Skip to content

EHP logoISEE logo

2013 Environment and Health - Basel

Abstract Number: 5138 | ID: O-2-17-06

Indoor exposure to SVOCs from consumer products and building materials: Empirical data validates air-dust partitioning models and informs measurement strategies

Robin, Dodson, Silent Spring Institute, United States; Ruthann, Rudel, Silent Spring Institute, United States

Background: Residential exposures can dominate total exposure for a variety of commercial chemicals of health concern, including flame retardants and some phthalates and pesticides. Yet methods for assessing household exposures to these and other semivolatile organic compounds (SVOCs) in large studies are limited.

Aims: 1) Validate Weschler and Nazaroff’s theoretical partitioning model relating indoor air and dust concentrations using empirical data for 83 SVOCs, and 2) recommend indoor exposure measurement strategies.

Methods: We simultaneously collected indoor air and dust samples in 170 homes and analyzed for 108 SVOCs, including phthalates, flame retardants, pesticides and PAHs. We use these data to validate a theoretical partitioning model relating indoor air and house dust concentrations. We then present factors to consider when selecting measurement strategies in exposure or health studies. We review strengths and weaknesses of dust wipes, vacuum dust, active air, and passive air sampling. Results: For model validation we used data on 47 SVOCs simultaneously detected in dust and air in at least one home and 22 detected in 50% of air and dust samples. Most were significantly positively correlated between air and dust. Ratios of measured dust and air concentrations span 6 orders of magnitude. As expected, compounds with higher log Koa have lower air concentrations relative to dust and lower detection frequencies in air. Predicted concentrations were reasonably correlated with measured (R2 ~ 0.8), with PAHs generally under-predicted whereas phthalates were more variable.


: Partitioning models allow air concentrations to be predicted from dust concentrations or the reverse. Taken together with expected concentrations, the models can be used to determine detection limits for sampling and to evaluate the potential utility of passive air sampling, which may be a promising exposure assessment strategy for large scale health studies because it can be deployed by participants and provides a more standardized measure than dust samples.