The characterisation of airborne particulate matter by automated mineralogy: the potential of the mineral liberation analyser for the monitoring of mine-derived emissions

Abstract

With its links to adverse environmental and human health impacts, air pollution is an increasing and ubiquitous global concern. Air pollution, especially the particulate matter component, has been linked to cancer, pulmonary and cardiac disease, neurotoxicity and pernicious impacts on fertility and pregnancy. It has also been observed to affect climate by altering the radiation and chemical balance of the atmosphere. Unlike other pollution vectors, airborne particles are not constrained by topography, enabling them to be rapidly dispersed over vast distances. Although international air quality guidelines have been implemented, more than 90% of the global population are exposed to levels higher than those recommended by the World Health Organisation.Airborne particulate matter (APM) is a physically and chemically complex mix of organic and inorganic substance and, to accurately assess the potential environmental and human health implications, a detailed physical and compositional characterisation at the single particle level is required. The ultimate goal of an analytical technique for APM is to quantitatively identify all the chemical species within each individual particle to predict potential impacts and develop successful mitigation strategies.Micro-beam techniques, such as energy dispersive scanning electron microscopy (SEM-EDS) have been used extensively to generate information on the physical, morphological and chemical properties of single atmospheric particles down to a nominal diameter of 0.1µm. Traditional SEM-EDS methods, however, are prone to operator error and bias and the manual data processing is time consuming and costly and therefore impractical for regular monitoring purposes. As a result, computer-controlled SEM (CCSEM) is rapidly gaining predominance. Although the use of automated SEM-EDS analysis is well documented, few studies to date have used automated mineralogy systems, such as the Mineral Liberation Analyzer (MLA).The MLA was designed to improve the efficiency of mineral processing plants and, although the technique has been extensively used in the mining industry since its inception, it is only recently that this technique has been applied to other environmental fields. To our knowledge this is the first time the MLA has been used in the application of ambient APM analysis and has the potential to be a powerful tool in the APM analytical arsenal.Prior to APM analysis, a suitable sampling and sample preparation technique was determined, a spectral reference library constructed and the accuracy and precision of the instrument tested by repeat analysis and comparison to certified reference materials (CRM), with mineral abundances (modal mineralogy) and repeat analysis showing relative standard deviations typically below 10%.Using the developed methodology, ambient APM was collected from four stations in the vicinity of a large iron-ore mining operation in Congonhas, Minas Gerais, Brazil, as well a control station lying outside the influence of the mine. To observe spatio-temporal variations, sampling was conducted over the dry, wet and transitional periods. The mineral phases observed strongly reflected the local geology, with clays and iron oxides contributing 70-80% of the particulates sampled. Particle size distributions reflected sources dominated by mechanical processes, with coarse particles (>2.5µm) accounting for 80-90%. A strong seasonality was observed with coarser particles more prevalent during the drier periods of May and August, with fine particles (<2.5µm) contributing to less than 10% of the total particulates sampled.Although the transport and ultimate fate of APM is largely controlled by particle size and shape, chemical species is critical in assessing the potential toxicity, as different species of the same elements can have different toxicological properties, bulk elemental concentrations may not be a true representation of bioaccessibility. As a result the ability to accurately identify potentially toxic elements in the atmosphere is essential to formulate a realistic risk assessment. Manganese bearing particles, predominantly Mn oxides and jacobsite, were detected in appreciable numbers, although only contributing to less than 1% of the total particulates sampled. The particle size distribution, however, indicated a geogenic, rather than anthropogenic, source. As deposition within the human lung is largely determined by size and density, the predominantly coarse particles and relatively insoluble mineral phases, suggested that the risk of inhalation toxicity for the exposed population is potentially relatively small.