recoverable stocks will emerge, while also helping prioritize envelope retrofit opportunities that reduce operational demand and protect salvage value. Traditional data collection methods for existing building stock analyses, such as plan reviews, surveys, and on-site audits, are labor-intensive, time-consuming, and often costly, which typically limits them to relatively small samples. In contrast, imaging sensors and street-level imagery offer a more cost-effective and repeatable means of observing the building stock, enabling the creation of spatially dense and extensive datasets. Computer vision enables machines to interpret visual data and is increasingly used to analyze street view images for understanding building characteristics. This study develops and evaluates a scalable computer vision workflow to extract façade material composition and window-to-wall ratio (WWR) from Google Street View (GSV) imagery, enabling neighborhood-scale mapping of building envelope characteristics that can support circularity-oriented planning. The framework is applied to a case study in Philadelphia's Fishtown neighborhood, known for its historic pre-1925 masonry rowhouses, using three data sources: (i) a “clean” dataset of approximately 400 cropped single-façade images, (ii) a larger “raw” streetscape dataset with over 5,000 uncropped Google Street View images showing multiple attached buildings and real-life obstructions, and (iii) city building footprint polygons and attributes that help identify candidate buildings and aggregate predictions at the building level.

A dual-paradigm framework is developed and evaluated. Initially, a supervised object-detection pipeline trained with the You Only Look Once (YOLO) computer vision model uses manually annotated façades, including labels for various cladding materials (such as brick, stucco, vinyl) and windows, to enable both material mixture estimation (fractional material shares per façade) and WWR calculation. Subsequently, a zero-shot vision–language pipeline leverages pre-trained foundation models with text prompts aligned to the same material classes to estimate material mixtures and WWR without needing task-specific training. A label-efficiency experiment measures labeling effort by training YOLO with different label budgets (e.g., 50, 100, 200, 300 images) and evaluating all YOLO versions, as well as the zero-shot approach, on a consistent test set (~100 images). Performance metrics include dominant-material accuracy, F1 scores, mean absolute error in material-share estimates, and errors in WWR.

Anticipated results include label-efficiency curves identifying the “break-even” labeling point at which supervised performance matches or exceeds zero-shot baselines for both material mixtures and WWR, and a recommended model selection strategy for similar façade-mapping efforts. For large-scale deployment, the best-performing YOLO configuration and the zero-shot pipeline are applied to over 5,000 raw GSV images to generate building-level façade material mixtures and WWR estimates across the neighborhood. Agreement and uncertainty are characterized through dominant-material consistency and vector-distance similarity (e.g., L1 distance) between model-derived mixtures, producing spatial confidence maps that highlight high-trust areas and prioritized locations for additional labeling, field verification, or workflow refinement. The resulting datasets and uncertainty-aware maps provide a foundational, transferable data layer for subsequent material stock estimation, enabling integration of envelope characteristics into circular economy workflows, including MFA/BMSA-driven material recovery prioritization and targeted retrofit screening.

Helium is a critical material and noble gas essential for superconducting technologies, semiconductor manufacturing, and advanced research. Helium is extracted from a small number of natural gas reservoirs mostly in United States, Qatar, and Algeria where it is present in economic quantities. Following separation from hydrocarbon natural gas, helium is stored and transported as cryogenic liquid (boiling point 4.2 K (-269° C)), then distributed as liquid or compressed gas. Given its tendency to diffuse, effuse, and boil off, helium logistics are complex and time constrained. Global helium production in 2019 was around 160 million cubic metres, amounting to merely 27 kt. Roughly 30% of refined helium finds end-use as a cryogenic liquid coolant in magnetic resonance imaging (MRI) machines in healthcare. Other major uses are in research laboratories and semiconductor manufacturing. Importantly, the industry is currently undergoing a global supply shift away from 100-years of USA dominance – and there have been several regional supply shortages. There is little extant research on the element, its resource criticality, life cycle assessment or environmental footprint. This study applies Material Flow Analysis (MFA) to map global helium stocks and flows, from extraction to end-use. We evaluate the sustainability of current usage patterns through the lens of industrial ecology, including substitution potential and material efficiency opportunities. As a noble gas, helium does not change its form or mix over its life cycle, thus providing ease of tracking through the global supply chain. To characterize the helium lifecycle, an original database was compiled covering the helium processes and the supply chain, including liquid and gaseous modes of supply and associated losses. Material flows to 8 end-use categories were quantified and stocks were modeled for helium resident in MRI machines and other smaller industries. Our analysis reveals that the helium economy follows a near-perfect linear consumption model. In liquid form, helium is dense and relatively efficient to transport, but in some pathways up to 50% of material is lost between extraction and use. Hence there is overall urgency to get liquid helium to end-users quickly. High-losses occur during liquid transfer and open-cycle venting. Technical interventions such as vacuum-jacketed transfer and closed-loop recovery systems could reduce dissipative flows. Technology substitution of helium is viable for some lifting and welding applications, but cryogenic uses like MRI exhibit near-zero elasticity of substitution, necessitating strict conservation measures. Increasing in-application recycling and localized in-facility recovery and reuse efforts could significantly extend the residence times of in-use helium in large systems. Overall, there is a trend toward additional new resource supply. Our findings provide new data on this under-studied critical material, its linear flows, and high loss rates, and inform the current transition of the helium economy to be more resilient.