Laboratory comparison of vacuum, OSHA, and HUD sampling methods for lead in household dust

The goals of this project were to evaluate and compare the efficiency and reproducibility of three methods for sampling lead-containing dust in homes. Lead-containing dust was generated in a 1-m3 chamber and uniformly deposited onto surfaces typically found in the home (painted wood, unpainted wood, varnished wood, linoleum, and carpet). Trials with three levels of lead concentrations were performed for each surface. Replicate, side-by-side, surface samples were collected using the Occupational Safety and Health Administration (OSHA) wipe method, the Department of Housing and Urban Development (HUD) wipe method, and a vacuum-filter method. Samples were digested with nitric acid and analyzed using graphite furnace atomic absorption spectroscopy per National Institute of Occupational Safety and Health Method 7105. Recovery for the HUD method was consistently the highest on most surfaces (linoleum, 89.9 to 108.9%; painted wood, 71.2 to 153.7%; unpainted wood, 25.3 to 76.0%; varnished wood, 8.7 to 165.6%). On carpet the vacuum method had a significantly higher recovery (26.2 to 47.8%). For all sampling methods the percent recovery depended on type of surface and lead concentration. The reproducibility of percent recovery for the HUD (pooled coefficient of variation [CV] = 0.22) and OSHA (pooled CV = 0.27) methods was lower than that of the vacuum method (pooled CV = 0.46), though not statistically significant. Reproducibility for all methods did not vary significantly over surface type or lead concentration. Overall, the HUD method yielded the most accurate measurements, with recoveries closest to 100%. It was also more durable than the OSHA method, where Whatman filters were observed to tear.