In this 16-month long investigation, we assessed the efficacy of engineering controls employed to reduce concentrations of airborne PCBs inside of a school. We found that improving ventilation with outdoor air and encapsulating PCB-containing caulk located on the inside face of the building envelope yielded reductions of PCBs levels in indoor air. Implementation of these engineering controls throughout the school also led to a reduction in the variability of indoor air PCB concentrations in classrooms. After adjusting ventilation and erecting physical barriers over the interior-facing PCB caulk, indoor air PCB levels were found to be directly related to temperature.
The mitigation methods employed in this school building reduced indoor air PCB concentrations by approximately 87% from initial concentrations of 299 - 1,800 ng/m3
. This reduction was meaningful as concentrations were brought below health protective guidelines for schools suggested by EPA [16
]. Moreover, the resulting concentrations were below the annual average guideline value of 230 ng/m3
, that was derived through a site-specific risk assessment for this school. The initial concentrations in the school were similar to airborne indoor PCB concentrations reported for other buildings constructed with PCB-containing caulk, e.g., 120 - 320 ng/m3
in a laboratory [11
], 715 - 2253 ng/m3
in unidentified types of buildings [22
], < 100 - 3,000 ng/m3
for public buildings in Switzerland [18
], 11 - 393 ng/m3
in a university building [8
] and 78 - 1406 ng/m3
in New York City schools [23
]. Consequently, the mitigation methods found to be effective here may also be efficacious elsewhere.
Minimal disruption of operations is an important objective of maintenance, construction, and remediation activities in buildings. The modification of ventilation systems and encapsulation of source materials described in this paper were conducted outside of school hours and therefore did not interfere with regular school activities. Common remediation methods for PCBs in buildings materials involve removal or chemical treatment of PCB caulk and surrounding materials such as brick, concrete, and mortar [24
]. Noise, dust, vibration, chemicals, and mechanical removal of building components associated with these abatement methods often require buildings to be evacuated for weeks or months. Our findings demonstrate that risks of PCBs can be managed without significant disruption of building operations. The benefits of mitigation of PCB-containing materials rather than abatement have been noted previously [25
]. Less disruption is expected to translate to lower costs of remediation although to our knowledge a rigorous comparison has not yet been made for costs of mitigation and abatement of PCBs in building materials.
After improving ventilation and encapsulating interior PCB-containing caulk throughout the school, concentrations of PCBs in indoor air did not vary significantly across rooms in the building but were found to vary significantly over time. Subsequent analyses demonstrated that variation in ambient temperature explained 79% of the temporal variability of school-wide average indoor air concentrations of PCBs. An association between PCB levels in indoor air and temperature has been reported previously. In one room of a prefabricated concrete office building sealed with PCB-containing caulk, indoor air PCB levels were reported to be associated with ambient temperature [12
]. Likewise, Kohler et al. [18
] found a 'slight trend' between indoor airborne PCB levels and room temperature, although the analysis could not account for other factors such as air exchange rate. Seasonal differences in indoor air concentrations of PCBs have been ascribed to temperature as well [7
In our investigation, PCB concentrations (natural log ng/m3) in indoor air decreased by 6.2 x103 for every 1/K decrease in ambient temperature above 20°C. According to the Clausius-Clapeyron equation and our estimated heat of evaporation of 82 kJ/mol for Aroclor 1260, natural log PCB vapor pressure would increase by -9.8 × 103 per 1/K decrease in temperature. The difference between the theoretical and observed slopes may result from differences between ambient temperature and the temperature of materials inside the building to which PCBs are adsorbed. When ambient temperatures were below 20°C and classroom temperatures were maintained by convective heating, PCB levels in indoor air were effectively constant. This observation indicates that airborne PCB concentrations in the building are only indirectly related to ambient temperature and instead directly a function of temperature in the indoor environment.
Additional investigation of residual sources of PCBs, including PCBs absorbed into other building materials over time and the effect of temperature on flux or concentration, may be useful in understanding ongoing PCB indoor air contamination following remediation. In this school, the continued presence of indoor air PCB levels above corresponding concentrations in outdoor air indicate that residual sources of PCB emissions remain within the building. Identification of ongoing PCB releases in the school was outside the scope of the present investigation. However, screening for potential source materials conducted at the outset of the investigation revealed PCBs at concentrations below 50 parts per million in bulk samples of ceiling tiles and mastic below floor tiles. These materials could be contributing to the residual PCBs observed in indoor air of the school.
The empirical findings from this longitudinal assessment have significant implications for management of PCB-containing building materials. First, relatively low cost engineering controls were found to reduce indoor air concentrations of PCBs substantially. Moreover, these mitigation methods were shown to be effective for at least 1-year. Management in-place methods such as these avoid the disruption and higher costs of demolition, disposal, and reconstruction required when PCB-containing building materials are removed from a school. Second, we found that short-term measurements, such as over a single day or season, may not be representative of long-term average or typical conditions in a building because ventilation and/or ambient temperature during sampling can have a strong influence on PCB levels in indoor air. The reliability of short-term measures to represent long-term average exposure is important in this situation because health risks of PCBs are considered to be a function of chronic exposure [26
]. Air sampling conducted when ventilation systems are not fully operational or when temperatures are elevated may lead to overestimates of risk. Conversely, air sampling conducted when ventilation rates are atypically high or temperatures are atypically low could lead to an incorrect determination that no actions are indicated based upon a measurement from a single point in time. In either case, risk assessment results based on short-term measurements may be erroneous and could lead to sub-optimal allocation of resources. To assess long-term average exposure more accurately, the ventilation system should be operated during testing in a manner that it is intended to operate in the future. Likewise, temperatures at the time of sampling should be similar to school-year average temperatures.