Household Dust: Loadings and PM10-Bound Plasticizers and Polycyclic Aromatic Hydrocarbons

Residential dust is recognized as a major source of environmental contaminants, including polycyclic aromatic hydrocarbons (PAHs) and plasticizers, such as phthalic acid esters (PAEs). A sampling campaign was carried out to characterize the dust fraction of particulate matter with an aerodynamic diameter smaller than 10 µm (PM10), using an in situ resuspension chamber in three rooms (kitchen, living room, and bedroom) of four Spanish houses. Two samples per room were collected with, at least, a one-week interval. The PM10 samples were analyzed for their carbonaceous content by a thermo-optical technique and, after solvent extraction, for 20 PAHs, 8 PAEs and one non-phthalate plasticizer (DEHA) by gas chromatography-mass spectrometry. In general, higher dust loads were observed for parquet flooring as compared with tile. The highest dust loads were obtained for rugs. Total carbon accounted for 9.3 to 51 wt% of the PM10 mass. Plasticizer mass fractions varied from 5 µg g−1 to 17 mg g−1 PM10, whereas lower contributions were registered for PAHs (0.98 to 116 µg g−1). The plasticizer and PAH daily intakes for children and adults via dust ingestion were estimated to be three to four orders of magnitude higher than those via inhalation and dermal contact. The thoracic fraction of household dust was estimated to contribute to an excess of 7.2 to 14 per million people new cancer cases, which exceeds the acceptable risk of one per million.

Studies on the concentration of contaminants in household dust have focused on the analysis of the total mass or of sieved fractions, where the smallest particle size obtained is generally 38 µm [48].
Considering the possibility that household dust is resuspendable and can become airborne, most methodologies have drawbacks when assessing human inhalation exposure. Such an approach requires measurements of the contaminant concentration in smaller particle sizes. The main goal of the present study was to determine household dust loadings and the respective PAH and plasticizer levels in the thoracic fraction (< 10 µm) of resuspendable material on the floor, which deposit anywhere within the lung airways. The in-situ resuspension chamber was previously devised and successfully applied to collect the deposited PM10 fraction from road pavements [49], but it was the first time this active sampling methodology was used to collect settled thoracic particles directly from the floor.

Methodologies
To determine and characterize dust loadings, a sampling campaign was carried out in four different houses located in the Spanish city of León (Table 1). In each housing unit, three rooms were investigated, including the kitchen, the living room and a bedroom. In each room, two samples were collected with, at least, a one-week interval. Samples in each house were performed one or two days before weekly cleaning. For dust collection, an in-situ resuspension chamber operating at an air flow rate of 25 L min -1 was used [49]. After vacuuming, PM10 was separated from the total dust through a Negretti stainless steel elutriation filter and collected onto 47 mm quartz fiber filters (Pallflex®), while particles with aerodynamic diameter > 10 μm were deposited in the methacrylate chamber and along the elutriation filter. Sampling was performed in surface areas of 1 m 2 for 30 minutes. Two to three different square meters were sampled using the same filter in order to ensure enough particulate mass for the subsequent gravimetric and chemical analyzes. Since some compounds may derive from the resuspension chamber itself or from ambient air that enters the system and passes through the filter during dust collection from the floor, air samples were vacuumed through the same system after sampling in each room. Background air filters were sampled for 30 minutes, as were PM10 floor dust samples. Due to the loss of a small filter fragment, the kitchen sample from house 2 obtained in the 2 nd week of sampling was discarded. After gravimetric determination, two punches (9 mm) of each filter were analyzed by a thermo-optical transmission technique to obtain the PM10 carbonaceous content (organic and elemental carbon, OC and EC). This method is based on the quantification of the CO2 released from the volatilization and oxidation of different carbon fractions under controlled heating by a non-dispersive infrared (NDIR) analyzer. The blackening of the filter is monitored using a laser beam and a photodetector, which enables separating the EC formed by pyrolysis [50]. The remaining portion of each filter was extracted by sonication for 15 minutes with three aliquots (25 mL each) of dichloromethane.
After filtration, the solvent was concentrated in a TurboVap system from Biotage and evaporated to dryness by a gentle nitrogen stream. All the extracts were analyzed by gas chromatography-mass spec-

Dust loadings
Huge differences in dust loadings between rugs and hard floorings were registered ( Figure 1) [58]. Dust loadings within the institution were in the range from 4.5 to 6.5 g m -2 . Masses of indoor settled dust were expectedly higher along high traffic, untarred roads and construction sites, as well as in older buildings. It should be borne in mind that dissimilar indoor dust sampling strategies (e.g., wipe versus vacuum methods) are used to measure loadings and amounts of toxicants per unit area, which renders comparisons between studies difficult. Lioy et al. [59] found that while loadings were substantially greater with wipe sampling, metal concentrations within the dust samples were similar for both methods of sampling. Vacuum cleaner sampling has its own series of problems, especially the variability in design and efficiency, and likely will not retain particles below showed the best reproducibility and correlation with other sampling techniques. The authors concluded that surface wipe sampling was the best method to measure accessible lead from carpets for exposure assessment, while vacuum sampling was most effective for providing information on total lead accumulation (long-term concentrations). In their review paper, Lioy et al. [6] state that although we have come a long way in determining the uses of house dust to identify sources of indoor contamination and to provide improved estimates of residential human exposure, one of the challenges still lies in the reliability of sampling techniques.

3.2.Carbonaceous, plasticizer and PAH particulate mass fractions
Total carbon accounted for 9.3 to 51%wt of PM10 with the highest mass fractions recorded in dust samples collected in the city center apartment ( Figure 2). An overwhelming proportion, always higher than 80%, of the total carbonaceous matter was composed of OC, whereas in many samples the EC was  Most likely due to its high volatility, DMP was the compound with the lowest mass fractions ( Table   2). On the other hand, DEHP, DNOP and DBP were the major phthalates in household dust. While high DBP values were observed in all parquet floor bedrooms, only samples from two living rooms showed detectable masses. Bamai et al. [61] also associated higher DBP levels in floor dust with compressed wooden floor. This type of flooring is usually composed of thin pieces, which are glued together and covered with wax, paint, and sometimes flame retardants. The surface applied products (gloss agents, plastic additives, paint, and varnish) contain DBP [61]. From quantitative and qualitative emission data on phthalates from different materials, Afshari et al. [62] reported that polyolefin covered with wax for floor polishing increased DBP concentration in chamber air by two-fold. DBP is also employed as a coalescing aid in latex adhesives, as well as a plasticizer in cellulose plastics and a solvent for dyes [61].
Furthermore, DBP has been reported to be largely present in cosmetic and personal care products [63]. Despite the concentration of DNOP in floor dust has been reported in only a very limited number of publications, the mass fractions of the present study are higher than those described in the literature.
There were no appreciable differences between the amounts found in PM10 of the various rooms of the houses. DNOP is used in carpet back coating, floor tile, and adhesives. It is also employed in cosmetics and pesticides. DEHP was present at higher concentrations in the bedroom and living room samples.
Although DEHP has been consistently described as one of the most abundant phthalates in settled dust, the levels have decreased over time, reflecting its phase-out in the EU. In Europe, the use of DEHP decreased drastically in 2001 and has to a large extent been replaced by DINP and DIDP, with their longer chains and lower volatility [64]. DEHP has been used in numerous consumer products, children toys, medical devices and building materials (e.g., vinyl flooring, furniture, paints, cables, wires, wall coverings, packaging materials) [65]. Bamai  DEHP and other phthalates are strongly sorbed to surfaces. A relatively small gas-phase concentration, such as 0.1 ppb, is enough for significant vapor transport of a PAE and its subsequent partitioning between the gas phase and indoor surfaces, including airborne particles and settled dust [68].   China. ∑15PAHs ranged from 1.2 to 280.4 µg g -1 , averaging 11.1 µg g -1 . Yadav et al. [74] investigated 7 the contamination level of EPA's priority PAHs in indoor dust from residential, educational, 8 commercial, public places and office premises in four major cities of Nepal. Concentrations of ∑16PAHs 9 ranged from 747 to 4910 ng g -1 (median 1320 ng g -1 ). The median concentration of ∑16PAHs quantified 10 by Mahfouz et al. [75] in dust retained in air-conditioning unit filters from 13 households in Greater 11 Doha, Qatar, was 218.0 ng g -1 , but a wide range of variation was registered (1.6-477.3 ng g -1 ). In 12 Palermo, Italy, Mannino and Orecchio [76]  with a geometric mean of 12.9 µg g -1 . These values were found to be comparable to those documented 19 in a previous review in which the total PAHs in samples collected from urban, rural, and suburban 20 homes ranged between 0.4-544 µg g -1 with a geometric mean of 4.5 µg g -1 [7]. High concentrations of 21 health risks, were observed for the smallest particles (< 43 μm). It must be borne in mind, once again, 29 that comparability between results of various works should be made with caution, as they concern 30 different surfaces, particle sizes, sampling and analytical methodologies, and list of compounds. A 31 summary of average PAH concentrations in settled house dust by country and year (data from 35 32 studies) can be found in a recent review article [80]. A major issue encountered when comparing these 33 studies was the variability in both the sampling methods employed and the dust particle size fractions 34 subjected to analysis. In 21 out of the 35 studies reviewed, the particle size cut-off points were either 35 150 μm or 63 μm. It has been suggested that particles > 150 μm do not easily and efficiently adhere to 36 hands or skin. Therefore, these sizes are less relevant when evaluating exposure via ingestion or dermal 37 pathways [81]. 38  (Table 3). While the medians 47 of the latter two compounds were higher in the PM10 sampled in the living rooms, pyrene showed higher 48 levels in the kitchens. PAH levels and speciation are highly dependent on the cooking methods [82], 49 biomass burning appliances and operating conditions [83], traffic fleet and meteorology in the outdoor 50 surrounding environment [84], among other factors.  Cdust represent the mean concentrations in dust (ng g -1 ); IngR is the ingestion rate of indoor dust (200 mg day -1 for children, 100 mg day -1 for adults); EF is the exposure frequency (180 days per year for both children and adults); ED is the exposure duration (6 year for children, 24 years for adults); CF is an unit conversion factor (10 -3 g mg -1 ); BW is the body weight (15 kg for children, 70 kg for adults); AT is the average time (2190 days for children, 8760 days for adults); InhR is the inhalation rate (7.6 m 3 day -1 for children, 12.8 m 3 day -1 for adults); PEF is the particulate emission factor 1.36×10 6 m 3 g -1 ); SA is the dermal exposure area (1150 cm 2 for children, 2145 cm 2 for adults); AFdust is the dust adherence factor (0.2 mg cm -2 day -1 for children, 0.07 mg cm -2 day -1 for adults); and ABS is the dermal adsorption fraction (0.001 for both children and adults, dimensionless).
Ingestion is the main pathway for intake of plasticizers from dust ( Table 4). Regardless of the route, household residents are exposed to higher intakes in the bedrooms, whereas the lowest doses are experienced in the kitchens. The daily intakes for children and adults (13-29 and 1.4-3.1 µg kg -1 day -1 , respectively) via dust ingestion were 3-4 orders of magnitude higher than those via inhalation (0.357-0.802 and 0.129-0.289 ng kg -1 day -1 , respectively) and dermal contact (14.9-33.0 and 2.05-4.62 ng kg -1 day -1 , respectively). Children are at higher risk of exposure to plasticizers than adults. The total DIing, DIinh and DIder for children were about 9.3, 2.8 and 7.1 times higher than those estimated for adults, respectively. DBP, DNOP and DEHP contributed the most to daily intakes. The dust ingestion intakes for these compounds were lower than the U.S. EPA maximum acceptable oral doses of 0.8, 0.01 and 0.02 mg kg -1 day -1 , respectively. However, the daily intakes of plasticizers through dust ingestion, inhalation and dermal contact of the present study are higher than those estimated for indoor dust from houses of several Chinese regions [39], and childcare facilities, salons, and homes across the USA [29].
Albar et al. [37] assessed human exposure to phthalates via dust ingestion for the worst-case scenario  where CS is the BaPTEQ concentration (mg kg -1 ); CSFing, CSFinh, and CSFder are carcinogenic slope factors of 7.3, 3.85 and 25 (mg kg -1 day -1 ) -1 , respectively. As shown in Table 5, the total cancer risk could be attributed almost entirely to ingestion and did not vary much with microenvironment.
Therefore, inhalation of resuspended particles through the mouth and nose or via dermal contact was almost negligible when compared with the ingestion route. Under most regulatory programmes, an ILCR between 10 −6 and 10 −4 indicates potential risk, an ILCR of 10 −6 or less is considered insignificant and an ILCR ≥ 10 −4 is taken as high risk. In the present study, the total risk of adult and children exposure to PAHs in dust via the three pathways ranged from 7.2 × 10 −6 to 1.4 × 10 −5 . This means that the resuspendable thoracic fraction of household dust can contribute to an estimated excess of 7.2 to 14 per million people new cancer cases. One cancer case per million people is usually used as a baseline level of acceptable risk.

Conclusions
This preliminary study provides a first insight on the occurrence of plasticizers and PAHs in PM10 from resuspended dust samples from Spanish households and adds to the growing evidence that nondietary exposure contributes to the total body burden. Considering that people spend most of their time indoors, exposure to these pollutants might lead to an increased human health risk. Although no appreciable differences between plasticizer and PAH levels in resuspended dust from the various residential microenvironments were observed, it was concluded that exposure through the ingestion route poses much higher risks compared to inhalation and dermal contact. This is of particular concern for infants due to their higher dust intake via frequent hand-to-mouth activities. Because of the small

Conflicts of Interest:
The authors declare no conflict of interest.