A size-selecting sampler for airborne dust.

Abstract

It has long been known that airborne dust often differs markedly in composition from dust which has settled on the ground, and this in turn differs in composition from the parent substance from which the dust was produced. This variation in composition is related to varia tion in particle size, and continues even within the size range of airborne dust. For instance, Drinker and Hatch (1936) quote an example where the free silica content of an airborne foundry dust varied from 72-3% for particles over 10 pi in diameter, 28-2% between 5 y. and 10 (jl, 22-7% between 2 (x and 5 \jl, down to only 3-3% below 2 u,. It has also been known for many years (McCrae, 1913 ; Watkins-Pitchford, 1916) that the dust retained in the lungs consists almost entirely of particles below about 7 \i in average diameter. More recent theoretical and experimental studies (Fin deisen, 1935 ; Davies, 1949) have shown that the failure of large particles to penetrate and be retained in the lungs is related to their high settling rate in air. It is evident then that any chemical analysis of airborne dust which is to be relevant to pneumo coniosis must be confined to that fraction of the dust which consists of particles of what may be called for convenience respirable size. This conclusion has been recognized by a number of workers, and they have attempted to comply with it by separating the respirable fraction from the sample after collec tion, either by liquid sedimentation or by air elutriation. The former method is unsatisfactory since the different densities of the various ingredients will cause a greater difference in settling rate in liquid than in air. The method is, moreover, open to another objection, which also applies to air elutria tion, namely that dust in the -air_is nearly always aggregated to some extent (Watson, 1949) and the aggregates will have a settling velocity very different from that of their constituent particles. If the dust is collected and then re-dispersed in liquid, it will be largely disaggregated, and even if re-dispersed in air its original state of aggregation will not be restored, so that it will not be separated into true respirable and non-respirable fractions. Apart from these objections, the process of separation after collection is laborious and to be avoided if possible, but until recently no easy and satisfactory method of separation while sampling was available. The most obvious device, a vertical elutriator, such as is used commercially for air separation of powders (Haultain, 1937) can be shown to be impracticable owing to the very large size of chamber needed for the volumetric sampling rate which is usually required (Walton, 1954). Another obvious possibility, the cyclone, is open to the objection that the dust is exposed to a process of attrition which may break down aggregates ; moreover, the behaviour of a cyclone is critically dependent on its design. Recently Walton (1954) has investigated the whole question of elutriation and has shown that a simple horizontal elutriator has many practical advantages and is moreover capable of giving as sharp a cut-off as any other elutriation process. The process of separation it uses is closely similar to that occurring in the respiratory tract, and although the shape of the cut-off curve is somewhat different from that currently attributed to the respiratory tract (Davies, 1952), the difference is not great and, owing to the general constancy of size distribution of dust particles in this size range (Wynn and Dawes, 1951), it is small in importance compared with the necessity of eliminating the bulk of the non-respirable dust. The fundamental principle of a horizontal elutriator is that dusty air is passed through a horizontal duct at such a velocity that particles which it is desired to capture will settle on the floor of the duct before they reach the end. Particles of lower settling rate will, to a greater or lesser extent pass through the duct ; those which were initially near the floor being captured, and those higher up passing through, the proportion doing so varying 284