Recommended flow rate of the aluminum cyclone for improved exposure assessment

Samples of the respirable fraction of airborne particles to which workers are exposed are an important component of many health protection programs, especially in those workplaces where there is a risk of health problems from exposure to respirable crystalline silica. The most common technique for size-selection of the respirable dust fraction is to use a miniature cyclone preselector, many of which are based on the "Higgens-Dewell" (HD) design. A variant of the HD cyclone, commonly referred to as the "aluminum (or aluminium) cyclone," was developed in Scandinavia. Early work showed that a flow rate of 2.2 l min-1 would be appropriate to meet the size-separation convention standardized under the International Organization for Standardization (ISO), while a later, non-peer-reviewed study suggested 2.5 l min-1 and this flow rate is currently recommended by the manufacturers. The International Sampler Comparison Group is working on revising the European Standard EN13205 on sampler performance testing for consideration as an ISO standard. In this work, 5 aluminum cyclone units were tested at 2.5 l min-1 and the cyclones were further tested at 2.3 and 2.2 l min-1 to determine the optimal flow rate. While the flow rate of 2.3 l min-1 had the lowest overall mean bias, a flow rate of 2.2 l min-1 gave bias <±10% over the whole area of size distributions of interest. This supports earlier findings and suggests that 2.2 l min-1 is the most accurate flow rate for sampling with the aluminum cyclone. However, 2.3 l min-1 also meets the specification of EN13205 in that the area of bias >±10% is minimal. The consequence of continuing to use the aluminum cyclone at 2.5 l min-1 is an underestimate of respirable particles when compared with the ISO convention.