Polychlorinated biphenyls (PCBs) and endosulfan, both semivolatile organic compounds (SOCs), exhibited large diurnal variations in concentration in Bloomington, IN (see Figure ). Four samples per day were collected serially over a 5-day period in September 1994. For both SOCs, air samples collected in the morning had twice the concentration of samples collected at midnight. The diurnal variations tracked atmospheric temperature, with the highest concentrations recorded at the highest temperatures. An important implication for sampling SOCs is evident from this study. If one is attempting to measure the daily concentration of an SOC, sampling should be in multiples of 24 h; sampling for more or less time could lead to inaccurate results.

Polychlorinated biphenyl (PCB) concentrations were measured in the indoor air of several public buildings in Bloomington, IN. The concentrations ranged from 5 to 300 times higher than outdoor concentrations, and the indoor air PCB levels were highest in buildings with the earliest construction dates; see Figure . These elevated indoor air PCB levels have caused contamination in sample preparation as well as artifacts in sample collection. PCBs in procedural blanks in a laboratory with an indoor air PCB concentration of 6 ng/m3 averaged 1 order of magnitude lower than those measured in a laboratory with an indoor air PCB concentration of 300 ng/m3. Air sampling conducted on the roof of a building with high indoor air PCB concentrations and roof-top ventilation gave PCB concentrations that were approximately five times above ambient levels. However, sampling on the roof of another building without roof-top ventilation gave PCB concentrations that were not artificially elevated. We recommend measuring the indoor air of all sample preparation and collection sites to ensure that contamination can be recognized and mitigated.

The indoor air concentrations of aldrin and dieldrin, two organochlorine termiticides, were monitored in the basement, main level, and upstairs area of a treated home. Aldrin and dieldrin concentrations were measured in the home between September 1987 and April 1995. Aldrin and dieldrin concentrations were highest in the basement, where the aldrin and dieldrin had been applied. Aldrin concentrations decreased with a half-life of 16 months in both the main level and basement of the home, but dieldrin concentrations remained the same over the study period. In 1995, we expanded our measurements to include the indoor and outdoor air concentrations of a set of organochlorine compounds used as pesticides. These compounds include aldrin, dieldrin, four chlordanes, pentachloroanisole, and the hexachlorocyclohexanes. All compounds except the hexachlorocyclohexanes had higher indoor than outdoor air concentrations, implying that their sources were in the home. In the indoor air, aldrin, dieldrin, and the chlordanes had higher concentrations in the basement than on the main level, indicating a basement source. Pentachloroanisole was evenly distributed throughout the indoor air, implying that it does not originate in the basement. Two remediation attempts were made to stop aldrin and dieldrin from infiltrating into the home. First, the basement walls were covered with tar and plastic to form a barrier against the foundation. Second, the basement air pressure was manipulated. While the tar and plastic appear to have had little effect, creating a positive indoor air pressure lowered the air concentrations of the two pesticides.

We have developed an automated, low-volume air sampler, capable of collecting multiple samples for semivolatile organic compound analysis in a short time (several hours). The new sampler uses small polyurethane foam plugs as the sampling media and a sampling carousel that has a 20 plug capacity. A computer controls the sampling protocol, and after the sampler is loaded with plugs, a sampling program is carried out with no operator intervention. The plugs are removed, and the analyte is extracted by supercritical fluid extraction. A series of quality control experiments were undertaken to demonstrate that the new sampler performed properly. Air volumes in the range of 100 m3 can be collected without exceeding the breakthrough volume of the foam plugs. Passive sampling and volatilization artifacts are minimal. Concurrent air measurements using our new sampler and a conventional high-volume air sampler over a week-long period gave statistically identical (at the 95% significance level) air concentrations for polychlorinated biphenyls and endosulfan.

No abstract