Effects of Surface Type and Relative Humidity on ... - ACS Publications

Apr 26, 2001 - Experiments were conducted in this system to investigate the effects of three surface types (Teflon, wallpaper, and carpet) and two lev...
0 downloads 0 Views 76KB Size
Download PDF
Environ. Sci. Technol. 2001, 35, 2200-2206

Effects of Surface Type and Relative Humidity on the Production and Concentration of Nitrous Acid in a Model Indoor Environment T H O M A S W A I N M A N , †,‡ CHARLES J. WESCHLER,§ P A U L . J . L I O Y , † A N D J U N F E N G Z H A N G * ,† Environmental and Occupational Health Sciences Institute, UMDNJ-Robert Wood Johnson Medical School and Rutgers University, Piscataway, New Jersey 08854

A nested chamber design was constructed for the purpose of studying parameters that affect indoor air chemistry. Experiments were conducted in this system to investigate the effects of three surface types (Teflon, wallpaper, and carpet) and two levels of relative humidity (50% and 70% RH) on the formation of gas-phase nitrous acid (HONO) through the heterogeneous reaction of nitrogen dioxide (NO2) with sorbed water vapor. The results of this investigation show that, compared with Teflon surfaces, carpet made of synthetic fibers increased the NO2 surface removal rate by nearly an order of magnitude and resulted in higher peak HONO concentrations. The results also suggest that the capacity of a surface to sorb water will determine if HONO is released from that surface after the NO2 source has been turned off and the heterogeneous reaction between NO2 and sorbed water is no longer significant. Vinyl-coated wallpaper was found to release HONO for prolonged periods of time after the NO2 source was turned off at both 50% and 70% RH whereas Teflon was found to do so only at 70% RH. The results of this investigation also demonstrate the utility of the nested chamber design in investigating indoor air chemistry.

Introduction It has been well-established that the use of unvented combustion appliances in homes, e.g., gas stoves and heaters, results in high indoor concentrations of nitrogen dioxide (NO2) (1-21). Nitrogen dioxide is of concern in indoor air due to its irritating effects on the respiratory tract (17, 21, 22). Time-averaged concentrations of indoor NO2 are often higher than time-averaged concentrations of outdoor NO2, especially when combustion appliances are present, and typically range from about 30 to 40 ppb with occurrences as high as 100 ppb. Real-time NO2 monitors placed near operating gas stoves in homes have measured NO2 concentrations as high as 1500 ppb (9). Most of the studies cited in which indoor NO2 measurements have been made have not considered the reactivity * Corresponding author telephone: (732)445-0158; fax: (732)4450116; e-mail: [email protected]. † UMDNJ-Robert Wood Johnson Medical School and Rutgers University. ‡ Present address: Oak Ridge National Laboratory, Oak Ridge, TN. § UMDNJ-Robert Wood Johnson Medical School. 2200

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001

of NO2 in indoor air. It has been established that both nitrous acid (HONO) and nitric acid (HNO3) are generated by the heterogeneous reaction of NO2 with water vapor (H2O) (23, 24). Chamber studies in which NO2 was introduced into humidified air have shown that HONO concentrations increase in a fashion that matches the decrease in NO2 concentrations (25, 26). Similar results were obtained when NO2 was introduced into a mobile laboratory either from a NO2 cylinder or through the operation of a gas stove (27, 28). There have been a number of chamber studies that have investigated the mechanism and kinetics of the heterogeneous reaction between NO2 and H2O (25, 26, 29-35). These studies have shown that, under the conditions investigated, the reaction is first order with respect to both NO2 and H2O. Using a mass balance approach to study this reaction, Febo and Perrino have suggested the following stoichiometry (36):

2NO2 + H2O T HONO + HNO3

(1)

Under the experimental conditions of Febo and Perrino, 98% of the HONO was released from the surfaces into the gas phase while all of the HNO3 remained adsorbed on surfaces To date, the data indicate that the formation of HONO occurs by the following steps: (i) diffusion of NO2 to the surface, (ii) reactive dissolution of NO2 into aqueous surface films, (iii) formation of reactive intermediate species, (iv) formation of HONO and HNO3, and (v) desorption of HONO to maintain a dynamic equilibrium between sorbed and gaseous phases. Step iii, the formation of reactive intermediates, is thought to be the rate-limiting step in this reaction sequence. Several studies have made simultaneous measurements of NO2 and HONO inside homes and other indoor environments (36-43). The results show that HONO concentrations are higher in indoor environments in which an unvented combustion source is operated. Time-averaged HONO concentrations are typically on the order of 5-15 ppb, but peak HONO concentrations as high as 100 ppb have been measured. Several investigations found that the indooroutdoor ratio of HONO exceeds unity even when a combustion source has not been active (37, 38, 43). These findings suggest that indoor HONO may be produced from ambient NO2 infiltrating into a house. The health effects of HONO exposure have been investigated in two published studies. In both studies, human subjects were exposed to HONO concentrations higher than typically present in indoor air. The results indicate that, for the exposure conditions investigated, HONO has some irritant effects on mucous membranes and also has some impact on lung function (44, 45). While HONO is known to photodissociate quickly in outdoor air, it will have a much longer lifetime in indoor environments where it is not exposed to direct sunlight. The dissociation of even a small fraction of indoor HONO to produce hydroxyl radicals would have a major impact on indoor air chemistry. The fate of HONO in indoor air remains to be investigated.

Experimental Section The purpose of the present study is to examine the effect of surface type and relative humidity on HONO formation in a model indoor environment. Two conditions of relative humidity (RH) and three types of surfaces were examined, resulting in eight possible combinations of relative humidity and surface type (Table 1). The four surface conditions examined were Teflon, Teflon with carpet, wallpaper, and 10.1021/es000879i CCC: $20.00

 2001 American Chemical Society Published on Web 04/26/2001

0.0056, 0.0069 -0.18, -0.40 0.89, 1.1 136, 147 13.7, 11 0.95 No data. a

with carpet

0.31 0.31 0.52 0.39 70% RH 5 6 7 8 without carpet

with carpet

0.29 0.29 0.29 0.39 50% RH 1 2 3 4 without carpet

expt conditions

expt no.

75 a 155 7.50 145 7.00 130, 110 11, 12

-0.03 -0.01 -0.07 1.2, 1.3

-0.0002 a -0.0001 -0.27 -0.0004 -0.34 0.0075, 0.0081 -0.18, -0.22

15

-0.43 -0.35 0.0006 0.0003 0.10 0.04 6.5 8.0 135 120 0.59 0.59

-0.78 -0.30 0.0075 0.0061 1.2 0.97 12 14 130 100 0.92 0.53

11 12 70% RH 13 14

-0.46 -0.25 0.0004 0.0004 0.06 0.06 5.5 7 140 135 0.59 0.59 50% RH 9 10

0.0009 0.03 0.0005 -0.06 0.0006 0.01 0.0075, 0.0069 a, 0.05 0.15 0.08 0.10 1.2, 1.1

expt no. HONO surface removal rate (h-1) NO2 deposition velocity (cm s-1) NO2 surface removal rate (h-1) peak HONO (ppb) peak NO2 (ppb) air exchange rate (h-1)

125 10.50 145 6.50 140 8.00 130, 130 12, 16.8

HONO surface removal rate (h-1) NO2 deposition velocity (cm s-1) NO2 surface removal rate (h-1) peak HONO (ppb) air exchange rate (h-1)

peak NO2 (ppb)

wallpaper-lined chamber Teflon-lined chamber

TABLE 1. Peak Concentration and Removal Rates for NO2 and HONO

wallpaper with carpet. Each surface condition was examined at 50% and 70% RH. The experiments were initiated by introducing NO2 into the inner chamber from a certified cylinder of 2400 ppm NO2 in air (Air Products, Lehigh Valley, PA) at a flow rate of approximately 15 mL/min until a nominal NO2 concentration of 140 ppb was reached. No measurement of the NO2 mass injected into the inner chamber was made. The NO2 in the cylinder was tested for HONO contamination by diluting it to 0.25, 1, and 5 ppm and monitoring for HONO as described in the Measurement of NO, NO2, and HONO section. There was no detectable HONO at these concentrations, indicating that any HONO contamination in the cylinder is negligible in comparison to the amount to be expectedly formed from the reaction of NO2 with water vapor. Data were acquired using a Macintosh Quadra 950 computer equipped with a National Instruments NB-MIO16L data acquisition board and a model NB-DSP-2300 math coprocessor. A data acquisition program was written in the National Instruments LABVIEW graphical programming language. A data point was collected once per second and averaged over 60-s intervals during the experiments. Design of the Model Indoor Environment. The experiments were conducted in a two-stage environmental chamber, consisting of a smaller chamber nested inside a larger one. This is a dynamic system in which air exchange occurs between the two chambers. The outer chamber is the Controlled Environmental Facility (CEF) located at the Environmental & Occupational Health Sciences Institute (EOHSI) at Rutgers University, Piscataway, NJ. The CEF is an exposure chamber designed for human studies and was utilized to maintain constant temperature and humidity for each experiment and to provide an activated carbon/HEPAfiltered air supply free from volatile organic contaminants and ozone. The CEF has a volume of 25 m3 and was operated at approximately 47 air exchanges per hour (ach) at 75 °F (24 °C). The inner chamber was constructed specifically for this and other indoor air quality experiments (46) and was operated at an air exchange rate below 1 ach. The inner dimensions of this chamber are 181 cm long by 120 cm wide by 115 cm high with a volume of 2.5 m3, a surface area of 11.1 m2, and a surface to volume ratio (S/V) of 4.4 m-1. To put this value in perspective, the typical S/V for a furnished room is in the range of 2.5-3.5 m-1. The inner chamber has a modular design for easy assembly and disassembly. The wall panels of the inner chamber are constructed from 3/8-in.thick gypsum wallboard. Strips of u-shaped polyethylene channel were placed over the edges of the panels to maintain their integrity. One side of the panels was covered with virgin electrical grade PTFE Teflon, 1/32 in. thick. The Teflon was attached to the panels by stapling it to the polyethylene channel around the perimeter of the panels. The other side of the panels was covered with vinyl wallpaper (Sunwall of America, Borden Inc.). The wallpaper was fastened directly to the panels with staples. No adhesives were used in the chamber construction. Both the Teflon and wallpaper surfaces were initially “conditioned” prior to the experiments by repeatedly exposing them to ozone in excess of 1 ppm for several hours. This conditioning should have removed any unreacted organics on the surfaces that may have potentially reacted with NO2. Each time the inner chamber was set up or the panels were turned around, the experimental surface was rinsed with distilled and deionized water and dried with lint-free paper. Experiments were conducted using either the Teflon surface or the wallpapered surface by simply turning the panels around. The panels were held in place by a frame constructed from 1/8-in.-thick aluminum (alloy 6063). The inner chamber was not constructed to be airtight; small gaps between the panels allowed air exchange with the outer chamber (the CEF). The air exchange rate depended upon how tightly the panels fit together after each VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2201

assembly. Four 1.5-in. square brushless 12VDC fans rated at 5.0 cfm were suspended from the ceiling of the inner chamber with stainless steel wire to mix the air within the chamber. The linear decay curves of spiked methane in the inner chamber indicate that the air in the inner chamber is reasonably well-mixed under the experimental conditions. Air sampling ports were installed in the side walls and ceiling by boring 3/8-in.-diameter holes through the panels with a cork borer. Teflon sample lines were inserted through the holes for sample collection. For the experiments in which carpet was a test surface, a 181 cm × 120 cm piece of level loop carpet (Liberty Carpet Co., Inc., Hi-Tech Collection) was placed on the chamber floor. The carpet is made of synthetic fibers from recycled carpet remnants. The carpet was “conditioned” prior to the experiments by repeatedly exposing it to ozone in excess of 1 ppm for several hours. This conditioning presumably removed most unreacted organics from the carpet backing, reducing the potential for side reactions with NO2. The premise behind utilizing a nested chamber approach is that the CEF will serve as a model for the ambient atmosphere and the inner chamber will serve as a model indoor environment. Measurement of Air Exchange Rate. The air exchange rate in the inner chamber was determined by measuring the decay of methane injected into it over time. The initial methane concentration was between 10 and 20 ppm. A GowMac model 23-500 total hydrocarbon analyzer was used to monitor the methane concentration on a real-time basis for a period ranging from 1 to 2 h. The air exchange rate was determined from the slope of the linear regression obtained by plotting the natural log of the methane concentration against time. An air exchange rate determination was made prior to and immediately after each set of experiments. Measurement of NO, NO2, and HONO. The analysis of gaseous nitrogen species [nitric oxide (NO), nitrogen dioxide (NO2), and nitrous acid (HONO)] was performed using two Monitor Labs model 8840 chemiluminescent nitrogen oxides (NOx) monitors, each providing a channel with a molybdenum (Mb) catalyst. Several investigations have shown that a chemiluminescent NOx monitor responds linearly and quantitatively to both HONO and HNO3 (47-49, 37, 39). NOx monitors have performed well in studies designed to compare measurement methods for HONO and HNO3 in ambient air (50, 51). NOx monitors can be used to measure HONO and HNO3 by placing denuder tubes that remove one or both of these gases from the sample airstream in the sample line. Denuder tubes coated with sodium chloride (NaCl) have been shown to be effective in removing HNO3 from the sample airstream; and denuders coated with sodium carbonate (Na2CO3) have been shown to remove both HONO and HNO3 with high efficiency (52-55, 37). When two NOx channels with Mo catalysts are operated simultaneously, HONO can be measured by placing a Na2CO3 denuder in one of the channels and a NaCl denuder in the other. The difference in the response of the two channels represents the HONO concentration. For this investigation, a denuder tube coated with 1% (w/v) Na2CO3 in 50/50 methanol-water was placed in one channel, and a denuder tube coated with 0.7% (w/v) NaCl in methanol solution was placed in the other. The two chemiluminescent NOx monitors were simultaneously calibrated prior to each experiment utilizing a Monitor Labs model 8550 dynamic calibrator and a certified gas standard containing 100 ppm (nominal) NO in N2 as per ASTM Method D3824-91. The detection limit for HONO and HNO3 was determined by monitoring the difference between the responses of the two Mo channels to NO2 over time. NO2 was injected into the Teflon-lined chamber and continuously monitored over a 40-min period. A data point was collected once per second, 2202

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001

and an averaged measurement was recorded every 30 s. In the experiments to determine detection limits, the denuders were not installed. The intent was for each of the Mo channels to have the same response. Any difference between the responses of the two channels under these conditions would be due to electronic noise. Since the HONO measurement is the difference in the response between the two Mo channels, the HONO detection limit will be determined by the difference in the response of these channels attributed to such noise. The NO2 underwent decay due to air exchange and reaction with water vapor, resulting in the same dynamic response by the Mo channels as experienced during the experiments. On the basis of the mean difference between the channels plus three standard deviations, a detection limit of 3.4 ppb has been calculated.

Results Peak Concentrations and Removal Rates. A total of 15 experiments was performed to investigate the differences in HONO formation among 8 possible combinations of surface type and relative humidity. The conditions of these experiments are shown in Table 1. Air exchange rates in the inner chamber ranged from about 0.3 to 0.8 ach. The air exchange rates for the experiments involving the wallpapered surfaces were consistently higher than for the experiments involving the Teflon surfaces. This is due to the fact that the panels did not fit together as tightly when the wallpaper side of the panels was faced in as compared to when the Teflon side was faced in. There were measurable HONO concentrations in all of the experiments in which HONO data was acquired (no data was obtained in experiment 5 due to an instrument malfunction). The NO2 injections resulted in peak NO2 concentrations ranging from 75 to 155 ppb with a mean of 130 ( 19 ppb. The resulting peak HONO concentrations ranged from 5.5 to 16.8 ppb. The peak NO2 and HONO concentrations are presented in Table 1. The data for experiments with two serial NO2 injections are separated by a comma; the first value represents the first injection, and the second value represents the second injection. The concentration-time profile of NO2 and HONO for experiment 1 is shown in Figure 1 and is a representative example. The slope of the linear regression of the natural log of the concentration against time was used to determine the overall removal rate for NO2 and HONO during the decay period in each experiment. The air exchange rate was subtracted from the overall removal rate to obtain the surface removal rate. The surface removal rates for NO2 and HONO are presented in Table 1. The removal rates were converted to deposition velocities by dividing them by the nominal surface area to volume ratio (S/V) of 4.44 m-1 and are also presented in Table 1. It is important to note that all of the information obtained from these experiments is related to how the two species, NO2 and HONO, were removed from the chamber after the NO2 concentration had peaked. The exact quantity of NO2 injected in each experiment to reach the peak NO2 concentration was not measured. NO2 Removal. For the experimental conditions tested, the major determining factor in the NO2 surface removal rate was the presence of the synthetic carpet. Without carpeting, the NO2 surface removal rate ranged from - 0.07 to 0.15 h-1, corresponding to deposition velocities of - 4 × 10-4 to 9 × 10-4 cm s-1. (The negative values may represent measurement errors or a small release from the surface). When the carpet was introduced into the chamber, the NO2 surface removal rate increased approximately an order of magnitude, ranging from 0.89 to 1.3 h-1 (deposition velocity from 56 × 10-4 to 81 × 10-4 cm s-1) with a mean of 1.12 ( 0.13 h-1 (deposition velocity (70 ( 8) × 10-4 cm s-1). The reason for this increase may be the increase in surface area

FIGURE 1. HONO formation in Teflon-lined chamber at 50% RH (experiment 1).

TABLE 2. Estimated NO2 Deposition Velocities Due to Carpet surface RH Teflon 50% RH 70% RH wallpaper 50% RH 70% RH

removal rate (h-1)

deposition velocity (cm s-1)

1.06 1.25

0.033 0.039

1.04 0.94

0.032 0.029

presented by carpet fibers. Estimates of the NO2 surface removal rate attributable to the carpet alone were made by subtracting the area corrected removal rate without the carpet, under the same experimental conditions, from the removal rate with the carpet. This difference was then divided by the area of the carpet ratio to the volume of the chamber to obtain the deposition velocity attributable to the carpet alone. These estimates are presented in Table 2 and are consistent under different RH and surface conditions. HONO Production/Removal. An examination of the data in Table 1 shows that the HONO concentrations were consistently higher when the synthetic carpet was placed in the model indoor environment. There are two plausible explanations for this observation. First, the carpet increased the surface area available for the reaction to occur, resulting in higher HONO concentrations. Second, since the NO2 deposition velocity was increased by an order of magnitude when the carpet was present, a larger mass of NO2 had to be injected into the chamber in order to achieve a peak concentration comparable to the experiments without the carpet. Injecting a larger mass of NO2 (reactant) would expectedly produce a larger HONO concentration. Since the mass of NO2 injected to achieve the desired peak concentration was not measured, the relative contribution of these factors to the increase in the HONO concentration cannot be determined. While the experiments with the carpet resulted in higher peak HONO concentrations, an examination of Table 1 shows that the carpet had no effect on the removal rate of HONO for any of the experimental conditions.

The data in Table 1 also shows that in 10 experiments, the HONO removal rate was slower than the air exchange rate, while in four experiments, the HONO removal rate was comparable to the air exchange rate. The four experiments were all conducted in the Teflon-lined chamber at 50% RH and suggest that, under these conditions, the rate of gasphase HONO production was much slower than the air exchange rate once the NO2 source was turned off. The results of the other 10 experiments, however, indicate that gas-phase HONO continued to be produced at a rate greater than the air exchange after the NO2 source was turned off.

Discussion Perhaps the most interesting aspect of the data collected is that in some of the experiments, HONO continued to be released from the surfaces after the NO2 concentration reached its peak and began to decay while in others it did not. A similar duality in the HONO removal rate was found in an indoor study conducted in two homes in which unvented combustion appliances were operated as a source of NO2 (37). The HONO removal rate was the same as the air exchange rate in one of the homes and less than the air exchange rate (indicating HONO emission) in the other. Another HONO investigation conducted in a research home, which introduced NO2 by both injection from a gas cylinder and through the use of unvented combustion appliances, also found the HONO removal rate to be about the same as the air exchange rate (39). There is evidence in the literature suggesting that some surfaces act as a reservoir for HONO and continue to release it into the gas phase even after the NO2 concentration has diminished. Two indoor HONO investigations measured the NO2 and HONO concentrations before, during, and after the operation of a gas stove (39, 36). In each of the studies, a window was opened after the gas stove was turned off, which allowed the indoor NO2 and HONO concentrations to decay to the much lower outdoor levels. After closing the window, the HONO concentration rapidly increased while the NO2 concentration did not. Both studies found persistent, longlived HONO concentrations in indoor air even when the NO2 concentrations were negligible. One of the studies also VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2203

FIGURE 2. HONO formation in wallpaper-lined chamber with carpet at 50% RH (experiment 12). included a laboratory investigation into the effect of wallpaper on HONO formation (36). An annular flow reactor was lined with wallpaper, and NO2 was passed through the reactor in a humidified stream of N2. At a low NO2 flow rate (100 mL min-1), all of the HONO remained sorbed by the wallpaper. Upon turning off the NO2 and passing N2 through the reactor at an increased flow rate (1500 mL min-1), HONO was quantitatively released from the wallpaper, and an exponential decay was observed. These studies clearly demonstrate the continued release of HONO from surfaces even after nearly complete decay of the NO2 concentration. A comparison of the experimental results for the TFElined chamber at 50% and 70% RH (Table 1) shows that there was a prolonged HONO release from the TFE at 70% RH but not at 50% RH. Svensson et al. showed that FEP Teflon film absorbs approximately three times as much water at 70% RH than at 50% RH (if the surfaces were perfectly smooth, about 10 monolayers vs 30 monolayers) (32). The PTFE Teflon utilized in the current investigation has about the same water uptake capacity as the FEP Teflon investigated by Svennson et al. (DuPont Corp.) and so should also have sorbed more water at 70% RH than at 50% RH. This suggests that the amount of water sorbed on a surface may be important in sustaining the release of HONO from that surface after the NO2 concentration has decreased to a negligible amount. While numerous investigations have shown that the rate of HONO formation is first order with respect to both water and NO2, little information exists concerning the prolonged release of HONO from surfaces after the NO2 source is turned off. It is the authors’ hypothesis that the amount of water sorbed to or condensed on a surface is the key factor in determining the prolonged release of HONO from that surface. The authors suggest that the following equilibrium relationships are involved in determining the amount of HONO released from a surface:

surface

2NO2(g) + H2O(aq) 798 HONO(aq) + H+ + NO3- (2) HONO(aq) T H+ + NO2-

2204

9

KA ) 5.1 × 10-4

(3)

HONO(aq) T HONO(g)

(4)

2HONO(aq) T NO2(g) + NO(g) + H2O(aq)

(5)

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001

where (g) and (aq) denotes the species in the gas and aqueous phases, respectively. While reaction 2 occurs too slowly in bulk solution to explain the rapid rate of HONO formation (56), it is generally accepted in the literature that this is a heterogeneous reaction that may involve surface catalysis (29-35, 26). Since HONO(aq) concentration is dependent on pH, as shown in reactions 2 and 3, the acidity of surfaces should play a role in determining prolonged HONO release via reaction 4. In accordance with Henry’s law, the maximum NO2 concentration in water sorbed on a surface will occur at the maximum NO2 concentration in the air. If a surface has only a small capacity to sorb water, such as Teflon at 50% RH, then the mass of dissolved nitrogen species (NO2 and nitrogen-containing products) will also be small. The data from the current and previous studies in which the HONO removal rate was the same as the air exchange rate suggest that the surfaces involved only had a small capacity to sorb water, at the experimental RH, and therefore retained only small quantities of dissolved nitrogen species. There may have been a prolonged release of HONO from the surfaces in these studies but at a rate that was much less than the rate of air exchange, which made the release undetectable. Surfaces that readily sorb water, such as certain wallpapers, will have larger quantities of dissolved nitrogen species that may continue to release HONO over prolonged periods of time at a rate that is greater than the typical rate of air exchange, making the release detectable. An isotopic study in which 18O-labeled water was reacted with NO2 lends strong support to the proposed mechanism (32). In addition to the exclusive formation of 18O-labeled HONO, 18O-labeled NO2 was also produced, consistent with the occurrence of reaction 5. Further support for the mechanism comes from a recent study which found that HONO formation results from the reaction of gas-phase NO2 with water adsorbed on surfaces. It was concluded that adsorbed water itself can participate in surface reactions (e.g., reaction 2) and is not just a medium for the adsorption and reaction of ionic species (29). The concentration of HONO in a room or a chamber at any time therefore will be dependent not only on its production via reaction 2 and removal by air exchange but also on its rate of release from surfaces. Surfaces with the ability to sorb sufficient quantities of water serve as HONO reservoirs and are responsible for maintaining elevated

HONO concentrations even after the NO2 concentration has decayed, as illustrated in Figure 2. In summary, the results of this investigation clearly show that synthetic carpet increases the NO2 surface removal rate by nearly an order of magnitude. Previous studies have shown the occurrence of a prolonged release of HONO from surfaces even after the NO2 concentration has decreased to near zero (39, 36), while other studies have not observed a prolonged release (37, 39). The results of the current study suggest that the capacity of a surface to sorb water, at a given relative humidity, will determine whether there is a prolonged release of HONO from that surface. The authors have proposed a reaction mechanism involving aqueous equilibria of nitrogen species to describe the occurrence of the prolonged release of HONO from surfaces and believe that there is enough supporting evidence in the literature to suggest that the reaction is surface catalyzed. The results of the experiments show that for surfaces with sufficient sorbed moisture, the HONO concentration will be dominated by its release from the surface once the NO2 concentration peaks and begins to decay. The results also suggest that the highest and most prolonged HONO concentrations occur during periods of elevated relative humidity in households that utilize unvented gas appliances and have surfaces that readily sorb water from the air. While controlled exposure studies have not revealed significant health effects resulting from high HONO exposures, the fate of HONO in indoor air warrants further investigation. Finally, the results of this investigation demonstrate the utility of the nested chamber design in investigating indoor air chemistry.

Acknowledgments Financial support for this research was provided by the Environmental and Occupational Health Sciences Institute (EOHSI). Special thanks to Dr. Mark Robson, Executive Director of EOHSI, for his friendship, advice, and support that made this research possible. J.Z. and P.J.L. are supported in part by the NIEHS Center Grant ES05022-10 to EOHSI.

Literature Cited (1) Wade, W. A.; Cote, W. A.; Yocom, J. E. J. Air Pollut. Control Assoc. 1975, 25, 933-939. (2) Palmes, E. D.; Tomczyk, C.; DiMattio, J. Atmos. Environ. 1977, 11, 869-872. (3) Goldstein, B. D.; Melia, R. W. J.; Chinn, S.; Florey, C.; Clark, D.; John, H. H. Int. J. Epidemiol. 1979, 8, 339-345. (4) Spengler, J. D.; Ferris, B. G.; Dockery, D.; Speizer, F. E. Environ. Sci. Technol. 1979, 13, 1276-1280. (5) Speizer, F. E.; Ferris, B.; Bishop, Y. M. M.; Spengler, J. Am. Rev. Respir. Dis. 1980, 121, 3-10. (6) Ryan, P. B.; Spengler, J. D.; Letz, R. Atmos. Environ. 1983, 17, 1339-1345. (7) Spengler, J. D.; Duffy, C. P.; Letz, R.; Tibbits, T. W.; Ferris, B. G. Environ. Sci. Technol. 1983, 17, 164-168. (8) Quackenboss, J. J.; Spengler, J. D.; Kanarek, M. S.; Letz, R.; Duffy, C. P. Environ. Sci. Technol. 1986, 20, 775-783. (9) Goldstein, I. F.; Andrews, L. R.; Hartel, D. Atmos. Environ. 1988, 22, 2127-2139. (10) Marbury, M. C.; Harlos, D. P.; Samet, J. M.; Spengler, J. D. J. Air Pollut. Control Assoc. 1988, 38, 392-398. (11) Petreas, M.; Liu, K. S.; Chang, B. H.; Hayward, S. B.; Sexton, K. J. Air Pollut. Control Assoc. 1988, 38, 647-651. (12) Ryan, P. B.; Soczek, M. L.; Treitman, R. D.; Spengler, J. D.; Billick, I. H. Atmos. Environ. 1988, 22, 2115-2125. (13) Brunekreef, B.; Houthuijs, D.; Dijkstra, L.; Boleij, J. S. M. J. Air Waste Manage. Assoc. 1990, 40, 1252-1256. (14) Melia, R. J.; Chinn, S.; Rona, R. J. Atmos. Environ. 1990, 24B, 177-180. (15) Adgate, J. A.; Reid, H.; Morris, R.; Helms, R. W.; Berg, R. A.; Hu, P.; Cheng, P.; Wang, O.; Muelenaer, P.; Collier, A. M.; Henderson, F. W. Arch. Environ. Health 1992, 47, 376-384.

(16) Lambert, W. E.; Samet, J. M.; Stidley, C. A. Atmos. Environ. 1992, 26, 2185-2192. (17) Samet, J. M.; Lambert, W. E.; Skipper, B. J.; Cushing, A. H.; Hunt, W. C.; Young, S. A.; McLaren, L. C.; Schwab, M.; Spengler, J. D. Am. Rev. Respir. Dis. 1993, 148, 1258-1265. (18) Levy, J. I. J. Air Waste Manage. Assoc. 1995, 48, 553-560. (19) Sohn, D. H.; Shin, W. K.; Jung, S. Y.; Jung, W. T. Jpn. J. Toxicol. Environ. Health 1995, 41, 127-133. (20) Ross, D. Environ. Technol. 1996, 17, 147-155. (21) Farrow, A.; Greenwood, R.; Preece, S.; Golding, J. Arch. Environ. Health 1997, 52, 189-194. (22) Berglund, M.; Bostroem, C. E.; Bylin, G.; Ewetz, G.; Gustafsson, L.; Moldeus, P.; Norberg, S.; Pershagen, G.; Victorin, K. Scan. J. Work, Environ. Health 1993, 19, 1-8. (23) Schwartz, S. E.; White, W. H. In Trace Atmospheric Constituents: Properties, Transformations and Fates; Advances in Environmental Science and Technology 12; Schwartz, S. E., Ed.; 1983, pp 1-116. (24) Lee, Y. N.; Schwartz, S. E. J. Geophys. Res. 1981, 86, 11971-1198. (25) Sakamaki, F.; Hatakeyama, S.; Akimoto, H. Int. J. Chem. Kinet. 1983, 15, 1013-1029. (26) Pitts, J. N.; Sanhueza, E.; Atkinson, R.; Carter, W. P. L.; Winer, A. M.; Harris, G. W.; Plum, C. N. Int. J. Chem. Kinet. 1984, 16, 919-939. (27) Pitts, J. N.; Biermann, H. W.; Tuazon, E. C.; Green, M.; Long, W. D.; Winer, A. M. J. Air Pollut. Control Assoc. 1989, 39, 13441347. (28) Biermann, H. W.; Pitts, J. N.; Winer, A. M. Advances in Air Sampling; American Conference of Governmental Industrial Hygienists; Lewis Publishers: Chelsea, MI, 1988; pp 265-289. (29) Goodman, A. L.; Underwood, G. M.; Grassian, V. H. J. Phys. Chem. 1999, 103, 7217-7223. (30) Brink, H. M. T.; Spolestra, H. Atmos. Environ. 1998, 32, 247251. (31) Kleffmann, J.; Becker, K. H.; Wiesen, P. Atmos. Environ. 1998, 32, 2721-2729. (32) Svensson, R.; Ljungstrom, E.; Lindqvist, O. Atmos. Environ. 1987, 21, 1529-1539. (33) Chan, W. H.; Nordstrom, R. J.; Calvert, J. G.; Shaw, J. H. Environ. Sci. Technol. 1976, 10, 674-682. (34) Jenkin, M. E.; Cox, R. A.; Williams, D. J. Atmos. Environ. 1988, 22, 487-498. (35) Kaiser, E. W.; Wu, C. H. J. Phys. Chem. 1977, 81, 1701-1706. (36) Febo, A.; Perrino, C. Atmos. Environ. 1991, 25A, 1055-1061. (37) Brauer, M.; Ryan, P. B.; Suh, H. H.; Koutrakis, P.; Spengler, J. D. Environ. Sci. Technol. 1990, 24, 1521-1527. (38) Brauer, M.; Koutrakis, P.; Keeler, G. J.; Spengler, J. D. J. Air Waste Manage. Assoc. 1991, 41, 171-181. (39) Spicer, C. W.; Kenny, D. V.; Ward G. F.; Billick, I. H. J. Air Waste Manage. Assoc. 1993, 43, 1479-1485. (40) Spengler, J. D.; Brauer, M.; Samet, J. M.; Lambert, W. E. Environ. Sci. Technol. 1993, 27, 841-845. (41) Weschler, C. J.; Shields, H. C.; Naik, D. V. Environ. Sci. Technol. 1994, 28, 2120-2132. (42) Allen, A. G.; Miguel, A. H. Atmos. Environ. 1995, 29, 3519-3526. (43) Lee, K.; Xue, J.; Spengler, J. D.; Ozkaynak, H.; Leaderer, B. P. In Indoor Air 99, Vol. 3; Raw, G., Aizlewood, C., Warren, P., Eds.; Construction Research Communications Ltd: London, 1999; pp 141-146. (44) Beckett, W. S.; Russi, M. B.; Haber, A. D.; Rivkin, R. M.; Sullivan, J. R.; Tameroglu, Z.; Mohsenin, V.; Leaderer, B. P. Environ. Health Perspect. 1995, 103, 372-375. (45) Rasmussen, T. R.; Brauer, M.; Kjaergaard, S. Am. J. Respir. Crit. Care Med. 1995, 151, 1504-1511. (46) Wainman, T. Use of a Two Tiered Dynamic Chamber to Investigate Indoor Air Chemistry. Doctoral Dissertation, Rutgers University, Piscataway, NJ, 1999. (47) Joseph, D. W.; Spicer, C. W. Anal. Chem. 1978, 50, 1400-1403. (48) Sanhueza, E.; Plum, C. N.; Pitts, J. N. Atmos. Environ. 1984, 18, 1029-1031. (49) Grosjean, D.; Harrison, J. Environ. Sci. Technol. 1985, 19, 862865. (50) Spicer, C. W.; Howes, J. E.; Bishop, T. A.; Arnold, L. H.; Stevens, R. K. Atmos. Environ. 1982, 16, 1487-1500. (51) Tanner, R. L.; Kelly, T. J.; Dezaro, A. D.; Forrest, J. Atmos. Environ. 1989, 23, 2213-2222. (52) Ferm, M.; Sjodin, A. Atmos. Environ. 1985, 19, 979-983. (53) Allegrini, I.; De Santis, F.; Di Palo, V.; Febo, A.; Perrino, C.; Possanzini, M.; Liberti, A. Sci. Total Environ. 1987, 67, 1-16. VOL. 35, NO. 11, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2205

(54) Sickles, J. E.; Hodson, L. L. Atmos. Environ. 1989, 23, 23212324. (55) Perrino, C.; De Santis, F.; Febo, A. Atmos. Environ. 1990, 24A, 617-626. (56) Sienfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change; John Wiley and Sons: New York, 1988; p 380.

2206

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 11, 2001

Received for review January 7, 2000. Revised manuscript received February 1, 2001. Accepted March 12, 2001.

ES000879I