United States Environmental Protection Agency Atmospheric Sciences Research Laboratory Research Triangle Park NC 27711 Research and Development .EPA-600/S3-84-114 Jan. 1985 &ERA Project Summary Hazardous Air Pollutants: Dry-Deposition Phenomena G.A. Sehmel, R.N. Lee, and T.W. Horst Dry-deposition rates were evaluated for two hazardous organic air pollutants in order to determine their potential for removal from the atmosphere by three building material surfaces. The organic air pollutants were nitrobenzene and perchloroethylene; the building materials used were cement, tar paper, and vinyl asbestos tile. Dry-deposition experi- ments were conducted in two stirred chambers that were originally designed and constructed for determining re- moval rates to vegetation in the environ- ment. Building materials were placed on the bottom of the stirred chambers, and removal rates were evaluated by in- troducing the organic air pollutant along with nondepositing SF6 tracer gas into a stirred chamber. Changes in air- borne concentrations with time were then monitored. Because the HAP removal rates were small, sulfur hexafluoride tracer gas was used to evaluate the rates of leakage from the stirred chamber in order to correct the deposition rates. When pollutant wall losses were assumed negligible, pollutant concentration decreases (corrected using the SF6 leakage rate) ranged from near 0 to 3%/h. Dry-deposition velocities were calculated from the rate-of-change over time of the airborne concentrations. All calculated deposition velocities were approximately 10~4 cm s'1 or less. This Project Summary was developed by EPA's Atmospheric Sciences Re- search Laboratory, Research Triangle Park, NC, to announce key findings of the research project that is fully docu- mented in a separate report of the same title (see Project Report ordering information at back). Introduction The role of dry-deposition in removing potentially hazardous air pollutants (HAPs) from the atmosphere is not well defined. This project was conducted to assess the magnitude of the process, especially deposition involving surfaces expected to be found in urban settings. The dry-deposition measurements were made in stirred chambers (SCs). The SC is an inert-surface box within which are placed selected deposition surfaces. The principal advantage of using the SC is that concentrations can be controlled within manageable limits throughout the duration of each experiment. During the experiments, the SCs were sealed to minimize leakage and charged with a preselected atmosphere of HAP and SFe tracer gas that was mixed continuously using an open fan. Concentrations within the SC were monitored intermittently for specified time periods depending on the concentration decay rate. Deposition rates were then calculated on the basis of concentration-decay rates. Seventeen experiments were conducted to determine the maximum removal rates for HAPs to large deposition-surface areas. Dry-deposition velocities of per- chloroethylene were evaluated during 12 experiments with cement, tile, and tar paper substrates. Deposition velocities of nitrobenzene were evaluated during five experiments with tile and tar paper sub- strates. The velocities were calculated from airborne concentration losses of HAP and SF6 tracer gas with time. The nondepositing SFe tracer gas was used to measure HAP leakage rates from the stirred chambers. These measurements of HAP dry-deposition velocities can be ------- usedjn the general predictive atmospher- ic transport, diffusion, and deposition model MPADD (Multi-Pollutant Atmos- pheric Deposition and Depletion), also developed as part of this effort. Procedure The experimental design was based on the assumption that HAP dry-deposition velocities would be comparable to those of other gases. A literature review of reported dry-deposition velocities for gases suggested that the deposition rate for many of the HAPs should range from 1CT4 to 1CT2 cm s"1. The loss rates of HAPs, as measured in SCs, were predicted to vary from 0.5% to >10% h~1 for the anticipated range of dry-deposition velocities. Thus, a stirred chamber measurement approach was selected for the experimentation. Deposition substrates were selected to represent typical construction materials: cement, tar paper, and vinyl asbestos tile. Cement was selected to represent construction materials in buildings, streets, and highways. Cement patio blocks were placed on the floor of the SCs in two configurations with different surface areas available for deposition, either 6.1 or 10 m2. Similarly, vinyl asbestos tile was selected because it is a floor covering used in industry and homes. Tar paper was selected because it has a high organic content, which might affect HAP removal rates. Both the vinyl asbestos tile and the tar paper were placed to cover the SC floor uniformly. After preliminary characterization tests with the SFe tracer were completed, experiments were begun using both the HAP and the SFe. Stirred chamber experiments were conducted .with either perchloroethylene (and SF6) or nitro- benzene (and SFe). At time zero, the chamber was charged with a preselected concentration of HAP and inert tracer gas. The concentrations within the chamber were monitored to provide a time series of the HAP and tracer decay. In some experiments, both SCs were used to conduct parallel runs to assess the reproducibility of the experiments. Results Figure 1 shows the concentration profiles of a typical experiment involving perchloroethylene and SFe. The data are normalized for presentation on the same graph by dividing each measured value by the starting concentration. Since the concentrations in the chamber tend to decrease exponentially with time, a plot of the log of the normalized concentration versus time should be linear. Loss rates of the HAP and tracer were calculated from the slope of the least-squares line which best fit the data. Both HAP and SF6 loss rates were significant and ranged from 1.44 to 7.19% h"1. Correcting the HAP loss rates for leakage (as measured by the tracer losses) the resultant HAP deposition losses ranged from near zero to 2.96% h"1. Deposition velocities were then calcu- lated from the HAP loss rates. For these calculations, HAP and tracer leak rates were assumed to be identical, and deposition in the SC to surfaces other than the selected deposition substrate (e.g., the Teflon walls) was assumed to be negligible. The results are summarized in Table 1. The HAP dry-deposition velocities were consistently small, approximately 10~4 cm s~1 or less. The observed HAP dry-deposition velocities are near the lower limit which was anticipated, based on values reported for other gases. The relatively low HAP dry-deposition velocities may have been caused by differences in substrate (the use of building materials in this study, rather than the more traditional vegetative canopies) or may result because organic gases deposit less rapidly than the inorganic gases for which most measure- ments have been made. Further research is needed to clarify these possibilities. Neither substrate area (either 6.1 or 10.0 m2 for cement) nor increased fan speed (experiments 4, 10, and 11) significantly affected HAP loss rates. Only two fan speeds were used: 520 rpm and a motor-limited maximum rpm of 600. Nearly all experiments were conducted at a fan speed of 520 rpm. Although losses tended to be exponen- tial with time, other time-dependent effects were apparent. Concentrations increased immediately after gas injection and continued to increase until a maximum concentration was observed. Subsequent- ly, concentrations tended to be exponen- tial with time as described by the equations from which dry-deposition velocities are calculated. Also, deviations from an exponential decrease were sometimes observed. These deviations 10° a 10' 10' Tracer Loss, %/h D SFe 2.04 A C2O, 2.40 I I I 40 80 Time, Hours 120 160 Figure 1. Perchloroethylene andSFe tracer gas concentration decreases for 6.1 rrf of cement block deposition surface. ------- Table 1. HAP-Dry Deposition Velocities Expt. HAP No. SC Chemical Name 1 2 4 6 7 8 9 10 11 15 16' 17 18 19 1 2 1 2 1 2 1 1 2 1 2 1 1 2 1 2 1 perchloroethylene perchloroethylene perchloroethylene perchloroethylene perchloroeth ylene perchloroethylene perchloroethylene perchloroethylene perchloroethylene perchloroethylene perchloroethylene perchloroethylene nitrobenzene nitrobenzene nitrobenzene nitrobenzene nitrobenzene Deposition Surface cement cement cement cement cement cement tile tar paper tile tar paper tile tar paper tar paper tile tar paper tile tar paper Surface Area (m*j 10.0 6.1 10.0 6.1 10.0 6.1 6.0 6.0 6.0 6.0 6.0 6.0 60 6.0 6.0 6.0 6.0 Fan Speed (rpm) 520 520 520 520 600 600 520 520 520 520 600 600 520 520 520 520 520 HAP Deposition] (%/h) 0.42 + 0.31 0.36 + 0.27 0.34 ± 0.35 0.75 + 0.12 0.72 + 0.17 0.08 + 0. 14 0.41 +0.14 0.48 + 0. 14 -002 + 0.53 0.06 ± 0.42 1.16+0.61 1.57 + 0.98 2.96 +3.60 0.01 ±1.06 0.70 +1.27 1.19 +0.69 0.68 + 0.72 Deposition Velocity] (cm/sec) 5.2 + 3.8 x 10~5 7.3 ± 5.5 x 10~5 4.2 +4.3 x 10~5 1.5 + 0.2 x 10'* 8.9 +2.1 x 10~5 1.6 + 2.8x JO'5 8.4+ 2.9 x 10~5 9.9 + 2.9 x 10's -0.0+ 1.1 x 70~" 1.2 + 8.6x 10~5 2.4 + 1.3 x 70~5 3.2 +2.0 x 10~4 6.1 +7.4 x 10'' 0.0 +2.2 x 10'* 1.4 +2.6 x 10'" 2.5+ 1.4x W" 1.4 +1.5 x 10~4 t Tabulated uncertainties represent the 90% confidence interval. might have been caused by temperature effects and sorption of gases on airborne particulates rather than on the test substrates. Conclusions and Recommendations The removal of gaseous organic air pollutants by dry deposition has not been studied extensively and few experimental data are available. Hence, obtaining dry- deposition velocities for nitrobenzene and perchloroethylene is a significant advance in the understanding of organic removal rates. Measured deposition velocities were significantly less than those usually reported for inorganic gases. Deposition velocities were consis- tently very low. Many calculated values were not significantly different from zero. Temperature effects were not intensive- ly investigated since all successful experiments were conducted in warm summer weather (ambient temperatures from 27 to 36°C.) Because the dry- deposition velocities measured for the two HAPs investigated were low, we can conclude that airborne concentrations of HAPs are not significantly reduced by dry deposition at temperatures characteristic of summer months, or at temperatures maintained within buildings. Vapor pressure, however, is anticipated to be a very important property influencing dry- deposition removal rates for organic gases. The SC approach for evaluating HAP dry-deposition velocities was adequate for the HAPs and substrates investigated. Leakage was always observed, and leakage rates tended to increase during the experimental series. To alleviate this problem, the SCs were periodically maintained by replacing critical seals or tightening the bolts holding sections of the chambers together to minimize exchange with outside air. One important advantage of the SC technique is that a large substrate surface area can be contained for the deposition substrate. Even though dry- deposition velocities might be low, they could be a function of spatial variations in substrate characteristics. Hence, by using the relatively large surface areas in an SC (6.0-m2 floor area), average dry- deposition velocities can be evaluated over these spatial variations. The authors recommend that the SC approach be used to evaluate additional HAPs and deposition substrates. However, other measurements techniques should be considered. Thermogravimetric tech- niques may be useful as a screening method to evaluate maximum removal rates for small surface areas on the order of 1 cm2. This technique may be essential for evaluating deposition velocities less than about 10~6 cm s~1. Flow-through chambers may be considered if the removal rates are significantly greater than those evaluated in the current experiments. Instrumentation with high sensitivity and rapid analysis techniques would be required if flow-through cham- bers were used. The authors also recommend that dry- deposition velocities be evaluated for additional construction materials, and for vegetative canopies and water surfaces. Currently, no data exist for predicting HAP dry-deposition velocities for vegeta- tion and soils. Dry-deposition velocities of HAPs onto construction materials appear lower than dry deposition velocities for other gases that were evaluated for vegetative canopies. ------- G. A. Sehmel. R. N. Lee, and T. W. Horst are with Pacific Northwest Laboratory, Rich/and. WA 99352. Larry T. Cupitt is the EPA Project Officer (see below). The complete report, entitled "Hazardous Air Pollutants: Dry-Deposition Phenomena."(Order No. PB 85-138 279; Cost: $11.50, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Atmospheric Sciences Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC 27711 * U.S. GOVERNMENT PRINTING OFFICE; 1985 — 559-016/7896 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Official Business Penalty for Private Use $300 OOOC329 PS II $ ENVIR PR0TECTIOK AGENCY REGION 5 LIBRARY 230 S DEARBORN STREIT CHICAGO IL ------- |