Characterization and estimation of health risks of PM10 road dust
Why do we care about non-exhaust emissions?
PM emissions from road vehicles include emissions from the tailpipe (exhaust emissions including fuel and lubricant combustion and fuel additives) and emissions due to wear and tear of vehicle parts such as brake, tyre and clutch and re-suspension of dust (non-exhaust emissions). Most of the research as well as policy action in the last few decades has largely focused on characterization and control of exhaust emissions, and a combination of stringent regulations and technological upgrades have resulted in a decline of the percentage contribution of vehicle tailpipe emissions to total ambient PM. As a result, the contribution of non-exhaust PM is becoming more important.
Non-exhaust PM comprises the various emissions that do not derive from the tailpipe of a vehicle including particles generated due to brake and tyre wear, road surface abrasion, wear and tear/corrosion of other vehicle components such as the clutch, and re-suspension of road surface dust. These particles often act as carriers of heavy metals, polycyclic aromatic hydrocarbons (PAHs), and thus, pose a risk to human health.
However, analysis of non-exhaust emissions is not very straightforward since emissions due to wear and tear of parts of vehicles (e.g. brakes, tyres, clutches) can deposit on the road surface, only to be re-suspended subsequently when vehicles are driven. As a result, road dust is often a complex mixture of coarse-sized particles derived from different sources such as wear and tear of vehicle components (brakes, tyres and clutches) and road surface, engine corrosion, tailpipe emissions, crustal dust and other emission sources.
What did we do?
Filter (PTFE) samples were collected next to a heavy traffic road in Birmingham (Bristol Road) using a road dust sampler (Amato et al., 2009) as well as tyre, brake pad and road dust and soil samples.
The filter samples were analyzed for aluminium (Al), silicon (Si), iron (Fe) and sulphur (S) using wavelength dispersive x-ray fluorescence (WD-XRF). After this, the filter samples were extracted using a reverse aqua regia solution (Allen et al., 2001) and analysed for copper (Cu), zinc (Zn), lead (Pb), barium (Ba), tin (Sn), antimony (Sb), calcium (Ca), vanadium (V), titanium (Ti), chromium (Cr) and manganese (Mn) using inductively coupled plasma mass spectrometry (ICP-MS).
The bulk soil (uncontaminated area) and road dust (Bristol Road) samples were dried, ground, sieved (2 mm sieve) and subsequently extracted using the reverse aqua regia method and analysed for the same subset of elements. Brake pad and tyre samples were frozen using liquid nitrogen (N2) and subsequently ground. The samples were passed through a 2 mm sieve and extracted and analyzed using WD-XRF.
What did we find?
- Mass loading for PM10 road dust was ~ 10 mg/m2
- Bulk road dust was correlated with PM10 road dust
- Very high concentrations of several elements in road dust including
- Brake pad dust was found to be rich in Fe, Ba and Ca and in case of tyre, Zn was the most abundant element.
- Oxidative Potential (OP) was higher for the bulk road dust compared to the PM10 road dust. However, while the OP was only found to be significantly associated with copper (Cu) in the bulk road dust, several elements were found to be associated with OP in case of PM10 road dust including copper, vanadium and sulphur among others. Copper, a marker for traffic emissions, was found to be associated with OP in both road dust fractions (bulk and PM10)
- More detailed analyses on brake and tyre composition
- Sampling across different site types (e.g. urban vs rural, heavy traffic vs. low traffic) to determine the variability of heavy metal concentrations
- Further tests for health risks using cellular assays (i.e. exposing human lung epithelial cells to the road dust fractions and brake and tyre wear particles to determine the effects)
I would like to thank the Simon Wolff Charitable Foundation and the University of Birmingham for the financial support for this study. Special thanks go to Barbara Macias Hernandez, Richard Johnson, Jamie Peart and Massimiliano Mascelloni from the University of Birmingham (UK) for assistance during sampling. Assistance from Dr. Stephen J. Baker and Dr. Jackie Deans, University of Birmingham is gratefully acknowledged.