Login 
Deploying the time and cost intensive techniques of aerosol measurement, this article looks to quantify sources contributing to airborne pollution at the Port Talbot steel works. Aerosols, otherwise known as particulate matter (PM) are solid or liquid droplets suspended in the air.
These particles are products of direct biogenic and anthropogenic activities, and may also be formed indirectly via chemical processes, as in secondary aerosols. Directly-emitted particles (primary aerosols) come from natural processes such as sea spray, volcanic eruption, windblown dusts, crustal matter erosion, forest fires or vegetation detritus, as well as human activities that emit PM into the atmosphere which include traffic, industry, agriculture, mining, coal and biomass combustion, and incineration.
Secondary particle formation involves chemical conversion of gases generated biogenically or anthropogenically into a particulate product. Examples of secondary aerosol are sulphate and nitrates, which are formed by oxidation of nitrogen (IV) and sulphur (IV) oxides respectively. Aerosol nomenclature derives from their size distribution. Particles may occur in three modes viz: ultrafine (aerodynamic diameter < 0.1 µm), fine (aerodynamic diameter, < 2.5 µm) and coarse (aerodynamic diameter, 2.5- 10 µm). Ultrafine and fine aerosols are usually formed by high temperature processes, often from human activities, while coarse particles are sourced mainly but not solely from natural processes.
Aerosols have direct and indirect impacts on human lives and are therefore worthy of study. They have a direct influence on the quality of life causing morbidity and mortality. A report from the UK Committee on the Medical Effects of Air Pollutants estimated that nearly 29,000 deaths annually in the UK are associated with anthropogenic particulate matter pollution. Toxicity of aerosols depends on many factors such as the particle’s size, number, mass, chemical composition and concentration. Their indirect impact on the quality of life includes the reduction of visibility through smog formation, and their influence upon climate change.
The understanding of aerosol properties is critically dependent upon available instruments for their study. Different instruments have evolved for researching aerosol properties and behaviour ranging from offline to online instruments. Our major objective in the Port Talbot air sampling campaign was to deploy different online and offline instruments for four weeks in order to monitor air quality and to quantify sources contributing to airborne concentrations. Port Talbot is an example of a major industrial setting in the United Kingdom. The main industry is the steel complex located on a south-west to north-west axis of the town. The activities of the steel works have been a major contributor to particulate matter pollution in the area.
Previous Automatic Urban and Rural Network (AURN) particulate matter monitoring data based on Tapered Element Oscillating Microbalance-Filter Dynamics Measurement System (TEOM-FDMS) instruments have shown periods of 24 hour exceedences of the European Union 24 hour Limit Value for PM10 in Port Talbot.
Aerosol measurement techniques
The two approaches employed for monitoring airborne particulate matter in our sampling campaign at Port Talbot were offline and online methods. Offline methods involve collecting the airborne particulate matter onto a substrate such as a filter paper. The filters would have been pre-weighed before sampling, and after sampling weighed again in order to calculate the gravimetric mass concentration of aerosol. The chemical composition of the aerosol is obtainable through laboratory analyses.
Aerosol may be analysed for organic (polycyclic aromatic hydrocarbons, hopanes, alcohols, organic acids or levoglucan) and inorganic (metals, water soluble ions, carbonaceous species) constituents depending upon the research objectives. The second approach is online measurement involving real time data collection. Online aerosol measurement is a high time resolution technique of analysis for obtaining information on particle mass/chemical composition, size and number/count.
Offline measurement techniques
One offline sampler deployed in the Port Talbot air monitoring campaign measured particle size distribution. We used a cascade impactor which has been applied for environmental pollution and visibility studies. The particular model used had cut points including 10, 5.6, 3.2, 1.8, 1.0, 0.56, 0.32 and 0.18 µm. A flow of air (30 L/min) enters through the inlet and PM size fractions are collected by impaction on the 47 mm Teflon filter substrates.
The last stage in the system collects the smallest size fraction of aerosol through filtration on a 37 mm- quartz fibre filter. The rotating ability of the instrument allows for uniform deposition of aerosol on the filter substrates.
Dichotomous air sampler
This is a sequential air sampler that measures two PM fractions; namely fine (PM2.5) and coarse (PM2.5-10) particles simultaneously on 47 mm filter substrates. It is a programmable sampler that can sample PM continuously for 16 days. There are two magazines (one for fine and the other for coarse) made up of 16 filter cassettes for sample collection and also two magazines for collection of exposed filters. Another feature of this sampler is incorporation of meteorological data collection.
The air flow of 16.1 L/min (PM10) passes through the inlet and is split into two fractions -15.0 L/min (PM2.5) and 1.7 L/min (PM2.5-10) by a virtual impactor. The coarse particle stream contains a small contribution of fine particles, for which a correction is applied.
High volume sampler
A high volume aerosol sampler can be loaded with 15 filter holders equipped with 150 mm quartz fibre filters. The one we used is different from the conventional high volume sampler, because it is programmable, and also equipped with meteorological measurement features. This sampler has similar features to a the dichotomous sampler, except that it samples only one PM fraction at a time. The air flow for the instrument is 100-1000 L/min.
Streaker sampler
A streaker is an air sampler that collects time and size-resolved samples. It samples on 82 mm filters (comprised of Nuclepore and Kapton) in a 180o pattern for 168 hours (seven days). The Nuclepore substrate collects the fine PM by filtration while the Kapton samples the coarse PM fraction by impaction. A 1 L/min flow of air passes through the inlet of the sampler through a non-rotating and subsequently a rotating impactor where two PM size ranges are collected.
The filters are analysed offline for metals with ion beam analysis by Proton Induced X-Ray Emission (PIXE). D’Alessandro et al (2003) reported a study of hourly analysis of elemental composition of urban aerosol in four towns in Italy with a streaker sampler. The sampler inlet we used was placed on the roof of our mobile laboratory.
Online air measurement – Aerosol Time of Flight Mass Spectrometer (ATOFMS)
Offline sampling is unable to differentiate aerosol internal and external mixing (e.g. internally mixed particles each contain a mixture of constituents while in externally mixed particles, each particle contains a specific compound or component, but there are many different types), and can suffer from loss of materials during sampling, transportation and analysis. Online sampling can overcome these problems and allows higher time resolution.
PM mass, chemical composition, size distribution and numbers can be determined in real time depending on the instrument’s response time. The ATOFMS is a single particle mass spectrometer designed for in-situ characterisation of the sizes and chemical composition of individual aerosol particles in real time drawn in at an air flow 0.9 L min-1. The system comprises three distinct regions: (i) an aerosol introduction interface; (ii) a light scattering region where particle and velocity/size are determined; and (iii) two reflectron-time-of-flight mass spectrometers for analysis of positive and negative ions.
The air enters through an aerodynamic lens which creates a beam of particles which are individually sized by measuring their time-of-flight between two low-powered (Nd:YAG) lasers. Each particle is then vaporised and ionised by a high powered laser, and composition is estimated from the mass spectral data collected.
Portable aerosol spectrometer
This is an online instrument that provides information on size-resolved particle numbers and mass concentrations in a real time using a light beam produced by a laser diode, but no chemical data. The unit we used classifies particles into 15 size bins as they enter the inlet mounted on top of our mobile laboratory. The instrument is connected to a computer where the output is displayed.
Chemical analyses
Aerosol measurement is both time and cost intensive, especially for offline sampling techniques. After field sampling, the filters on which aerosols are collected are transferred to the laboratory for assays.
The chemical analyses applied to the samples include: • Ion chromatography for water soluble ions (sulphate, nitrate, chloride, phosphate, oxalate, sodium, calcium, potassium, magnesium) • Spectrometry (Graphite Furnace Atomic Absorption Spectrometry (GFAAS) and Inductively Coupled Plasma Mass Spectrometry (ICPMS) for analysis of metals • Proton Induced X-Ray Emission (PIXE) for metals/elements in samples collected by the Streaker sampler • Thermal/optical transmission analysis for carbonaceous species (organic and elemental carbon) • Gas Chromatography/Mass Spectrometry (GCMS) for organic constituent analysis
For this project, the main areas of interest are analysis of water soluble ions and metals in the exposed filters.
Results
Preliminary data for air quality parameters from the AURN site, Port Talbot (uk-air.defra.gov.uk/data/data_selector?q=73274#mid) are presented in Figure 1. The hourly concentrations of air pollutants (PM10, PM2.5, SO2, CO, NOx, O3) show wide variations. The 24 hour average of 50 µg/m3 for PM10 was not exceeded except for two episode days where PM10 concentrations of 79 and 52 µg/m3 were observed. The mean concentrations of PM10 and PM2.5 during the one month sampling were 22.16±14.85 µg/m3 and 11.35±4.32 µg/m3, respectively.
The PM2.5/PM10 ratio for the one month sampling period is 0.51, indicating almost equal contributions of fine and coarse particles. This is a low PM2.5/PM10 ratio reflecting the contributions of marine particles at this coastal site and dust from the industrial site to coarse particle concentrations. In the traffic polluted environment of Marylebone Road, a PM2.5/PM10 ratio of 0.67 has been reported. Another observation from Figures 1 and 2 is the higher peaks of air pollution parameters shown in week two and week four. The trend is spread across the time series plots for all the pollutants except for ozone, for which maritime air is the main source.
Particulate and gaseous pollutants show higher concentrations towards the southwesterly sector where the steelworks complex is located (Figure 2a-f). SO2, CO and NOx are products of fuel combustion. Elevated concentrations of particulate pollutants observed at the AURN site in this sector serve to confirm the industrial contribution. Charron and Harrison (2005) have shown during periods of high wind speed, there is a possibility of dilution of PM2.5 as well as resuspension of coarse particles. Consequently, as a result of a high wind speed, the ratio of PM2.5/PM10 is altered.
The report of AQEG (2005) indicated that PM2.5 in Port Talbot is dominated primarily by long range transportation from the east. This is clearly shown in the PM2.5 pollution rose in which apart from the industry contribution, another emission source from the easterly wind sector is clear. PM2.5 in the UK has been reported to be dominated by secondary aerosol. There are other possible emission sources of pollutants affecting the AURN site apart from the steel industry, as seen in the pollution roses for PM2.5, NOx and ozone. Traffic remains a significant factor for NOx and PM2.5 emissions.
Concentrations of ozone did not exceed the 100 ppb of eight hour maximum daily exposure given by WHO (2006), despite the fact that there were instances of hourly exceedences. Ground-level ozone at higher concentration has direct and indirect detrimental effects on public health.
Source apportionment
Further laboratory analyses of offline filter samples is ongoing. All of the data will be used in a receptor modelling study designed to quantify the contributions of the various source categories to measured pollutant concentrations (referred to as source apportionment).
Examples of source apportionment models that have been used for source identification and apportionment are: Chemical Mass Balance (CMB), multivariate techniques such as Positive Matrix Factorization (PMF) and Principal Component Analysis (PCA), and Pragmatic Mass Closure. Possible sources that could have contributed to high PM concentrations in Port Talbot include the Irish Sea, steel industry, traffic, railway and crustal materials.
It is therefore of importance to screen the air pollution data through a receptor model to infer the contributions by each of these factors mentioned. In this case, it is planned to use the PMF model, developed by Paatero and Tapper (1994).
Published: 01st Sep 2012 in AWE International
