*3.4. Morphology*

The SEM analysis revealed the presence of well-formed minerals and irregular aggregates, with abundant silicate minerals and an un-sorted mixture of geogenic and anthropogenic particles (Figure 2). Particles with irregular and subangular to angular shapes, including plate like morphology, with variable chemical composition that comprises Fe, Cu, Zn, S, Al, Ti and Sb, sugges<sup>t</sup> abrasion processes on their formation, such as tyre, break and pavement wear and vehicle corrosion interfaces. Rounded, longish and plate-like particles were also found. Although some of the particles present a larger size, they are formed by several smaller particles, usually <10 μm, which are composed of a mixture of anthropogenic and natural substances. It is known that brake abrasion generates particles containing Zn, Cu, Ti, Fe, Cu, and Pb, and other specific compounds such as sulphate silicate and barium sulphate. Particles from tyre abrasion comprise Cd, Cu, Pb, and Zn [63,64]. Silicates and Fe oxides/hydroxides tied to Cl/S can also be associated with traffic or resuspension. Particles with spherical morphology are produced in high-temperature processes (e.g., asphalt and industry/metallurgy). Silicate and iron plerospheres and cenospheres in fly ashes were also found, with typical Si-Al-Fe-Cu-Ca and Fe-Si-Al-Na compositions. Although carbon could not be calculated due to the SEM analysis technique, its presence was confirmed by the XRD analyses. Asphalt paving, or bituminous, materials are mainly made of carbonaceous components (e.g., saturated hydrocarbons, aromatics, asphaltenes, etc.) that are mixed with mineral aggregates. In addition to metals, particles from brake-pad and brake-disc abrasion consists of carbon fibres and graphite, while particulate material from tyre wear comprises various carbonaceous constituents (organic compounds such as natural rubber copolymer, organotin compounds, and soot) [64]. Particle size indicates that road dust might be an important source of resuspended atmospheric particulate matter associated with non-exhaust emissions (e.g., brake, tyre, and road wear), as suggested in previous studies [65]. The size of most of the non-exhaust particles is commonly much larger than that of exhaust particles, since its formation comprises processes such as corrosion, crushing and mechanical abrasion.

**Figure 2.** SEM images and composition of anthropogenic and geogenic road dust particles: (**<sup>a</sup>**–**f**) plerospheres, cenospheres, aggregates, plate like particles and fibrous steel with considerable plastic deformation and fragmentation; (**g**) irregular aggregates; (**h**) porous Ca-Si-Fe rich particle; (**i**,**j**) well-formed minerals.

#### *3.5. Grain Size Distribution*

The grain size distribution of road dusts is presented in Figure 3. Results show a similar pattern for both suburban (S1) and urban areas (S2 and S3) with a marked unimodal distribution. The mass volume of particles peaked in the range from 10 to 106 μm, although small modes, barely noticeable, were observed below 5 μm. Particles < 100 μm can easily be resuspended in the wake of passing traffic or by the blowing wind and might enter the mouth and nose while breathing. Particulate matter of 10 and 2.5 μm or less in diameter (PM10 and PM2.5, respectively) can ge<sup>t</sup> deep into the lungs and some may even ge<sup>t</sup> into the bloodstream. Urban sample S3 presented a higher percentage of PM10, with S3PM10 37.8% > S1PM10 27.2% > S2PM10 25.0%, while PM2.5 represented a smaller fraction, with 4.5% (S3) > 3.0% (S1) > 2.7% (S2) of the total mass volume. A literature review by Grigoratos and Martini [66] documented unimodal mass size distributions of brake wear PM10, with a mass weighed mean diameter of 2–6 μm. On the other hand, tyre wear PM10 often exhibits a bimodal distribution with one peak lying within the fine particle size range and the other one within the coarse range (5–9 μm). It is estimated that almost 40%–50% by mass of generated brake wear particles and 0.1%–10% by mass of tyre wear particles is emitted as PM10. In terms of mass, more than 85% of diesel particulate exhaust emissions are below 1 μm. Gasoline vehicles emit an even higher proportion of smaller particles than diesel vehicles [67].

**Figure 3.** Size distribution (%) of road dust particles, Ø < 106 μm.

#### *3.6. Human Health Exposure Assessment*

The hazard quotient (*HQ*), or non-carcinogenic effects, sugges<sup>t</sup> that ingestion is the major route of children's exposure to road dusts, with *HQing* ≈ *HI* (Figure 4), both by hand-to-mouth common habits and by resuspended particles. Dermal and inhalation routes can be considered negligible. Fraction F1 revealed a higher probability to induce non-carcinogenic health effects in children. The *HI* values for adults were approximately an order of magnitude lower than those for children. For both children and adults, Zr is the element that most contributes to possible non-carcinogenic effects. Adverse health effects may occur mostly by ingestion of resuspended particles (children: F1 *HQi*ng-Zr ≈ *HI* ranging from 27.51 to 553.10; F2 *HQ*ing-Zr ≈ HI ranging from 5.66 to 8.71; adults: F1 *HQ*ing-Zr ≈ HI ranging from 2.52 to 5.13; F2 *HQ*ing-Zr ≈ HI ranging from 0.38 to 0.69). Studies sugges<sup>t</sup> that Zr and its compounds represent a risk for pulmonary health effects (benign) potentially associated with short-term exposure [68].

**Figure 4.** Non-carcinogenic chronic hazard quotient (*HQ*) of Al, As, Cu, F, Fe, Mn, Rb, V and Zr by ingestion in children and cumulative non-carcinogenic hazard index (*HI*). Logarithmic scale.

The probability of an individual to develop any type of cancer over lifetime (Risk) by As (RiskAs) content in fraction F1 of the suburban sample is 1.58 × 10−<sup>4</sup> (Table 2), a Risk above the acceptable target of 1 × 10−<sup>4</sup> proposed by USEPA [24], so the adoption of local measures is suggested. Fraction F2 of the same sample showed a RiskAs = 1.58 × 10−5. RiskAs values of 3.07 × 10−<sup>5</sup> and 2.46 × 10−<sup>5</sup> were obtained for fraction F1 of road dust from urban streets, while the corresponding values for fraction F2 were 1.84 × 10−<sup>5</sup> and 1.05 × 10−5. These cancer risks are in the range for which managemen<sup>t</sup> measures are required. The dermal risk for As in all samples and fractions was also within this range. Arsenic and its inorganic compounds are classified as carcinogenic to humans since 2012 [69]. The Pb content in fraction F1 of sample S3 is also indicative that, by the ingestion pathway, it may pose a risk (2.24 × <sup>10</sup>−5) to human health.


**Table 2.** Estimated human health risk for elements As and Pb.

F1—fraction <0.074 mm; F2—fraction >0.074 mm and <1 mm; ing—ingestion; inh—inhalation; drm—dermal.
