Edmond Kauffer

Personal thoracic CIP10-T sampler and its static version Cathia-T

A specific version of the personal aerosol sampler CIP 10 was designed, named CIP10-T, for sampling the conventional CEN thoracic fraction. A static sampler, named CATHIA, was also designed. It uses the same sampling head, but the size selected particles are collected onto a filter. The combined particle efficiency of the aspiration slot and the selector was measured in a horizontal wind tunnel at low air velocity, close to 16 cm s-1. The flow rate of both samplers was fixed at its nominal value, i.e., 71 min-1. Two different methods were used: the former was based on the Aerodynamic Particle Sizer (TSI); the latter used the measurement of particle size distribution of the collected samples by the Coulter technique. For the CIP10-T sampler, the particle collection efficiency onto the rotating cup was also measured. For both samplers bias and accuracy maps have been calculated, following the recommendations of a new CEN standard about sampler performance. The bias does not exceed 10% in absolute value for both samplers, within a large range of particle size distribution of the total aerosol. For the CIP10-T sampler, the accuracy map exhibits a large area where the accuracy is better than 10%, corresponding for example to 4 microns < or = MMAD < or = 14 microns for GSD = 2. For the same geometric standard deviation, the accuracy is still better than 20% for 15 microns < or = MMAD < or = 21 microns. For the CATHIA-T sampler, the accuracy map can be roughly divided into two parts. The accuracy remains better than 10% for MMAD < or = 12 microns, and it remains between 10 and 20% for coarser aerosols, with 13 microns < or = MMAD < or = 20 microns, provided GSD > or = 2.

Site Comparison of Selected Aerosol Samplers in the Wood Industry

Several samplers (IOM, CIP 10-I v1, ACCU-CAP, and Button) were evaluated at various wood industry companies using the CALTOOL system. The results obtained show that compared to the CALTOOL mouth, which can be considered to be representative of the exposure of a person placed at the same location under the same experimental conditions, the concentrations measured by the IOM, CIP 10-I v1, and ACCU-CAP samplers are not significantly different (respectively, 1.12, 0.94, and 0.80 compared to 1.00), the Button sampler (0.86) being close to the ACCU-CAP sampler. Comparisons of dust concentrations measured using both a closed-face cassette (CFC) and one of the above samplers were also made. In all, 235 sampling pairs (sampler + CFC) taken at six companies provided us with a comparison of concentrations measured using IOM, CIP 10-I v1, ACCU-CAP, and Button samplers with concentrations measured using a CFC. All the studied samplers collected systematically more dust than the CFC (2.0 times more for the IOM sampler, 1.84 times more for the CIP 10-I v1 sampler, 1.68 times more for the ACCU-CAP sampler, and 1.46 times more for the Button sampler). The literature most frequently compares the IOM sampler with the CFC: published results generally show larger differences compared with the CFC than those found during our research. There are several explanations for this difference, one of which involves CFC orientation during sampling. It has been shown that concentrations measured using a CFC are dependent on its orientation. Different CFC positions from one sampling session to another are therefore likely to cause differences during CFC-IOM sampler comparisons.

THE USE OF A NEW STATIC DEVICE BASED ON THE COLLECTION OF THE THORACIC FRACTION FOR THE ASSESSMENT OF THE AIRBORNE CONCENTRATION OF ASBESTOS FIBRES BY TRANSMISSION ELECTRON MICROSCOPY

A new static device, the CATHIA sampling head, based on the collection of the thoracic fraction is proposed for the assessment of the airborne concentration of asbestos fibres by transmission electron microscopy. By comparison with a standard aerosol sampling head, it has been shown that this sampler reduces the total mass concentration, but does not introduce any change in the most common index used to characterize an asbestos aerosol fibre, that is the concentration of fibres with length greater than 5 microns, diameter less than 3 microns and length to diameter ratio greater than 3. The homogeneity of the deposited dust on the collection filter favours the use of this sampling head with both the indirect and direct preparation methods.

Producing Samples for the Organization of Proficiency Tests. Study of the Homogeneity of Replicas Produced From Two Atmosphere Generation Systems

This article describes two atmosphere generation systems used for the production of replicas. The first, the Sputnic system, is based on the Sputnic air sampler developed by the National Institute of Occupational Health in Oslo (Norway). It is used to generate asbestos fibres or silica particles and allows the simultaneous production, by means of sampling on filters, of up to 114 replicas. The second is a multipurpose system that allows dust sampling on foams used with the CIP 10-R device. Twenty samples can be taken simultaneously. In total, 120 series of samples allowed characterization of the variability of the two generation systems used for the production of replicas loaded with asbestos fibres or silica dust. The coefficients of variation characterizing the dispersion of the filter loading in the Sputnic system are <10% for high densities asbestos fibre or silica dust samples. The coefficient of dispersion is on average higher when the asbestos fibre density is lower. The differences observed between the measurements taken on the different crowns of the Sputnic system are low and <2%. The results obtained with the multipurpose system show that replica dispersion is on average equal to 4%, which will allow proposal in the near future of a proficiency test dedicated to the quantitative analysis of crystalline silica on foams sampled with the CIP 10-R device.

Validation of the Analysis of Respirable Crystalline Silica (Quartz) in Foams Used with CIP 10-R Samplers

Sampling the respirable fraction to measure exposure to crystalline silica is most often carried out using cyclones. However, low flow rates (<4 l min(-1)) and continuing improvement in workplace hygiene means less and less material is sampled for analysis, resulting in increased analytical uncertainty. Use of the CIP 10-R sampler, working at a flow rate of 10 l min(-1), is one attempt to solve current analytical difficulties. To check the ability of the analysis of quartz sampled on foams, known amounts of quartz associated with a matrix have been injected into foams. The results obtained show that the proposed protocol, with prior acid attack and ashing of the foams, satisfies the recommendations of EN 482 Standard [CEN. (2006) Workplace atmospheres-general requirements for the performance of procedures for the measurements of chemical agents. Brussels, Belgium: EN 482 Comité Européen de normalization (CEN).], namely an expanded uncertainty of <50% for quartz weights between 0.1 and 0.5 times the 8-h exposure limit value and <30% for quartz weights between 0.5 and 2 times the 8-h exposure limit value, assuming an exposure limit value equal to 0.1 mg m(-3). Results obtained show that the 101 reflection line allows a quartz quantity of the order of 25 μg to be satisfactorily measured, which corresponds to a 10th of the exposure limit value, assuming an exposure limit value of 0.05 mg m(-3). In this case, the 100 and 112 reflection lines with expanded uncertainties of ~50% would also probably lead to satisfactory quantification. Particular recommendations are also proposed for the preparation of calibration curves to improve the method.

Effect of Using a Cowl When Measuring the Fiber Number Concentration by the Membrane Filter Method

This article compares samples taken with three different sampling heads: (1) open-faced sampling head, (2) open-faced sampling head with stainless-steel extension cowl, and (3) open-faced sampling head with graphite-impregnated extension cowl. Sampling was performed in three factories producing man-made mineral fibers (alkaline silicate fibers, refractory ceramic fibers, glass fibers). Flow rate was varied (1 L/min or 2 L/min). The average densities measured on the sampling filter for fibers of <3 μm in diameter varied from 19 to 91 fibers/mm 2 . No significant difference in fiber density was observed in relation to the nature of the sampling head for fibers with a diameter of <3 μm. The deposits noted on the internal walls of the cowl were small and much less than that documented in the literature for man-made mineral fibers until now. They were greater for fibers with a diameter of >3 μm than for those with a diameter of <3 μm. For large-diameter fibers, it would appear that cowl deposit can be reduced by increasing the sampling flow rate.