Grantee Research Project Results

Final Report: A Study of Ozone Concentration Gradients in Large Buildings, Including an Examination of Indoor Chemistry, Ventilation, Occupant Health Effects and Effects on HVAC Systems

EPA Grant Number: R824796
Title: A Study of Ozone Concentration Gradients in Large Buildings, Including an Examination of Indoor Chemistry, Ventilation, Occupant Health Effects and Effects on HVAC Systems
Investigators: Spengler, John D.
Institution: Harvard University
EPA Project Officer: Chung, Serena
Project Period: October 1, 1995 through September 30, 1998
Project Amount: $425,708
RFA: Indoor Air Quality in Large Office Buildings (1995) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Air

Objective:

The objective of this research project was to study ozone concentration gradients in large buildings, including an examination of indoor chemistry, ventilation, occupant health effects, and effects in the heating, ventilation, and air conditioning (HVAC) systems.

Summary/Accomplishments (Outputs/Outcomes):

Progress has been made in five areas: (1) the single pass HVAC system model, (2) the return air HVAC system model, (3) the longitudinal study, (4) the cross- sectional study, and (5) the ozone-terpene reaction experiments.

Single Pass HVAC System Model. A model was developed to predict ozone loss in a single pass through a functioning ventilation system. During the daytime operating period, less than 20 percent of the ozone entering the HVAC is lost, as ozone contaminated air is brought from outdoors into an occupied space. Most of the ozone loss occurs as the air travels through the filtering and cooling components of the HVAC and the air handling unit (AHU); we estimate a loss of about 13 percent in our model. In the air transmission or duct system, the amount of ozone loss is not significant; we estimate about 2 ppb in our model. The high airflow velocities create small surface-to-air boundary layers, greatly reducing the amount of ozone available to react with the duct surface. At night, the conditions inside the AHU and the ducting component result in more ozone loss because of increased residence time caused by reduced air flows. The nighttime models predict that 34 percent of the ozone entering the AHU is lost, and that 21 percent of it is lost as it travels through 115 ft of ducting.

Outdoor temperature and humidity combined were found to have a small but significant effect on the loss of ozone inside the AHU; however, determining the effect was the temperature and humidity terms combined into an enthalpy term. The change in enthalpy to the air caused by the functioning of the AHU was included in the model. This factor improved the predictive ability of the model by 0.79 ppb, compared to the model with no environmental variables.

Return Air HVAC System Model. For a return air handling system, whose daytime mode of operation was to provide 20 percent constant fresh air, the return air stream contained 16 percent of the ozone of the outdoor air stream. Outdoor temperature and humidity were found to have a small but significant effect on the mean return air ozone concentration. Every 10°C decrease in outdoor temperature resulted in 0.07 ppb increase in the mean return air ozone concentration. Every 10 percent increase in relative humidity resulted in a 0.9 ppb decrease in the mean return air ozone concentration.

Although outdoor ozone concentrations during the daytime hours are higher than the nighttime concentrations, the nighttime indoor ozone concentration may be higher than the daytime hours because of a number of factors. Buildings have increased ventilation associated with economizer cycles on HVAC units, and elevated nighttime ozone periods occurring as a result of air transport from upwind urban centers (i.e., classic transport, conditions from New York City metropolitan area lead to hours of higher exposure later in the nighttime as air passes over Connecticut Valley, and western and central Massachusetts). This finding is significant for cities downwind of large urban centers, in which ozone produced during the day upwind arrives in the nighttime.

Longitudinal Study. In the longitudinal study, ozone concentrations were monitored during a 6-week period. The monitoring took place in four separate locations within a single building, all served by the same ventilation system. In total, four buildings in two cities were monitored; two in Atlanta and two in Boston. Samples also were collected at the fresh air intake, and in the supply and return air stream, as well as diffusers serving the study areas. Analysis of the Atlanta data showed indoor concentrations to be approximately 8.4 (STD 4.8) percent of outdoor air ozone concentrations. Mean indoor/outdoor ozone ratios did not significantly change through time. This indicates that HVAC mode of operation is the principle factor controlling indoor/outdoor ozone ratios. A model was developed to predict ozone concentrations within a building based on outdoor, HVAC diffuser, and return air measurements. The model was tested on data collected in the cross-sectional study. Analysis of the Boston data is ongoing.

Cross-Sectional Study. Ozone concentrations were measured in 12 buildings in 4 cities: Atlanta, Los Angeles, Baltimore, and Nashville, following the same protocol as the longitudinal study. The mean indoor/outdoor ozone ratio was 7.4 (STD 5.6) percent for the 12 buildings in the cross-sectional study. This was comparable to the values obtained in the longitudinal study. The purpose of the study was to test the model developed in the longitudinal study with the cross-sectional data from the different cities. The model effectively predicted the concentration of ozone indoors between 2 ppb and 6 ppb. The bias at the low end likely is because of the difficulties in measuring ozone concentrations below 2 ppb. The 12 buildings studied in the cross-sectional study were sampled during U.S. Environmental Protection Agency Base Building evaluations. We are evaluating whether other environmental factors collected during the measurements will improve the predictive ability of the model.

Ozone-Terpene Reaction Experiments. In a set of chamber experiments designed to determine if ozone, in combination with VOCs commonly found in buildings (terpenes), can have a significant effect on the level of ultra-fine particles in the indoor air. From the data generated in the chamber experiments, two models were developed. The first model predicts the particle number concentration at steady state for a given concentration of ozone. The implication for the first model is that ozone concentrations typically encountered in buildings are much lower than the concentrations needed to generate any noticeable particles. However, particle generation may still occur in the nighttime hours, where ozone levels may be higher than in the daytime hours, and when most of the cleaning solvents containing terpenes are used. The second model developed predicts the time to steady state given a certain ozone concentration. For buildings where ozone levels are much lower than 100 ppb, exposures of more than 1 hour are required before particle generation reaches steady state.

Supplemental Keywords:

air, ozone, indoor air, particulates, pathogens, monitoring, New York City, Massachusetts, MA., RFA, Air, Scientific Discipline, Geographic Area, Ecological Risk Assessment, EPA Region, State, particulate matter, Environmental Chemistry, tropospheric ozone, indoor air, Atmospheric Sciences, Environmental Monitoring, indoor air quality, occupant health, hvac, building related illness, monitoring, ozone concentration gradients, office buildings, particulates, ventilation rates, indoor VOC compounds, ozone, Region 1 , cross sectional study, large buildings, longitudinal studies, Massachusetts (MA), building construction

Progress and Final Reports:

Original Abstract

1996

1997