Organic Compounds in Dust from Rural and Urban Paved and Unpaved Roads Taken During the San Joaquin Valley Fugitive Dust Characterization Study

Soil samples

The road dust was collected from each road by Ashbaugh and coworkers (2003) as part of the FDCS. Samples from unpaved roads were obtained by sweeping loose surface material into a dustpan, and paved road samples were collected by vacuum sweeping. Further, dust samples from a staging area and construction sites were also available. A staging area is a designated area for farm equipment to gather, for example, before and after fieldwork and for loading crops during harvest. Surface soil samples from two construction sites were also collected that had been earlier impacted by dozers and other earth moving equipment; for more details, see Ashbaugh et al. (2003). Although portions of these samples were size-segregated using a resuspension chamber (Ashbaugh et al., 2003) into PM10 and PM2.5, only the total road dust samples were available to us for analysis. The samples were shipped frozen overnight to our FIU laboratory (Miami, FL), where they were stored on arrival in a freezer at <−21°C. To obtain representative results, samples of 5 g dust each were prepared for extraction. The sample codes, descriptions, and locations are given in Table 1.

Extraction and instrumental analysis

The sample extraction method applied here is identical to that used to extract surface dust samples collected at feedlots and dairies as part of the SJV FDCS; therefore, for a detailed description of extraction and instrumental analysis using gas chromatography–mass spectrometry (GC-MS), the reader is referred to Rogge and coworkers (2006).

Compound identification and quantification

Data were acquired and processed with the HP-Chemstation software. Individual compounds were identified by comparison of mass spectra with literature and library data, comparison of mass spectra and GC retention times with those authentic standards and/or interpretation of mass spectrometric fragmentation patterns, see Rogge et al. (2006). Relative response factors (RRF) were determined using 1-phenyldodecane as injection standard. Identifiable compounds were quantified using the MS-data system. Ion counts were converted to compound mass using the area counts of the internal recovery standards (five deuterated standards added to the samples before extraction, including three perdeuterated n-alkanes (C₂₀D₄₂, C₂₃D₅₀, and C₃₀D₆₂) and two perdeuterated n-alkanoic acids (n-decanoic-D19 acid, n-tetradecanoic-D27 acid). Measurement uncertainty was determined using replicate solvent extractions, GC-MS analyses and the standard deviation of each of the RRFs obtained for authentic standards, replicate sample analysis, and error propagation. For compounds for which no authentic standards were available, the RRFs and measurement uncertainties were assumed to be identical to those of available standard compounds with similar chemical structures, polarities, and molecular weights; see footnote to Table 2.

Aliphatic lipids

n-Alkanes found in dust samples from paved roads in urban and rural areas showed distributions that are indicative of a mixture derived from biogenic (natural) and fossil fuel sources, with the higher plant contributions less dominant in urban areas (see CPI values, Table 2). Similarly, dust samples from agricultural and public unpaved roads had n-alkane distributions originating from fossil fuel products (C₁₄–C₂₆) and biogenically derived n-alkanes (C₂₅–C₃₅) with odd-carbon number predominance, peaking at C₃₁. Further, tire-wear abrasion with its distinct higher molecular weight n-alkane levels of up to C40 and higher (Rogge et al., 1993c) is especially pronounced in road dust collected from an urban paved road (FDPVR1), which revealed overall highest concentrations, followed closely by the other rural paved road dust samples; see Fig. 4a. Not surprising, urban paved road dust (FDPVR1) extracts revealed a large unresolved complex mixture (UCM) of branched and cyclic hydrocarbons (Table 2, Figs. 1a and 3a), a typical indicator for fossil fuel-derived inputs (Simoneit, 1982, 1985; Rogge et al., 1993b). The soil samples analyzed for the staging area showed an even more pronounced fossil fuel input (Fig. 3c). Possibly, dripping of engine oil from the farm equipment added to the UCM.

OPEN IN VIEWER

FIG. 4.

Total concentrations for biogentic, fossil fuel, and traffic-derived organic compounds. (a) n-Alkanes, n-alkanoic acids, and n-alkanals. (b) Tire-wear debris marker, benzothiazole and fossil fuel markers, hopanes. PAH are indicative of contributions from incomplete combustion. PAH, polycyclic aromatic hydrocarbons.

n-Alkanoic acids (C₈–C₃₆) belong to the most prominent compound class detected in road dust, staging area, and construction sites. Dust samples from unpaved roads presented a bimodal n-alkanoic acid distribution, the homologs >C22 are derived mainly from higher plant waxes and those <C20 are ubiquitous in origin. Overall, the highest total fatty acid and n-alkanes concentrations are found in paved road dust (see Fig. 4a). Not surprising, consider that the grinding motion of tires over a hard paved surface is more effective in pulverizing vegetative detritus that contains n-alkanes and n-alkanoic acids as part of their surface waxes (Rogge et al., 1993c, 1993d). Further, n-alkanoic acid content of samples from paved roads is enriched by lower molecular weight fatty acids, likely of microbial and fungal origin (Rogge, 2000), which is more prevalent where mechanical grinding of vegetative detritus allows enhanced microbial decay. At the urban paved road site, the C₁₈ fatty acid dominates, whereas in the rural paved samples, the fatty acid distribution is dominated by C₁₆. The presence of the monounsaturated C₁₈ fatty acids in both paved and unpaved road samples (Fig. 2c) is representative of fresh and unoxidized lipids at these sites. In all samples, including the staging area, vascular plant wax inputs were apparent, that is, n-alkanoic acids >C₂₂ with a strong even carbon number predominance and C_max at 28 or 30 (Table 2).

n-Alkanols could not be identified in dust from the urban paved road, but occurred at low levels in the public unpaved road and staging area samples (Table 2, Fig. 2b). n-Alkanols peaked at C28 for the rural paved and agricultural unpaved road sites, typical of vascular plant wax (Rogge et al., 1993d; Simoneit, 1999). The chromatograms from dirt roads show strong contributions of crude oil type products (see UCM in Figs. 1a, and 3a, c); therefore, potentially minor n-alkanol contributions from natural sources are simply masked and not detectable.

Similarly, n-alkanals were not detectable in both the urban paved road and staging area samples, see Fig. 4a. On the other hand, dust samples from paved roads going through rural areas and unpaved roads used by agricultural equipment and the public contain fatty aldehydes with an even-to-odd carbon number preference and C_max at 28 or 30 indicative of recent plant litter input. However, for the unpaved roads used for agricultural purposes only and those for private use, a continuous n-alkanal series is observed with little carbon number preference for the lower carbon numbers, less than C₂₀. n-Alkan-2-ones, which were postulated to be the direct result of microbial β-oxidation of n-alkanes (Allen et al., 1971; Rieley et al., 1991), presented C_max=33 in the rural paved road sample FDPVR4, different than the peak observed for the n-alkanes (C_max=31). However, the public unpaved road samples revealed a very similar distribution for the higher molecular n-alkanes and n-alkan-2-ones, similar to the urban (FDPVR1) and rural (FDPVR3) paved roads. Possibly, the mechanical pulverization of litter on paved roads by car tires aids the microbial β-oxidation of n-alkanes.

The presence of alkylcyclohexanes is particularly pronounced in paved and unpaved roads and in the staging area and further supports the input from vehicular emissions, as their distribution ranging from C₂₀ to C₂₈ and C_max at 23 or 24 falls within the values observed for diesel truck particulate emissions (Schauer et al., 1999). Therefore, not surprising, the levels of alkylcyclohexanes determined collaborates with the findings for hopanes (pentacyclic triterpanes), polycyclic aromatic hydrocarbons (PAH), and benzothiazoles that are associated with traffic-related exhaust and brake dust emissions (Rogge et al., 1993b, 1993c).

Steroids

Steroids are typical constituents in soil organic matter (Stevenson, 1966; Jaffé et al., 1996; Rogge et al., 2007) and are the most abundant compound group in soils from dairies and feedlots (Rogge et al., 2006). The prominent phytosterols (C₂₈ and C₂₉ sterols) derived from terrestrial plants are sitosterol, stigmasterol, and campesterol (Manitto, 1981; Volkman, 1986), and these compounds were observed in the paved rural and unpaved road samples (e.g., Fig. 2d). Sitosterol was also detected in the staging area at a relatively high concentration (141 ng g⁻¹). Cholesterol is the major sterol in body tissue of animals, but is also found at a lesser degree in plants (Heftmann, 1968; Manitto, 1981), and at very low levels in bacteria (Heftmann, 1968). Cholesterol was observed for all road samples but in the construction sites. Cholestane, one possible tracer for fossil fuel and fossil fuel combustion products (Rogge et al., 1993b; Simoneit, 1999), was present in all samples with higher concentrations in dust from paved roads and staging area. These cholestane concentrations seem to reflect the usage of the roads by vehicles and other machinery that are operated by internal combustion engines.

Terpenoids and isoprenoids

Only a few triterpenoids were detected in these road dust samples. The pentacyclic triterpenoids α-amyrin, β-amyrin, ursolic, betulinic acids, and taraxerol, common constituents of higher terrestrial plants (Chandler and Hooper, 1979; Pushpa and Rastogi, 1979) were identified only in the rural paved road sample and may reflect plant detritus on the ground (Rogge et al., 2007). Dehydroabietic acid, a biomarker of conifers (Simoneit, 1977; Rogge et al., 1998), was observed only in the staging area dust sample, possibly due to wood storage or biomass burning at or near this site. Amyrin acetates were found at low levels for the public unpaved road site (Table 2, Fig. 3a). The production of acetates in soil seems to be related to the presence of anaerobic microorganisms, which utilize this compound as an intermediate during the turnover of carbon in habitats subject to steep oxygen gradients (Küsel et al., 1999).

Phytol in soil is an indicator of fresh and recent plant litter and was detected at a higher concentration in the rural paved road sample (Table 2). Phytol was below the detection limit in dust samples from the urban paved road, staging area, and construction sites. Its degradation product, 6,10,14-trimethylpentadecan-2-one, was observed in all road dust samples, except for the staging area. Identical to phytol, its highest concentration was for the dust samples from rural paved roads. Squalene, the precursor of all steroids, showed the highest concentration for the construction sites and was not detected in the paved road dust.

Benzothiazoles and sulfur

Benzothiazole is derived from the chemical breakdown of vulcanization accelerators used during tire manufacturing and was already detected in tire wear particles and fine particulate dust from residential roads (Rogge et al., 1993c). In the present study, benzothiazole was found in dust from sites that are used by vehicles, that is, paved and unpaved roads (Fig. 4b). 2-Hydroxybenzothiazole was detected only in paved road dust samples, where also the highest benzothiazole concentrations were observed. In fact, at those sites, a major UCM was found (e.g., Fig. 1a, Table 2), and for the rural paved road samples, a considerable amount of sulfur was also detected (Fig. 1c). This supports inputs from vehicular emissions to those locales, as UCM is an indicator of fossil fuel inputs, similar to hopanes with the 17α(H), 21β(H) configuration, and C_max at 30 (Fig. 3d); and sulfur is a raw material for vulcanizing in tire manufacture.

Polycyclic aromatic hydrocarbons

PAH are formed during the incomplete combustion of fossil fuels and biomass burning (Rogge et al., 1993a, 1993b, 1993c; Rogge et al., 1998; Simoneit et al., 2005). These compounds are also introduced via road construction materials such as asphalt and other tar containing products (Rogge et al., 1993c; Srogi, 2007) that may add PAH to the road dust when asphalt surface is fresh and tire frictions result in asphalt particles abrasion or later when weathered asphalt surface becomes brittles and breaks up (Lee and Dong, 2010). Direct, vehicular exhaust is the major contributor of PAH found in road dust from paved and especially unpaved roads, but also tire wear particles, lubricating oils, and organic brake lining wear particles add PAH to the road dust (e.g., Rogge et al., 1993c; Srogi, 2007). Indirectly, via atmospheric fallout, PAH can be added to the road surface originating from virtually any source that generates PAH. In addition to the genotoxicity and carcinogenicity, PAH may induce immunomodulatory and cytotoxic effects on humans (e.g., Strickland and Kang, 1999; Janssen et al., 2003; Jalava et al., 2007). Due to wind and weather, PAH associated with road dust can be injected back into the atmosphere, a process that can occur multiple times, and expose especially people living and working near roadways (Tiitanen et al., 1999). Along roads that lead through agricultural areas, PAH deposition onto plants decreases quickly with the distance from the road; nonetheless, PAH are observed in forages that are consumed by farm animals and are ultimately associated with food products (Rogge et al., 1993d; Killian et al., 2000; Crépineau et al., 2003; Srogi, 2007; Dan-Badjo et al., 2008).

PAH ranging from three-ring phenanthrene to six-ring coronene have been detected and quantified in the road dust samples (Fig. 4b). Highest PAH concentrations were found in paved road dust from a rural road (FDPVR3) with 1443.4 ng g⁻¹ and urban road (FDPVR1) with 924 ng g⁻¹. In contrast, the other rural paved road (FDPVR4) reveals very little PAH, with only 35 ng g⁻¹ (>40 times less than determinant in FDPVR3) indicating that PAH content in paved road dust can vary substantially, depending on the usage frequency. Among the unpaved roads, the roads used exclusively by agricultural equipment (FDUPR1–3) show the lowest total average PAH concentration with just 4.7 ng g⁻¹. Instead, publicly used unpaved roads (FDUPR4–6) show a 17 higher total PAH concentration, that is, with 83 ng g⁻¹ still far below concentration levels observed for paved rural and urban roads. The unpaved staging area (FDSTA1) shows the highest PAH concentrations with 740 ng g⁻¹ for any of the roads or areas used by farm equipment. Construction sites (FDCON1–2) are impacted by heavy earth moving equipment for a short time only, and, therefore, the PAH content in surface dust is of recent origin only. Nonetheless, the average total PAH concentration is with 113 ng g⁻¹ much higher than that found in dust from unpaved roads. PAH released with the exhaust from vehicles from internal combustion engines show a linear relationship between alkylcyclohexanes, PAH, and hopanes (following discussed) emissions, a trend that is observed here as well (see Table 2) which is indicative of the origin of most PAH found here to be derived from fossil fuel-powered internal combustion engines.

Pentacyclic triterpanes

Pentacyclic triterpane hydrocarbons, mainly hopanes, are associated with fossil fuel that have been matured over millions of years and are today, similarly to steranes, used to assess the maturity of crude oils. Although not found in gasoline and diesel fuel since they belong to the higher boiling fraction of crude oil, they are associated with lubricating oils and are consequently found in the exhaust of gasoline as well as diesel fuel-powered vehicles and are widely used as marker compounds for vehicular emissions (e.g., Rogge et al., 1993b, 1993c; Simoneit, 1985, 1999). Here, the major eight hopanes have been quantified, with the highest concentrations of 4163 ng g⁻¹ for dust collected from the rural paved road (FDPVR3), followed by the urban paved road FDPVR1 with 2998 ng g⁻¹ and unpaved staging area (FDSTA1) with 3305 ng g⁻¹. The lowest hopane concentrations are found for the unpaved roads, paralleling cholestane, a sterane used as well as a marker compound for vehicular emissions and the concentrations for alkylcyclohexanes and PAH as just discussed.

Saccharides

The extracts of dust samples from rural and urban paved roads have distinct saccharide distribution patterns. The rural paved roads are marked by the presence of high concentrations of monosaccharides (α- and β-glucose, α- and β-fructose, Fig. 1d), whereas the reduced sugars (inositols, xylitol) occurred only in the urban paved road sample (Fig. 1b). The latter may be related to fungal metabolic products from intracellular osmoregulators (mainly under stress conditions) (Loos et al., 1994). The disaccharides sucrose and mycose (or trehalose) were detected in both paved and unpaved roads (see Figs. 1b, d, 2a, and 3b). Mycose is a fungal metabolite (Martin et al., 1988); whereas sucrose is the predominant sugar in the phloem of plants (Bieleski, 1995). The occurrence of sucrose/mycose ratios lower than 1 was observed for rural paved and agricultural unpaved road samples (e.g., Fig. 2a), indicating more pronounced fungi/yeast residues from microbial activity near those sites at that time. Similar saccharide distribution profiles were observed for dust from unpaved road samples. Both agricultural and public road locales had α- and β-glucose, sucrose, mycose, and glycerol as the most common sugars found. In fact, those saccharides occurred in most dust samples analyzed here, and were the dominant saccharide tracers of all agricultural soil samples from the SJV (Rogge et al., 2007). Levoglucosan, a tracer for biomass burning (Rogge et al., 1998; Simoneit et al., 1999), was detected at low levels only in the dust sample from an urban paved road (Fig. 1b).

Pesticides and plasticizers

Two different pesticides were found in the agricultural unpaved road samples. DDE, a degradation product of the already banned pesticide DDT, was detected at low concentrations (e.g., 13 ng g⁻¹). The defoliant Fosfall was identified at a higher level (428 ng g⁻¹, e.g., Fig. 2a). This indicates that some spillage of agricultural products has occurred along the roads bordering fields. Dactual (DAC-893) was present in the sample from the staging area at 238 ng g⁻¹. Plasticizer contamination was detected in dust from urban paved and public unpaved road samples (Fig. 3a), as well as in the staging area. These plasticizers are dominant components of plastics and major compounds in smoke from burning plastics (Simoneit et al., 2005).

Diphenylamines

Diphenylamine and derivatives are widely used as an antioxidant in the rubber and elastomer industry (as antiager), in the production of dyes and lubricating oils, as a stabilizer for nitrocellulose-containing explosives and propellants and diphenylamine is used to prevent postharvest deterioration of apples and pears (Drzyzga, 2003). Butyl-octyl diphenylamine, particularly, can control the increase of oil viscosity in internal combustion gasoline engine. Some ecotoxicological studies demonstrated the potential hazard of various diphenylamines to the aquatic environment and to bacteria and animals (Drzyzga, 2003). This compound class was found at the staging area site only, reinforcing the significant input of anthropogenic-derived organic components at this site.