This is the author s version of a work that was submitted/accepted for publication in the following source:

This is the author’s version of a work that was submitted/accepted for

pub-lication in the following source:

Ayoko, Godwin,

Lim, Mckenzie, &

Morawska, Lidia

(2003) Elemental

Com-position of Airborne Particles in Traffic-Affected Brisbane Areas. In Winch,

B (Ed.) Proceedings of National Clean Air Conference, 23 - 27 November

2003, Newcastle City Hall, New South Wales.

This file was downloaded from:

http://eprints.qut.edu.au/24340/

Notice: Changes introduced as a result of publishing processes such as

copy-editing and formatting may not be reflected in this document. For a

definitive version of this work, please refer to the published source:

ELEMENTAL COMPOSITION OF AIRBORNE PARTICLES IN

BRISBANE

McKenzie C. H. Lim, Godwin A. Ayoko and Lidia Morawska

International Laboratory for Air Quality and Health School of Physical and Chemical Sciences

Queensland University of Technology 2 George Street

Brisbane QLD 4001

Summary

Three urban sites (Woolloongabba, ANZ stadium and QUT M block) were selected to study the elemental composition of Total Suspended Particles (TSP) and PM2.5 airborne particles. The results of the particles elemental concentrations were submitted for multi-criteria decision making methods, which showed that all sites were influenced by vehicular emission, local industrial activities or crustal matter. Woolloongabba was associated with elements from vehicle emission (Fe, Cd, Cu, and Zn) and local industrial activities (Co). ANZ was associated with crustal matter (Si) and QUT with elements from oil combustion (Zr, V and Mo), crustal matter (Al) and construction activities (Sr and Mn). ANZ is the least polluted of the sites. Keywords: urban airborne particles, elemental compositions, multivariate analysis,

source apportionment.

1. Introduction

According to the Australian Bureau of Statistics (2002), a total of 12.8 million motor vehicles were registered in Australia by the end of March 2002. Among the states of Australia, Queensland recorded the highest annual increase in the registered motor vehicles in the past five years. Due to the wide spread use of metallic compounds as oil additive and anti-wear agents, a significant proportion of airborne elements in urban areas are from traffic related emissions. Therefore, there is a growing concern about traffic-generated pollutants in general and airborne elements in particular. To address this, ongoing evaluations of air quality are being undertaken by different government agencies throughout Australia (Hinwood and Di Marco 2002).

Among the pollutants concerned, fine particles in the size range 0.1 to 2.5 µm (PM2.5) have been targeted recently because of their potential health effects (Weckwerth 2001) and ability to be transported over long distances (Wilson 1996). Airborne particles of the 0.01 to 100 µm of diameter size range hereafter referred to as Total Suspended Particles (TSP) are also of interest because of their organic and inorganic constituents.

Airborne particles carry metals with potential adverse health and ecotoxicological effects. Therefore, the toxic properties of particles are in part due to the biochemical activities of the metals and it is important to monitor the presence of airborne metals in the environment (Miller 1979). Verrall et. al. (1986), Chan et. al. (1997) as well as Thomas and Morawska (2002a) have carried out independent investigations of the elemental composition

of aerosols in Brisbane. A common feature of their findings is that crustal matter; industrial dust, marine aerosol and vehicular emissions are the main contributors of pollutants to ambient air in Brisbane. However, no elemental composition analysis of samples from the vicinity of ANZ stadium or Woolloongabba busway, which was constructed in Year 2000 has been carried out to date. Chan et. al. (1997) acknowledged that various methods including chemical mass balance, multivariate analysis, enrichment factor analysis and reconstructions of major components from observed elemental composition have been used for source apportionments but employed only the last method, which led to considerable overestimation of crustal elements. Since multi-criteria decision making methods have not been previously employed to rationalize air quality in Brisbane, this study reports the concentrations of a variety of elements in PM2.5 and TSP samples from three representative urban sites in Brisbane and the application of the multi-criteria decision making procedures, PROMETHEE (Preference Ranking Organisation Method for Enrichment Evaluation) and GAIA (Geometrical Analysis for Interactive Aid) (Espinasse et. al. 1997) to air quality in Brisbane. These procedures have previously been applied to analytical and environmental problems (Espinasse et. al. 1997).

2. Experimental Methods

2.1. Description of the Sampling Sites

Samplings were performed in the months of June to July, 2002 during which ambient temperatures ranged between 15 – 28oC in the day times and fell down to 10oC during the night. The south east coast of Australia usually received a southwesterly wind of 11 to 20 km/hr speed during this period.

Woolloongabba Concourse Area was on the street level

at a height of 7 metres away from passing buses and opposite the main street. Samplings were conducted at the peak hours in the morning and late afternoon. Samples were obtained for TSP and PM2.5 on five consecutive days. Sampling lines were positioned on a portable trolley at the height of 0.8 metre from the concrete floor. High flow sampling system of 0.02 m3/min flow rate was used for the collection of TSP and PM2.5. In general, Woolloongabba urban is surrounded by a very busy traffic flow and a range of business activities like panel beating, cement processing and car paint spraying. The Southeast Freeway runs through it to the north and south while several busy urban roads perpendicular to the Freeway connect Woolloongabba to the eastern suburbs.

ANZ stadium is located near a busy intersection between

Kessel Road and Main Road. Sampling was carried out at about 20 m from the intersection close to the ANZ stadium with a forest reserve in its background. The samplers were mounted at a height of 1.8 m high on a sampling chamber standing on a lawn field adjacent to a drive way. Each sample of the TSP and PM2.5 was obtained every four consecutive days. Up to 2000 vehicles per hour travel at peak hours between 6:30 am and 5:30 pm every day. Otherwise, less than 500 vehicles per hour were recorded. The percentages of the types of vehicles recorded on a particular day are 88.7% (light vehicles), 3.5% (small trucks), 3.9% (medium trucks), 3.0% (large trucks) and 0.8% (unclassified).

QUT M Block is located on the 6th level of a building

facing the Southeast Freeway in the Queensland University of Technology (QUT) Gardens Point Campus. The site is within the Brisbane City and is at a distance of 210 m from the Southeast Freeway. It is surrounded by a Botanical Garden and Brisbane River to the north and east respectively. Samples were obtained over five consecutive days.

2.2. Sample Preparation and Analysis

TF-200 Teflon membrane filters of diameter 47 mm and nominal pore-size 0.2 µm were obtained from Pall Corporation and used as received for the collection of total suspended particle (TSP) and PM2.5. A simple sampling system using Pall Corporation 47 mm in-line polycarbonate filter holder and Dekati PM10 cascade impactor was used to collect the TSP and PM2.5

respectively. The TSP and PM2.5 filters were weighed on a Mettler Toledo analytical balance in a room of constant humidity (50%) and temperature (23oC). Collected samples were stored in a glass vial at -15oC. Each vial was wrapped with aluminium foil.

Each filter was digested in a Teflon digestion vessel by 70% Aristar grade nitric acid and analytical reagent grade concentrated hydrofluoric acid in a ratio of 3:1. The digestion method is an adaptation of the nitric acid digestion (Thomas and Morawska 2002b) and an aqua-regia acidic mixture (Wang et. al. 1998) methods. All samples were placed in Teflon digestion vessels with the acidic mixture. Digestion was carried out in a CEM microwave oven (Matthews, NC, USA) at 75 psi and 60% of 650 W power for 30 min and followed by 100% power for 15 min. A further 60% power at 20 psi for 30 min digestion was used to ensure complete dissolution of the elements. After digestion, the solution was diluted to 25 mL in a standard flask and stored in a polyethylene vial. The elemental analysis was performed using Thermal Inductively Coupled Plasma Atomic Emission Spectrometry (ICPAES). EM Science standard element diluted solutions at 0, 0.5, 1, 2.5 and 5 ppm were used for calibration according to the optimized instrumental parameters described below. The carrier gas, coolant gas and plasma gas was argon gas at a pressure of 35 Psi. Its monochromator was calibrated by ICP profile solution of 10 ppm of sulfur, copper and potassium to give 8400 steps. The same instrumental calibration was done on polychromator to give about 10070 steps for sulfur and 5630 steps for copper. A correlation coefficient close to unity was observed for linear calibration curves of all the elements. The detection limits of all the elements were taken as three times the standard deviation of the field blank samples, which ranged from 3 ng/sample to 186 ng/sample for the elements studied.

To validate the digestion method, a National Institute Standards and Technology Standard Reference Material 1648 Urban Particulate Matter was digested as described above after overnight drying at 120oC and analyzed together with the samples. The blank was subtracted from the obtained elemental concentrations in the air samples and by taking into account the sampling time and flow rates. Mass concentrations in ngm-3 were obtained.

3. Data Analysis

The statistical correlations were studied by using linear regression method in Microsoft XP excel. Pairwise comparison of the TSP and PM2.5 elemental concentrations of all sites were conducted. Only detected elements between the paired sites were used in the regression method. The chemometrics analysis was performed using a multi-criteria decision making system, PROMCALC V3.2. The PROMETHEE II method is coupled with GAIA in the software. The former provides a full ranking of all objects from the best to the worst based on their net outranking flows while the latter

displays PROMETHEE results visually as PCA biplots and facilitates the interpretation of the global performance of each object with reference to a decision axis, π which appears in the biplot. Thus, useful information about the underlying trends in the data matrix such as clustering of objects and variables and characterization of outliers may be obtained from GAIA biplots. The longer the projected vectors of the variables, the more variance they contain. Vectors oriented in the same direction are correlated whereas vectors oriented in the opposite directions are conflicting. When the decision axis is long, the best objects are located in the direction of the axis and as far as possible from the origin of the GAIA biplots (Espinasse et. al. 1997).

4. Results and Discussion

The mass concentrations of TSP and PM2.5 are discussed in comparison with local and overseas data. The correlations between sites with respect to their PM2.5 and TSP mass concentrations are explained. The significance of the data by means of multivariate analysis is discussed below.

4.1. TSP and PM

mass concentrations

This study reports PM2.5 average of (18.9 ± 4.9) µgm-3 from both ANZ stadium and QUT sites. Compared to an annual average in Milan of 45.5 µgm-3 (Marcazzan et. al. 2001), the PM2.5 average concentration for Brisbane sites is lower and is comparable with that found by Chan et. al. (1997). In comparison with a developing country like India (Negi et. al. 2002) and Taiwan (Wang et. al. 1998) with a mean range of 22 µgm-3 to 600 µgm-3, the TSP average concentration (16.6 ± 8.0 µgm-3) of the present study was much lower.

Although the present mass concentrations of PM2.5 may not sufficiently warrant an elaborate action for the reduction of particle mass concentrations, several government agencies have taken steps to continuously monitor its level over the last few years. For instance, nearly a decade ago, the Environmental Protection Agency (EPA) in Victoria reported that PM2.5 in Brisbane averaged at an annual concentration of 6 µgm-3 (Robinson 1999) which is roughly equivalent to two years average of 7.3 µgm-3 measured by Chan et. al. (1997) from December 1993 to November 1995. Unfortunately, monthly measurement performed by Commonwealth Scientific and Industrial Research Organization (CSIRO) in 1996 showed that there is an increase of 46% in the mass concentration of PM2.5 in Brisbane, although this was about 24% below national monthly averages of other state capital cities (Robinson 1999). The mass concentration of PM2.5 continues to climb over the years until recently which is probably contributed by the increase in the vehicular registration in South East Queensland. It is well known that since 1997, the Australian registrations of all motor vehicle types has increased between 3% and 18%. Queensland

alone has an increase of 3.9% registered motor vehicles from March 2001 to March 2002. Together with the present on-road vehicles, there were a total of 12.8 million registered motor vehicles around Australia (Australian Bureau of Statistics 2002).

Currently, there is no Australian standard for PM2.5. As a result, several government agencies like CSIRO, EPA, National Environment Protection Council (NEPC) and other institutions have conducted studies and reviewed the overseas standards as part of the Australian guidelines. For instance, the United State of America and Canada developed daily average standards for PM2.5 of 65 µgm-3 and 30 µgm-3 respectively. Recently, California proposed an annual average standard of 12 µgm-3 and New Zealand also proposed an interim guideline of 25 µgm-3 daily average (NEPC 2002). Due to potential health problems caused by PM2.5, special efforts made by various state agencies to obtain better understanding of the fluctuations in the mass composition of the PM2.5 should continue.

A review by NEPC found that the PM10 could serve as a surrogate for PM2.5 (NEPC 2002). The ratio of PM2.5 to PM10 reflects a wide range of particle emission sources of which PM2.5 is generally associated with anthropogenic sources like motor vehicle exhaust and combustion of fuels and other natural materials. A ratio approaching unity will quite accurately pinpoint the types of emission sources. By adopting the same approach, it was predicted that the majority of particles must have originated from anthropogenic sources especially at QUT site where the ratio very close to one (0.94). The larger ratio for the ANZ site (1.24) indicates that there are sources other than traffic related emissions.

4.2. Source apportionment

4.2.1. Elemental relationship between sites Because all three sites are within the distance of 12 km, it was expected that they might have the same source of pollution. Therefore, correlations between elements at different sites were analyzed and presented in Table 1. Generally, strong positive correlations were observed between elemental levels at different sites especially for TSP contents. This indicates that these sites experienced fairly similar sources of pollution. However, the PM2.5 correlations between the elemental levels at Woolloongabba and ANZ on one hand and Woolloongabba and QUT on the other hand are very low suggesting that the sources of the pollutants at each of these sites were not the same. The correlations of PM2.5 bound elements at ANZ and QUT sites are strong suggesting that they are influenced by relatively similar pollution sources. Considering that the correlations of Woolloongabba with QUT and ANZ are weak, it appears

Table 1: Regression for the paired sites

Elemental Correlations TSP PM2.5 Woolloongabba Concourse & ANZ 0.93 0.19 Woolloongabba Concourse & QUT 0.99 0.10

ANZ & QUT 0.93 0.75

that the ambient air in Woolloongabba might not be only associated with the vehicular emission but other types of anthropogenic sources.

4.2.2. General assignment of the sources of elements

The three most abundant elements (Fe, Si and Al) have been associated with crustal matter. This association is particularly reasonable because the TSP Fe/Si ratios are found to be 0.01 to 0.08 with the average of 0.04 which is an indication that local rock influences the chemistry of the particles (Negi 2002). However, their presence in smaller concentrations in PM2.5 fraction might indicate that there are other sources of elements in agreement with the finding by Chan et. al. (1997) that Fe was also related to combustion activities. Similarly, Al is a prominent element present in the metal compartment within the vehicles engines (Bigard 1969, Brewer and Belzer 2001) and Si might arise from frictional grinding on the road (Weckwerth 2001).

The sources of other elements could also come from the presence of various local industries such as paint spraying, cement processing and other aerosol producing activities during the course of samplings could also affect the elemental concentrations considerably. Given the large variation in Figure 1, the mean values of each element in the TSP and PM2.5 from the sites are not statistically significant. As the ANZ sampling site was positioned close to the traffic flow, its PM2.5 elemental levels could be reasonably attributed to the neighbouring traffic pollution. Similar level was also observed at all other sites. The opposite is observed for the TSP elemental levels in which its level at Woolloongabba concourse area is much higher than those at QUT and ANZ. Therefore, it could be affirmed that the Woolloongabba urban is not just affected by the busy traffic but by other sources of pollutions from its surrounding business centre.

One of the local activities in the vicinity of the Woolloongabba site is the car paint spraying which was located just about half kilometer away from the Woolloongabba sampling point. This might have caused

Figure 1: Total average elemental concentrations for three sites.

the elevated elemental levels at this site.

Whittaker et. al. (2003) listed a range of inorganic chemicals like cadmium pigments, chromium oxide, iron oxide, manganese pigments, hydrated alumina, magnesium oxides and zinc sulfide as pigmentation agents and colour enhancers in paints. In addition, other activities like coal-fired cement works and vehicular emission that generate a high amount of airborne particles that could contribute to the relatively high elemental concentrations at Woolloongabba site (Chan et. al. 1997).

4.2.3. PROMETHEE and GAIA analysis of the PM2.5 air quality of the sites.

Figure 2 and 3 show the respective GAIA scores and loading plots obtained for the PM2.5 data matrix that consisted of 10 objects and 16 variables (elements). The scores plot shows that the three sites (Woolloongabba, QUT and ANZ) are well separated. The Woolloongabba objects are on negative PC1 axis while the QUT objects are on positive PC1 axis. ANZ objects form a tight cluster on both sides of PC1 axis. The decision axis points towards the ANZ site showing that it is less polluted than Woolloongabba and QUT sites. In keeping with this, PROMETHEE II ranked Woolloongabba and QUT objects as having the higher elemental pollution.

Figure 2: PM2.5 score plot of matrix of 10 objects and

16 variables for Woolloongabba concourse, QUT and ANZ

In Figure 3, the long vectors of Zr, Mo, V, Al, Mn and Sr are oriented in the same direction and opposite QUT objects, indicating they are mostly found at QUT site. The first three elements could possibly come from the 0 5 10 15 20 Wolloongabba Concourse QUT ANZ TSP P M 2.5 -1.5 -1 -0.5 0 0.5 1 1.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 PC1 (35.9%) PC2 (26.9%) π ANZ QUT Woolloongabba Concourse

engine components and combustion fuels (Moldovan et. al. 1999, Fukui et. al. 2001, Weckwerth 2001) whereas the rest might originate from geological source or construction dust (Cadle et. al. 1999, Qin et. al. 1997, Brewer and Belzer 2001). Similarly, long vectors are observed for the variables like Fe, Cd, Co, Sn, Cu, Zn and Mg. Vehicular emission is a possible source contributor for Cu, Zn and Mg (Fukui et. al. 2001, Weckwerth 2001) and the rest could be from local industrial activities (Cadle et. al. 1999, Kumar et. al. 2001 and Wang et. al. 2001). Since the vectors for these elements are oriented in the opposite direction to Woolloongabba objects, these objects have high concentrations of Fe, Cd, Co, Sn, Cu, Zn and Mg in accord with the proposition that elemental pollution at Woolloongabba is influenced by vehicular emission and local industrial activities. Significant concentrations of Si and Fe are found at ANZ site, both of which are crustal elements. Apparently, ANZ site is largely influenced by the geological source over the vehicular source in its fine particle as Negi et. al. (2002) and Kumar et. al. (2001) have confirmed that the geological origin of these elements. This is probably due to that the sampling station was in the vicinity to a bushy background.

Figure 3: PM2.5 loading plot for Woolloongabba

concourse, QUT and ANZ score plot of matrix of 10 objects and 16 variables

5. Conclusion

The TSP average elemental content showed that the traffic-related emission source was not probably the sole contributor to the ambient air at Woolloongabba urban. There was an indication of the influence of other types of anthropogenic sources from its surrounding commercial centres. The PM2.5 elemental concentrations at all the sites were either influenced by the vehicular emission, local industrial sources or geological source.

The above statement is further confirmed by the results of the chemometrics analysis. The analysis of the PM2.5 elemental data showed that the vehicular emission and local industry were the two most important emission sources at the Woolloongabba site whereas QUT is largely influenced by anthropogenic sources with a prominence of vehicular emission source in addition to the geological source possibly due to long range particle transportation. Therefore, QUT site has the highest air pollution in its PM2.5. Generally, the air quality at the ANZ site was reasonably under the influence of geological source, in addition to vehicular emission.

Overall, inorganic species contribute significantly to the ambient air quality around Brisbane but they are not of special environment concerns at the levels they were found. Many of the elements have multiple sources. Therefore, it was not possible to apportion them to particular sources unequivocally. Future studies on these and other sites are planned.

Acknowledgements

This work was performed with a Polaris Q GCMS instrument donated by Thermo-Finnigan. The authors would like to thank Maricela Yip and Pierre Madl of the University of Salzburg, Austria for their assistance during the fieldwork.

References

Australian Bureau of Statistics 18 November 2002, 9309.0 Motor Vehicle Census, Australia.

Bigard H.M. 1969, ‘Aluminium in the Passenger Car’, Metallurgia: the British Journal of Metals, 179-83. Bilos C., Colombo J. C., Skorupka C. N. & Rodriguez

Presa M. J. 2001, ‘Sources, distribution and variability of airborne trace metals in La Plata City area, Argentina’, Environmental Pollution, 111:149-158.

Brewer R. & Belzer W. 2001, ‘Assessment of metal concentrations in atmospheric particles from Burnaby Lake, British Columbia, Canada’, Atmospheric Environment, 35:5223-5233.

Cadle S. H., Mulawa P. A. & Hunsanger E. C. 1999, ‘Composition of light-duty motor vehicle exhaust particulate matter in the Denver, Colorado area’, Environmenal Science and Technology, 33:2328-2339.

Chan Y. C., Simpson R. W., McTainsh G. H., Vowles P. D., Cohen D. D. & Bailey, G. M. 1997, ‘Characterisation of chemical species in PM2.5 and PM10 aerosols in Brisbane, Australia’, Atmospheric Environment, 31(22):3773-3785.

Espinasse B., Picolet G. & Chouraqui E. 1997, ‘Negotiation support systems: A multi-criteria and multi-agent approach’, European Journal of Operational Research, 103:389-409.

Fukui M., Sato T., Fujita, N. & Kitano, M. 2001, ‘Examination of Lubricant Oil Components Affecting

-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 PC1 (35.9%) PC2 (26.9%) Zr Mg Zn Si Fe Cr Mn Sr Al V Co Sn Cd Pb Mo Cu

the Formation of Combustion Chamber Deposit in a Two-Stroke Engine’, Japanese Society of Automotive Engineers Review, 22:281-285.

Hinwood A. L. & Di Marco P. N. 2002, ‘Evaluating hazardous air pollutants in Australia’, Toxicology,

181-182:_361-366.

Kumar A. V., Patil R. S. & Nambi, K. S. V. 2001, ‘Source apportionment of suspended particulate matter at two traffic junctions in Mumbai, India’, Atmospheric Environment, 35:4245-4251.

Lorranger S., Tetrault M., Kennedy G. & Zayed, J. 1996, ‘Manganese and other trace elements in urban snow near an expressway’, Environment Pollution,

92_(2):203-211.

Marcazzan G. M., Vaccaro S., Valli G. & Vecchi, R. 2001, ‘Characterisation of PM10 and PM2.5 particulate matter in the ambient air of Milan (Italy)’, Atmospheric Environment, 35:4639-4650.

Miller F.J.G., Graham J. A. & Lee R. E. 1979, ‘Size Considerations for Establishing a Standard for Inhalable Particles’, Journal of Air Pollution Control Association, 29(6):610-615.

Moldovan M.G., Gomez M. M. & Palacios M. A. 1999, ‘Determination of platinum, rhodium and palladium in car exhaust fumes’, Journal of Analytical Atomic Spectrometry, 14:1163-1169.

Negi B. S., Jha S. K., Chavan S. B., Sadasivan S., Goyal A., Sapru M. L. & Bhat C. L. 2002, ‘Atmospheric dust loads and their elemental composition at a background site in India’, Environmental Monitoring and Assessment, 73:1-6.

NEPC, National Environmental Protection (Ambient Air Quality) Measure 8 February 2002, Dimension paper Qin Y., Chan C. K. & Chan, L. Y., 1997.

‘Characterisatics of chemical compositions of atmospheric aerosols in Hong Kong: spatial and seasonal distributions’, The Science of the Total Environment, 206:25-37.

Robinson D. L. 1999, Fine Particulates (PM2.5) Air Pollution Australia, LEAD Action News, 7:3, ISSN 1324-6011. Armidale Air Quality Group, NSW. setting PM2.5 standard in Australia.

Thomas S. & Morawska L. 2002a, ‘Size-selected particles in an urban atmosphere of Brisbane, Australia’, Atmospheric Environment, 36:4277-4288. Thomas S. & Morawska L. 2002b, A simple nitric acid

extraction for the determination of ultratrace metals in submicrometer aerosols’, personal communication. Verrall K. A., Muller W. A., Kingston P. A., Frazer B.,

Rose H. & Ratery A. 1986, ‘Deteminations of sources contributing to suspended particular matter in Brisbane’, Air Pollution Control Branch, Queensland. Wang C. F., Chin C. J. & Chiang P. 1998, ‘Multielement

analysis of suspended particulates collected with a beta-gauge monitoring system by ICP atomic emission spectrometry and mass spectrometry’, Analytical Sciences, 14:763-768.

Wang C. X., Zhu W., Peng A., & Guichreit R. 2001, ‘Comparative studies on the concentration of rare

earth elements and heavy metals in the atmospheric particulate matter in Beijing, China and in Delft, the Netherlands’, Environment International, 26:309-313. Weckwerth G. 2001, ‘Verification of traffic emitted aerosol components in the ambient air of Cologne (Germany)’, Atmospheric Environment, 3:5525-5536. Whelan J. 1998, Brisbane Busways: public transport

miracle, white elephant or environmental catastrophe?, Queensland Conservation Council. Whittaker, Clark and Daniels, Inc., Jun 2003, Glossary

of terms – paints and coatings, www.wcdinc.com/samples.html.

Wilson R. & Spengler J. D. 1996, Particles in our air: concentrations and health effects, Physico-chemical properties and measurement of ambient particles, Harvard University Press.