Road traffic, especially in urban and congested areas, will be one of the most important sources for particulate matter (PM) in the future. Until today there is a lack of coarse PM measurements and its contribution to the PM₁₀ particle load. There are only few papers dealing with physical properties and chemical characteristics of atmospheric particles derived from road traffic environments. Quantification of the mass concentration of non exhaust particles and its attribution to specific sources is still a difficult task. In the last years European environmental regulations have been enforced in order to reduce exhaust particle emissions from road traffic. The most important achievement of these policies is a reduction of particle emissions from diesel engines. Only little attention, however, was paid to reducing non exhaust particle emissions and therefore, an efficient reduction for the non exhaust particles could not been realized in the same way until now. These non exhaust particles are mostly due to abrasion and are mainly derived from brakes, tyres, and road surface, but also comprise re-suspended material accumulated on the road surface.

Der Fachvortrag zur Veranstaltung ist im Volltext verfügbar. Das PDF enthält alle Bilder und Formeln.

Introduction

Road traffic, especially in urban and congested areas, will be one of the most important sources for particulate matter (PM) in the future. Until today there is a lack of coarse PM measurements and its contribution to the PM₁₀ particle load. There are only few papers dealing with physical properties and chemical characteristics of atmospheric particles derived from road traffic environments. Quantification of the mass concentration of non exhaust particles and its attribution to specific sources is still a difficult task. In the last years European environmental regulations have been enforced in order to reduce exhaust particle emissions from road traffic. The most important achievement of these policies is a reduction of particle emissions from diesel engines. Only little attention, however, was paid to reducing non exhaust particle emissions and therefore, an efficient reduction for the non exhaust particles could not been realized in the same way until now. These non exhaust particles are mostly due to abrasion and are mainly derived from brakes, tyres, and road surface, but also comprise re-suspended material accumulated on the road surface.

At present, there are no EU regulations specifically designed to control and/or limit the non exhaust PM emissions from road traffic, except for some European countries, which have forbidden the use of studded tyres. The fact that some established vehicle and road-surface noise regulations (e.g. “Flüsterasphalt”) will also lead to a reduction of non exhaust particle emissions is still largely unknown. The latest studies predict that in central Europe the contribution of non exhaust PM to total traffic emissions will increase to 80-90% by the end of this decade (Fig. 1) In combination with projected future climate scenarios (e.g. global warming, climate change with an expected increase in heat waves with droughts) and with the assumption of no additional reduction in particle emissions, the Intergovernmental Panel on Climate Change (IPCC) predicts a general decrease in air quality in the future (IPPC, 2007). The estimated particle diameters of the non exhaust PM emissions are larger than exhaust emissions (Thorpe, A. et al., 2008). However, with respect to the size range this relatively coarse PM is still able to enter the respiratory system, where it can impart negative impacts on human health (Brunekreef & Forsberg, 2005). Therefore, more information with new measurement strategies is required to monitor and quantify the additional particle load by non exhaust particles, so that new air pollution abatement strategies can be developed.

Background

Ambient airborne PM covers a wide range of particle sizes, ranging from particle diameters in a range from a few nanometers to approx. 100 micrometers (Grobéty et al., 2010). For operational air pollution monitoring and particle reduction policies, there are at this time two particle metrics definitions that are commonly applied: the so-called PM₁₀ and PM_2.5 particle size fractions. Airborne particles with an aerodynamic diameter of less than 10 μm belong to the PM₁₀, whereas PM_2.5 refers to particles smaller than 2.5 μm, which is also called the fine particle fraction. The coarse fraction of PM₁₀ comprises particles with diameters in the size range of 2.5 μm to 10 μm (AQEG, 2005; USEPA, 2005). Therefore we have to consider that operational PM₁₀ particle measurements consist of a "mixture" of fine and coarse particles, which complicates the unequivocal identification of possible air pollution sources and makes it sometimes nearly impossible. The need for a separated measurement and monitoring program (PM₁₀ and PM_2.5) is described more in detail by Kaminski et al. 2013.

According to the formation processes, airborne PM generated by road traffic is categorised in three main categories (Fig. 2): The nucleation mode, the accumulation mode (secondary particles) and the coarse particle mode (primary particles). Particles emitted directly from combustion sources belong to the nucleation mode. These particles have a diameter smaller than approx. 50 nm. Because these particles are quite small, their proportion in the PM₁₀ mass concentration is very low, but they could be present in large numbers. The residence time in the atmosphere can be a few hours, and therefore they could be transformed by coalescence and condensation processes into larger accumulation-mode particles. Particles in the size range between around 50 nm and 2.5 μm belong to the accumulation mode and could have a significant contribution of the PM₁₀ mass concentration, with a residence time in the atmosphere up to tens of days, depending on the meteorological conditions. Particles with a size > 2.5 μm belong to the coarse particle mode. This particle size fraction includes wind-blown crustal matter and PM generated during abrasion processes. Coarse particles have significantly shorter residence times in the atmosphere in comparison to accumulation-mode particles, but could contribute substantially to the PM₁₀ mass concentration.

Within the context of our study, it is important to note that combustionderived particles are formed by chemical conversion of gases via condensation, nucleation and coagulation (e.g. in chain aggregates of diesel soot) and thus, they are fine particles, which belong mainly to the category of secondary particles. In contrast, the primary particle fraction (coarse particles) with a number of different formation and release mechanisms into the atmosphere originate mainly from mechanical abrasion and corrosion, which leads to a substantial contribution to the PM₁₀ mass concentrations at traffic-related sampling sites and is significantly influenced by meteorological conditions (e.g. wind, precipitation and temperature).  

Particle sampling and subsequent analysis

According to the needs for a more detailed study of coarse particles, DWD and BASt started in 2013 in cooperation with the Universities of Strasbourg and Freiburg a long-term monitoring program for particle sampling and analysis at highly frequented motorways in Germany. One of the goals of the program is the measuring and monitoring of PM within the framework of the prediction that the contribution of non exhaust PM to total traffic emissions will increase to 80-90% in central Europe by the end of this decade (Rexeis et al., 2009). Furthermore the monitoring program is focused on a possible impact of projected future climate scenarios (keyword: climate change), with an expected increase of heat waves in combination with drought periods. Such conditions could have an additional effect, namely an increase of the PM contribution from road traffic.

A variety of sampler technologies exists for ambient aerosol measurement to collect PM with different dimensions, aerodynamic performance and sample efficiency (e.g. high volume and low volume). The different samplers can be assigned to two main categories: active and passive. Active samplers require an electrically driven pumping device to generate an airflow through the sampling head (air intake). In contrast, the passive samplers depend on the ambient wind field (atmospheric turbulence). Passive samplers are cost effective, able to be operated without any power supply and are mainly free from the need for maintenance. In the last years this technology received increased attention and is therefore recommended for operational field measurements, e.g. within the framework of long-term monitoring programs. In a first approach, a twelvemonth pilot study, the passive sampler Sigma-2 (VDI-Guideline 2119, 2013) has been installed at two different motorway sections in North Rhine Westphalia. One sampling location was set up at the motorway BAB A 555 close to Cologne (approx. 17 km south of Cologne). The second one is located at the BAB A 4, close to the motorway intersection Köln-Gremberg, about 6 km east of Cologne.

The passive sampling method Sigma-2 collects coarse particles on an acceptor surface, which is suitable for optical single-particle analysis (TLM=transmitted-light microscope) as well as for single-particle analysis via scanning electron microscopy (SEM). The TLM analysis allows for the determination of a number settling rate of the “total” atmospheric particle load (d_p 2.5-80 µm, geometric equivalent diameter), used for the calculation of the total ambient aerosol mass concentration (VDI 2119, 2013). Furthermore it allows for the differentiation by particle type (e.g. opaque, transparent) and particle size (d_p 2.5-10 µm, d_p 10-80 µm geometric equivalent diameter), used for the calculation of the size-fractionated mass concentration of these particles (Dietze et al., 2006). Results from this method provide information on the distribution ratio of different particle sizes over a given period and sampling site and permit distinction between and first estimation of anthropogenic and natural particles in the sampling area. Furthermore, samples of special interest can be selected for subsequent chemical and mineralogical analysis by SEM.

First results

The TLM analysis of atmospheric particles > 2.5 µm distinguishes transparent particles of natural origin (e.g. mineral dust, spores, pollen) and carboncontaining opaque particles (e.g. coarse combustion-derived particulates), as well as typical abrasion particles of anthropogenic origin (e.g. tire wear, metal fragments from brake abrasion processes). Object-specific features of the individual particles, such as projected area and optical density, were analysed in the 2.5 - 80 μm size range with the aid of an automated computer controlled image processing system.

Figures 3 and 4 show the average size distribution over one year (June 2013 – May 2014) of the two different sampling sites from the motorway sections BAB A 555 and BAB A 4. The maximum of the total particle mass concentrations of both sampling sites is located in the size interval from 10 to 20 µm, and has values of 7.2 µg/m³ for BAB A 4 and 5.2 µg/m3 for BAB A 555. For the opaque particle fraction we find the maximum of particle mass concentration for both sampling sites in the same size interval, but it is more prominent. The concentrations for the one-year mean values are 1.9 µg/m³ (BAB A 555) and 2.3 µg/m³ (BAB A 4). The concentration maximum for opaque particles, which in this size fraction are considered to be dominated mainly by abrasion processes (tire wear), is more significant in contrast to the total particle size maximum. It is noticeable that the maximum of coarse opaque particles at these sampling locations heavily influenced by road traffic is typically between 10 and 20 µm. The microscopic analysis shows for these abrasion particles a typical shape of compact rolls, which are morphologically significantly different to exhaust combustion residues (e.g. coarse soot and ash).

At both sampling sites, the size distributions of the transparent particles (e.g. mineral and biogenic particles excluding pollen) differ significantly from the distribution of the opaque (e.g. combustion residues, tyre abrasion) particles. Surprisingly, the shape of the size distributions for the transparent particles has a right-skewed tendency. We will find the highest particle mass concentrations in the small particle sizes, which means that the concentration is decreasing with increasing diameter. This is in contrast to previous passive sampler measurements and single-particle analysis performed by our group at other roadside locations. However, it should be pointed out, that the current results are the first acquired data along highly frequented motorways and represent only a sampling period of one year.

Summary and outlook

Two main sampling sites at highly frequented motorways (BAB A 555 and BAB A 4) have been set up in Germany with Sigma-2 passive sampler devices for subsequent automated optical single-particle analysis and calculation of ambient aerosol mass concentration for particles in the size from 2.5-80 µm. This one-year pilot study revealed for the BAB A 555 location a particle mass concentration that is on average approx. 25% lower than that at the BAB A 4 sampling site. The opaque particle fraction shows for both sampling sites the maximum concentration of PM in the 10 to 20 µm size interval. This is typical for sampling locations under the direct influence of road traffic and is in good agreement with passive sampler measurements from the past at different roadside locations. However, the transparent particle fraction at both motorways exhibits a right-skewed size distribution, which is unusual for a roadside station.

Further analysis of the data, in combination with meteorological parameters (e.g. temperature, wind, precipitation) and the time of the year (spring, summer, autumn, winter), is ongoing. Additionally, some samples of interest have been selected for a detailed chemical and mineralogical single-particle analysis by SEM with energy-dispersive X-ray (EDX) spectroscopy.

On a longer term, our study will focus on, validate and, if possible, quantify the possible increase in non exhaust emissions, which we anticipate as a result of climate change.

References

Brunekreef, B., Forsberg, B. Review – Epidemiological evidence of effects of coarse airborne particles on health. Eur Respir J 2005; 2: 309-318; DOI: 10.1183/09031936.05.00001805

Dietze, V., Fricker, M., Goltzsche, M., Schultz, E. Air quality measurement in German health resorts – Part 1: Methodology and verification, Gefahrstoffe – Reinhaltung der Luft. Air Quality Control. 66(1):45–53 (2006).

Grobéty, B., Gieré, R., Dietze, V., Stille, P. Airborne Particles in the Urban Environment. Elements, VOL. 6 (2010), PP. 229–234, August 2010 (ISSN 1811 - 5209).

IPPC 2007a. Summary for policymakers. The physical science basis. Working Group I contribution to the Intergovernmental Panel on Climate Change Fourth Assessment Cambridge, U.K.: Cambridge University Press (2007).

Kaminski, U., Fricker, M., Dietze, V. The PM 2.5 Fine Particle Background Network of the German Meteorological Service-First Results. Meteorol Zeitschrift Vol. 22 No. 2 (2013).

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.

Thorpe, A., Harrison, R.M. Sources and properties of non exhaust particulate matter from road traffic: A review. Sci. Tot. Environ. 2008, 400, 270-282.

VDI 2119. Ambient air measurements. Sampling of atmospheric particles > 2,5 µm on an acceptor surface using the Sigma-2 passive sampler. Characterization by optical microscopy and calculation of number settling rate and mass concentration. ICS: 13.040.01. Beuth Verlag, Berlin (2013).

Fig. 1 The contribution of non exhaust particulate matter (e.g. brake dust, tire wear, as well as suspension and re-suspension from the road surface) to total traffic emissions is predicted to increase in central Europe to 80 – 90% until the end of this decade (after Rexeis & Hausberger (2009): Atmospheric Environment 43).

Fig. 2 Idealized schematic of atmospheric aerosol size distributions including sources, sinks, particle modes and particle formation as a function of particle diameter (from Whitby and Cantrell 1976).

Fig. 3 Size distribution (d_p 2.5 – 80 µm) of the average particle mass concentration (opaque, transparent and total particle fraction excluding pollen) at the BAB A 555, 17 km south of Cologne (31.05.2013 – 30.05.2014).

Fig. 4 Size distribution (d_p 2.5 – 80 µm) of the average particle mass concentration (opaque, transparent and total particle fraction excluding pollen) at the BAB A 4, close to motorway intersection Köln-Gremberg, 6 km east of Cologne (31.05.2013 – 30.05.2014).