Particle size distribution and perchlorate levels in settled dust from urban roads, parks, and roofs in Chengdu, China

Abstract

A total of 27 settled dust samples were collected from urban roads, parks, and roofs in Chengdu, China to investigate particle size distribution and perchlorate levels in different size fractions. Briefly, fine particle size fractions (<250 μm) were the dominant composition in the settled dust samples, with mean percentages of 80.2%, 69.5%, and 77.2% for the urban roads, roofs, and the parks, respectively. Perchlorate was detected in all of the size-fractionated dust samples, with concentrations ranging from 73.0 to 6160 ng g⁻¹, and the median perchlorate levels increased with decreasing particle size. The perchlorate level in the finest fraction (<63 μm) was significantly higher than those in the coarser fractions. To our knowledge, this is the first report on perchlorate concentrations in different particle size fractions. The calculated perchlorate loadings revealed that perchlorate was mainly associated with finer particles (<125 μm). An exposure assessment indicated that exposure to perchlorate via settled road dust intake is safe to both children and adults in Chengdu, China. However, due to perchlorate mainly existing in fine particles, there is a potential for perchlorate to transfer into surface water and the atmosphere by runoff and wind erosion or traffic emission, and this could act as an important perchlorate pollution source for the indoor environment, and merits further study.

---

Environmental impact

Perchlorate contamination has been a hot topic recently, especially for its high level of indoor pollution. However, no known indoor source has been found for perchlorate contamination. Our current research found that median perchlorate levels increased with decreasing particle size, and perchlorate mainly existed in fine outdoor settled particles, indicating outdoor settled fine particles could act as an indoor perchlorate contamination source when re-suspended by wind erosion or traffic emission. To our knowledge, this is the first report on the perchlorate levels in different particle size fractions of settled outdoor dust, which could help us to deeply understand perchlorate pollution in dust samples and achieve indoor perchlorate contamination source apportionment.

1. Introduction

Urban road dust (URD) is a complex mixture and has recently increasingly raised concerns, because it could act as both sink and source for many anthropogenic pollutants, such as PAHs, perfluorinated compounds, and heavy metals.^1–3 Humans could be exposed to URD and suffer an adverse health risk via ingestion, inhalation, and dermal contact, especially for toddlers and children, who play more on the ground and have a higher frequency of hand to mouth activities. Fine particles have lower densities, higher surface areas, and thus contain higher levels of most contaminants.^1,3,4 What’s more, the efficiency of conventional sweeping decreases with decreasing particle size, particularly for those particle sizes smaller than 250 μm.^5,6 Settled fine URD could be re-suspended by wind erosion or traffic emission and affect human health. In addition, previous studies documented that fine URD has more potential to transfer into lakes and rivers through rainfall runoff, and poses direct effects on downstream aquatic and lake ecosystems.^1,7 Furthermore, the re-suspended outdoor settled fine dust could act as an important pollution source for the indoor environment, especially for those pollutants with a lack of indoor sources, e.g. perchlorate.⁸ Therefore, it is necessary to understand the grain size distribution of URD and size dependent levels of pollutants given the reasons stated above.

Perchlorate contamination has been a hot topic recently, mainly due to the fact that it is found ubiquitously in various environmental media and human samples,^9–13 and also poses adverse health effects in humans, such as affecting thyroid function by inhibiting the uptake of iodine.¹⁴ Perchlorate has high water solubility, mobility, and stability, and is widely used in solid rocket propellants, munitions, fireworks, roadside flares, air bag inflation systems, and pharmaceutics.¹⁵ Besides anthropogenic sources, perchlorate also has a natural origin.^16,17 Quite recently, perchlorate was found to widely occur in outdoor and indoor dust from some Asian countries,^{8,10,12,13,18} European countries,^12,19 and the U.S.,¹² with the concentrations ranging from below detection to 5300 mg kg⁻¹ in outdoor dust, while it ranged from below detection to 821 mg kg⁻¹ in indoor dust. Yao et al.⁸ conducted an important study and found that indoor PM_2.5-associated perchlorate might mainly originate outdoors, and perchlorate was primarily distributed in fine airborne particles. However, little is known regarding grain size dependent perchlorate levels in settled road or outdoor dust. As stated above, settled fine dust could be re-suspended by wind erosion or traffic emission easily, and needs further study.

Therefore, the primary objectives of this study were (1) to investigate the particle size distribution of settled dust in roads, parks, and roofs from Chengdu, China; (2) to determine and compare perchlorate levels in different particle size fractions and its related loading for individual fractions in settled road, park, and roof dust; and (3) to evaluate human exposure to perchlorate via outdoor settled dust ingestion, inhalation, and dermal contact. To our knowledge, this is the first report regarding perchlorate concentrations in different particle size fractions in settled outdoor dust, which could help us to deeply understand perchlorate pollution in dust samples.

2. Materials and methods

2.1. Materials

Perchlorate was purchased from Sigma-Aldrich (St. Louis, MO, USA). The ¹⁸O-labeled perchlorate used as an internal standard (IS) was obtained from Cambridge Isotope Laboratories (Andover, MA, USA). HPLC-grade methanol was obtained from J. T. Baker (Phillipsburg, NJ, USA). Milli-Q water was used throughout the study.

2.2. Study site

Chengdu (12 400 km²) is the capital of the Sichuan province, and is the largest typical inland city in southwest China with a population of 14.2 million. It is the home of the panda, and it currently has the most rapid economic development in China. Chengdu has three primary roads, the 1^st ring road, 2^nd ring road, and 3^rd ring road, and they are 19.4, 28.3, and 51.4 km long, respectively. These three main roads undergo regular daily hand-sweeping or mechanical sweeping, and have a high traffic intensity. The roofs (higher than 18 floors) and urban parks were randomly selected in southern, western, eastern, and northern Chengdu, and the roofs are rarely swept, but the parks undergo regular daily hand-sweeping.

2.3. Sample collection

A total of 27 outdoor settled dust samples were collected using a vacuum cleaner (Dyson DC34, UK) from the 30^th of March to the 2^nd of April in 2015, including 4 samples from the 1^st ring road, 5 samples from the 2^nd ring road, 10 samples from the 3^rd ring road, 4 samples from 4 different parks, and 4 samples from 4 different roofs (higher than 18 floors) in Chengdu, China. The weather of the last two weeks before the sampling campaign was cloudy, and the average temperature was approximately 22 °C. Road dust samples were collected according to the U.S. EPA standard methods.^20,21 Settled dust samples in the parks were collected along the center of the main path, and deposited dust samples on each roof were directly collected using a vacuum cleaner. All the samples were preserved in sealed polyethylene packages separately to avoid contamination, and subsequently the samples were transported to the laboratory and stored at −20 °C until pretreatment.

2.4. Sample preparation

Dust samples were sorted into particle size fractions of <63 μm (silt and clay), 63–125 μm (very fine sand), 125–250 μm (fine sand), 250–500 μm (medium sand), and 500–1000 μm (coarse sand) using nylon sieves according to sedimentology.²² Each sorted fraction was weighed and then subsequently extracted using the method developed in our previous study,¹⁰ and each sample was treated in duplicate. The total perchlorate level (C_total) in each dust sample was calculated using eqn (1).where C_i (mg kg⁻¹) is the perchlorate concentration in the particle size fraction i; M_i (kg) is the weight of the particle size fraction i.

2.5. Instrumental analysis

Analysis was performed using LC-MS/MS (Agilent Technologies, USA) using the method described in our previous study.¹⁰

2.6. Quality assurance and quality control

Quantification was conducted using internal calibration, and the internal standard concentration was 10 ng mL⁻¹. Calibration standards were injected after each batch of 100 injections, and the RSD of the slope was within 3%. The limit of quantification (LOQ) was 0.15 ng mL⁻¹, and defined as a value corresponding to a signal-to-noise ratio of 10. Recoveries of native perchlorate spiked into each fraction in triplicate were higher than 90%, and the results are presented in Table S1 in the ESI.† After analyzing each batch of 20 samples, a group of solutions was injected to ensure accuracy and check for any cross-contamination and memory effect during sample pretreatment and instrumental analysis, including low (1.0 ng mL⁻¹), middle (10 ng mL⁻¹), and high (100 ng mL⁻¹) levels of perchlorate standards, procedural blank, and reagent blank. The precision and accuracy of the analysis, reported as recoveries and relative standard deviation (RSD), are shown in Table S1.† No detectable perchlorate level was found in the blanks.

2.7. Exposure evaluation

The daily intake of perchlorate via dust ingestion, inhalation, and dermal contact was calculated for children and adults using the equations suggested by the U.S. EPA,^23,24 and is given in Section 1 in the ESI.† In the case of dust ingestion and dermal contact, the total perchlorate level was used, while for dust inhalation, the finest particle associated perchlorate concentration (C_{<63 μm}) was employed. Two different exposure Scenarios (A and B) were considered in this study. Briefly, in the case of Scenario A, the median total perchlorate concentration and mean dust ingestion rate were used, and for dermal contact, the median perchlorate concentration was used, which represented the mean exposure scenario. For Scenario B, the 95^th percentile perchlorate concentration and high dust ingestion rate were used, and for dermal contact, the 95^th percentile perchlorate concentration was used, which represented the high exposure scenario. All of the parameters for exposure evaluation used in this study are shown in Table S2 in the ESI.†

2.8. Statistical analysis

Statistical analysis was performed using the SPSS 21.0 software program. Statistical tests were considered significant when p < 0.05.

3. Results and discussion

3.1. Particle size distribution

As seen in Fig. 1, the fine particle size fractions (<250 μm) were the dominant composition in the URD samples, with mean percentages accounting for 78.8%, 80.4%, 81.5%, 69.5%, and 77.2% of the URD samples from the 1^st ring road, 2^nd ring road, 3^rd ring road, roofs, and the parks, respectively, suggesting the importance of understanding the contaminant levels in the fine particles. This is consistent with previous studies, which suggest that both infrequent sweeping (roofs) and the low removal efficiency of street sweeping might result in this phenomenon.^1,25 There were some differences in grain size distribution among each sample type (Fig. 1). Briefly, the 63–125 μm fraction was the most abundant in the 2^nd and 3^rd road dust samples, followed by the <63 μm and 125–250 μm fractions. However, in the case of the 1^st road dust sample, no significant difference was found between the 63–125 μm and <63 μm fractions. The relatively lower average vehicle speed and traffic density on the 1^st ring road (40–60 km h⁻¹) of Chengdu city as compared to the 2^nd (60–80 km h⁻¹) and 3^rd (80–100 km h⁻¹) ring road might lead to this result.²⁶ For the dust samples from the roofs, no differences were obtained between each size fraction, and this might be ascribed to infrequent sweeping and little anthropogenic activity. Due to regular sweeping, coarse particles were significantly less present than fine particles in the dust samples from the parks.

Fig. 1 Particle size distribution of each sampling site (values followed by the same letter (a–e) in each sampling site are not significantly different at the 0.05 probability level).

3.2. Perchlorate in size-fractionated and bulk settled URD

This is the first report on perchlorate levels in grain size-fractionated URD samples. Perchlorate was detected in all of the size-fractionated dust samples (Table 1), with concentrations ranging from 73.0 to 1500 ng g⁻¹, from 80.3 to 2170 ng g⁻¹, from 148 to 3020 ng g⁻¹, from 198 to 4630 ng g⁻¹, and from 310 to 6160 ng g⁻¹ in the 500–1000 μm, 250–500 μm, 125–250 μm, 63–125 μm, and <63 μm fractions, respectively. Generally, the median perchlorate levels increased with decreasing particle size, and the perchlorate level in the finest fraction was significantly higher than the >125 μm fractions, with the exception of the 2^nd ring road samples, in which the concentration of perchlorate in the finest fraction was only significantly higher than that in the coarsest fraction. Previous studies indicated that finer particles have a greater potential to transfer into surface waters by urban runoff, and might have direct effects on aquatic systems,^1,4 in addition, finer particles are more likely re-suspended by wind erosion or traffic emission and could have more adverse health effects on humans via dust inhalation. It should be noted that perchlorate levels at the grain size <125 μm were comparable to or relatively higher than those of indoor dust from Chengdu.¹³ No known indoor source has been found for perchlorate, and hence, outdoor fine particles might be an important source for indoor perchlorate pollution. Therefore, high levels of perchlorate in the finer particle fractions requires concern and more investigation. In the case of the bulk samples, the median perchlorate concentration of the 1^st ring road samples was 1450 ng g⁻¹, and was significantly higher than those of the other samples, while no significant differences were found among the other samples. The 1^st ring road is located in the center of Chengdu city, where high traffic density, lower average traffic speed, and great anthropogenic activities might result in higher perchlorate levels, and similar results were obtained in the finer particle fractions (<125 μm) of settled dust samples from the different sampling sites in this study. The median concentration of perchlorate observed in this study was lower than those in the dust fall samples of Malta,¹⁹ and also lower than that in the outdoor dust from Sichuan, China around Chinese Traditional Spring Festival.¹⁰

Table 1 Perchlorate in size-fractionated settled dust and bulk dust samples (ng g⁻¹)^a

Sampling site		500–1000 μm	250–500 μm	125–250 μm	63–125 μm	<63 μm	Total
a The same letter (a–c) in each sampling site means not significantly different at the 0.05 probability level.
1^st ring road	Mean (SD)	588 (625)	856 (893)	1260 (1200)	2320 (1630)	3440 (1890)	2200 (1750)
	Median	355	534	840	1740	2870	1450
	Range	138–1500	193–2170	338–3020	1170–4630	1850–6160	1070–4810
	Difference	a	a	a	ab	b	—
2^nd ring road	Mean (SD)	412 (347)	513 (501)	725 (683)	942 (633)	1230 (779)	861 (650)
	Median	256	279	445	829	778	592
	Range	104–962	132–1340	209–1900	348–1990	663–2490	352–1960
	Difference	a	ab	ab	ab	b	—
3^rd ring road	Mean (SD)	285 (206)	388 (257)	465 (288)	711 (506)	1430 (860)	795 (537)
	Median	241	371	434	617	1024	614
	Range	73.0–724	80.3–859	148–834	198–1500	392–2850	241–1850
	Difference	a	a	a	a	b	—
Roofs	Mean (SD)	151 (60.2)	235 (85.8)	337 (93.8)	702 (512)	1650 (1430)	719 (526)
	Median	123	200	315	524	1456	593
	Range	117–241	177–361	259–461	309–1450	370–3320	253–1440
	Difference	a	a	a	ab	b	—
Parks	Mean (SD)	391 (304)	448 (353)	743 (817)	1090 (839)	1960 (1690)	1180 (978)
	Median	322	317	384	920	1887	1176
	Range	109–812	191–967	241–1960	334–2180	310–3740	255–2120
	Difference	a	a	ac	ab	bc	—

---

3.3. Perchlorate loads

To determine the contribution of each particle size to the overall perchlorate contamination of the URD, perchlorate loads in each grain size fraction (GSF_load) were calculated for each individual sample using eqn (2), and the results are shown in Fig. 2.where C_i is the perchlorate level in the URD with a particle size fraction of i; GS_i is the mass percentage of the particle size fraction i; and m is the number of particle size fractions.

Fig. 2 Perchlorate loads in each particle size fraction (values followed by the same letter (a–e) in each sampling site are not significantly different at the 0.05 probability level).

For the URD samples collected from the 1^st ring road, 2^nd ring road, 3^rd ring road, and the parks, the finer particle size fractions (<125 μm) have a significantly higher contribution to perchlorate levels in the URD samples, with mean loads of 82.8%, 72.1%, 79.0%, and 73.5%, respectively, while the coarser fractions (>125 μm) accounted for less than 30.0% of the perchlorate in the dust samples. In the case of the roofs, the perchlorate loads varied more than the other samples, with a mean percentage of 62.6 ± 30.4% for size fractions less than 125 μm. Little anthropogenic activity and rare cleaning might cause this phenomenon. However, the finest size fraction also contributed significantly more to the total perchlorate concentration in the URD samples from the roofs. Based on the obtained perchlorate loads in this study, perchlorate was mainly in the particle size fractions <125 μm, indicating daily conventional sweeping could not remove it efficiently, and thus, fine particle associated perchlorate could easily transfer into surface water and the atmosphere by runoff or wind erosion and traffic emission.^1,26,27 In addition, fine particles have a higher tendency to undergo intake by humans via dermal contact and inhalation, therefore, fine particle associated perchlorate could pose more potential adverse health effects, especially for those who work in outdoor environments, and merits more attention.

3.4. Exposure evaluation

Daily human intake of perchlorate via URD ingestion, inhalation, and dermal contact was evaluated for children and adults in Chengdu, China. As mentioned above, perchlorate mainly existed in fine particles based on the results of this study, therefore, the particle size associated perchlorate level was considered during calculation, and the results are given in Table 2. Briefly, dust ingestion was the primary exposure route for both children and adults in both exposure Scenarios A and B, and accounted for more than 98% of the daily perchlorate intake via URD based on the obtained data, followed by dermal contact and inhalation. Due to a higher frequency of hand to mouth activities and more time to play on the ground, the daily perchlorate intakes via URD for children were at least 5 times higher than those for adults. The total perchlorate intakes via URD for children were 2.99 and 49.7 ng kg⁻¹ d⁻¹ based on the exposure Scenarios A and B, respectively, while those for adults were 0.32 and 7.14 ng kg⁻¹ d⁻¹, respectively. Both of them were orders of magnitude lower than the reference dose suggested by the U.S. EPA,²⁸ indicating exposure to perchlorate via settled URD is safe for local residents based on the current study. However, as perchlorate was mainly associated in fine particles, settled URD could act as a pollution source for the indoor environment when re-suspended by wind erosion or traffic emission. Previous studies implied that outdoor dust might be an important source for indoor perchlorate contamination, and perchlorate could accumulate in the indoor environment.^8,19 Therefore, it is necessary to conduct more investigations on fine particle associated perchlorate levels in settled outdoor dust to understand the indoor source of perchlorate, which need to go down to dust size levels like 30 μm or so, which are more likely to re-suspend into air.

Table 2 Daily perchlorate intakes (ng kg⁻¹ d⁻¹) of children, and adults via ingestion, inhalation and dermal contact of outdoor dust in Chengdu, China

Exposure scenario	Children		Adults
	A	B	A	B
Ingestion	2.96	47.8	0.32	7.11
Inhalation	0.002	0.009	0.0004	0.002
Dermal contact	0.03	0.13	0.004	0.02
Total	2.99	47.9	0.32	7.14

---

4. Conclusions

Due to the low removal efficiency of conventional street sweeping, fine particle size fractions (<250 μm) were the dominant composition in the URD samples. Particle size associated perchlorate levels were investigated in the URD samples from Chengdu, China. To our knowledge, this is the first report on the concentration of perchlorate in different particle size fractions in URD samples. The results revealed that the median perchlorate levels increased with decreasing particle size. Combined with the calculated perchlorate loads, perchlorate was mainly associated with a grain size of <125 μm, suggesting that perchlorate has the potential to transfer into the aquatic environment and atmosphere via urban runoff and wind erosion or traffic emission, and pose direct effects on ecosystems or humans. Exposure evaluation indicated that human direct expose to perchlorate via URD intake is safe to local residents based on the current study, but as perchlorate primarily existed in fine particles, URD might act as an important indoor perchlorate contamination source, and merits more study.

Acknowledgements

This study was supported by the Scientific Research Starting Foundation for Young Teachers, Sichuan university (No. 2015SCU11024), and the Experiment Technology Project, Sichuan university (No. 2015-111).

References

1. H. Zhao, X. Li, X. Wang and D. Tian, Grain size distribution of road-deposited sediment and its contribution to heavy metal pollution in urban runoff in Beijing, China, J. Hazard. Mater., 2010, 183(1), 203–210.
2. D. Lorenzi, J. A. Entwistle, M. Cave and J. R. Dean, Determination of polycyclic aromatic hydrocarbons in urban street dust: implications for human health, Chemosphere, 2011, 83(7), 970–977.
3. M. Murakami and H. Takada, Perfluorinated surfactants (PFSs) in size-fractionated street dust in Tokyo, Chemosphere, 2008, 73(8), 1172–1177.
4. S. Roger, M. Montrejaud-Vignoles, M. Andral, L. Herremans and J. Fortune, Mineral, physical and chemical analysis of the solid matter carried by motorway runoff water, Water Res., 1998, 32(4), 1119–1125.
5. R. Pitt, Demonstration of nonpoint pollution abatement through improved street cleaning practices, EPA-600/2-79-161, U.S. Environmental Protection Agency, Cincinnati, OH, 1979.
6. G. Bender and M. Terstriep, Effectiveness of street sweeping in urban runoff pollution control, Sci. Total Environ., 1984, 33(1), 185–192.
7. L. Maltby, D. M. Forrow, A. Boxall, P. Calow and C. I. Betton, The effects of motorway runoff on freshwater ecosystems: 1. Field study, Environ. Toxicol. Chem., 1995, 14(6), 1079–1092.
8. L. Yao, L. Yang, J. Chen, K. Toda, X. Wang, J. Zhang, D. Yamasaki, Y. Nakamura, X. Sui and L. Zheng, Levels, indoor–outdoor relationships and exposure risks of airborne particle-associated perchlorate and chlorate in two urban areas in Eastern Asia, Chemosphere, 2015, 135, 31–37.
9. P. Brandhuber, S. Clark and K. Morley, A review of perchlorate occurrence in public drinking water systems, J.–Am. Water Works Assoc., 2009, 101, 63–73.
10. Z. Gan, H. Sun, R. Wang and Y. Deng, Occurrence and exposure evaluation of perchlorate in outdoor dust and soil in mainland China, Sci. Total Environ., 2014, 470, 99–106.
11. R. E. Tarone, L. Lipworth and J. K. McLaughlin, The epidemiology of environmental perchlorate exposure and thyroid function: a comprehensive review, J. Occup. Environ. Med., 2010, 52(6), 653–660.
12. Y. Wan, Q. Wu, K. O. Abualnaja, A. G. Asimakopoulos, A. Covaci, B. Gevao, B. Johnson-Restrepo, T. A. Kumosani, G. Malarvannan and H.-B. Moon, Occurrence of perchlorate in indoor dust from the United States and eleven other countries: implications for human exposure, Environ. Int., 2015, 75, 166–171.
13. Z. Gan, L. Pi, Y. Li, W. Hu, S. Su, X. Qin, S. Ding and H. Sun, Occurrence and exposure evaluation of perchlorate in indoor dust and diverse food from Chengdu, China, Sci. Total Environ., 2015, 536, 288–294.
14. J. Wolff, Perchlorate and the thyroid gland, Pharmacol. Rev., 1998, 50(1), 89–106.
15. E. T. Urbansky, Perchlorate chemistry: implications for analysis and remediation, Biorem. J., 1998, 2(2), 81–95.
16. P. K. Dasgupta, P. K. Martinelango, W. A. Jackson, T. A. Anderson, K. Tian, R. W. Tock and S. Rajagopalan, The origin of naturally occurring perchlorate: the role of atmospheric processes, Environ. Sci. Technol., 2005, 39(6), 1569–1575.
17. E. Urbansky, S. Brown, M. Magnuson and C. Kelty, Perchlorate levels in samples of sodium nitrate fertilizer derived from Chilean caliche, Environ. Pollut., 2001, 112(3), 299–302.
18. T. Zhang, X. Chen, D. Wang, R. Li, Y. Ma, W. Mo, H. Sun and K. Kannan, Perchlorate in Indoor Dust and Human Urine in China: Contribution of Indoor Dust to Total Daily Intake, Environ. Sci. Technol., 2015, 49(4), 2443–2450.
19. A. J. Vella, C. Chircop, T. Micallef and C. Pace, Perchlorate in dust fall and indoor dust in Malta: an effect of fireworks, Sci. Total Environ., 2015, 521, 46–51.
20. US EPA, Procedures for laboratory analysis of surface/bulk dust loading samples, Appendix C.2, http://www.epa.gov/ttn/chief/ap42/appendix/app-c2.pdf.
21. US EPA, Procedures for sampling surface/bulk dust loading, Appendix C.1, http://www.epa.gov/ttn/chief/ap42/appendix/app-c1.pdf.
22. R. L. Folk, Petrology of Sedimentary Rocks, Hemphill Publishing Company, Austin, TX, 1974.
23. US EPA, Risk Assessment Guidance for Superfund. Volume I: Human Health Evaluation Manual (Part A), http://www.epa.gov/oswer/riskassessment/ragsa/.
24. US EPA, Guidelines for exposure assessment, http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=15263.
25. R. Khanal, H. Furumai and F. Nakajima, Toxicity assessment of size-fractionated urban road dust using ostracod Heterocypris incongruens direct contact test, J. Hazard. Mater., 2014, 264, 53–64.
26. F. Amato, M. Pandolfi, A. Alastuey, A. Lozano, J. C. González and X. Querol, Impact of traffic intensity and pavement aggregate size on road dust particles loading, Atmos. Environ., 2013, 77, 711–717.
27. M. A. Laidlaw and G. M. Filippelli, Resuspension of urban soils as a persistent source of lead poisoning in children: a review and new directions, Appl. Geochem., 2008, 23(8), 2021–2039.
28. US EPA, Perchlorate and perchlorate salts, http://www.epa.gov/iris/subst/1007.htm.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5em00435g

---