Aggregates and naturally occurring asbestos: the need of a correct analytical approach

Aggregates (sand, gravel and crushed stone) characterized by good mechanical properties and no undesired reactivity, are used in huge amounts in many industrial sectors, especially in construction (e.g. concrete, asphalt, paving). Sand and gravel extracted from alluvial or glacial deposits are typically rounded and well selected, whereas crushed stone is angular and suitable for certain applications (e.g. railway ballast). Use of offshore deposits is mostly restricted to beach erosion control and replenishment. Demand for aggregates is governed essentially by markets, and sources of supply need to be situated close to each other, because of transportation costs. The most common rock types (depending on geology) are represented by basalts, porphyries, orthogneisses, carbonatic rocks and “green stones” (serpentinites, prasinites, amphibolites, metagabbros). Especially “green stones” may contain traces, and sometimes appreciable amounts of asbestiform minerals (chrysotile and/or fibrous amphiboles). For example in Italy, the chrysotile asbestos mine in Balangero (Turin) produced over 5 Mt railroad ballast (crushed serpentinites), which was used for in northern and central Italy, from 1930 up to 1990. The legal threshold for asbestos content in track ballast is established in 1000 ppm: if the value is below this threshold, the material can be used, otherwise it must be disposed of as hazardous waste, with very high costs. The presence of asbestiform minerals must be first assessed by preliminary geological and mineralogical surveys in quarry areas, both for glacial – alluvial deposits and “massive” rock mass (crushed stone). The quantitative asbestos determination in rocks is a very complex analytical issue: although techniques like TEM-SAED and micro-Raman are very effective in the identification of asbestos minerals, a quantitative determination on bulk materials is almost impossible or expensive and time consuming. Another issue is represented by the discrimination of asbestiform minerals (e.g. chrysotile, asbestiform amphiboles) from the common acicular – pseudo-fibrous varieties (lamellar serpentine, non-asbestiform amphiboles). Also, the correct sampling is of crucial importance, considering the size of rock fragments (sand, gravel or silt) and the geological variability within the quarry. In this work, more than 400 samples from the main Italian quarry areas were characterized by a combined use of XRD and an up to date sample preparation and quantitative SEM-EDS analytical procedure. The first step consists in the recognition of “green stones” (presence of serpentine and/or amphiboles) by means of macroscopic petrography (gravel) or XRD (sand, silt). The second step is represented by the “self-grinding” of the rock fragments (Los Angeles rattle test for gravel), and the quantitative SEM-EDS analysis of the “fine” fraction (< 2 mm). The third and last step consists in the complete grinding of the bulk sample and following SEM-EDS quantification. The results show a great variability for serpentinite-rich samples, with a wide asbestos concentration range; on the other hand, metabasites (prasinites, amphibolites) are generally less critical, because the presence of asbestiform amphiboles (especially tremolite - actinolite) is rarer and more occasional. As regards the samples deriving from alluvial and glacial deposits, the fibers tend to concentrate in the fine fraction (<2 mm).