Introduction

Rising plastic production and use have led to widespread plastic pollution in all environmental compartments, contributing to ecological damage1. Weathering and degradation of plastics through physical, chemical, and biological processes generates microplastics (MPs), small plastic particles of various shapes that are <5 mm in size, and which can have deleterious impacts on numerous organisms2,3. Whereas MPs have been extensively studied in marine4, freshwater5, and soil ecosystems6, less attention has been paid to MPs as an atmospheric pollutant, particularly in indoor settings, and research on the determination of the fate and transport of MPs in the air as well as their health effects are needed7,8. Current research has shown that, due to their small size and buoyancy, many MPs are readily suspended and transported by wind and air currents becoming an inhalation hazard9,10. Thus, research on the determination of the fate and transport of MPs in the air as well as their health effects are needed8.

The primary route of exposure associated with airborne MPs is inhalation, which may lead to inflammation and other adverse effects11. While the health effects related to environmental MP exposure are not yet fully understood, research on workers in the plastic production industry indicates the potential for significant health risks, with workers displaying increased rates of lung disease and cancer12. Although the general population is likely exposed to much lower concentrations of airborne MPs, the rising use of plastic in indoor environments increases the potential for exposure. Sources of these airborne MPs include textile clothes, anthropogenic activities, and fragmentation of larger plastics7,13.

It is reasonable to expect that exposure to MPs is likely to vary among different indoor settings; i.e., among different rooms within a building. In bedrooms there are typically numerous sources of polymer textile products, such as clothing, bedding, and furniture that can release MPs14. Compared to other rooms within a house, kitchens tend to have more plastic products, including containers, utensils, and packaging materials that could contribute to MPs in the room. The abundance of MPs indoors is also influenced by the number of people and their activity level in the space15. Over time clothing, carpets, and other textiles undergo repeated stretching and abrasion, leading to fiber breakage and the subsequent release of MPs16. Thus, the release of MPs is expected to be greater in areas of high activity (e.g., group offices) than in areas of lower human activity (e.g., personal offices).

Airborne MPs are sampled using passive or active methods. Passive sampling typically collects fallout particles deposited onto surfaces or into containers, and studies employing this approach have been reported from China17, USA18 and the United Kingdom19 in indoor air. In contrast, airborne MPs captured actively using a pump have been reported from France20, Spain21 and China22. The active sampling technique can provide data over a shorter time-frame and has the potential to include more of the MPs in the air mass that stay aloft longer, but requires more resources20. Various methods are employed for the pretreatment of indoor MPs, such as sieving, digestion, density separation, filtration, and drying23. The objective of the pretreatment is to segregate MPs from other particulate matter and facilitate the identification of MPs23,24. However, pretreatment processes may introduce contamination, result in MP loss, or alter the surface characteristics of MPs, which can affect their analysis and interpretation23. Further, there is currently a lack of established or standardized methods for identification and quantification of airborne MPs, which is needed to better understand their occurrence, characteristics, and environmental and health impacts10.

In this study, we investigated the abundance, compositions, sources, sizes, morphology, and distributions of the MPs deposited on silicon in residential homes and an academic chemistry building. Samples were collected in bedrooms, kitchens, living rooms as well as in individual and group offices over the course of a month, and were analyzed directly by µ-FTIR and optical microscopy. In addition, a subset of samples was analyzed by SEM-EDX to study the surface and elemental composition of select MPs. The objective of the study was to (1) evaluate the feasibility of using silicon collection and direct analysis of particles by µ-FTIR in transmission mode, and (2) compare the abundances, types, sizes, shapes, sources of the MPs by room types. We tested the following hypotheses regarding differences among room types. We hypothesized that MPs deposition rates would be higher in bedrooms compared to other rooms, higher in kitchens than in living rooms, and higher in group offices than in personal offices. Overall, this work seeks to further analytical methods and provide empirical data on the characteristics of airborne MPs in different indoor environments, both of which are needed to assess the potential human health risks associated with inhalation of MPs.

Results and discussion

Comparative analysis of sampling methods

Typically, indoor air fallout is collected into a glass containers such Petri dishes17,19. The containers are covered and brought back to the laboratory where the particles may be subjected to digestion and density separation to concentrate and isolate MPs before analyses25. However, certain digestion protocols can induce degradation, fragmentation, or dissolution of MP particles, thereby exerting a considerable influence on the subsequent analytical outcomes26. Density separation techniques are also commonly used to segregate MPs from other constituents by utilizing solutions of defined densities. However, the efficiency of isolating smaller MPs can be reduced if their density is similar to that of the surrounding matrix or if the other constituents are insufficiently dense to sink in the selected solution27. To avoid these issues, in this study, we collected MPs fallout onto silicon and analyzed them without sample preparation. This sampling approach allows for direct analysis using µ-FTIR, negating potential losses and contamination due to sample preparation and speeding analyses.

To assess the potential for particle loss during sample preparation steps, a comparative analysis was carried out between two methods: fallout collected on silicon filters followed by direct µ-FTIR analysis and fallout collected using a more traditional approach into glass vials, followed by transfer onto filters for analysis. The number of MPs observed by the direct analysis approach was significantly higher than that for the traditional approach, yielding more MPs across all size ranges (Fig. 1 and Supplementary Table 1). Furthermore, sampling with glass vials has several potential drawbacks. MPs may adhere strongly to the walls of vials, and some may not be dislodged during rinsing, resulting in incomplete transfer during sample preparation. Additionally, small lightweight particles can be lost during the transfer and filtration steps.

Fig. 1: Comparison of indoor microplastic deposition rates (by size) collected using two sampling approaches.
figure 1

Blue bars are for fallout on three silicon square filters (inside open Petri dish), followed by direct µ-FTIR analysis, and red bars are for fallout collected into three adjacent glass vials (see photo inset), followed by transfer to silicon filters and subsequent µ-FTIR analysis.

Full size image

Our results suggest that direct analysis may provide a more accurate measure of MPs in atmospheric fallout, provided the particles are not exposed to any wind or air currents that would physically blow the particles off the filter surfaces. Thus, the direct sampling and analysis approach would not be suitable for outdoor studies or for indoor studies in areas near windows or vents. Hereafter, we only report results for MP particles deposited on the silicon filters.

Microplastic deposition rates by location and room type

In total, we identified 6134 MPs (>25 µm) from 45 samples collected across the 15 sampling sites. MPs were identified in all samples, with abundances on each filter ranging from 100 to 1186 MPs. Overall deposition rates (MPs/m2/day ± 1 SD) varied by location, ranging from 18.0 ± 3.6 (×103) in a personal (work) office to 159.2 ± 58.7 (×103) in a house bedroom (Table 1). Average deposition rates varied by room type, with bedrooms generally having the highest average rates (104.4 ± 49.7 (×103)), followed by the kitchen (59.5 ± 52.5 (×103)), group offices (42.5 ± 9.6 (×103)), living room (39.4 ± 29.5 (×103)), and personal offices (30.9 ± 11.2 (×103) (Fig. 2, Supplementary Table 2).

Table 1 Microplastic deposition rates (103 MP/m2/d) across sampling sites
Full size table
Fig. 2: Mean microplastic deposition rates (MPs/m2/day ± 1SE) in residential homes and an academic chemistry building determined by direct µ-FTIR analysis of particles on silicon filters.
figure 2

Deposition rates were influenced by location and activity (see text).

Full size image

Overall MP deposition rates were generally similar to rates reported for twenty family homes in Jiangsu, China, where averages ranged from 93.8 × 103 to 311 × 103 MPs/m2/day17. However, both these rates are higher than those reported elsewhere (Table 2). For example, MP deposition rates were an order of magnitude lower in a dormitory and office in Shanghai, China, ranging from 2100 to 29,000 and 600 to 4500, respectively15. In the UK, rates averaged 3735 ± 1343 in homes and 3177 ± 1860 in workplaces19, and in northern New Jersey, the abundance of MPs deposited on surfaces in an office was 643018. Additionally, MP deposition rates in government areas in Sydney, Australia ranged from 0.2 to 6.2 MPs/m2/day28, and households in the Humber region, UK showed deposition rates averaging 6.2 ± 3.1 MPs/m2/day with a range of 0.9–14.6 MPs/m²/day29. Only our study employed direct analysis of deposited particles. All other studies had some level of sample preparation, such as chemical pretreatment, density separation (flotation), and filtering, all steps that could result in particle losses. The cutoff size for particle analysis also plays a role when comparing data. In the current study, we measured particles down to ~25 µm in size, whereas Zhang et al.15 and Soltani et al.30 only measured down to 50 µm. Yao et al.18 utilized a confocal Raman microscope with a lower size limit of 1 µm, Ageel et al.28 applied a µ-FTIR with a 10 µm cutoff. Additionally, Jenner et al.29 applied a stereomicroscope with a lower size limit of 5 µm.

Table 2 Comparison of studies reporting microplastic deposition rates for indoor environments using passive collection
Full size table

Our hypothesis that MPs deposition rates would be higher in bedrooms than in other rooms was supported (Contrast F1,10 bedroom vs. other = 7.69, p = 0.02). This finding underscores the potential of inhalation of MPs, particularly in bedrooms14. Our hypothesis that kitchens, with their numerous plastic products, including containers, utensils, and packaging materials, would have higher MPs deposition rates than in living rooms was not supported (Contrast F1,10 kitchen v. living room = 0.52, p = 0.49). It is worth noting that the kitchen in house 3 showed relatively high MP deposition rates (Table 1), which may be attributed to its relatively small size and its numerous plastic items. Our hypothesis that group offices, with their high human visitation and activity, would have higher MPs deposition rates than in personal offices was not supported (Contrast F1,10 group office v. personal office = 0.17, p = 0.69). One possible reason for this is that personal offices are smaller and contain a higher density of furniture, which could contribute to MPs level being comparable to those in group offices. Additionally, variables such as building materials, cleaning practices, and ventilation systems may impact the levels of MPs contamination23.

Sizes and shapes of microplastic fallout

The overall abundance of MPs deposited on the silicon filters increased with decreasing size of the particles (Fig. 3), with particles >100 µm accounting for 6.3% of the total, 75–100 μm particles accounting for 7.7%, 50–75 μm particles accounting for 29.4%, and 25–50 μm particles, the smallest size fraction measured, accounting for 55.64%. The majority (>90%) of the MPs were between 25 µm and 100 μm across all sites. Smaller MPs pose a higher risk of inhalation because their tiny size allows them to become airborne, generally stay aloft longer than larger particles, and when inhaled penetrate deeper into the lungs31. Once inhaled, the particles can cause oxidative stress, inflammation, and respiratory problems, with the smaller particles having the potential to translocate via the bloodstream32. MPs have been identified within human lung tissue samples with one study having a mean particle length of 223 μm (range 12–2475 μm)33. In the current study, MPs concentrations increased with decreasing size, and as the 25–50 μm size fraction was the most abundant, it is clear that people are exposed to inhalable MPs during daily life.

Fig. 3: Size distribution of microplastics deposited on silicon filters in the indoor environments studied.
figure 3

The plot includes all data combined and shows that small spherical particles are predominant.

Full size image

Particles were also classified by morphology, though it can be challenging to do so from image data, particularly when distinguishing between spherical and irregular shaped particles, both of which can be considered as fragments. Here the form factor and aspect ratio shape descriptors were used distinguish fibers (aspect ratio > 3), spherical particles (form factor > 0.6), and irregular particles (form factor ≤ 0.6)34. Fragments were the dominant morphology among MPs at all sites, with spherical particles accounting for 88.1% of total MPs (Fig. 3, Supplementary Table 3). Fibers accounted for only about 2.8%. Fragments dominated in the 25–50 µm (96.8%), 50–75 µm (91.2%) and 75–100 µm (83.4%) fractions of MPs in indoor air. In contrast, fibers constituted the majority of larger (>100 μm) size classes (Fig. 4). This distribution of MP shapes is consistent with previous studies that focused on atmospheric deposition samples collected in Germany35, northern New Jersey, USA18, Jiangsu, China17 and Wenzhou, China36. However, there are other studies that report fibers being the dominant type of airborne MPs in Birmingham, UK19, Southeast Queensland, Australia37, Russia38, India39, and Sri Lanka40. The disparities in the results may be attributed to variations in the locations and methods employed for sample collection, as well as the diverse approaches used for separation and analysis. These differences complicate the direct comparison of study outcomes, given that the quantity, size, shape, and type of MPs detected are variable and largely contingent upon the detection techniques utilized41.

Fig. 4: Proportion of shapes in different size categories for particles deposited on silicon filters in indoor environments (all data combined).
figure 4

Fibers are predominate in the larger size categories, spherical particles predominant in the lower size categories, and irregular-shaped particles predominant in between the two.

Full size image

Chemical composition of microplastic fallout

The proportion of each type of MP deposited on the silicon filter is shown in Fig. 5, with associated particle counts given in Supplementary Table 4. For both fragments and fibers, polyamide (PA) and polyarylamide (PARA), a partially aromatic polyamide were major polymers. Notably, fibers show a wider variety, with Polyvinyl Chloride (PVC), Polyethylene Terephthalate (PET) and Polyolefin (PO). Thus, the predominant polymers identified were those belonging to the nylon family, particularly PARA and PA, which constituted 61.6% and 22.5% of the MPs, respectively. These results are consistent with previous studies that detected MPs in urban homes in Jiangshu, China17 and Barcelona, Spain21. Nylon has numerous consumer applications, and MPs originating from these materials may derive from textiles (e.g., clothing, carpets) and home furnishings (e.g., curtains, sofas)42.

Fig. 5: Chemical composition of microplastics by morphology.
figure 5

Polyarylamide (PARA) and polyamide (PA) were the most common polymer identified regardless of particle shape. Fibers had a wider distribution of polymer compositions by percent.

Full size image

The composition of MPs in indoor environments varies around the world. For example, in Shanghai, polyester and rayon were the dominant types of MPs15. PP was found to be a major type of MP in apartments and offices in Paris20. Polyethylene (PE) and Polypropylene (PP) are most common in South Korea41. Studies in the United Kingdom frequently detect polyethylene terephthalate (PET) in indoor dust, suggesting contributions from carpets and upholstery29. Meanwhile, PET and acrylic fibers are prevalent in indoor environments in the United States, likely due to clothing, furniture, and textiles18. Some of this variation among studies may also be attributed to the manual identification of randomly selected particles. In the current study, an automated particle find feature coupled with µ-FTIR with single particle analysis provides a more accurate accounting of particles.

The composition of MP fallout varied by room (Fig. 6, Supplementary Table 5). Whereas nylon (PARA and PA) was the main polymer across room types, the levels varied with bedrooms having the most (94.0%) and kitchens the least (68.0%). These differences may be attributed to bedrooms typically having a higher density of textiles, such as carpets, rugs, curtains, and bedding, which are often composed of nylon17. In contrast, kitchen environments including more of polymers such as resin (7.9%), rubber (5.1%), and PVC (5.8%), likely originating from cookware, utensils, cabinets, and equipment43.

Fig. 6: Polymer composition of MPs in different room types.
figure 6

While the percent distribution of polymers varied by location, the relative distribution was consistent for the top several individual polymers, with polyarylamide (PARA) detected the most followed by polyamide (PA) and resin.

Full size image

Matching microplastics in fallout with potential sources

FTIR spectra were collected from potential sources, including synthetic carpets, blankets, laundry lint, and a dog bed, at several of the sampling sites (Fig.7). The representative MPs FTIR spectra are shown in Supplementary Fig. 1, including PA, PARA, and PET. The source material exhibiting similar spectra (polymer composition) to MP fallout at the sites, with PARA and PA common for carpets. Additionally, PET was identified as one of the major fiber sources, displaying compositions consistent with those of blankets found in the surveyed residential areas. The primary constituent of dryer lint was identified as cotton. Overall, we deduce that textiles are the main source of MPs in indoor environments.

Fig. 7: Main types of textile products in indoor environment and their corresponding spectra.
figure 7

Legend: A Carpet-1 (house living room); B Carpet-2 (house bedroom); C Carpet-3 (group office); D Blanket-1 (house living room); E Blanket-2 (house living room); F Blanket-2 (house living room); G Dog bed (house living room); H. Lint from a laundry home dryer.

Full size image

Given its abundance in the indoor air fallout, nylon is of particular concern. Nylon MPs can pose several potential health risks, including respiratory issues from inhalation44, accumulation in the digestive tract29, chemical exposure from additives45, hormonal disruption46, and inflammation and immune reposes47. As noted, smaller airborne MPs pose a higher potential risk as their size allows them to be easily inhaled, penetrating deep into the lungs31. Additionally, their higher surface area relative to volume leads to greater adsorption of toxic chemicals and pollutants12, as well as the potential release of toxic additives48. Thus, minimizing inhalation and ingestion of MPs is important for health.

Microplastic surface characteristics and elemental analysis

The surface morphology and the elemental composition of the select indoor MPs are shown in Fig. 8. The MPs exhibited surface fractures, cracks, and tears. This surface damage is likely due to mechanical wear and tear from use, as well as aging from exposure to environmental factors such as temperature fluctuations, humidity, UV light, interactions with household chemicals over time, which make them more susceptible to damage49.

Fig. 8: SEM micrographs of particles collected from indoor samples.
figure 8

Legend: A a fiber, B two fragments, and C a film or fiber. Four spots on each micrograph were analyzed using EDX, and the O/C ratios were calculated and presented in the graphs.

Full size image

SEM-EDX analysis showed that the MPs were mostly composed of carbon and oxygen. For example, one fiber observed was composed of an average of 65.1% carbon and 34.9% oxygen (Fig. 8A). The frayed end of the fiber (spot 4) had a higher O/C ratios compared to other parts (spot 1 and 3). Spot 2 had the lowest O/C ratio, but it appears to be an attached particle and not the fiber itself. Others have shown that there can be uneven degradation on the surface from same MP18. Wearing and peeling were observed on the surfaces of two fragments, which were composed of an average 49.4% carbon and 50.6% oxygen (Fig. 8B). On the surface of one fragment, the O/C ratios ranged from 0.7 to 1.6. A film-like particle was also observed, characterized by the presence of pits, grooves, and irregular edges, with carbon at 66.4% and oxygen at 33.6% (Fig. 8C). Compared with the other three spots, the edge of the film (spot 12) shows obvious curling and the highest C/O ratio.

While the O/C ratio is a potential indicator for assessing the oxidation level of MPs, it is important to collect multiple points on the same sample for a comprehensive evaluation50, as evidenced by the varying O/C ratios of our measured particles. While SEM-EDX could help identify MPs and provide information on surface morphology, elemental composition, and aging of MPs, the significant variation in the O/C ratio data on individual MPs, the cost of the analysis, and the limited number of MPs that can be examined, limits this technique’s usefulness.

Methods

Sampling and treatment

Three silicon square filters (1 cm × 1 cm; pore size: 1 μm) were deployed at three residential homes in living rooms, bedrooms, and kitchens, as well as three personal offices and three group offices in the Department of Chemistry and Biochemistry at the University of Mississippi in Oxford, Mississippi (n = 45 total samples). The sampling filters and containers were pre-cleaned as described in Section 2.5. The silicon filters were set out in open Petri dishes at a height between ~1.5–2 m and left undisturbed for 28 days between 21 May 2024 and 21 June 2024. Particles were passively collected on the filters over the course of the month. After the collection period the dishes were covered with their glass covers and carefully transported to the laboratory for µ-FTIR analyses as described below.

Since the silicon filter surface lacks adhesion, and since the traditional method of collecting directly into glass containers requires subsequent sample preparation steps that can result in MPs being lost or that can introduce contamination, a comparison study was performed. Fallout was collected on silicon filters (as before) and collected directly into small (20 mL) glass vials with a 2-cm diameter opening. Sampling was conducted in one of the group offices. After sampling, the silicon filters in the Petri dish were covered (as before), and the vials were capped. In the clean room the vials were rinsed (three times) with ultrapure (0.22 µm filtered) water. The rinse solution was filtered through the same type of (pre-cleaned) silicon filters used for direct collection of fallout. The filters were air-dried in a laminar flow hood in the clean room before analysis as described below.

For blanks, silicon filters (in Petri dishes) and glass vials were set out and uncovered briefly (a few seconds) at the sampling site before being covered for the duration of the sampling period. The 1 µm pore size of the filters is well below the size range of MPs particles reported herein (>25 µm), and it is plausible that the filter pores may help retain or secure MPs if some portion of the MPs penetrate into them.

µ-FTIR analysis, particle size, and morphology

Silicon filters were gently removed from Petri dishes using tweezers and placed on the sampling stage of a LUMOS II FTIR microscope system (Bruker Corp, Billerica, MA, USA). Particles on the filters were analyzed directly using the instruments automated find particle feature and the liquid nitrogen-cooled mercury cadmium tellurium (MCT) point detector. Spectra were collected in transmission mode between 650 and 4000 cm–1 using 16 scans and compared with the system’s database libraries (OPUS v8.7.4 software). These libraries include the Bruker-KIMW ATR-IR Polymer Library, the Bruker-KIMW ATR-IR Additives Library, the BPAD-Bruker Optics ATR-Polymer Library, the Bruker General IR Library, the FTIR Library of Plastic Particles (FLOPP)51, and the FTIR Library of Plastic Particles Sourced from the Environment (FLOPP-E)51. A particle was identified as plastic if the hit quality index (HQI) exceeded a value of 200 and the identification matched that of a plastic.

Particle size and morphology were determined from optical images collected by the LUMOS II and analyzed by Image J as described elsewhere52. Briefly, the RBG image was binarized into a black and white image, and the Particles8 plugin of the software’s morphology macro was used to calculate various size and shape descriptors for each individual particle. To determine the morphology, both the form factor and aspect ratio shape descriptors were used. Particles with an aspect ratio of 3 or greater were classified as fibers, particles with a form factor of greater than 0.6 were classified as spherical particles, and particles with a form factor of 0.6 or less were classified as irregular particles34.

Surface characterization

After µ-FTIR analysis select samples were analyzed by field-emission SEM-EDX (JSM-7200FLV, JEOL Ltd., Japan) for more detailed analysis of the select particles’ morphology and surface texture, along with their elemental composition. Particles on the silicon filter were transferred onto SEM stubs covered with carbon tape by touching the stub to the filter. Before analysis, the samples were sputter coated with Gold-Palladium (60:40). An electron energy of 20 kV was used as it was determined to be optimum.

Statistics

Deposition rates of MPs were analyzed to determine the effect of different room types. A linear mixed-effects model was used, with room type as the fixed effect and replicate (i.e., house or office) as the random effect. Specific hypotheses for differences among room types were tested using planned orthogonal contrasts. These contrasts included: bedroom versus all other room types; kitchen versus living room, and group offices versus personal offices. These contrasts are based on the hypotheses that MP deposition rate would be greater in bedrooms than in other rooms, greater in kitchens than in living rooms, and greater in group offices than in personal offices. A p-value of less than 0.05 was considered statistically significant in all analyses. Analyses of deviance of MPs deposition rate used Type II sums of squares to obtain Wald F statistics using the Kenward-Roger method of approximating degrees of freedom. The linear mixed models were fitted using the lmer function and the lmerTest package associated with the R Studio software53.

Contamination mitigation and blanks

To avoid microplastic contamination, cotton laboratory clothes were worn during sample analysis. All Petri dishes were rinsed three times with milliQ filtered (0.22 µm) water prior to use and dried in laminar flow hood (AirClean 6000 Workstation) located within a HEPA filtered clean room environment. Silicon filters were first sonicated in acetone for three minutes, then in methanol for five minutes to remove any particles. Following this, the filters were optically inspected under the microscope to ensure cleanliness before sampling. Three silicon filters in a petri dish covered with a lid were placed in the laboratory as a blank but no MPs were found on them after the sampling time.