The potential inhalation hazard posed by dioxin-contaminated soil

Chemosphere, Voi.23, Nos.ll-12, pp 1719-1729, Printed in Great Britain

1991

0045-6535/91 $3.00 + 0.00 Pergamon Press plc

THE POTENTIAL INHALATION HAZARD POSED BY DIOXIN-CONTAMINATED SOIL Dennis J. Paustenbach*, Tibor T. Sarlos, Brent L. Finley, David A. Jeffrey, and Michael J. Ungs ChemRisk, A Division of Mcl.arert/Hart, 1135 Atlantic Avenue, Alameda, California 94501

ABSTRACT Conservative models were used to estimate the airborne concentrations of 2,3,7,8 tetrachlorodibenzo-p-dioxin (TCDD) vapor and particulates originating from soil containing 100 ppb TCDD. The upper-bound estimates were 3.25 pg/m3 of airborne TCDD vapor on-site and 0.51 pg/m3 for TCDD vapor 100 meters downwind. The TCDD air concentration on-site due to suspended particulate is estimated to be 1.4 pg/m3, based on a TSP level of 0.07 mg/m3. Assuming 70 years of continuous exposure to these concentrations, the upper-bound cancer risks determined from the Jury model were estimated to be 9.4 x 10-6 to 1.1 x 10-4 and 1.5 x 10-6 to 1.7 x 10-5 for inhalation of on- and off-site vapor, respectively, and 4.1 x 10-6 to 4.6 x 10-5 for dust inhalation. Since few sites have average soil concentrations as high as 100 ppb TCDD, this worstcase analysis indicates that inhalation will rarely, if ever, be a significant route of exposure to TCDD-contaminated soil. Experimental results support this claim and point to much lower risk estimates (8.4 x 10-9 to 9.9 x 10-8), suggesting that the parameters used in the Jury model are likely to overestimate the actual airborne levels of TCDD at contaminated sites.

INTRODUCTION Although 2,3,7,8-TCDD has been shown to have a negligible vapor pressure (1.5 x 10.9 film Hg at 25°C; Schroy et al., 1985), several risk assessments have claimed that airborne concentrations of TCDD from contaminated soil could be sufficient to pose a significant health risk (USEPA, 1985). Other reports have suggested that the inhalation hazard posed by TCDD vapors and particulates is negligible and that it may be dismissed (Paustenbach, et al., 1986). To quantitatively assess this issue, the highest plausible airborne concentration of TCDD (vapor and particulate-bound) originating from a site with contaminated soil was predicted and the associated theoretical cancer risks were estimated.

1719

1720

METHODS In this analysis, one acre of soil was assumed to be uniformly contaminated with 100 ppb TCDD to a depth of 10 cm. It was also assumed that only 50% of the soil is exposed. The other half is paved or is covered by buildings, driveways or private residences. Since it is known that diffusion of chemical vapors through cement or other hard building surfaces is negligible compared to the potential escape through actual cracks (Zapalac, 1983), and since the percent of cracks or imperfections in the surface is only 0.01 - 0.1% of the total floor area (Grismund et al., 1983), it was assumed that those paved portions of the hypothetical site do not contribute significant amounts of TCDD vapor. Potential TCDD vapor emissions from site soils were estimated using the model of Jury et al. (1983), assuming a constant soil temperature of 20oC. The Jury model predicts the time-weighted average vapor flux from contaminated soil, based on the soil characteristics and the physical properties of the chemical. The model accounts for the contaminant's volatility (vapor pressure), strength of adsorption to soil particles (KD), and upward leaching in the soil column due to evapotransporation (wicking). It also accounts for the continual loss of contaminant from the soil; that is, the soil vapor flux constantly decreases. The values for the parameters used in this analysis are shown in Table 1 (Marple et al., 1987; Freeman and Schroy, 1985,1986). The results of the Jury model were compared with actual field/laboratory data. The details of the calculations are presented in the Appendix. The Jury model relies on a knowledge of a number of chemical- and soil-specific parameters. "KD", the soil desorption coefficient, is a measure of a substance's tendency to remain sorbed to soil versus existing as a free unbound liquid. "H", Henry's Constant, reflects the tendency of the desorbed liquid to enter the vapor state. Thus, the model treats soil-bound contaminant emissions as a two-stage process. Other parameters needed for the Jury model include foc, Pa, Pw, PT, Pb, and Jw (defined in Table 1). All of them, with the exception of "Jw", are related to the characteristics of the soil. A "box model' approach was used to predict the maximum plausible levels of TCDD vapor on-site. This approach assumes that the air above the surface soils is contained in a "building" having two open walls and a 2 meter high ceiling, with wind blowing through it. The choice of a 2-meter ceiling is to represent the approximate breathing zone of an individual walking over exposed site soils. This box-model approach is depicted in Figure 1. The base of the box is the entire site (one acre), since individuals may be located anywhere within the 1-acre extent of soil contamination. However, as was previously discussed, only 1/2 acre emits TCDD vapor into this box due to the buildings and other paved sections covering large portions of the site. The box model assumes that the TCDD vapor travels only a very short distance from the soil prior to inhalation. Therefore, the model accounts for removal of TCDD vapor by a uni-directional windstream but ignores dispersion effects and photodegradation (these are important when estimating the risks to persons off-site). Airborne contaminant concentrations determined from the

1721

box model are therefore really only relevant for maximally exposed individuals close to the emitting surface.

To this extent, the box model produces upper-bound estimates of the airborne

contaminant concentration, since most individuals.will be further removed from the emitting surface, where air dispersion effects are important. The SCREEN air dispersion model (USEPA, 1988) was used to estimate a worst-case (downwind) air concentration of TCDD at a distance of 100 meters off-site. As its name implies, SCREEN is a model which generally produces upper-bound estimates of the vapors emitted from an area source. It accounts for horizontal and vertical dispersion during worst-case meteorological conditions (that is, extremely slow, stable winds). For the purposes of this study, the annual average air concentration was conservatively estimated to be 10% of the predicted maximum hourly concentration. This assumption is recommended in several regulatory guidelines (USEPA, 1982; CAPCOA, 1987). A receptor height of 2 meters was used with a wind speed of 2 meters/second (the same parameters used for the box model analysis). Flat terrain was assumed. For industrial areas or farm land, regions where TCI)D spills/contaminationare likely to occur, the assumption of a fiat terrain is appropriate. To estimate the hazard due to airborne TCDD bound to total suspended particulate (TSP) originating from the contaminated site, a n average TSP concentration of 0.07 rag/m3 was used (Paustenbaeh et al., 1986). This concentration is similar to TSP measured in some rural areas (Trijonis et al., 1980; USEPA, 1985a).

Since there are many potential non-soil sources of

airborne particulate matter, the total TSP (0.07 mg/m3) was adjusted to account for the fraction likely to originate from the TCDD-contaminated soil. It was assumed for the purposes of this analysis that about 20% of the total TSP was derived from the site soil. Several studies have shown that only about 15-25% of airborne TSP is comprised of soil particles emitted from any one particular site (Lioy and Daisey, 1986), The air concentration of TCDD bound to particulates, then, is simply the product of the concentration of TSP due to suspension of site soils (0.07 x 20% = 0.014 mg/m3) and the average TCDD concentration in the soil (100 ppb), o r l . 4 pg/m3 TCDD. The EPA's unit risk value (URV) of 3.3 x 10-5 (pg/m3)-l, and a lower URV value of 2.9 x 10 .6 (pg/m3)-l, based on the recent histopathological re-analysis of the same dataset used by the EPA (Keenan et al., 1990 a,b), were used to estimate the excess lifetime cancer risk. The URV contains all the standard exposure factors (breathing rate, exposure duration, etc.), so that by multiplying the URV by the contaminant air concentration, a plausible theoretical excess lifetime cancer risk is estimated. For the purposes of this analysis, it was conservatively assumed that exposure to TCDD vapors and particles occurs continuously for a 70-year lifetime.

RESULTS OF JURY MODEL Using Jury's model and the parameters in Table 1, the predicted vapor flux was 7.47 X 10-14 mg TCDD/sec-cm2. Combining this flux with a wind speed of 2 meters/see and a receptor height of 2 meters, the box model estimated an average on-site TCDD vapor concentration of 3.25 pg/m3.

1722

This concentration is associated with a lifetime cancer risk of 1.1 x 10 -4 (URV of 3.3 x 10-5 pg/m3) or 9.4 x 10-6 (URV of 2.9 x 10-6). Using the SCREEN model, the upper-bound average annual concentration 100 meters downwind off-site was estimated at 0.51 pg/m3. This concentration corresponds to an increased lifetime cancer risk of 1.7 x 10 -5 (URV of 3.3 x 10-5) or 1.5 x 10-6 (URV of 2.9 x 10-6), This result was obtained using a SCREEN wind class of "6", the most stable meteorological setting. Since actual wind patterns at the site will probably be less stable, with more changing wind directions over time, the SCREEN result is quite conservative. An increased cancer risk of 4.6 x 10-5 (URV of 3.3 x 10-5) or 4.1 x 10-6 (URV of 2.9 x 10-6) was estimated due to resuspended soil. In this analysis, all particles were conservatively assumed to be respirable.

REASONABLENESS OF RESULTS Yanders and co-workers, in a series of papers published between 1986 and 1989 (Palausky et al., 1986; Yanders et al., 1989; Kapila et al., 1989), described the results of experiments designed to determine the fate and mobility of TCDD in soil. This information was used to validate the reasonableness of the model predictions. Low concentrations of both 2,3,7,8-TCDD and 1,2,3,4TCDD in various solvent media (including waste oil itself) were applied to soil samples obtained from uncontaminated areas of Times Beach, Missouri (Yanders et al, 1989). These experimental soil columns were regularly watered to simulate local precipitation patterns. Twelve-hour light/twelve-hour dark cycles were also used to simulate day and night. The data from these experiments shows a fairly dramatic loss of TCDD from the top 5 mm of soil one month following initial application (Kapila et al, 1989). However, this "toss" was accounted for by a corresponding increase in the TCDD levels measured at lower depths (which suggested loss was not due to volatilization). It was found that essentially 100% of the original TCDD applied to the soil was detected 12 months after TCDD application, although the dioxin moved to lower depths (Kapila et al, 1989). Although these soil column experiments were conducted using waste oil as an application medium, to simulate the Times Beach scenario, and the oil is expected to stabilize TCDD and decrease its volatilization, the negligible vapor loss of TCDD suggested by these experiments can still be understood in terms of the low vapor pressure of TCDD. Based on these studies, it seems reasonable to suggest that the volatilization or environmental halflife of TCDD in soil could be 50 to 100 years, or longer (Gough, 1990). From this, it is possible to calculate the TCDD air concentration and compare it to the predicted Jury calculation, using the same box model. For example, one acre of soil contaminated with 100 ppb TCDD x 5 mm depth will emit 0.5 x 105 ug TCDD vapor over a 50 year period. This emission rate, when coupled with the box model, yields a TCDD vapor concentration of 1.6 pg/m3. This is reasonably close to the result obtained from the Jury model (3.25 pg/m3). Another approach to predicting the airborne TCDD levels from these soil column experiments is to

1723

consider the amount of airborne TCDD emitted from soil. To simulate wind flow over the surface soils, which is expected to contribute to TSP generation, an air inlet and outlet were fitted to the otherwise closed chamber. In order to protect the laboratory workers from any TCDD that might volatilize off the surface soils and ultimately escape the chamber, glass fiber filters and XAD-2 traps were used to capture TCDD particulates and vapor after they left the chamber (Yanders, 1990). At the end of the 12-month experiment, these filters and traps were analyzed for total emitted TCDD; no TCDD was detected in any of them. The limit of detection for TCDD in the filter/traps was approximately 10 ng (Yanders, 1990). This figure corresponds to the total amount of dioxin emitted from the soil over the 12 months. Since 100 L/rain. of air passed through the chamber, about 52,000 cubic meters of air passed over the surface soils, the maximum possible air concentration would be 10 ng/52,000 m3, or 0.02 pg/m3. Based on this limit of detection, and knowing the amount of TCDD applied to the soil columns, it is possible to compute the fraction of TCDD that was lost due to volatilization or particulate emission. The half-life for the photochemical decomposition of TCDD in air can conservatively be estimated to be 10 minutes; the actual photochemical hail-life for TCDD is likely to be considerably longer (Crosby and Wong, 1977; USEPA, 1985). Taking this 10-minute half-life into account, 0.0042% of the total TCDD available would be an upper limit of annual TCDD loss due to volatilization and particulate emissions. If one assumes that it is possible that about one-half the LOD of the chemical was present on the filter, then about 0.0021% of the TCDD present in the soil was lost each year. Multiplying 0.0021% by 70 gives the total percent loss of TCDD (vaporization and particulate emission) expected after 70 years, or 0.147% of the initial amount. The actual 70-year percent loss is less than this, since with each year that passes, the amount of TCDD remaining in the soil is less than the preceding year(s). Assuming 100 ppb dioxin contamination of a one acre plot down to a depth of 10 cm (the same values used in the Jury model), about 4 x 107 ug TCDD is initially present in the soil. 0.147% of this mass emitted over 70 years yields an emission rate of 840 ug/year. After normalizing to a one square centimeter area source, converting ug to rag, and converting years to seconds, the flux is 6.7 x 10 -17 mg/sec-cm2. This value is about 1000-fold lower than that predicted by the Jury model. Using this lower flux, 6.7 x 10-17mg/sec-cm2, and the box model, the inhalation cancer risk could be as high as 9.9 x 10-8 (for URV of 3.3 x 10-5 pg/m3), and as low as 8.4 x 10-9 (for URV of 2.9 x 10-6 pg/m3) for persons on-site. For exposed persons offsite, the risks are even smaller. These risks arc approximately 1,000 fold lower than estimates predicted by the Jury model or assuming a 50-year soil half-life.

DISCUSSION The Jury model is a popular method for estimating the vapor flux of soil-bound contaminants. However, like all models, the results are only as good as the parameter values. In this evaluation of TCDD emitted from soil, it appears the values for some of the parameters may be inappropriate

1724

since experimental data suggest the actual dioxin flux to be significantly less:than the flux predicted by the Jury model. In the Jury calculations, both the water solubility and the vapor pressure of TCDD are critical parameters. A water solubility of 8 x 10 -6 mg]L, suggested by both Marple etal. (1987) and Freeman and Schroy (1986), is a very low value whose determination is subject to significant experimental error. The same can be said for the vapor pressure of TCDD, reported to be only 8.6 x 10-8 pascals at 20°C (Freeman and Schroy, 1986). TCDD's vapor pressure in soil, especially in the presence of co-contaminants, has not yet been evaluated. Because the Jury model uses fundamental parameters like water solubility and vapor pressure more than once in the mathematical formulae in the model, any errors in these particular parameter values will tend to be magnified in the final result (flux). There are other reasons why the Jury model over-predic~:l the actual flux rate of dioxin vapor from soil. First, we assumed the fraction of soil consisting of organic matter, or foc, was assumed to be 1%. This value is probably low for most soils. This factor can significantly affect the flux rate, since the organic matter within the soil is responsible for the tenacious binding of TCDD. Second, the soil porosity was assumed to be 0.45 (45%). Greater soil porosity enhances vapor diffusion and ultimately contaminant migration to the surface. This value is probably too high for most soils. Fate and transport models are commonly used to predict the movement of chemicals in the environment. Most models accepted by the regulatory community are designed to skew the results in modeling to the high side, making the likelihood of overestimating the concentration at a receptor location much greater than that of underestimating it. Most of these models were meant to be used only in screening assessments. Such assessments provide the risk assessor with a rapid, inexpensive method for initially separating contaminated sites into two categories: those that clearly do not warrant concern, and those that may require a more in-depth, realistic analysis. Thus, it's important to keep in mind that results obtained using approximation methods such as the box model, the SCREEN air dispersion model, or even perhaps the Jury model are not meant as substitutes for either good experimental data or more rigorous modelling approaches. The risk estimates derived from the experimental data presented in this analysis are very low and for most purposes negligible. Although the risk calculation based on a volatilization half-life of 50 years does show risks that might be considered significant, the second calculation, based on the LOD, clearly shows a very low level of risk. Since the volatilization half-life calculation is based on a crude estimate of the tit2 (which is probably considerably greater than 50 years), the LOD calculation is probably more accurate. Additionally, this low LOD-based risk level was calculated conservatively assuming a photochemical half-life of 10 minutes for TCDD in air. For a longer, probably more realistic half-life, photochemical degradation in the experimental soil column chamber would be negligible, and the maximum loss of TCDD vapor and particulates would be even less than the very small amount originally calculated.

1725

The results of this analysis indicate that inhalation of TCDD vapor or particulates will generally not be a significant exposure pathway where the soil levels are less than 100 ppb. Although only cancer risks were evaluated, the cancer h~-a_rdis far more significant than any other toxicological endpoint for dioxin. Therefore, the (experimental) data described in this analysis indicate that all other health hazards associated with inhalation exposure to TCDD vapors and particulate matter must also be negligible.

REFERENCES CAPCOA. (1987). Toxic Air Pollutant Source Assessment Manual for California Air Pollution Control District and Applications for Air Pollution Control District Permits. Volumes 1 and 2. California Air Pollution Control Officers Association. Sacramento, CA.

Crosby, D.G. and A.S. Wong. (1977). Environmental Degradation of 2,3,7,8Tetrachlorodibenzo-p-dioxin (TCDD). Sciene~ 195, pp. 1337-1338. Freeman, R.A. and J.M. Schroy. (1985). Environmental Mobility of Dioxins. American Society for Testing and Materials. p. 421-439. Philadelphia, PA. Freeman, R.A. and J.M. Schroy. (1986). Modeling the Transport of 2,3,7,8-TCDD and other low volatility chemicals in soil. Environmental Prot, ress 5(1~:28-33. Gough, M. (1990). A Svmnosium on the I nnb Guidelines. Washington, DC.

Resources for the Future;

Grismund, D.T, M.H. Sonderegger and R.C. Sherman. (1983). A Framework of a Construction Quality Survey for Air Leakage in Residential Buildings. In: Procedures of Thermal Performance 9fExternal Envelopes ofBuildines. ASHRAE. 422-452. Jury, W.A., W.F. Spencer, and W.J. Farmer. (1983). Behavior Assessment model for trace organics in soil: I. Model Description. ~ 12: 558-564. Kapila, S., Yanders, A.F., Orazio, C.E., Meadows, J.E., Cerlesi, S. and T.E. Clevenger. (1989). Field and laboratory studies on the movement and fate of tetrachlorodibenzo-p-dioxin in soil. Chemosphere 18f 1-6~:1297-1304. Keenan, R.E., D.J. Paustenbach, R.J. Wenning and A.H. Parsons (1990a). Pathological ReEvaluation of the Kociba et al. Bioassay of 2,3,7,8-TCDD; Implications for a Risk Assessment. J. (Submitted). Keenan, R.E., R.J. Wenning, A.H. Parsons and D.J. Paustenbaeh (1990b). A Re-evaluation of the Tumor Histopathology of Kociba et al. 1978 Using 1990 Criteria; Implications for the Risk Assessment of 2,3,7,8-TCDD Using the Linearized Multi-Stage Model. Proceedings of the 10th International Conference on Dioxins and Chlorinated Organic Compounds; Dioxin '90. Lioy, P.J. and J.M. Daisey (1986). Airborne toxic elements and organic substances. Environ. Sci. Technol. 20(1):8-14. Marple, L., Brunck, R., Berridge, B. and L. Throop. (1987). Experimental and Calculated Physical Constants for 2,3,7,8-Tetrachlorodibenzo-p-dioxin. Solving Hazardous Waste Problems. pp. 105-113. American Chemical Society, New York, NY. Palausky, J., Kapila, S., Manahan, S.E., Yandcrs, A.E., Malhotra, R.K. and T.E. Clevcngcr. (1986). Studies on vapor phase transport and role of dispersing medium on mobility of 2,3,7,8TCDD in soil. Chcmosnhcrc 15(9-12~:1389-1396. Paustenbach, D.J., H.P. Shu, and F.J. Murray. (1986). A critical examination of assumptions used in risk assessments of dioxin contaminated soil. Re~ul. ToxicoL Pharmacol. 6:284-307.

1726

Schroy, J.M., F.D. Hilc~an, and S.C. Cheng. (1985). Physical/chemical properties of 2,3,7,8TC'DD, ~ 14(6/7~:8T7=880. Trijonis, J., J. Eldon, J. Gins and G. Berglund. (1980). Analysis of the St. Louis RAMS Ambient Particulate Data, EPA Report 450/4-80-C06a. Produced by Technology Service Corporation under EPA Contract 68-02-2931 for the ~ e of Air, Noise, and R~d_i~tion of the U.S. Environmental Protection Agency, Washington, D.C. USEPA. Pierce, T.E., D.B. Turner, J.A. Catalano, and F.V. Hale. (1982). PTPLU - A Single Source Gaussian Dispersion Algorithm. EPA-600/8-82-014. U.S. Environmental Protection Agency. Washington, DC. USEPA. (1985). Health assessment document (HAD) of polychlorinated dibenzo,p-dioxins. Office of Health and Environmental Assessment. E P A = ~ I - 0 1 4 F . NTIS PB86-122546. Environmental Oitcria and Assessment Office, Cincinnati, Ohio. USEPA. (1988). SCREEN. Screening Procedures for Estimating the Air Quality Impact of Stationary Sources (DRAFT). U.S. Environmental Protection Agency Publication. EPA-450/488-010. Yanders, A.F. (1990). Unpublished results (private communication). Yanders, A.F., Orazio, C.E., Purl, R.K. and Kapila, S. (1989). On translocation of 2,3,7,8Tetrachlorodibenzo-p-dioxin: Time dependent analysis at the Times Beach Experimental Site. ChemosT)here 19(1-6~:429-432. Zapalac, G.H. (1983). A Time-Dependent Method for Characterizing the Diffusion of 222Rn In Concrete. Health Physics45. (2):377-383.

1727

APPENDIX EXPERIMENTAL DETERMINATION OF TCDD VAPOR CONCENTRATIONS

I.

Mass Balance Calculation Assumntions:

1) 2)

Half-life of volatiliTation = 50 years TCDD at soil depths greater than 5 rmn is stationary

1/2 acre x 4 x 103 m2/acre x 1 x 104 cm2/m2 x 0.5 cm = 1 x 107 cm3 (volume of TCDD soil contamination) 100 I~g/kg x 1 x 107 cm3 x 1 kg/1000 g x 1 x 10-6 g/~tg x 1.3 gtrdcm3 = 1.3 g (mass of TCDD available for volatilization) If qt2 (volatilization) = 50 years, 0.65 g/50 years = Initial Emission Rate For box 2 meters high with a square base of 1 acre (4,000 m2), volume of air blown through box by 2 n~ter~sec wind over 50 years is: 2 m x (4,000 m2)1t2 x 2 m/see x 3.15 x 107 see/year x 50 years = 4 x 1011 n# (total volume of air blown through box over 50 years) Therefore, 0.65 g/4 x 1011 m3 = 1.6 x 10-12 g/m3, or 1.6 pg/m3 TCDD vapor II.

L i m i t o f Detection C a l c u l a t i o n

Assumotions: 1)

Limit of detection of TODD collected on filter (vapor and particulate) = 10ng

2)

Photolytic half-life = 10 minutes

3)

Duration of experiment = 12 months

Initial amount of T C D D (before soil chamber experiment) = 42 soil columns x 15 g TODD appfied/column = 630 g However, since soil samples are continually being removed during the experiment: Average mass TODD available for volatifization]particulate loss during course of experiment = 630 g/2 = 315 g At 1 m3 = 1,000 L chamber size, and airflow = 100 L/rain, air clearance time = 10 minutes. Since ti/2 (photolysis) = 10 rain., expect 50% of the TCDD leaving the surface soils during daylight hours to be photochemically degraded and not detected in the trap or filter. Assume 0% photochemical degradation during dark hours. For 12-hour daylight/12-hour night cycle, expect total photochemical degradative loss = 25% for airborne TCDD: 10 ng LOD + 10 ng (25%/75%) = 13.3 ng = mass of TCDD which can leave the soil surface without being detected. 13.3 ng/315 g = 13.3 ng/315,000 ng = 0.000042, or 0.0042% = maximum possible annual percent loss of TCDD via vaporization and particulate emission. At 1/2 LOD, 0.0042%/2 = 0.0021% annual loss. At 0.0021% loss/year x 70 years, the total loss of TCDD due to volatilization/particulate emission = 0.147%. This is probably too high since the vapor loss from soil is probably much faster during the first year (the basis of this calculation) than any subsequent year.

1728

TABLE 1. VALUES USED IN MODEL TO ESTIMATE THE VOLATILIZATION OF TCDD FROM SOIL

Parameter Ko~

Description

Value

Source/Rationale

Organic soil carbon-liquid partition coefficient

2.78 x 106 ml/g

Marple et al., 1987

S

Water solubility

7.96 x 10-6 mg/P

Marple et al., 1987; Freeman and Schroy, 1986

H

Henry's constant

3.44 x 105 atmm3/mole

Determined from p, MW, and S

h

Dimensionless Henry's constant

1.43 x 10"3

Determined from H

p

Vapor pressure

8.61 x 10-s pascals b

Freeman and Schroy, 1986

MW

Molecular weight

321.974 g/mole

KD

Solid-liquid partition coefficient

2.78 x 104 ml/g

Determined from Koc, and fo~

fo~

Organic soil carbon fraction

0.01

Assumed

Air diffusion coefficient

0.045 cm3/cm-sec

Freeman and Schroy, 1985

Water diffusion coefficient

5.6 x 10-6 cma/cm sec

Freeman and Schroy, 1985

Dgair

DI water

Pd

Soil particle density

2.65 g/cm 3

Assumed

Pa

Volumetric air fraction of soil

0.35

Assumed

Pw

Volumetric water fraction of soil

0.15

Assumed

PT

Total soil porosity

0.50

PT = 1-Pb/Pd

Pb

Dry soil bulk density

1.3 gm/cm 3

Assumed

Jw

Water evaporation rate

10 in/year

Assumed

Soil concentration of TCDD

100/ag/kg (wet soil)

Initial assumption

t

Time-span for average flux calculation

70 years

Initial assumption

Z

Depth of contamination

10 cm

Initial assumption

[TCDD]

aValue for 22°C

bValue for 20°C

FIGURE 1.

BOX MODEL USED TO ESTIMATE OUTDOOR ON-SITE TCDD INHALATION EXPOSURE

EMITTING AREA SOURCE

T T T

CROSS-SECTIONAL AREA WIND PASSES THROUGH

J

WIND

2 METER HEIGHT