The influence of complex matrices on method performance in extracting and monitoring for microplastics.

Previous studies have evaluated method performance for quantifying and characterizing microplastics in clean water, but little is known about the efficacy of procedures used to extract microplastics from complex matrices. Here we provided 15 laboratories with samples representing four matrices (i.e., drinking water, fish tissue, sediment, and surface water) each spiked with a known number of microplastic particles spanning a variety of polymers, morphologies, colors, and sizes. Percent recovery (i.e., accuracy) in complex matrices was particle size dependent, with ∼60-70% recovery for particles >212 µm, but as little as 2% recovery for particles <20 µm. Extraction from sediment was most problematic, with recoveries reduced by at least one-third relative to drinking water. Though accuracy was low, the extraction procedures had no observed effect on precision or chemical identification using spectroscopy. Extraction procedures greatly increased sample processing times for all matrices with the extraction of sediment, tissue, and surface water taking approximately 16, 9, and 4 times longer than drinking water, respectively. Overall, our findings indicate that increasing accuracy and reducing sample processing times present the greatest opportunities for method improvement rather than particle identification and characterization.

Quantitative assessment of visual microscopy as a tool for microplastic research: Recommendations for improving methods and reporting.

Microscopy is often the first step in microplastic analysis and is generally followed by spectroscopy to confirm material type. The value of microscopy lies in its ability to provide count, size, color, and morphological information to inform toxicity and source apportionment. To assess the accuracy and precision of microscopy, we conducted a method evaluation study. Twenty-two laboratories from six countries were provided three blind spiked clean water samples and asked to follow a standard operating procedure. The samples contained a known number of microplastics with different morphologies (fiber, fragment, sphere), colors (clear, white, green, blue, red, and orange), polymer types (PE, PS, PVC, and PET), and sizes (ranging from roughly 3-2000 µm), and natural materials (natural hair, fibers, and shells; 100-7000 µm) that could be mistaken for microplastics (i.e., false positives). Particle recovery was poor for the smallest size fraction (3-20 µm). Average recovery (±StDev) for all reported particles >50 µm was 94.5 ± 56.3%. After quality checks, recovery for >50 µm spiked particles was 51.3 ± 21.7%. Recovery varied based on morphology and color, with poorest recovery for fibers and the largest deviations for clear and white particles. Experience mattered; less experienced laboratories tended to report higher concentration and had a higher variance among replicates. Participants identified opportunity for increased accuracy and precision through training, improved color and morphology keys, and method alterations relevant to size fractionation. The resulting data informs future work, constraining and highlighting the value of microscopy for microplastics.

Crack Patterns of Environmental Plastic Fragments.

Secondary microplastics usually come from the breakdown of larger plastics due to weathering and environmental stress cracking of plastic wastes. In the present study, 5013 plastic fragments were collected from coastal beaches, estuary dikes, and lake banks in China. The fragment sizes ranged from 0.2 to 17.1 cm, and the dominant polymers were polypropylene and polyethylene. Cracks were observed on the surfaces of 49-56% of the fragments. Based on the extracted crack images, we proposed a general crack pattern system including four crack types with specific definitions, abbreviations, and symbols. The two-dimensional spectral analysis of the cracks suggests that the first three patterns showed good regularity and supported the rationality of the pattern system. Some crack metrics (e.g., line density) were closely correlated with the carbonyl index and additives (e.g., phthalate esters) of fragments. For crack investigation in field, we proposed a succinct protocol, in which five crack ranks were established to directly characterize the degree of cracking based on the line density values. The system was successfully applied to distinguish the differences in crack features at two representative sites, which indicates that crack pattern is a useful tool to describe the morphological changes of plastic surfaces in the environment.

Linking the physical and chemical characteristics of single small microplastics or nanoplastics via photolithographic silicon substrates.

Small microplastics and nanoplastics are of growing concern as they pose more risks to ecological and human health than larger particles. However, characterization of the morphological and chemical features of single particles faces a major challenge if instrumental combination is not available. Here, we developed a marker system via computer aided design and crafted it on a silicon substrate (8 × 8 mm) via direct-write lithography. We dripped 20 µL of a solution containing tiny particles extracted from a highly weathered plastic fragment onto a silicon substrate. After the solution was oven-dried, the polymer composition of particles down to 895 nm was located via multiple markers and identified using micro-Raman. The lithographic substrate was then transferred to a scanning electron microscope with energy dispersive spectroscopy capability, and the surface morphology and element distribution were captured for the same particle. Similarly, the morphology and surface elevation were characterized using a scanning electron microscope and an atomic force microscope. The average retrieval rate for particles reached 86% if all characterization experiments were conducted within one week. Our results suggest that photolithographic silicon substrates provide a novel and economical way to link the physicochemical characteristics of small microplastics and parts of nanoplastics.

Microfiber fallout during dining and potential human intake.

The pervasiveness of microfibers, including fibrous microplastics indoors and outdoors, has drawn attention. However, some places such as the dining environment that are closely related to human diet and health have been neglected. Here, we characterized short-term microfiber fallout in different dining spots and conducted long-term monitoring in a college cafeteria. The results showed that the microfiber abundance of restaurants during the peak hour of dinnertime (75 ± 19 MFs/plate/meal) was approximately two times that of households (36 ± 23 MFs/plate/meal). The high microfiber abundance was positively correlated with strong human activities (i.e., sitting rate of people) in restaurants, which was verified by the kinetics data of the cafeteria (R2 =0.871, p = 0.000). Cotton (63%), polyester (17%), and rayon (14%) were the top three detected microfibers via µ-FTIR, and cloth friction can aggravate fiber shedding significantly. Moreover, high hairiness and short staple yarn style were likely to increase the formation of microfibers. Additionally, room structure can obviously influence microfiber abundance that households without separate dining rooms showed three times higher microfiber abundance (66 MFs/plate/meal) than those (21 MFs/plate/meal) with separate dining rooms, because partition walls were verified to effectively reduce fiber transport. Collectively, microfiber fallout during dining deserves our great attention, which may induce human intake of 63-232 MFs/person/d.

Separation and enrichment of nanoplastics in environmental water samples via ultracentrifugation.

Nanoplastics are an emerging contaminant in aquatic environments. However, analytical methods for the separation, concentration, and identification of nanoplastics, which are essential to assess nanoplastic presence in the environment, are lacking. Here, we developed a new and easy-to-use method to separate and enrich nanoplastics in field water samples with ultracentrifugation. River water was spiked with polystyrene fragments (< 1000 nm) at an environmentally relevant concentration (108-109 particles/L). The polystyrene fragments were successfully separated and enriched by a factor of nearly 50 times with a high recovery rate (87.1%) after undergoing our process. Particles were then characterized using UV-vis spectroscopy, scanning electron microscopy (SEM), and enhanced darkfield microscopy with a hyperspectral imaging (HSI) spectrometer. These techniques are non-destructive and allow the assessment of plastic concentration, morphology, and polymer type. Our method can potentially be applied to other water samples to supply clean, enriched nanoplastic samples that can facilitate their identification in environmental samples.

Microplastic quantification affected by structure and pore size of filters.

Filters of various structures (filter by pore depth or pore width) and pore sizes are used to extract microplastics (<5 mm) in researches. In present study, we demonstrate that filters with different structures and pore sizes can lead to different outcomes in microplastic filtering. Our results showed that when filtering large-sized microplastics, nylon filter (double-layer-hole type) retained nearly 100% of fibers, while polycarbonate filter (single-layer-hole type) only retained 61.7%. Polycarbonate filter retained the most fragments (80.8%), while cotton fiber filter (multilayer-hole type) retained the least (54.4%). Pellets were retained on different layers of nylon and cotton fiber filters, and could not be quantified accurately. Additionally, the sizes of some fibers and fragments captured were not within the expected ranges by lattice-knitting filters. Large fiber (3568.0 µm) was not filtered out after 1000 µm pore-size filtration. Small fragment (37.2 µm) was found on 50 µm pore-size filters. To validate laboratory results, filed waters containing microplastics (∼90% in form of fibers) were filtered through different pore-size filters. As expected, the relationship between abundance and pore size followed a same trend as that in laboratory fiber samples. Thereby, our results indicated that filter structure and pore size could affect the abundances of microplastics with different shapes. To obtain more accurate abundance of microplastics in a wide size range, and to consider filtration duration, size limitation of observation, and spatial resolution of identification instrument, we recommend that water samples should be filtered using 20 µm pore-size filters with a double-layer-hole type of structure.

Microplastics in take-out food containers.

Microplastics have been detected in various media including water, sediment, and seafood, whereas there are few studies focusing on microplastics in take-out containers. In this study, we collected take-out containers made of common polymer materials (polypropylene, PP; polystyrene, PS; polyethylene, PE; polyethylene terephthalate, PET) from five cities in China. Microplastics in the containers were analyzed after different treatments (direct flushing and flushing after immersing with hot water). Our results showed that microplastics were found in all take-out containers and abundance ranged from 3 to 29 items/container. The highest abundance occurred in PS containers with rough surface. The polymer types of some detected particles were the same as those of original containers, accounting for 30% of the total microplastics; other types included polyester, rayon, acrylic, and nylon. Treating the containers with hot water did not influence microplastic abundance. Our study indicates that microplastics in take-out containers come from atmospheric fallout and flakes from container's inner surfaces. Under slight mechanical force, loose structure and rough surface of PS containers can flake off microplastics, entering water more easily. Based on the microplastic abundance in take-out containers, people who order take-out food 4-7 times weekly may ingest 12-203 pieces of microplastics through containers.

Microplastic Fallout in Different Indoor Environments.

Microplastics in the air have gradually attracted our attention in recent years; however, temporal and spatial trends of microplastics in indoor air are rarely discussed. In the present study, we tracked microplastic fallout in a dormitory, an office, and a corridor on both workdays and weekends for three months. In addition, an air conditioner was used to understand airflow influence on microplastic resuspension in the dorm. Among the three sampling sites, the highest average microplastic abundance appeared in the dormitory (9.9 × 103 MPs/m2/d), followed by the office (1.8 × 103 MPs/m2/d) and the corridor (1.5 × 103 MPs/m2/d). In the dormitory, the average MP abundance on weekends (1.4 × 104 MPs/m2/d) was approximately three times of that on weekdays (5.8 × 103 MPs/m2/d). In the office; however, the abundance on weekends (1.2 × 103 MPs/m2/d) was 50% of that on weekdays (2.4 × 103 MPs/m2/d). Microplastic fallout existed mostly in the form of fibers and showed similar polymer compositions to the textile products used in indoor environments. The airflow tests using an air conditioner suggested that airflow turbulence increased resuspension of microplastics. Taken together, we conclude that indoor environments are prone to serious microplastic pollution, but microplastic level varies greatly due to different characteristics of indoor setting. Our results also indicate that textile quantity is one of the main factors affecting microplastic abundance in indoor air, whereas air conditioner-induced airflow turbulence can cause microplastic migration in indoor environments.

Superimposed microplastic pollution in a coastal metropolis.

The mitigation of microplastic pollution in the environment calls for a better understanding of the sources and transportation, especially from land sources to the open ocean. We conducted a large-scale investigation of microplastic pollution across the Greater Melbourne Area and the Western Port area, Australia, spanning gradients of land-use from un-developed catchments in conservation areas to more heavily-developed areas. Microplastics were detected in 94% of water samples and 96% of sediment samples, with abundances ranging from 0.06 to 2.5 items/L in water and 0.9 to 298.1 items/kg in sediment. The variation of microplastic abundance in sediments was closely related to that of the overlying waters. Fiber was the most abundant (89.1% and 68.6% of microplastics in water and sediment respectively), and polyester was the dominant polymer in water and sediment. The size of more than 40% of all total microplastics observed was less than 1 mm. Both light and dense polymers of different shapes were more abundant in sediments than those in water, indicating that there is microplastic accumulation in sediments. The abundance of microplastics was higher near coastal cities than at less densely-populated inland areas. A spatial analysis of the data suggests that the abundance of microplastics increases downstream in rivers and accumulates in estuaries and the lentic reaches of these rivers. Correlation and redundancy analysis were used to explore the associations between microplastic pollution and different land-use types. More microplastics and polymer types were found at areas with large amounts of commercial, industrial and transport activities. Microplastic abundances were also correlated with mean particle size. Microplastic hotspots within a coastal metropolis might be caused by a combination of natural accumulation via hydrological dynamics and contribution from increasing anthropogenic influences. Our results strongly suggest that coastal metropolis superimposed on increasing microplastic levels in waterbodies from inland areas to the estuaries and open oceans.

A practical approach based on FT-IR spectroscopy for identification of semi-synthetic and natural celluloses in microplastic investigation.

In previous studies of marine debris or microplastics (<5 mm), various types of semi-synthetic celluloses (e.g. rayon) are ubiquitous in some field investigations. However, it is hard to distinguish semi-synthetic and natural celluloses clearly even using the spectroscopic method. In this study, 8 semi-synthetic and 4 natural celluloses were employed as the test materials to simulate the environmentally relevant samples. Our results showed that these original commercial products exhibited obvious physical (e.g., color) and chemical (e.g., spectra) changes after UV weathering and agent (H2O2 and KOH) digestion treatments. The changes of 4 characteristic bands (1735, 1425/1419, 1105, 1060-1053/1030-1027 cm-1) were evaluated. We found that the band at 1105 cm-1 which is assigned to the CO antisymmetric in plane stretching band only existed in natural fibers even after the weathering and digestion treatments. The mixture of semi-synthetic and natural fibers from the real field samples was also easily distinguished using the characteristic band at 1105 cm-1. Our results suggest that the characteristic band at 1105 cm-1 could be an ideal reference to distinguish natural and semi-synthetic fibers in field microplastic investigations. We also proposed a practical method to enhance the library of polymer spectra and improve the accuracy of semi-synthetic microplastic identification.

Comparison of microplastic pollution in different water bodies from urban creeks to coastal waters.

Although freshwater and estuary systems are recognized as origins and transport pathways of plastics to the oceans, there is a lack of comparison of microplastics in different water bodies or river networks. In the present study, the spatial distribution of microplastics was compared across different water bodies, including city creeks (Shanghai), rivers (Suzhou River and Huangpu River), an estuary (Yangtze Estuary) and coastal waters (East China Sea) in the Yangtze Delta area. Significant spatial differences of microplastic abundances were revealed across the sampling areas. The results showed that the abundance of microplastics was higher (1.8-2.4 items/L) in freshwater bodies than that in estuarine and coastal water (0.9 items/L). In the Suzhou River and the Huangpu River, microplastics showed trends of increasing abundance downstream, where the peak of microplastic pollution is closer to the city center and the estuary. In respect of abundance, microplastics are likely to be transported from pollution sources to sink areas via river networks. The proportion of fibers was the highest in city creeks (88%), followed by the Suzhou River (85%), the Huangpu River (81%), the Yangtze Estuary (66%) and the East China Sea (37%). Similarly, polyesters dominated in city creeks and rivers. The results suggest that both the abundance and properties of microplastic pollution varies across different water bodies. Microplastic pollution in small freshwater bodies is more serious than in estuarine and coastal waters. Therefore, we support prioritization of water monitoring for microplastics within entire river networks, instead of single water body surveys.