Article pubs.acs.org/est
Estimation of the Contributions of Brake Dust, Tire Wear, and Resuspension to Nonexhaust Traffic Particles Derived from Atmospheric Measurements Roy M. Harrison,*,†,§ Alan M. Jones,† Johanna Gietl,† Jianxin Yin,† and David C. Green‡ †
National Centre for Atmospheric Science, Division of Environmental Health & Risk Management, School of Geography, Earth & Environmental Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, United Kingdom ‡ Environmental Research Group, King’s College London, Franklin-Wilkins Building, 150 Stamford Street, London, SE1 9NH, United Kingdom S Supporting Information *
ABSTRACT: Size-fractionated samples of airborne particulate matter have been collected in a number of campaigns at Marylebone Road, London and simultaneously at background sites either in Regents Park or at North Kensington. Analysis of these samples has enabled size distributions of total mass and of a number of elements to be determined, and roadside increments attributable to nonexhaust emissions arising from traffic activity have been calculated. Taking a novel approach, the combined use of size distribution information and tracer elements has allowed the separate estimation of the contributions of brake dust, tire dust, and resuspension to particle mass in the range 0.9−11.5 μm aerodynamic diameter and mean contributions (±s.e.) at the Marylebone Road sampling site are estimated as resuspended dust 38.1 ± 9.7%, brake dust 55.3 ± 7.0%, and tire dust 10.7 ± 2.3%, (accounting for a total of 104.1% of coarse particle mass in the traffic increment above background).
1. INTRODUCTION During recent years environmental regulations have resulted in substantial reductions in the exhaust emissions from road traffic. These reductions in combustion products have not been accompanied by similar reductions in nonexhaust emissions, i.e. the abrasive emissions from brake, road, and tire wear, and the resuspension of materials from the highway surface, which, as a result, make up a similar proportion of the airborne particulate matter (PM) resulting from vehicle use as exhaust emissions.1,2 Rexeis and Hausberger3 predict that in central Europe, the contribution of nonexhaust PM to total traffic emissions will increase to 80−90% by the end of this decade. While the aerodynamic diameters of nonexhaust particulate emissions tend to be larger than those of exhaust emissions,4 they are still within the size range that may enter the respiratory system where they may lead to adverse effects upon health.5 Many studies have sought to distinguish between the different nonexhaust particulate emissions by identifying particular chemical components.4 While the constituents of brake material may vary among manufacturers and over time, iron, copper, antimony, and barium have been associated with the particulate matter released from brake operation.6 Road surfaces are generally composed of either concrete or aggregate with a bituminous binder and abrasion of such a surface is likely to result in particulate matter of mineral origin.7 Tire wear is © 2012 American Chemical Society
likely to result in predominantly carbonaceous particles, although small quantities of metals, in particular zinc which is used as a vulcanization activator, may be present.8 Material resuspended from the road surface may include all types of vehicle abrasion debris, in addition to material from nonroad sources which has been deposited on the road surface. This may include mineral dust from the local environment, typically including silicon, aluminum, calcium, and iron9 particularly in arid locations. Within the United Kingdom, winter maintenance procedures involve spreading deicing salt on roadways, but the practices that are common in some northern countries of sanding roadways10 and fitting studded tires11 are not adopted. Particulate matter is present in a range of sizes in the atmosphere. Combustion processes generally result in smaller particles, which over time will agglomerate into the accumulation mode. Abrasive processes result in particles with aerodynamic diameters larger than the accumulation mode which are lost from the atmosphere largely by sedimentation. Resuspension of material from the land or road surface by high wind speeds tends to result in particulate matter with Received: Revised: Accepted: Published: 6523
March 6, 2012 May 25, 2012 May 29, 2012 May 29, 2012 dx.doi.org/10.1021/es300894r | Environ. Sci. Technol. 2012, 46, 6523−6529
Environmental Science & Technology
Article
Table 1. Campaign Mean Concentrations and Curbside to Background Ratios for the Main Analytes background site mass Al Ca Si Zn Fe Cu Sb Ba V Ni Ti Cr
μg ng ng ng ng ng ng ng ng ng ng ng ng
−3
m m−3 m−3 m−3 m−3 m−3 m−3 m−3 m−3 m−3 m−3 m−3 m−3
curbside site
2007
2009
2010
2011
22.2
15.8 39.6 163.2
21.3 238.2
36.6 303.8
20.9 276.6 9.4 1.3 4.5 2.1 1.6 2.1 2.1
14.0 190.4 410.6 796.4 36.2 342.2 11.6 1.9 6.1
7.2 3.6
2007
2009
2010
2011
ratio
28.8
22.3 56.0 235.5
29.3 379.6
57.7 1418.0 42.6 5.2 20.7 3.3
23.0 1127.5 45.2 6.1 22.7 1.9 2.1 3.4 4.2
17.1 221.7 494.5 1020.4 37.7 1453.9 46.5 8.1 30.9
1.3 1.4 1.3 1. 6 1.3 4.4 4.3 4.1 4.2 0.9 1.3 1.6 2.0
804.9 399.9
1484.1 1788.0
Simultaneous background samples of particulate matter were collected within an urban park at Regents College, approximately 0.4 km north of Marylebone Road, in 2007. During the subsequent years approximately coterminous background samples were collected at the AURN urban background site at North Kensington. This site is located on school premises on the western side of a residential road some 4.1 km to the west of Marylebone Road. Seven samples were collected at both sites in each campaign, with the exception of the 2011 campaign during which one background sample was lost. The 2007 and 2011 campaigns were conducted in February and March, while the 2009 and 2010 campaigns took place in May and June. During the 2011 campaign, trials were being conducted of dust suppression by application of calcium magnesium acetate (CMA), which may have influenced results for this campaign. 2.2. Sampling and Analysis. The size-segregated particulate matter was collected by Micro-Orifice Uniform Deposit Impactor (MOUDI) model 100 (MSP Corporation) sampling through a 2-m vertical stainless steel tube with a rain cover. These instruments were operated to sample particles in 10 size fractions of typically >21.4, 11.8−21.4, 7.4−11.8, 3.7−7.4, 2.2− 3.7, 1.2−2.2, 0.66−1.2, 0.39−0.66, 0.21−0.39, and 0.83). In each case the gradients of the regression lines were significantly different from zero while the intercepts were not. The plots of curbside versus background concentrations of aluminum and iron show distinct differences between the two metals. Whereas, with a few exceptions curbside aluminum concentrations are only slightly greater (median difference = 38 ng m−3) than the background concentrations (Figure 1a), curbside iron concentrations are consistently at least twice the background concentration (Figure 1b) and are generally more scattered. 4.1. Effect of Wind Direction. Previous work on vehicle exhaust emissions27 has shown higher concentrations at the Marylebone Road site when the wind direction is aligned with the highway axis and from the crossing and junction to the west, and when the direction is normal to the axis and rotating air in the street canyon brings material from the highway to the measurement position. Lower concentrations were found when the wind was normal to the highway axis and the rotating airflow in the street canyon brought clean air from above to the measurement position, and when the flow was aligned with the highway and from the opposite direction to the junction. As the highway is aligned on an approximately 260° to 80° axis the number of hours in each sample during which the wind was from four equal angled sectors (centered on directions normal to, and aligned with, the highway) was derived by classifying the hourly measured wind directions at Heathrow as shown in Table S2. The necessity to sample for several days in order to achieve measurable quantities of the less abundant elements means that there can be considerable variation in wind direction during the course of a sample. The data were, however, divided into three groups by wind directionsamples with >80% N and E hourly wind directions, samples with >80% S and W hourly wind directions, and other samplesand these are plotted separately for iron in Figure S2. It is evident that relatively higher curbside concentrations of iron are associated with winds from the south and westthe directions of traffic pollutant dominance found in the study of vehicle exhaust, while lower relative concentrations are associated with winds from the north and east, confirming that the major source of iron at the curbside site is associated with traffic. The mean spectra for iron and aluminum (taken to be representative of brake and nonbrake metals) were calculated for samples when >80% hourly wind directions were from the south and west, when >80% hourly wind directions were from the north and east, and for other periods, and are shown in Figure 3. The spectra for iron show curbside concentrations significantly lower when the wind was from the north and east compared to “other” directions, and significantly higher when from the south and west compared to “other” wind directions. The aluminum spectra show a less significant difference between the curbside concentrations at different wind directions, but still show an enhancement associated with S and W wind directions. From this figure it is clear that while both iron and aluminum share a mode at 3 μm, aluminum (and siliconnot shown) also have a mode at a larger diameter which appears as a shoulder. 4.2. Estimation of Tire Dust Concentration. Zinc has often been taken as a tracer of tire rubber emissions although it does have other sources related to traffic.4 The mean size distribution of zinc at Marylebone Road and the background sites appears in Figure 4, and at Marylebone Road is clearly
Figure 3. Influence of predominant wind direction upon normalized size distributions at curbside and background sites for (a) aluminum and (b) iron (wind 80% N + E; S + W, dm/dlog(Dp) spectra).
Figure 4. Mean size distribution of zinc at Marylebone Road and the background sites (all data dm/dlog(Dp) spctra).
trimodal. The submicrometer mode, which appears at very similar concentration at the background sites, is most probably from regional transport of aerosol, and consistent with a contribution of 47% of UK atmospheric emissions of zinc arising from metal production.28 The other modes are, however, strongly elevated at Marylebone Road and the inventory attributes 33% of zinc emissions to automobile tire and brake wear.28 It is, however, unclear as to whether the two supermicrometer modes relate separately to these two sources or whether zinc contamination of road dust may also contribute. 6527
dx.doi.org/10.1021/es300894r | Environ. Sci. Technol. 2012, 46, 6523−6529
Environmental Science & Technology
Article
empirical figure determined in the UK which may not be applicable elsewhere due to the wide variations in brake pad composition. It has been assumed that the main component of resuspended particles is soil and other crustal material. This assumption is supported by the very close similarity of the size distributions of silicon and aluminum and the ratio of the two which is very close to the ratio in average crustal material of 3.4.32 Using the fact that silicon represents 27.7% of average earth crustal material, the silicon concentrations have been scaled by a factor of 3.6 in order to represent the mass of resuspended particles. Tires, brake pads, and road surface wear particles may also contribute silicon, so this may lead to an overestimate of the road dust contributions, while dilution with other noncrustal materials would have the opposite effect. A better estimate could be obtained by sampling the PM10 fraction of road dust, but such data were not available to this study. Zinc is taken as a tracer of tire dust. If it is assumed that the zinc content of average brake pads and tires is similar in the UK to that in the U.S. at the time of the work conducted by Schauer et al.,29 then it may be assumed that tire wear and brake wear make a broadly similar contribution to emissions from road traffic. Road dust also makes a small contribution (see above). Kreider et al. 33 measured the chemical composition of particles abraded from a tire, reporting a zinc content of 0.9%. As a simplification, and taking account of other literature data,4,8 we have assumed that 50% of the measured zinc in the supermicrometer particles arises from tire wear and have assumed a zinc content of tire rubber of 1%, hence using a factor of 50. Figure 5 shows the separate estimates of the three components as well as their combined mass compared with the measured mass increment at Marylebone Road showing a surprisingly good agreement. Based upon these data, the combined reconstructed mass accounts for 104.1% of the measured mass in the size range 0.9−11.5 μm and within this, the resuspended dust contribution is 38.1 ± 9.7%, brake dust is 55.3 ± 7.0%, and tire dust is 10.7 ± 2.3% (mean ± s.e.). Overall, the nonexhaust emissions appear to account for all of the >0.9 μm particles and hence 77% of the road traffic increment. This is higher than earlier estimates based upon a less full appreciation of the emissions,1,2,18 but refers only to the material which is not volatile in the MOUDI. The combination of particle size distributions and chemical tracers, even though these are often contributed by more than one source, does appear to allow a realistic reconstruction of the mass of coarse mode particles contributed by nonexhaust traffic sources. Quantitative estimates have been arrived at for the contribution of brake dust, tire dust, and resuspended particles but it should be recognized that not only are these site specific, they are also subject to considerable uncertainties which have yet to be quantified. It would be expected that in southern Europe and other hotter, drier parts of the world, the contribution of resuspended road dust would be substantially greater. On the other hand, in Scandinavia, the use of studded tires and road sanding is common practice in the winter months leading once again to a quite different pattern in contributions to nonexhaust traffic-related particulate matter in the atmosphere.
Schauer et al.29 have carried out measurements in road tunnels along with direct analysis of brake dust, tire dust, and exhaust emissions to identify sources of metals emitted from motor vehicles. They found that emissions of zinc arose from all three sources, i.e., brake wear, tire wear, and exhaust emissions, and used a Chemical Mass Balance Model to resolve the respective contributions. This was made more difficult by the fact that gasoline tailpipe and tire wear emissions were strongly colinear and consequently could not be separated by the mass balance model. However, since the abundances of zinc in gasoline and diesel tailpipe emissions were similar and the CMB model attributed only a very small proportion of zinc (0− 11%) to diesel tailpipe emissions, it seems likely that the vast majority of the emissions of zinc attributed to combined gasoline tailpipe and tire wear are actually due to tire wear. This factor accounted for 20−69% of total zinc emissions while brake wear accounted for 17−56%, and resuspended road dust accounted for 7−21%. According to European data (from Spain and Switzerland) derived by Amato et al.30 from PMF analysis, tire wear contains about 1% Zn, and road dust in the range 1572−2183 μg g−1, which is insufficient for the road dust estimated on the basis of the silicon content to appreciably influence airborne zinc at our site. Amato et al.30 attributed zero zinc to either motor exhaust or road wear. Exhaust emissions of zinc are in the submicrometer size range31 and therefore negligible at the Marylebone Road site given the insignificant increment in this size range above the background site at North Kensington. 4.3. Accounting for the Mass of Coarse Particle Emissions. We have postulated that three sources are the predominant contributors to the supermicrometer particle increment seen at the Marylebone Road site, i.e., brake dust, tire dust, and particle resuspension. A fourth source of road surface abrasion may also contribute but we have no unique tracer available for this source. Carbonaceous particles from the exhaust are thought to be largely limited to the submicrometer fraction.19 Figure 5 shows an attempt to account for the traffic increment (i.e., the difference between concentrations at Marylebone Road and the background site) by the three sources above. In the case of brake dust, the barium concentrations have been scaled by a factor of 91 in response to the fact that Gietl et al.6 estimated that barium represented about 1.1% of the mass of brake dust particles. This is an
■
ASSOCIATED CONTENT
S Supporting Information *
Figure 5. Reconstruction of supermicrometer particle mass through scaling of tracer elements. Error bars are standard errors of the means (all available data).
Further details; Table S1: Locations and types of sites; Table S2: Selection of wind sectors; Figure SI: Relationship of iron 6528
dx.doi.org/10.1021/es300894r | Environ. Sci. Technol. 2012, 46, 6523−6529
Environmental Science & Technology
Article
(13) Hays, M. D.; Cho, S.-H.; Baldauf, R.; Schauer, J. J.; Shafer, M. Particle size distributions of metal and non-metal elements in an urban near-highway environment. Atmos. Environ. 2011, 45, 925−934. (14) Ham, W. A.; Ruehl, C. R.; Kleeman, M. J. Seasonal variation of airborne particle deposition efficiency in the human respiratory system. Aerosol Sci. Technol. 2011, 45, 795−804. (15) Salma, I.; Ocskay, R.; Raes, N.; Maenhaut, W. Fine structure of mass size distributions in an urban environment. Atmos. Environ. 2005, 39, 5363−5374. (16) Waheed, A.; Li, X.; Tan, M.; Bao, L.; Liu, J.; Zhang, Y.; Zhang, G.; Li, Y. Size distribution and sources of trace metals in ultrafine/fine/ coarse airborne particles in the atmosphere of Shanghai. Aerosol Sci. Technol. 2011, 45, 163−171. (17) Allen, A. G.; Nemitz, E.; Shi, J. P.; Harrison, R. M.; Greenwood, J. C. Size distributions of trace metal elements in the atmospheric aerosols in the United Kingdom. Atmos. Environ. 2001, 35, 4581− 4591. (18) Charron, A.; Harrison, R. M. Fine (PM2.5) and coarse (PM2.5−10) particulate matter on a heavily trafficked London highway: Sources and processes. Environ. Sci. Technol. 2005, 39, 7768−7776. (19) Harrison, R. M.; Beddows, D. C. S.; Dall’osto, M. PMF analysis of wide-range particle size spectra collected on a major highway. Environ. Sci. Technol. 2011, 45, 5522−5528. (20) Liu, Y.-J.; Harrison, R. M. Properties of coarse particles in the atmosphere of the United Kingdom. Atmos. Environ. 2011, 45, 3267− 3276. (21) Manning, A. J.; Nicholson, K. J.; Middleton, D. R.; Rafferty, S. C. Field study of wind and traffic to test a street canyon pollution model. Environ. Monit. Assess. 2000, 60, 283−313. (22) Arciszewska, C.; McClatchey, J. The importance of meteorological data for modelling air pollution using ADMS-Urban. Meteorol. Apps. 2001, 8, 343−350. (23) Ayers, G. P. Comment on the regression analysis of air quality data. Atmos. Environ. 2001, 35, 2423−2425. (24) Warton, D. I.; Wright, I. J.; Falster, D. S.; Westoby, M. Bivariate line-fitting methods for allometry. Biol. Rev. 2006, 81, 259−291. (25) Jones, A. M.; Harrison, R. M. Interpretation of particulate elemental and organic carbon concentrations at rural, urban and kerbside sites. Atmos. Environ. 2005, 39, 7114−7126. (26) Birmili, W.; Allen, A. G.; Bary, F.; Harrison, R. M. Trace metal concentrations and water solubility in size-fractionated atmospheric particles and influence of road traffic. Environ. Sci. Technol. 2006, 40, 1144−1153. (27) Harrison, R. M.; Jones, A. M. Multisite study of particle number concentrations in urban air. Environ. Sci. Technol. 2005, 39, 6063− 6070. (28) NAEI. UK Emissions of Air Pollutants 1970 to 2008; 2010. (29) Schauer, J. J.; Lough, G. C.; Shafer, M. M.; Christensen, W. F.; Arndt, M. F.; DeMinter, J. T.; Park, J.-S. Characterization of Metals Emitted from Motor Vehicles, Research Report; Health Effects Institute, 2006. (30) Amato, F.; Pandolfi, N.; Moreno, T.; Furger, M.; Pey, J.; Alastuey, A.; Bukowiecki, N.; Prevot, A. S. H.; Baltensperger, U; Querol, X. Sources and variability of inhalable road dust particles in three European cities. Atmos. Environ. 2011, 45, 6777−6787. (31) Fushimi, A.; Saitoh, K.; Fujitani, Y.; Hasegawa, S.; Takahashi, K.; Tanabe, K.; Kobayashi, S. Organic-rich nanoparticles (diameter: 10− 30 nm) in diesel exhaust: Fuel and oil contribution based on chemical composition. Atmos. Environ. 2011, 45, 6326−6336. (32) Asimov, I. The elementary composition of the Earth’s crust. J. Chem. Educ. 1954, 70−72. (33) Kreider, M. L.; Panko, J. M.; McAtee, B. L.; Sweet, L. I.; Finley, B. L. Physical and chemical characterization of tire-related particles: Comparison of particles generated using different methodologies. Sci. Total Environ. 2010, 408, 652−659.
concentrations to NOx at Marylebone Road; Figure S2: Influence of predominant wind direction upon concentrations of iron. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +44 121 414 3494; fax: +44 121 414 3708; e-mail: r.m.
[email protected]. Present Address §
Also at Department of Environmental Sciences/Center of Excellence in Environmental Studies, King Abdulaziz University, PO Box 80203, Jeddah, 21589, Saudi Arabia. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was funded by Defra and the National Centre for Atmospheric Science. Meteorological data was provided by the UK Meteorological Office via the British Atmospheric Data Centre. Road traffic data at Marylebone Road was provided by Transport for London.
■
REFERENCES
(1) Harrison, R. M.; Yin, J.; Mark, D.; Stedman, J.; Appleby, R. S.; Booker, J.; Moorcroft, S. Studies of the coarse particle (2.5 − 10 μm) component in UK urban atmospheres. Atmos. Environ. 2001, 35, 3667−3679. (2) Querol, X.; Alastuey, A.; Ruiz, C. R.; Artinano, B.; Hansson, H. C.; Harrison, R. M.; Buringh, E.; ten Brink, H. M.; Lutz, M.; Bruckmann, P.; Straehl, P.; Schneider, J. Speciation and origin of PM10 and PM2.5 in selected European cities. Atmos. Environ. 2004, 38, 6547− 6555. (3) Rexeis, M.; Hausberger, S. Trend of vehicle emission levels until 2020 − Prognosis based on current vehicle measurements and future emission legislation. Atmos. Environ. 2009, 43, 4689−4698. (4) Thorpe, A.; Harrison, R. M. Sources and properties of nonexhaust particulate matter from road traffic: A review. Sci. Total Environ. 2008, 400, 270−282. (5) Brunekreef, B.; Forsberg, B.. Epidemiological evidence of effects of coarse airborne particles on health. Eur. Respir. J. 2005, 26, 309− 318. (6) Gietl, J. K.; Lawrence, R.; Thorpe, A. J.; Harrison, R. M. Identification of brake wear particles and derivation of a quantitative tracer for brake dust at a major road. Atmos. Environ. 2010, 44, 141− 146. (7) Tervahattu, H.; Kupiainen, K. J.; Raisanen, M.; Makela, T.; Hillamo, R. Generation of urban road dust from anti-skid and asphalt concrete aggregates. J. Hazard. Mater. 2006, 132, 39−46. (8) Wik, A.; Dave, G. Occurrence and effects of tyre wear particles in the environment − A critical review and an initial risk assessment. Environ. Pollut. 2009, 157, 1−11. (9) Pakbin., P.; Ning, Z.; Shafer, M. M.; Schauer, J. J.; Sioutas, C. Seasonal and spatial coarse particle elemental concentrations in the Los Angeles area. Aerosol Sci. Technol. 2011, 45, 949−963. (10) Norman, M.; Johansson, C. Studies of some measures to reduce road dust emissions from paved roads in Scandinavia. Atmos. Environ. 2006, 40, 6154−6164. (11) Gustafsson, M.; Blomqvist, G.; Gudmundsson, A.; Dahl, A.; Jonsson, P.; Swietlicki, E. Factors influencing PM10 emissions from road pavement wear. Atmos. Environ. 2009, 43, 4699−4702. (12) Orza, J. A. G.; Cabello, M.; Lidon, V.; Martinez, J. Contribution of resuspension to particulate matter inmission levels in SE Spain. J. Arid Environ. 2011, 75, 545−554. 6529
dx.doi.org/10.1021/es300894r | Environ. Sci. Technol. 2012, 46, 6523−6529
