800R88100 DC 90440 May 1988 SCD# 17 Wittr INTERIM SEDIMENT CRITERIA VALUES FOR NONPOL AR HYDROPHOBIC ORGANIC CONTAMINANTS ------- INTERIM SEDIMENT CRITERIA VALUES FOR NONPOLAR HYDROPHOBIC ORGANIC COMPOUNDS INTRODUCTION Toxic contaminants in the bottom sediments of lakes, rivers and coastal waters can degrade the environment. Available data indicate many locations where existing sediment contaminant concentrations are now causing significant adverse environmental effects on aquatic life, even when water column contaminant concentrations comply with established water quality criteria (Malins et al. I960, 1982). Since 1985, the Criteria and Standards Division of EPA has been pursuing the Equilibrium Partitioning (EP) approach for estimating sediment quality criteria for nonpolar and metal contaminants. .In anticipation of favorable review of the approach by EPA's Science Advisory" Board, interim sediment criteria values based on the EP approach for selected nonpolar, hydrophobic organic compounds were developed. These interim criteria values can be used to evaluate the appropriate applications of sediment criteria in existing regulatory programs. This report describes how the interim numbers were developed and briefly how the criteria values can be used to evaluate the extent of sediment contamination. Preparation of this report has resulted in a great deal of discussion regarding choice of partition coefficients and methods for determining the uncertainty in the interim criteria values. Therefore, it is very likely that the final values that EPA will recommend will differ from these values although not substantially. Any user of these numbers should be aware that these numbers are indeed interim and not final criteria values. The main part of this report describes the development of interim criteria values for nonpolar organic contaminants for which chronic water quality 1 ------- criteria have been generated. In Appendix E of the report, additional interim criteria for selected PAHs are given. EQUILIBRIUM PARTITIONING APPROACH Before describing how the interim criteria were estimated, the technical approach that forms the basis for the sediment criteria development effort will be discussed. The approach that is being pursued by the Criteria and Standards Division for establishing sediment quality criteria, on the recommendation of participants in the technical workshops and steering committees, is the EP approach (Neff 1985, Cowan 1986, Cowan 1987). The EP approach is based on two interrelated assumptions. First, that the interstitial water concentration of the contaminant is controlled by partitioning between the sediment and the water at contaminant concentrations well below saturation in both phases. Thus, the partitioning can be calculated from the quantity of the sorbent(s) on the sediment and the appropriate sorption coefficient(s). For nonpolar organic contaminants, the primary sorbent is the organic carbon on the sediment; therefore, the partition coefficient is called the organic carbon normalized partition coefficient, KOC< Second, the toxicity and accumulation of the contaminant by benthic organisms is correlated to the interstitial, or pore water concentration and not directly to the total concentration of the contaminant on the sediment. When the EP approach is used to estimate sediment quality criteria, chronic water quality criteria (WQC) are used to establish the "no-effect" concentration, in the interstitial water. Chronic water quality criteria are used to protect benthic organisms from effects due to their long-term exposure to low ambient concentrations in the sediment. The use of WQC assumes that the sensitivities- ------- of water column and benthic species to a compound are similar. This assumption is being evaluated. This interstitial water concentration (Cw) is then used with the partition coefficients (KQ(.) and the following equation Csed * *oc to calculate the concentration of the contaminant on the sediment (C .,) that sed at equilibrium will result in this interstitial water concentration. This concentration on the sediment will be the numerical criteria value (SQC). For compounds when, chronic water quality criteria are not available, the EP approach can still be useful. For example, using upper-bounds effects concentrations will give comparable (i.e, upper-bounds effects) sediment concentrations. The interpretation of such sediment values is analogous to the interpretations of the comparable water column values used in their derivation. DEVELOPMENT OF INTERIM NUMBERS To estimate interim sediment criteria values, two sets of data are needed for each compound for which criteria values are required. These data are the water quality criteria and the partition coefficients. WATER QUALITY CRITERIA VALUES Water Quality Criteria (WQC) concentrations are available for 17 nonpolar organic chemicals (Hansen 1987). The criteria values are summarized in Table 1 The procedures for deriving these criteria are described in Appendix A. The WQC concentrations consist of the Criteria Maximum Concentration (CMC) and ------- the Criteria Continuous Concentration (CCC). The CMC is not applicable for derivation of SQC concentrations because it protects aquatic life from acutely lethal effects of a chemical. The CCC is the lower of the Final Chronic Value (FCV), the concentration protecting aquatic life from chronic toxicity, and the Final Residue Value (FRV), the concentration protecting uses of aquatic life. These uses include marketability of aquatic life based on FDA or other action levels or consumption of aquatic life by wildlife. The CCC is the appropriate value to use in deriving SQC because it protects aquatic life from effects due to long-term exposure to contaminated sediments. Both the FCV and FRV are presented in Table 1. Important limitations of Table 1 should be mentioned. First, the WQC concentrations for acenaphthene, aniline, diethylhexylphthalate, methyl- parathion, phenanthrene, and 1,2,4-trichlorobenzene must be used with caution • because they are preliminary values until criteria documents have been peer reviewed and accepted. Second, the PCB criteria is based on the FDA action level of 5 mg/kg and bioaccumulation factors measured in th> laboratory. Since 1980 when this criteria was developed, the FDA action level has been changed to 2 mg/kg. Furthermore, the residue values do not account for bioconcentration in the food chain which results in bioaccumulation factors for fish at least 10 times higher than those measured in the laboratory. This bioconcentration has been shown to be important for DDT and PCB. PARTITION COEFFICIENTS For estimating the interim sediment quality criteria values presented here, it is assumed that the sediment organic carbon partition coefficient, ------- K , can be accurately calculated from the octanol-water partition coefficient, K , using the following equation (DiToro 1985): ow Log10(Koc) » 0.00028 + 0.983*LoglO(KQw) (2) The K values used in the regression analysis were carefully screened to remove data for experiments that were conducted at high particle concentrations and to ensure that only nonpolar organic compounds were included. This screening is important because particle interactions at high particle concentrations can ,-esult in errors in the K values (DiToro 1985). This relationship is chosen to calculate K values rather than using tabulated oc Koc values because KQw values have been determined by more researchers and the procedure for determining K values is simpler than that used for ow determining K values because interferences caused by dissolved organic carbon and particle effects do not have to be considered or accounted for in the experimental design and data analysis. Because K is used to estimate K , and ultimately the interim sediment OW OC quality criteria, it is important that both an estimate of the mean K and a quantification of its uncertainty be determined. To provide a preliminary estimate of the K values and their uncertainty, for each compound in Table 1. the following alternative methods were used. • Review of all measured values and calculation of the geometric mean and standard deviation of the mean from the data • Determine recommended value from Leo-Hanch database • Estimate log K from correlations with aqueous solubility • Estimate log K from structure-activity relationships. OW ------- The results of the four methods are presented in Table 2. The estimates derived from the four methods compared for overall consistency. The review of all measured values was conducted using the database developed by Envirosphere for the report by Pavlou et al. (1987). Because this database has been updated since that report was prepared the values in Table 2 may differ slightly from those presented in that report. The recommended value for the log KQw from the Leo and Hanch Log P Database (Leo 1984) was also tabulated for comparison. The methods used for estimating the log KQw values from the solubility and from the chemical structure of the compound are outlined in Appendices B and C, respectively. Estimates of KQW based on aqueous solubility, which were corrected for solids melting point (Bowman and Sans 1983), were calculated f»om solubility and melting point data reported in the Arizona Database of Aqueous Solubility (Yalkowsky et al. 1987). The log mean K for solubilities measured in the range of 15 to 25°C is reported in Table 2. PCB Aroclor KQW values reported in the "measured" column in Table 2 were calculated using the median of the log mean K values for the homologs and ow the Aroclor homolog composition as illustrated in Appendix D. Log mean K OW values for the homologs were compiled from six sources (Rapaport and Eisenrich 1984, Rapaport and Eisenrich 1985, Shiu and MacKay 1986, Woodburn et al. 1984, Miller et al. 1984, Chiou et al. 1977) and the median of the reported values determined. The Aroclor homolog composition was taken from Verschueren (1983). The K values in Table 2 for the four methods do in general agree; however, for some compounds the values range over several orders of magnitude. For these compounds, the wide range in uncertainty, the disagreement between the recommended value of Leo-Hanch and the geometric mean of all reported ------- values, or lack of confirmatory data makes it difficult to chose a definitive log KQw value. Based on the review of all the data, log K values for 11 of the 17 nonpolar compounds, for which there are WQC, are considered acceptable at this time and are use<3 to calculate interim criteria values. Further review is ongoing to determine the most acceptable mean and uncertainty values to represent the Kx values for these compounds. The accepted mean, standard 0" deviation (S.D.), and the 95% confidence intervals for the log K value; and WWf the mean, S.D., and 95V confidence intervals for the log K value for each of the 11 compounds are presented in Table 3. The 95% confidence intervals for the K values were calculated assuming a t statistic of 1.96, which is the value for large sample size, rather than the statistic for the specific sample size used to estimate the mean and S.D. for the compound. The mean log K values in this table were estimated using Equation (2) and the mean log K value. The S.D. of the log K value was estimated using the following ow oc equation: S.D. - y7 (S.D. of log Knjz + (0.3)2 OW where 0.3 represents the standard error due to the regression relationship. This estimate of the standard error of the regression is large because the least squares regression method assumes that all uncertainty is in the log K value. As part of the ongoing review, the most appropriate methods for estimating the mean and S.D. of the log KQw values are being examined. Also, alternative regression methods will be used to determine the most appropriate value for the standard error due to the regression. As a result of this review, the final log K values and the final criteria values may change slightly OW from those presented in this report. ------- INTERIM SEDIMENT QUALITY CRITERIA Table 4 summarizes the interim sediment criteria values calculated using Equations (1) and (2) and the FCV and FRV criteria, respectively, for freshwater (Table 1) as the basis of the interim criteria. Table 5 summarizes the criteria- values using FCV and FRV criteria, respectively, for saltwater (Table 1). When both FCV and FRV values are available for a compound, the FCV concentrations should only be used to calculate SQC when they are lower than the corresponding FRV concentrations. However, SQC values derived from both FCV and FRV values are presented to permit the user to determine what end use is being protected for. Estimates of the SQCs are shown for the mean and 95% confidence interval of the log KQC values. The confidence interval is reported to illustrate the uncertainty in the interim criteria values and to permit the user to estimate the likelihood that the sediment does or does not exceed the criteria value. The confidence interval represents the range within which with 95% certainty the sediment criteria value will fall. The lower value of the confidence interval represents the concentration which with 97.5% certainty will result in protection from chronic effects or of uses depending on the WQC value used in the SQC derivation. Any contaminant in a sediment at concentrations less than this value would not be of concern; however, the sediment can not be considered "safe" because the sediment may contain other contaminants above safe levels but for which criteria do not exist. The upper value of the confidence interval represents the concentration which with 97.5% certainty will result in hazardous long-term impacts on the benthic fauna. Thus, any sediments with concentrations above this level are considered hazardous. Concentrations within the confidence intervals can be considered either "safe" 8 ------- or hazardous with respect to that compound with certainties between 2.5 and 97.5%. APPLICATION OF INTERIM NUMBERS To determine if the sediment concentration of a nonpolar contaminant exceeds the sediment criteria values, the concentration of the contaminant and the organic carbon content of the sediment must both be known. The analytical methodologies for measuring the concentration of nonpolar organic compounds and the organic carbon content in sediments are described in Cowan and Riley (1987). Because the sediment criteria values are presented as normalized to organic carbon content (i.e., presented on a per organic carbon weight basis), the normalized sediment concentrations of the contaminants must be calculated. These normalized concentrations can then be directly compared with the interim values in Tables 4 and 5. Alternatively, the sediment criteria values could be multiplied by the lowest organic carbon content and the total concentrations compared with these criteria values. To facilitate this second type of comparison, Tables 6 and 7 contain the sediment criteria values for specific organic carbon contents of 1 and 10% for fresh and saltwater, respectively. These organic carbon contents represent the average range over which the EP approach has been examined (Karickhoff 1984, DiToro 1985). SAMPLE CALCULATION To illustrate the use of the interim sediment criteria values, an example calculation is presented. ------- For example, consider a site where previous analyses have indicated that DOT is present in the freshwater sediments at a concentration of 0.1 mg/kg of sediment and that the organic carbon (fQC) content is 2% or 0.02 kg of C/kg of sediment. To calculate the normalized sediment concentration in terms of organic carbon content, the formula is as follows: Normalized Concentration « Sediment Concentration/^ For this specific example, Concentration (mg/kg C) - (0.1 mg/kg)/(0.02 kg C/kg) » 5 mg/kg C Comparing this value to the values in Table 4, the normalized concentration exceeds the criteria values based on the FRV for freshwater. Alternatively, the criteria value in Table 4 could be multiplied by the organic carbon content of the sediment to calculate the criteria value for a specific organic carbon content. The formula is as follows: Sediment Concentration - (SQC)(foc) The calculation for this same case (i.e., 2% organic carbon) using the lower confidence interval value would be SQC at 2V O.C. (mg/kg) - 0.183 mg/kg C x 0.02 kg C/kg • 3.66E-3 mg/kg Comparison of this value with the measured concentration indicates that the criteria value is again exceeded. Using the upper confidence interval value also indicates that the criteria are exceeded. The first calculation method would be most appropriately used when several contaminant concentrations are available across a site that varies in organic carbon content. In that case, the calculation of the organic carbon normalized values and contours of concentration could be used to indicate the approximate" 10 ------- area or sampling sites that are above the criteria value. The second calculation method would be most appropriate when several contaminant concentrations are available, but the organic carbon content of the sediment is constant. 11 ------- REFERENCES Bowman, B. T. and W. W. Sans. 1983. "Determination of Octanol-Water Partitioning Coefficients (K ) of 61 Organophosphorus and Carbamate Insecticides and Their Relationship to Respective Water Solubility (S) Values." J. Environmental Science and Health B18(6): 667-683. Call, D. J., L. T. Brooke, M. L. Kasuth, S. H. Poirier, and M. 0. Hogland. 1985. "Fish Subchronic Toxicity Prediction Model for Industrial Organic Chemicals that Produce Narcosis." Environ. Tox. and Chem. 4:335-341. Chiou, C. T., V. H. Freed, D. W. Schmedding and R. L. Kohner. 1977. "Partition Coefficient and Bioaccumulation of Selected Organic Chemicals." Environmental Science and Technology 11:475-478. Cowan, C. E. 1986. Updated Sediment Criteria Integrated Work Plan: May 1986. Prepared by Battelle, Pacific Northwest Laboratories, Richland, Washington. For Criteria and Standards Division, Environmental Protection Agency, Washington, D.C. Submitted by Battelle, Washington Environmental Program Office, Washington, D.C. Cowan, C. E. 1987 Updated Sediment Criteria Integrated Work Plan: September 19B/. Prepared by Battelle, Pacific Northwest Laboratories, Richland, Washington. For Criteria and Standards Division, Environmental Protection Agency, Washington, D.C. Submitted by Battelle, Washington Environmental Program Office, Washington, D.C. Cowan, C. E. and R. G. Riley. 1987. Guidance for Sampling of and Analyzing for Organic Contaminants in Sediments. Prepared by Battelle, Pacific Northwest Laboratories, Richland, Washington. For Criteria and Standards Division, Environmental Protection Agency, Washington, D.C. Submitted by Battelle, Washington Environmental Program Office, Washington, D.C. DiToro, D. M. 1985. "A Particle Interaction Model of Reversible Organic Chemical Sorption." Chemosphere 14: 1503-1508. Hansen, D. J. 1987. Memorandum to Organizers of Presentation to the SAB [Science Advisory Board] on SQC [Sediment Quality Criteria]. Karickhoff, S. W. 1984. "Organic Pollutant Sorption in Aquatic Systems." J. Hydraulic Eng. 110(6):707-735. Leo, A. 1984. Medicinal Chemistry Project. Pomona College, Pomona, CA. Lyman, W. L., Reehl, W. F. and D. H. Rosenblatt. 1982. "Handbook of Chemical Estimation Methods." McGraw-Hill Book Co. pp. 16-25. Mai ins, D. C., B. B. McCain, D. W. Brown, A. K. Sparks and H. 0. Hodgins. 1980. Chemical Contaminants and Biological Abnormalities in Central and Southern Puget Sound. NOAA Technical Memo OMPA-2, National Oceanic and Atmospheric Administration, Boulder, CO. 12 ------- Malins, 0. C.r B. B. McCain, 0. W. Brown, A. K. Sparks, H. 0. Hodgins and S. L. Chan. 1982. Chemical Contaminants and Abnormalities in Fish and Invertebrates from Puget Sound. NOAA Technical Memo OMPA-19, National Oceanic and Atmospheric Administration, Boulder, CO. Miller, M. M., S. Ghodbane, S. P. Wasik, Y. B. Tewari and 0. E. Martire. 1984. "Aqueous Solubilities, Octanol-Water Partition. Coefficients, and Entropies of Wetting of Chlorinated Benzenes and Biphenyls." Journal of Chemical Engineering Data 29:184-190. Miller, M. M., Wasik, S. P., Huang, G-L., Shiu, W-Y., and D. McKay. 1985. "Relationship Between Octanol-Water Partition Coefficient and Aqueous Solubility." Environmental Sci and Technol 19(6): 522-529. Neff, J. M. 1985. Sediment Criteria Integrated Work Plan. Prepared by Battelle. For Criteria and Standards Division, Environmental Protection Agency, Washington, D.C. Submitted by Battelle, Washington Environmental Program Office, Washington, D.C. Pavlou, S., R. Kadeg, A. Turner, and M. Marchlik. 1987. Sediment Quality Criteria Methodology Validation: Uncertainty Analysis of Sediment Normalization Theory for Nonpolar Organic Contaminants. Prepared by Envirosphers Company, Bellevue, Washington. For Criteria and Standards Division, Environmental Protection Agency, Washington, D.C. Submitted by Battelle, Washington Environmental Program Office, Washington, D.C. Rapaport, R. A. and S. J. Eisenrich, 1984. "Chromatographic Determination of Octanol-Water Partition Coefficients (K *s) for 58 Polychlorinated Biphenyl Congeners." Environmental Science and Technology 18:163-170. Rapaport, R. A. and S. J. Eisenrich. 1985. "Corrections." Environmental Science and Technology 19:376. Shiu, W. Y. and D. Mackay. 1986. "A Critical review of Aqueous Solubilities, Vapor Pressures, Henry's Law Constants and Octanol Water Partition Coefficients of the Polychlorinated Biphenyls." Journal of Physical Chemistry, submitted for publication. Stephan, C. E., D. I. Mount, D. J. Hansen, J. H. Gentile, G. A. Chapman, and W. A. Brungs. 1985. Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organ-isms and Their Uses. National Technical Information Service, Springfield, VA. PB85-227049. 98pp. Veith, G. D., Call, D. J., L. T. Brooke. 1983. "Structure-Toxicity Relationships for Fathead Minnow, Pimeohales promelas; Narcotic Industrial Chemicals." Can J. Fish and Aquatic Sci. 40:743-748. Verschueren, K. 1983. "Handbook of Environmental Data on Organic Chemicals." Van Nostrand Reinhold Company, New York. 13 ------- l!f!!J2HAnASvS!L0: ^h"1"9*^ 1986' "SMILES' A Modern Chemical «>d Language Information System." Chemical Design Automation News. Woodburn, K. B., W. J. Doucette and A. W. Andren. 1984. Generator Column Determination of Octanol /Water Partition Coefficients for Selected C°ngeners-" Environmental Science and Yalkowsky, S. H. S. C. Valvani, W. Y. Kuu (eds). 1987. "Arizona Database of Aqueous Solubility," College of Pharmacy, University of Arizona 14 ------- TABLE 1. Final Residue Values (FRV) and Final Chronic Values (FCV) for Nonpolar Organic Compounds Freshwater /*g/L Saltwater Compound (Pub. Date) C. Acenaphthene* Aniline* Chlordane (1980) Chlorpyrifos (1986) DDT (1980) Dieldrin (1980) Di ethyl hexy 1 phthal ate* Endosulfan(1980) Endrin (1980) Ethyl Parathion (1986) Hepatachlor (1980) Hexach 1 orocycl ohexane (1980)*** Methyl Parathion* Phenanthrene* Polychlorinated Biphenyl (1980) Toxaphene (1986) 1 ,2,4-Trichlorobenzene* A.S. Number 83-32-9 62-53-3 57-74-9 2921-88-2 50-29-3 60-57-1 117-81-7 115-29-7 72-20-8 56-38-2 76-44-8 608-73-1 298-00-0 85-01-8 s 8001-35-2 120-82-1 FCV 57 7.2 0.17 0.041 - 0.29 360 0.056 0.045 0.013 - 0.080 0.15 6.3 - <0.039 23 FRV - — - 0.004 - 0.0010 0.0019 - . 0.0023 - 0.0038 • - - 0.014 0.0002** • FCV — 27 0.0064 0.0056 - 0.084 360 0.0087 0.0093 - . - 0.076 4.6 - 0.21 ™ FRV . _ 0.004 . 0.0010 0.0019 . 0.0023 - 0.0036 - - . 0.030 0.0002** • * Draft criteria documents. ** See criteria document for explanation of this residue-based value. *** Also known as Lindane 15 ------- TABLE 2. K Values for Compounds in Table 1 ow LOG K CHEMICAL (PUB DATE) ACENAPHTHENE ANILINE CHLORDANE(IMB) CHLORPYRIFOS(1986) 001(1980) DiaORIN(1980) OinHYLHEXYLPHTHAUTE ENDOSULFAH(IMI) OKMIN(I980) ETHYL PARAIHION(1966) HEPTACHLOR H£XACHLOROCYCLOHEXANE(l«al)* METHYL PARATHIOM PHENANTHRENE POLYCHLORINATED BIPMENVLS: AROCLOR 1242 AROCLOR 12S4 AROCLOR 1261 TOXAPHfNE(1906) 1.2.4-TRICHLQROBENZENE C.A.S. Ho •3-32-0 82-63-3 17-74-9 2921-00-2 60-29-3 60-IM 117-01-7 116-29-7 72-21-8 S0-38-2 78-44-8 008-73-1 296-U-l 86-11-8 6001 36-2 120-02-1 LEO-HANCH MEASURED 3 92 • 90 - 4 96 • 37 4 32 - 3 83 - 3 83 - 3 72 2.86 4 46 6.10 a 73 694 - 4.02 LEO-HANCH PAVLOU ET AL (1987) FRAGMENT ANALYSIS 4.07 0.92 I 14 4.69 6.91 2 92 6 66 4 87 2 92 3 47 4 61 37$ 2.79 4.49 - - - - 4.26 AQUEOUS GEOMETRIC SOLUBILITY MEAN 3 72 0 63 4 76 (.11 6 16 4 40 (70 - 3 01 3 94 6 18 3.41 3 34 4 14 - - - 4.77 3 30 4 18 090 4 81 4 98 6.62 4.92 4 44 3 86 4.64 3 36 4.42 0 26 . 3.26 SO .0032 .03(0 .407 0141 .1479 490 0 366 0 196 8.329 0.063 0.116 0.190 0.1263 9UCI 4 02 0 91 • Ob 4 96 (.73 3 96 3 74 0 47 S 89 3 23 4 19 (.87 3 26 - 4 34 - 1 16 - 4 97 - 6 11 - 6 31 - ( 88 - 6 14 - 4.26 - 6 19 - 3 47 - 4 86 - 6.63 - 4.77 * Also knoMi •• LindMM ------- TABLE 3. K and K Values for Selected Nonpolar Organic Compounds ow oc Log K ow Compound Acenaphthene Aniline Chlorpyrifos DDT Dieldrin Endrin Ethyl-parathion Heptachlor Lindane Phenanthrene PCB (1254) Mean(a) 4.18 0.98 4.98 6.02 4.92 4.44 3.86 4.54 3.35 4.42 6.25 S.D. 0.0832 0.0350 0.0141 0.148 0.490 0.355 0.197 0.329 0.063 0.116 0.196 95% (b) Confidence 4.02 0.91 4.95 5.73 3.96 3.74 3.47 3.90 3.23 4.19 5.87 Interval 4.34 1.05 5.01 6.31 5.88 5.14 4.25 5.18 3.47 4.65 6.63 Mean 4.11 0.96 4.90 5.92 4.84 4.36 3.79 4.46 3.29 4.35 6.14 S.O.(c) 0.311 0.302 0.300 0.334 0.574 0.465 0.359 0.445 0.306 0.322 0.358 95% (b) Confidence 3.50 0.37 4.31 * 5.26 3.71 3.45 3.09 3.59 2.69 3.71 5.44 Interval 4.73 1.56 5.49 6.58 5.97 5.29 4.51 5.34 3.90 4.98 6.85 (a) Geometric Mean* (b) 95% confidence interval calculated assuming log normal distribution of K values and 1.96 for t(P95%). °" (c) Standard error of log K calculated using following formula: OC S.O. = V(S.D. of log KOW)^^0.3F where 0.3 *s the standard error from regression relationship ------- TABLE 4. Sediment Quality Criteria Values for Selected Nonpolar Organic Compounds for Freshwater Sediment Quality Criteria (/»g/gC) Sediment Quality Criteria (/*g/gC) Freshwater Compound FCV Acenaphthene Aniline Chlorpyrifos DDT Dieldrin Endrin Ethyl Parathion Heptachlor Lindane Phenanthrene PCB (1254) 57 7.2 0.041 0.29 0.045 0.013 0.08 6.3 Mean 732 0.0662 3.22 19.9 1.04 0.0810 0.157 139 95% Confidence Interval 180 0.0169 0.831 1.49 0.128 0.0160 0.0394 32.6 3,030 0.262 12.7 273 8.68 0.416 0.636 605 Freshwater FRV 0.001 0.0019 0.0023 0.0038 0.014 Mean 0.828 0.130 0.0532 . 0.110 19.5 95% Confidence Interval 0.183 0.00976 0.00654 0.0148 3.87 3.80 1.79 0.443 0.840 99.9 ------- TABLE 5. Sediment Quality Criteria Values for Selected Nonpolar Organic Compounds for Saltwater Sediment Quality Criteria (/ig/gC) Sediment Quality Criteria (/ig/gC) Compound Acenaphthene Aniline Chlorpyrifos DDT Dieldrin Endrin Ethyl Parathion Heptachlor Lindane Phenanthrene PCB (1254) Saltwater FCV 27 0.0056 0.084 0.0093 4.6 Mean 0.248 0.440 5.77 0.215 102 95% Confidence 0.0635 0.114 0.431 0.0264 23.8 Interval 0.984 1.73 79.2 1.793 442 Saltwater FRV 0.001 0.0019 0.0023 0.0038 0.030 Mean 0.828 0.130 0.0532 0.104 " 41.8 95% Confidence 0.183 0.00976 0.00654 0.0140 8.29 Interval 3.80 1.79 0.443 0.796 214 ------- TABLE 6. Sediment Quality Criteria Values for Selected Nonpolar Organic Compounds for Freshwater at 1 and 10% Organic Carbon Content (all criteria with units of ppm or mg/kg) FCV FRV Compound Acenaphthena Am 1 in* Chlorpyrifoa DDT Dialdrin Endrin Elhyl Par»thion Hcptachlor Lindan* Phenanthrena PCB (12S4) Uean 7.33 1 66E-3 1 1322 1 199 1 111 I.616E-3 1 57E-3 1.39 11 961 CI MI 1 169E-3 I.31E-3 1 Ilk 1 ME 3 9 180E-3 1 394E-S 1.326 si a J.63E-I 1.127 2 73 1 1667 4.16E-3 6 38E-3 6. IS Heart 73.3 6 61E-S 1.322 1 M 1 1114 a.iK-s 1.117 13 9 101 951 U.I l.ME 3 I.M3 1.149 1 II2B l.«K-3 3 ME 3 3.26 CI 313 1.1262 1.27 27.3 1 666 1 1416 1.6636 61. i Ha an 6 JBE 3 1 30E-3 1 S33E-3 1.10E-3 I 19S 11 9SICI 1.63E3 1.1366 1 697BE-3 1 1179 1 6664E-3 4.43E-3 I.14IE-3 8 46E-3 1.1387 1.999 Mean 1 1626 1 1131 1.326E-3 1 I116E-3 1 8S 101 961 CI I. 1112 1 976E-3 1 664E-3 1 466E-3 1 367 1366 1.179 1.1443 6 6841 9 99 ------- TABLE 7. Sediment Quality Criteria Values for Selected Nonpolar Organic Compounds for Saltwater at 1 and 10% Organic Carbon Content (all criteria with units of ppm or mg/kg) FCV FRV 'expound II III Item 9SI CI HIM 961 C Acenaphthcna AniliM 2.46E-3 1 63SE-S 9 ME-9 1.1248 6.3SE-3 Chlorpynfo* 4.48E-3 1.14E-3 1.1173 1 1441 11114 DOT Diddrin 8.8577 4 31E-3 0 792 I.t77 1 1431 fndrin J 1SE 3 0 M4E-3 01179 I.I2U 2 64E-3 ithyl Parathion Heptachlor I indan* Phenanthr.n. 1.02 0.231 4.42 11 2 2 36 PCB (12(4) [ bean 0 OM4 0.173 • ME 3 792 1 30E 3 0 179 I H3E-3 1 04E-3 44 2 0 418 11 9SI CI 1.I3E-3 0 M76E 3 I.MS4E 3 0.14K-3 0 M29 1 0380 1 0179 4 43E-3 7.96E-3 2.14 atean 1 M2I 0 1131 » 328E-3 O.III4E-3 4.11 1M 9Ui I 1112 0 976C-3 0 6S4E-3 1 40E-3 1 829 CI • 388 8 179 9 9443 8 8790 21 4 ------- Appendix A. Water Quality Criteria Derivation Procedures The water quality criteria listed in Table 1 were obtained from published or draft aquat'ic-life water quality criteria documents. These numerical water quality criteria concentrations were derived using the "Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organisms and Their Uses" (Stephan et al. 1985). These "National Guidelines" specify minimum data requirements and data synthesis procedures which allow calculation of numerical criteria concentrations to protect the presence and uses of aquatic life. Water quality criteria concentrations to protect the presence of aquatic life are derived using the following procedure. First, all available data on the toxicity of the chemical are collected, reviewed for acceptability, and sorted by test type. If minimum database requirements for acute and chronic toxicity as specified in the "National Guidelines" are met, Criteria Maximum Concentrations (CMC) and Criteria Continuous Concentrations (CCC) are calculated to protect important aquatic life from acute and chronic toxicity, respectively. If the toxicity of the chemical is dependent on a water quality characteristic, then the CMC or CCC values are derived as a function of the appropriate characteristic. Because aquatic ecosystems can tolerate some stress and occasional adverse additions, protection of all species at all times and places is not necessary. Therefore, criteria derivation procedures result in criteria concentrations intended to protect most species most of the time but not all of the species all of the time. 22 ------- Water quality criteria concentrations to protect the uses of aquatic life are derived using the following procedure. A Criteria Continuous Concentration to limit residues in aquatic life can be derived if maximum permissible tissue concentrations and data on bioaccumulation or bioconcentration factors are available. Maximum permissible tissue concentrations are based on either (a) a FDA action level for fish oil or edible portions of fish or shellfish, (b) a maximum acceptable dietary intake based on a wildlife feeding study, or (c) residue-effects data for aquatic life. Either bioconcentration or bioaccumulation factors are required to calculate the water concentrations expected to limit chemical uptake by organisms to below the permissible tissue concentration. Bioconcentration factors, the concentration of the chemical in the organism divided by the concentration In the exposure water, are calculated from laboratory studies where steady-state conditions are achieved. Bioaccumulation factors, the concentration of chemical in the organism from all sources (e.g., food, water) divided by the concentration in the water, are calculated from data obtained in field studies. It is important to note that if food-chain transfer is an important uptake route, criteria concentrations derived using bioconcentration factors, such as those for DDT and PCBs, will probably be underprotective. The water quality criteria statement contains a concentration limit, averaging period, and return frequency and is stated as follows: "The procedures described in the "Guidelines for Deriving National Water Quality Criteria for the Protection of Aquatic Organisms and Their Uses" indicate that, except possibly where a locally important species is 23 ------- very sensitive, (1) aquatic organisms and their uses should not be affected unacceptably if the four-day average concentration of (2) does not exceed (3) ng/l more than once every three years on the average and if the one-hour average concentration does not exceed (4) /ig/L more than once every three years on the average." In this statement insert either "freshwater" or "saltwater" at (1), the name of the chemical at (2), the lower of the chronic-effect or residue-based concentrations as the Criteria Continuous Concentration at (3) and the acute effect-based Criteria Maximum Concentration at (4). The Criteria Continuous Concentration based on either chronic effect or residue data can be used to derive sediment quality criteria for nonpolar organic chemicals using the equilibrium partitioning approach. Detailed knowledge of the National Guidelines and the water quality criteria document as a basis for understanding the water quality criteria that are used in the derivation of sediment quality criteria is highly recommended. 24 ------- Appendix B. Aqueous Solubility and Octanol-Water Partition Coefficient Estimates of KQW from solubility are based on a thermodynamically inspired correlation developed for 61 organophosphorus and carbamate insecticides (Bowman and Sans 1983). The relationship includes a melting point (MP) correction for chemicals that are solids at room temperature. The regression analysis using K and the estimated aqueous solubility of the liquid (or supercooled liquid- for solids) yields the following relationship (Bowman and Sans 1983): Log KQw = 0.280 - 0.839 log S$cl where S , is the molar aqueous solubility of the supercooled liquid. The relationship between the molar solubility of the solid, S i-d, and the supercooled liquid, S ^, is given by the following relationship (Bowman and Sans 1983, Miller et al . 1985): W Tm ( . , where AHf"/Tm « AS, the entropy of fusion, is reported to be 13.5 cal/mole a°K for most low melting point compounds and R « 1.987 cal/mole °K. 25 ------- Appendix C. Estimating K using CLOGP ow The CLOGP program, based on Leo's Fragment Constant Method (Lyman et al. 1982), estimates Tog K using fragment constants (fj) and structural factors (F-j) that have been empirically derived for many molecular groups. The estimated K is obtained from the sum of constants and factors for each of the molecular subgroups comprising the molecule as follows: "ow ' t The method assumes that log KQw is a linear additive function of the structure of the solute and its constitutive parts, and that the most important structural effects are described by available factors. The structure of the compound is specified using the Simplified Molecular Interactive Linear Entry System (SMILES) notation (Weininger and Weininger 1986). The notation uniquely describes the empirical formula and molecular structure of the compound of interest. 26 ------- Appendix 0. Example Calculation of Aroclor KQW Aroclor Aroclor 1242 Homo log Homo log Homolog Fraction log K ow mono 0.03 4.33 di 0.13 5.12 tri 0.28 5.57 tetra 0.30 5.84 penta 0.22 6.35 hexa 0.04 7.05 For Aroclor 1242: log KQw » log1Q [0.03(104'33) + 0.13(105*12) + 0.28(105'57) + 0.30(105'84) + 0.22(106'35) + 0.04(107'05) - log1Q [1,270,684] log KQw * 6.10 27 ------- Appendix E. Interim Sediment Criteria Values for Polycyclic Aromatic Hydrocarbons In the main part of this report, interim sediment criteria values are shown for 11 nonpolar organic compounds for which chronic water quality criteria have been generated. During visits to regional Superfund offices to discuss the application of these interim criteria values, the need for interim criteria for several of the polycyclic aromatic hydrocarbons (PAHs) was identified. The PAHs are Fluoranthene, Pyrene, Benzo(a)pyrene and Benzo(a)anthracene. This appendix describes how those interim criteria values were developed and provides all supporting data for their calculation. The freshwater chronic water quality criteria values for the four PAHs given in Table E.I were determined by Mr. Anthony (Ron) Carlson of EPA-Duluth using the computer automated method system developed by that laboratory. The calculated criteria values for Phenanthrene and Acenaphthene are given for comparison. Accepted chronic criteria for these compounds are 0.006 mg/L and 0.057 mg/L, respectively. The computerized method which uses the K and solubility values for the compound to calculate both the acute and chronic water quality criteria is based on the work of Veith et al. (1983) and Call et al. (1985). The values given are based on toxicity to chronically exposed fathead minnows. This method is estimated to provide acute and chronic toxicity values that are within a factor of 3 and 5, respectively, of the actual values for approximately 80% of the known industrial compounds. The log K values for the four PAHs given in Table E.2 were developed 3 ow as described in the main part of the report. The log KQC values and 28 ------- Table E.I. Predicted Fathead Minnow Toxicity Values Compound Name Phenanthrene Acenaphthene Fluoranthene Pyrene Benzo(a)pyrene Benzo(a)anthracene I* * A* J • Nuaber 85018 83329 206440 129000 50328 56553 Mode of Action(a) NN NN NN NN NN(?) NN(?) L°9 Kow 4.49 4.07 4.95 4.95 6.12 5.66 Solubility (mg/1) 1.26 4.44 0.24 0.13 0.004 0.014 Acute Chronic (mg/1) 0.6 1.3 0.25 0.25 0.025 0.061 0.035 0.086 0.013 0.013 0.0012 0.0030 a) NN = Nonpolarnarcosis. The question nark indicates that the mode of action is not known with certainty. ------- the 95% confidence interval for the log KOC values given in Tabl$ E.3 were also estimated using the same methods and assumptions described previously. The final criteria values are given in Table E.4 on an organic carbon normalized basis and for 1% and 10% organic carbon in Table E.5. These interim criteria values can be used in the same way as the interim values in the main part of the report to determine if a sediment sample exceeds or does not exceed the criteria values. 29 ------- Table E.2. Log K values for the four PAHs in Table 1 ow Log Kow Compound Fluoranthene Pyrene Benzo(a)pyrene Benzo( a) anthracene Leo-Hanch Measured 5.2 4.88 5.97 Fragment Analysis 4.95 4.95 6.12 5.66 Aqueous Solubil ity 4.61 4.44 5.81 6.34 Geometric Mean 5.25 5.09 6.05 5.74 S 0 0 0 0 .0. .139 .187 .168 .264 9f Confidence 4.98 4.72 5.72 5.22 >V Interval 5.52 5.46 6.38 6.26 ------- Table E.3. Log K and log K values for the for PAHs. ow oc Compound Geometric Mean Log Kow s.o. 95% Confidence Interval Log Koc Mean S.D. 95% Confidence Interval Muoranthene 5.25 Pyrene 5.09 Bcnzo(a)pyrene 6.05 Benzo(a)anthracene 5.74 0.139 0.187 0.168 0.264 4.98 4.72 5.72 5.22 5.52 5.46 6.38 6.26 5.16 5.00 5.95 5.64 0.33 0.35 0.34 0.40 4T51 4.31 5.27 4.86 5.81 5.70 6.62 6.43 ------- Table E.5. Interim Sediment Quality Criteria values for four PAHs for 1 and 10% Organic Carbon Contents. (All criteria with units of ppm or mg/kg.) 1% 10% Compound Fluoranthene Pyrene Benzo(a)pyrene Benzo ( a) anthracene Mean 18.8 13.1 10.6 13.2 95% Confidence Interval Mean 4.24 2.66 2.25 2.17 83.8 64.6 50.2 80.0 188 131 106 132 95% Confidence Interval 42.7 26.6 22.5 21.7 838 646 502 800 ------- Table E.4. Interim Sediment Quality Criteria values for four PAHs. Compound Fluoranthene Pyrene Benzo(a)pyrene Benzo(a) anthracene WQC 13 13 1.2 3 Sediment Quality Criteria (ug/g C) Mean 95% Confidence Interval 1,883 423 8,375 1,311 265 6,465 1,063 225 5,018 1,317 217 7,999 33 ------- CONTRIBUTORS Dr. Oom DiToro, Mr. Paul Paquin, and Mr. Benjamin Wu of HydroQual, Inc., tabulated Kow and K0c values for the nonpolar contaminants, with assistance from Or. Spyros Pavlou and Mr. Roger Kadeg of Envirosphere Co, Inc. Mr. David Hansen of EPA's Narragansett Laboratory, and Mr. Nelson Thomas and Mr. Ron Carlson of EPA's Duluth Laboratory provided water quality criteria for nonpolar contaminants, the QSAR based criteria for PAHs, and the appendix on criteria development procedures. All of the above people as well as Or. Herb Allen of Drexel University, Ms. Alexis Steen and Dr. James Fava of Battelle, and Mr. Chris Zarba, the EPA Work Assignment Manager, provided valuable comments on the draft of this report. ------- |