A Preliminary Assessment of Size-Fractionated Microplastics in Indoor Aerosol—Kuwait’s Baseline

The omnipresence of microplastic (MP) in various environmental samples, including aerosols, has raised public health concerns; however, there is presently very limited information on MPs in indoor aerosol. This paper presents a unique dataset where smaller MPs have been sampled using a six-stage cascade impactor from indoor environments in Kuwait. The MP concentration in the indoor air varied between 3.2 and 27.1 particles m−3, and the relative MP concentration decreased linearly from the lowest to the highest size fraction. A significant effect of location was observed for the total number of MPs (F2,14 = 5.80, p = 0.02) and the inhalable fraction (F2,14 = 8.38, p = 0.005), while location had no effect on the respirable fraction (F2,14 = 0.54, p = 0.60). A significant effect of the type of air conditioning used was also observed for the total number of MPs (F2,19 = 5.58, p = 0.01) and the inhalable fraction (F2,19 = 6.45, p = 0.008), while location had no effect on the respirable fraction (F2,19 = 1.30, p = 0.30). For the total number of MPs and the inhalable fraction, the concentration was significantly higher for the split unit air-conditioning as compared to the central air-conditioning plants. The presence/absence of carpets had no significant effect on the MP concentrations (total: F1,19 = 4.08, p = 0.06; inhalable: F1,19 = 3.03, p = 0.10; respirable: F1,19 = 4.27, p = 0.05). The shape was dominantly fibers, with few fragments in lower size fractions. These datasets represent the first baseline information for Kuwait, and the smaller MPs in all the samples further underscore the need to develop standardized protocols of MP collection in the ≤2.5 µm fraction that can have more conspicuous health implications.


Introduction
The persistent nature and omnipresence of microplastic (MP) in the aquatic environment has attracted massive attention from the scientific community. Several thousand publications on MP in the aquatic environment have been published since 2004, when the term was first introduced. In contrast, research on MP in aerosols remains less explored, with less than two dozen studies on outdoor air  and only a few on indoor air [1,14,[23][24][25][26][27][28].
The presence of microplastics in the air has been related to release from clothing, furnishings, synthetic tires, and degraded plastics, among other causes [1][2][3]6,17,[29][30][31][32][33][34]. Aerosols can be a significant pathway for transferring MPs to humans via inhalation [4,23,[35][36][37][38][39]. The ecological concern from microplastics emanates from the fact that they can be inhaled by humans and can potentially lead to adverse health effects, such as localized inflammation [40], genotoxicity [4], and the development of oxidative stress and cytotoxicity [36]. For this reason, it is prudent to look at the finer MPs that can be inhaled [41]. Considering the lower size cutoff for microplastics i.e., 1 µm [42], the fine MPs have greater potential to be transferred via aerosolization into the human respiratory system [17]. A few studies on airborne microplastics gained more attention when they suggested human health risks due to MP inhalation, most importantly, the respirable (PM 10 ) and inhalable (PM 2.5 ) fractions that can reach deep in the lungs and may be taken up by both macrophages and epithelial cells [43,44]. Some studies reported that fibrous MPs up to 250 µm in size avert the lung's clearance mechanisms [45]. Several health issues were reported, including reduced lung capacity in work-related conditions, coughing, and breathlessness [46][47][48]. Two recent reviews have highlighted the potential effects on human health [38,39].
A study reported that microplastics deposition is more likely to occur in the upper airway tract (i.e., nose, mouth, throat) and can reach the gut when swallowed [31]. Inhaled or ingested finer MPs are believed to be able to translocate to the circulatory system and other organs [40]. The plastic additives, dyes, and pigments could lead to reproductive toxicity, carcinogenicity, and mutagenicity [49,50]. Over 4000 chemicals are currently used in the plastic food packaging industry itself [51], and can provide a scale of the chemical toxicity they can induce. Most of the additives added to the plastic during processing are of small molecular size and often not chemically bound to the polymeric materials, which make them susceptible to leaching into the surrounding environment [36]. There is voluminous literature available on the sorption ability of MP [3,35,[52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70]. It has been reported that polycyclic aromatic hydrocarbons (PAHs) and metals released from traffic emissions, as well as microorganisms, adhere to MP surfaces and may be transferred directly to the human lung [4].
It has been estimated that about 89% of modern human activities are conducted in the indoor environment [71]; hence, it is prudent to monitor indoor environments. Some studies have reported a higher indoor concentration of suspended and deposited MPs compared to the outdoors [1,13]. Unfortunately, there have only been seven studies conducted so far on MP in indoor aerosol; four of these studies have sampled indoor air using active samplers [1,23,27,28] and the other three collected indoor dust from noncarpeted apartments and the East China Normal University [14,[24][25][26].
The present study is the first attempt to generate data on MP in Kuwait's indoor aerosols, which addresses the larger issue of the paucity of MP data in indoor air and considers the fact that due to the extremely hot and arid climate in Kuwait, people spend most of their time indoors in an air-conditioned atmosphere where the fresh air input and exchange is very limited.

Sample Collection and Preparation
In this study, a six-stage ambient viable cascade impactor from Tisch Environmental, model TE-10-800, was used for sample collection. The air was drawn at 30 L min −1 for 360 min. The samples were collected directly onto the impactor plates without using any substrate and collected samples were microscopically identified and quantified with further verification using hot needle and micro-Raman spectroscopy.
The sample preparation of microplastic in aerosol samples is a critical step, since the aerosol samples include various inorganic (mineral grains) and organic matter (pollen, fungi, microbes, soot, etc.) that might be difficult to segregate. The removal of MP from the other particulate aerosols is a crucial step for accurate identification and spectrographic characterization [72]. However, in this study, we have used a slightly different approach of collecting samples directly on an aluminum alloy, pre-cleaned cascade impactor, without using any intervening substrate. This implies that the likely loss of material or leaching of particles from the collecting substrate was minimized. Once the samples were collected, each of the collected impactor plates was subjected to a microscopic examination, all the particles were meticulously collected using an electrostatically charged ultrafine glass rod and in some cases using forceps, and placed on a pre-cleaned glass slide. This was done in a lamellar flow cabinet. The blank and control samples were also processed along with collected samples.

Identification of Microplastics in Aerosol Samples
The microplastics on these slides were identified using a multi-tier process. An initial visual examination was carried out using a fluorescence stereomicroscope (Leica DM2500 LED) at 40/0.75X to 1000X magnification, sorting out particles based on the absence of cellular structure and thickness consistency along with their length, relatively homogenous coloring, and transparency. These particles were counted and their size was measured using ImageJ software, however, there was no measurement made for particles below 5 µm, which were transferred directly to another glass slide using an electrostatic charge to be observed under scanning electron microscope later.
The second tier of identification was done after these particles were strained with Nile red and MP identification was done under a UV stereomicroscope using a hot needle technique [73,74]. The hot needle test works effectively as the plastic pieces curl and deform on touching, while the other non-plastic materials will not. About 50% of the samples were subjected to the hot needle test. As part of the quality assurance procedure, a blank sample and a positive control were also examined with each collection.
We also attempted polymeric characterization of the MPs on~10% of samples selected randomly. The sample fraction used for polymeric characterization was taken from stages 1-5, and micro-Raman spectroscopy was used [15]. Micro-Raman was preferred as it can be used to identify microplastics up to 1 µm in size [16] and smaller MPs are more relevant for human health assessments. However, we have not reported the polymeric data in this communication as we believe it requires further processing to remove concurrent fluorescence interference that we believe is due to coloring agents. The spectra observed using 785 nm excitation laser showed that most of the larger fibers were polyester and nylon, and fragments were acrylic and polyurethane.

Microplastic in Indoor Air
An initial assessment of the microplastic in indoor aerosol in Kuwait was done by sampling several different types of sites, including public/government buildings, residential dwellings of different types spread over the city, a hospital, and a mosque. Sampling was carried out over an approximately 9 month period from January to the beginning of October, 2021. Air was drawn through a six-stage cascade and the orifice dimensions and size fractions for each stage are given in Table 1. The aerosol was drawn at 30 L min −1 for 360 min, resulting in collection of 10.8 m 3 of aerosol. Table 2 presents the number of MPs in size-fractionated aerosols and their concentrations per m 3 of aerosol. The focus of this study was to establish a baseline on size-fractionated MPs in aerosol, since most of the human health assessments are based on the size fractions rather than chemical composition. Blank 1 (exposed to indoor air-passive) The MP concentrations in the indoor aerosol samples of residential dwellings were much higher; the carpeted flats had MP concentrations of 10.8-27.1 MP m −3 , while in houses with low occupancy (four persons) it was 6.3-13.0 MP m −3 . The mosques in Kuwait are centrally air-conditioned with a thick carpeted floor and due to COVID-19 restrictions, everyone was required to bring their own prayer mat. The MP concentration in the mosque was 14.3 MP m −3 . The MP concentration in hospital air was also observed to range between 3.9-4.4 MP m −3 . We believe the use of PPEs, including disposable coveralls, facemasks, and hospital sheets are also contributing to the MP load in hospitals.
The size of MPs observed and measured under microscope were found to vary from 0.45 µm to 2800 µm. The size fractionation showed that the most dominant size class by enumeration was in the two size fractions >7 and 4.7-7.0 µm. These are quite different from other studies, because none of the other studies have used a cascade impactor with these cut-off sizes. The presence of ultrafine MPs is certainly a matter of huge concern from the human health perspective, more specifically for people in Kuwait who spend most of their time indoors throughout the year, but also for other countries where the indoor-outdoor air exchange is limited, especially during the winter months.
Statistical analyses were performed using SAS. Differences between categories (location, presence of carpet, and type of air conditioning) were tested using an ANOVA model, followed by Scheffe's post hoc tests. All data are presented as mean ± standard error of mean. The relative MP number concentration decreased linearly from the lowest to the highest size fraction (Figure 1). houses with low occupancy (four persons) it was 6.3-13.0 MP m −3 . The mosques in Kuwait are centrally air-conditioned with a thick carpeted floor and due to COVID-19 restrictions, everyone was required to bring their own prayer mat. The MP concentration in the mosque was 14.3 MP m −3 . The MP concentration in hospital air was also observed to range between 3.9-4.4 MP m −3 . We believe the use of PPEs, including disposable coveralls, facemasks, and hospital sheets are also contributing to the MP load in hospitals. The size of MPs observed and measured under microscope were found to vary from 0.45 µm to 2800 µm. The size fractionation showed that the most dominant size class by enumeration was in the two size fractions >7 and 4.7-7.0 µm. These are quite different from other studies, because none of the other studies have used a cascade impactor with these cut-off sizes. The presence of ultrafine MPs is certainly a matter of huge concern from the human health perspective, more specifically for people in Kuwait who spend most of their time indoors throughout the year, but also for other countries where the indoor-outdoor air exchange is limited, especially during the winter months.
Statistical analyses were performed using SAS. Differences between categories (location, presence of carpet, and type of air conditioning) were tested using an ANOVA model, followed by Scheffe's post hoc tests. All data are presented as mean ± standard error of mean. The relative MP number concentration decreased linearly from the lowest to the highest size fraction (Figure 1).  A significant effect of the type of air conditioning was also observed for the total number of MPs (F2,19 = 5.58, p = 0.01) and the inhalable fraction (F2,19 = 6.45, p = 0.008), while location had no effect on the respirable fraction (F2,19 = 1.30, p = 0.30). For the total number of MPs and the inhalable fraction, the concentration was significantly higher for the split unit as compared to the central plant (Figure 3). A significant effect of the type of air conditioning was also observed for the total number of MPs (F 2,19 = 5.58, p = 0.01) and the inhalable fraction (F 2,19 = 6.45, p = 0.008), while location had no effect on the respirable fraction (F 2,19 = 1.30, p = 0.30). For the total number of MPs and the inhalable fraction, the concentration was significantly higher for the split unit as compared to the central plant ( Figure 3). A significant effect of the type of air conditioning was also observed for the total number of MPs (F2,19 = 5.58, p = 0.01) and the inhalable fraction (F2,19 = 6.45, p = 0.008), while location had no effect on the respirable fraction (F2,19 = 1.30, p = 0.30). For the total number of MPs and the inhalable fraction, the concentration was significantly higher for the split unit as compared to the central plant ( Figure 3).     No significant linear relationships were observed between the occupancy and the MP concentration (total: F1,20 = 0.23, p = 0.64; inhalable: F1,20 = 0.08, p = 0.78; respirable: F1,20 = 0.11, p = 0.74; Figure 5). Studies have reported a higher indoor concentration of suspended and deposited MPs compared to outdoors [1,13]. A limited number of studies have been conducted for assessing MPs in indoor environments, however a direct comparison between these studies will not be meaningful as they have used different sampling strategies and sample processing techniques ( Table 3).
The shape, size, color, and polymer type of MPs in the indoor environment have been reported by most of the prior studies. The concentration of MPs in the indoor samples have been highly variable; 0.4-59.4 MP m −3 was reported from Paris, France, whereas 1583 Studies have reported a higher indoor concentration of suspended and deposited MPs compared to outdoors [1,13]. A limited number of studies have been conducted for assessing MPs in indoor environments, however a direct comparison between these studies will not be meaningful as they have used different sampling strategies and sample processing techniques (Table 3).     Diverse shapes, including fiber, foam, fragments, and film, have been detected in the atmospheric microplastics, with fibers being the dominant shape (Table 4). Fiber was the most dominant shape in indoor samples in France, 39 cities in China, Australia, Portugal, and in this study. However, in another study from Wenzhou, China, the indoor aerosol had 80% fragments and 10% fibers, similar to the situation in Aarhus, Denmark, where MPs in indoor aerosols were predominantly fragments (87%), with fibers constituting only 13%. In Hamburg, more than 90% of MPs detected were fragments and less than 10% were fibers, (Klein and Fischer 2019). MPs in the air from Chinese research reported 67−80% fibers, <30% fragments and <3% granules. Details of the shapes are also provided in Table 4.
The microplastics consisted of different colors, with most dominant ones being red, orange, yellow, white, grey, blue, black, green, and transparent. The use of color to identify potential sources of plastic debris is sometimes practiced [42], however, this can be quite misleading. Several studies have reported the colors of the identified MPs and these are summarized in Table 4. Blue and red MPs were reported from Paris [2], while black, blue, red, transparent, brown, green, yellow, and grey particles were reported from Shanghai, China [11].
A study done in Paris reported much higher indoor concentrations, ranging from 1 to 60 fibers m −3 , as opposed to significantly lower outdoor concentrations ranging between 0.3 and 1.5 fibers m −3 [1]. In any case, exposure to microplastics concentrations has been shown to be higher on average in indoor environments than outdoor ones due to the former incurring more sources that allow several factors (e.g., ventilation and airflow) to influence MPs' behavior and elevate their levels [36,75]. By contrast, MP concentrations in the latter environment are subjected to dilution from outside air, and therefore, exposure to lower MP levels is expected [1]. In addition, people spend 70-90% of their time indoors, which enhances exposure levels. Interestingly, microplastics generated indoors can frequently contaminate the environment outdoors, whereas only 30% of particulate matter produced outdoors can penetrate the indoor environment [75]. This underscores the importance of the indoor environment as the main exposure source of airborne microplastics [36]. 12 Countries Indoor dust PET-concentrations ranged between 29-120,000 µg/g. Concentrations are arranged in decreasing order of concentration as: South Korea (25,000 µg/g), Japan (23,000 µg/g), Saudi Arabia (13,000 µg/g) Greece (9700 µg/g), Romania (9100 µg/g), United States (8900 µg/g), Kuwait (8600 µg/g), Vietnam (3900 µg/g), China (3700 µg/g) (one sample had 120,000 µg/g, which was 12% of the total mass of dust), Pakistan (1900 µg/g), India (1600 µg/g), Colombia (1000 µg/g). Free TPA median concentrations in dust samples ranged from 2.0 µg/g (Pakistan) to 34 µg/g (Japan). The highest TPA concentration was found in the sample from India (200 µg/g

Conclusions and Future Perspectives
This study confirms the presence of microplastic in the indoor air-conditioned buildings across Kuwait, a hyper-arid country, where most of the activities are indoor. The study also adds to the limited data on MPs in indoor aerosols; however, the concentrations vary across different types of buildings, depending on type of air conditioning. This study also provides an insight into the MP distribution within the inhalable and respirable fractions of aerosols, considering the 50% cut-off size for aerodynamic size fractions of 2.5 µm, and it provides evidence for a much higher inhalable fraction, roughly a factor of three more than the respirable fraction. The need for <10 and ≤2.5 µm data for aerosols was highlighted as being a potentially important dataset for human health assessment [34].
The data from studies looking at MPs cannot be directly compared as each one used a very different approach. In spite of the un-harmonized methodologies employed within all these studies, some reasonable observations can still be inferred. It could be summarized that fibers and fragments are the predominant shapes of MPs in indoor aerosols, while transparent and black were the most prevalent colors. It is quite evident that due to lack of standardized methodologies, the atmospheric microplastics research certainly lacks sufficient comparable data. With the use of active and passive sampling strategies, the reporting units are very different and often cannot be compared. The passive sampling also provides insufficient information for inhalation risk assessment. Our study provides a first dataset on size-fractionated MPs in indoor aerosols.
Another important discussion we would like to bring up is that regarding the polymeric characterization of MPs-there have been many concerns raised on the methodologies followed and amount of information provided for assessing the data quality [77]. On the other hand, we would also like to question the relevance of the generated information regarding polymers, when the health risk assessments are not using the polymer type but the size, hence suggesting that just assuming that the particle detected is plastic is sufficient? [31].
Based on the experience gained in MP research we would like to highlight that some important points to consider for future work are: (1) Active and passive sampling techniques should be used jointly for better assessment of both short-term and long-term atmospheric MP accumulation rates, respectively, particularly when the aspect of human health risk is investigated.
(2) Concentrations of 5-15% H 2 O 2 or KOH ought to be utilized as opposed to 30% H 2 O 2 for extruding organic matter from the collected samples in order to avoid the significant deterioration of polymers' physical and chemical properties.
(3) A microscopic identification should be preferred instead of density separation using, for example, NaCl, NaI, etc. for particles >5 µm.
(4) The choice of filter is critical, and we recommend that it is better to use a cascade impactor without a filter.
(5) Polymer characterization should be taken up for 5-10% of the samples, depending on the size of MPs, one of the techniques, i.e., attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR), micro-Fourier transform infrared spectroscopy (µ-FTIR), and/or micro-Raman spectroscopy (µ-RAMAN) should be used. However, whether this information will be useful for human health risk assessments is not obvious. (6) The microplastics in aerosol fractions should be reported for the total number of particles per unit volume of air (number of MP m −3 ) and the number of MPs in each of the aerodynamic classes. Such volumetric measurements will be more useful for human health assessments and the estimation of inhalation doses. (7) There is a need to harmonize and standardize the methodology for sample collection, preparation, identification, and reporting of atmospheric microplastics. Moreover, research on the health risk implications to humans is an essential step that can be further accomplished by understanding the interactions between contaminants and microplastics, and their pathways of transfer and eventual exposure to humans. Institutional Review Board Statement: Not applicable for this study. No humans or animals were involved.
Informed Consent Statement: Not applicable for this study.
Data Availability Statement: Information is available in the manuscript.