v°xEPA
United States
Environmental Protection
Agency
EPA-530-R-14-003
October 2014
Risk Assessment of Spent Foundry Sands
In Soil-Related Applications
Evaluating Silica-based Spent Foundry Sand
From Iron, Steel, and Aluminum Foundries
Prepared by:
U.S. EPA Office of Resource Conservation and Recovery
Economics and Risk Assessment Staff
U.S. Department of Agriculture-Agricultural Research Service
The Ohio State University
and
RTI International under EPA contract number: EP-W-09-004
UJ
o
Agricultural
Research
Service
T • H • E
OHIO
SIAIE
UNIVERSITY
-------
[This page intentionally left blank.]
-------
Disclaimer
Disclaimer
This document was prepared by staff from the Office of Resource Conservation and
Recovery (ORCR), U.S. Environmental Protection Agency (EPA), the U.S. Department of
Agriculture-Agricultural Research Service (USDA-ARS), and The Ohio State University (OSU).
This document was subsequently reviewed by the EPA Office of Solid Waste and Emergency
Response (OSWER), USDA-ARS, and OSU, as well as externally peer reviewed. Any opinions,
findings, conclusions, or recommendations do not change or substitute for any statutory or
regulatory provisions. This document does not impose legally binding requirements, nor does it
confer legal rights, impose legal obligations, or implement any statutory or regulatory provisions.
Mention of trade names or commercial products is not intended to constitute endorsement or
recommendation for use. This document is being made available to the public. Any questions or
comments concerning this document should be addressed to Timothy Taylor, U.S.
Environmental Protection Agency, Office of Resource Conservation and Recovery, 1200
Pennsylvania Ave. N.W., Washington, DC 20460 (email:Taylor.Timothy@epa.gov).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Acknowledgments
The U.S. Environmental Protection Agency (EPA), U.S. Department of Agriculture-
Agricultural Research Service (USDA-ARS), and The Ohio State University (OSU) are
extremely grateful to the following individuals and organizations for their substantial
contributions to this document.
Particular thanks are due to Dr. Robert Dungan and Dr. Rufus Chaney of USDA-ARS for
their initial study of spent foundry sands and for their subsequent writing and editing of
substantial portions of the document.
Thanks are also due to Dr. Libby Dayton and Dr. Nick Basta of OSU for their intensive
study and analysis of the sands, contributions to writing the document, and their helpful review
of intermediate drafts of the document.
The risk assessment work would not have been possible without the support of RTI
International, especially Steve Beaulieu and Donna Womack.
Finally, EPA would like to thank the dedicated staff of the Office of Solid Waste and
Emergency Response, Office of Resource Conservation and Recovery, many of whom played an
important part in structuring, conducting, and writing portions of the risk assessment, without
whom, this document would not have been possible in its current form.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
List of Abbreviations
Table of Contents
Section Page
Disclaimer i
Acknowledgments ii
List of Abbreviations ix
Executive Summary 1
1. Introduction 1-1
1.1 Purpose 1-3
1.2 Major Features of the SFS Evaluation 1-4
1.3 Roadmap to this Report 1-6
2. Background and Characteristics of Spent Foundry Sand 2-1
2.1 Foundry Sand Characteristics 2-1
2.2 Molding and Core Sands 2-1
2.2.1 Green Sands 2-1
2.2.2 Chemically Bonded Sands 2-2
2.3 Reclamation and Disposal 2-3
2.4 Collection and Analysis of U.S. SFSs 2-4
2.4.1 Spent Foundry Sand Collection 2-4
2.4.2 PAHs and Phenolics 2-6
2.4.3 Dioxins and Dioxin-like Compounds 2-6
2.4.4 Trace Elements 2-6
2.4.5 Leach Tests 2-7
2.5 Constituents and Properties of Spent Foundry Sand 2-7
2.5.1 Properties Important to Soil Quality and Function 2-7
2.5.2 Metals and Metalloids 2-10
2.5.3 Organics 2-13
2.5.4 Constituent Leaching Potential 2-21
2.5.5 Plant Uptake of Trace Metals from Spent Foundry Sands 2-28
2.5.6 Potential to Impact Soil Biota 2-29
3. Problem Formulation 3-1
3.1 Scope of the SFS Risk Screening 3-1
3.1.1 Types of SFSs 3-1
3.1.2 SFS Characteristics 3-2
3.1.3 Beneficial Uses of SFS 3-3
3.1.4 Conceptual Models 3-3
3.1.5 Assumptions Behind the Risk Screening 3-6
3.2 Analysis Plan 3-8
3.2.1 Analysis Phase I: Identifying Constituents of Concern 3-9
3.2.2 Analysis Phase II: Risk Modeling 3-12
4. Analysis Phase I: Identification of COCs for Modeling 4-1
4.1 Purpose 4-1
Risk Assessment of Spent Foundry Sands in Soil-Related Applications iii
-------
List of Abbreviations
4.2 Groundwater Exposure 4-2
4.2.1 LeachateData 4-2
4.2.2 Selection of Constituents 4-3
4.2.3 Comparisons to Screening Levels and Regulatory Levels 4-3
4.2.4 Results 4-6
4.3 Inhalation Exposure 4-6
4.3.1 Scenario 4-7
4.3.2 Selection of Constituents of Potential Concern 4-7
4.3.3 Deterministic Modeling 4-9
4.3.4 Results 4-13
4.4 Soil Pathways Exposure 4-14
4.4.1 Remove SFS Constituents that are Nondetects 4-15
4.4.2 Remove SFS Constituents without Benchmarks 4-16
4.4.3 Remove SFS Constituents by Comparing to SSLs andEco-SSLs 4-18
4.4.4 Results 4-23
4.5 Analysis Phase I Results 4-23
5. Analysis Phase II: Risk Modeling of COCs 5-1
5.1 Overview of Phase II Probabilistic Modeling 5-1
5.2 Screening Probabilistic Modeling of the Groundwater Ingestion Pathway 5-2
5.2.1 Groundwater Model Inputs 5-3
5.2.2 Groundwater Model Outputs 5-5
5.2.3 Results 5-6
5.3 Refined Probabilistic Modeling of the Soil/Produce and Groundwater Ingestion
Pathways 5-6
5.3.1 Modeling Framework Overview 5-8
5.3.2 Exposure Scenario—Use of SFS in Home Gardens 5-10
5.3.3 Potential Release Pathways and Receptors 5-11
5.3.4 Source Modeling 5-13
5.3.5 Fate and Transport: Refined Groundwater Modeling 5-16
5.3.6 Fate and Transport: Produce Modeling 5-19
5.3.7 Human Exposure Modeling 5-20
5.3.8 Ecological Exposure Modeling 5-30
5.3.9 Human Health Effects Modeling 5-33
5.3.10 Ecological Effects Modeling 5-39
5.3.11 Calculating Modeled SFS-Specific Screening Levels 5-40
5.3.12 Results: Comparing Screening Values to SFS Constituent
Concentrations 5-42
6. Risk Characterization 6-1
6.1 Overview of the Risk Characterization 6-1
6.2 Key Risk Assessment Questions 6-3
6.3 Overarching Concepts 6-3
6.3.1 Background Concentrations 6-3
6.3.2 Chemical Reactions in Soil 6-4
6.3.3 Soil-Plant Barrier 6-5
6.3.4 Interactions Among Constituents 6-6
6.3.5 Highly Exposed Populations 6-7
Risk Assessment of Spent Foundry Sands in Soil-Related Applications iv
-------
List of Abbreviations
6.4 Spent Foundry Sand Product Risks 6-8
6.5 PAHs, Dioxins, Furans, and Dioxin-Like PCBs in SFS 6-8
6.5.1 PAHs 6-8
6.5.2 PCDDs, PCDFs, and Dioxin-like PCBs 6-9
6.6 Phenolics in SFS 6-10
6.7 Metals and Metalloids in SFS 6-11
6.7.1 Antimony 6-11
6.7.2 Arsenic 6-14
6.7.3 Chromium 6-3
6.7.4 Cobalt 6-5
6.7.5 Copper 6-8
6.7.6 Iron 6-11
6.7.7 Manganese 6-13
6.7.8 Nickel 6-16
6.7.9 Other Metals 6-20
6.8 Uncertainty Characterization 6-21
6.8.1 Risk Screening Modeling 6-21
6.8.2 State-of-the-Science on SFS 6-25
7. Findings and Conclusions 7-1
7.1 Beneficial Use of SFS (Chapter 1) 7-1
7.2 Characterization of SFS (Chapter 2) 7-1
7.3 Exposure Scenarios Examined (Chapters) 7-1
7.4 Screening of Exposure Pathways (Chapter 4) 7-2
7.5 Modeling of Exposures from Home Gardening (Chapter 5) 7-2
7.6 Characterization of Risks Associated With SFS Beneficial Use (Chapter 6) 7-3
8. Agency Policy on the Beneficial Use of Silica-Based Spent Foundry Sands from Iron,
Steel and Aluminum Foundries 8-1
9. References 9-1
List of Figures
Figure 1-1. Framework for the SFS assessment 1-6
Figure 2-1. Dehydrogenase activities at (a) week 4, (b) week 8, and (c) week 12 in
Sassafras sandy loam soil amended with 10%, 30%, and 50% (dry wt.) spent green
sand from iron, aluminum, or brass foundries 2-32
Figure 2-2. Dehydrogenase activities at (a) week 4, (b) week 8, and (c) week 12 in
Sassafras sandy loam soil amended with 10%, 30%, and 50% (dry wt.) fresh core
sand made with either phenol-formaldehyde, phenolic urethane, or furfuryl alcohol
based resins 2-33
Figure 2-3. Adult earthworm survival after 28 days in the SFS blends 2-35
Figure 3-1. Conceptual model: the use of SFS in roadway subbase 3-4
Figure 3-2. Conceptual model: the blending site 3-5
Figure 3-3. Conceptual model: the use of SFS-manufactured soils in home gardens 3-6
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
List of Abbreviations
Figure 3-4. Analysis Plan for the risk assessment of SFS uses in soil-related applications 3-10
Figure 5-1. How the Monte Carlo approach addresses uncertainty 5-2
Figure 5-2. Conceptual Cross-Section View of the Modeled Subsurface 5-3
Figure 5-3. Basic Monte Carlo looping structure for the home garden 5-9
Figure 5-4. Model stability 5-10
Figure 5-5. Meteorological regions and SFS use areas 5-11
Figure 5-6. Conceptual model for modeling the home gardener 5-12
Figure 5-7. Analysis of Home Range Sizes for the Short Tailed Shrew 5-32
Figure 6-1. Concentration distributions of antimony in SFS (top) and U.S. and Canadian
soils (bottom) 6-13
Figure 6-2. Concentration distributions of arsenic in SFS (top) and U.S. and Canadian
soils (bottom) 6-1
Figure 6-3. Concentration distributions of chromium in SFS (top) and U.S. and Canadian
soils (bottom) 6-4
Figure 6-4. Concentration distributions of cobalt in SFS (top) and U.S. and Canadian
soils (bottom) 6-7
Figure 6-5. Concentration distributions of copper in SFS (top) and U.S. agricultural soils
(bottom) 6-10
Figure 6-6. Concentration distributions of iron in SFS (top) and U.S. and Canadian soils
(bottom) 6-12
Figure 6-7. Concentration distributions of manganese in SFS (top) and U.S. and Canadian
soils (bottom) 6-15
Figure 6-8. Concentration distributions of nickel in SFS (top) and U.S. and Canadian
soils (bottom) 6-18
List of Tables
Table ES-l: Phase I Results - SFS Constituents Requiring Further Evaluation 4
Table ES-2. Comparing SFS Concentrations to Various Screening Values (mg kg'Mry
weight, unless otherwise noted) 6
Table 2-1. Description of the U.S. SpentFoundry Sands 2-5
Table 2-2. Particle Size Distribution, USDA Textural Class, and Bulk Density for 43
SFSs 2-9
Table 2-3. Total Metal Concentrations in the Spent Foundry Sands as Determined by
EPA Method 3050B 2-11
Table 2-4. Metal Concentrations in 39 of 43 SpentFoundry Sands (June 2005 Samples)
as Determined by EPA Method 305 lAa 2-12
Risk Assessment of Spent Foundry Sands in Soil-Related Applications vi
-------
List of Abbreviations
Table 2-5. Concentrations of the PAHs in Spent Foundry Sands 2-15
Table 2-6. Concentrations of Phenolics in Spent Foundry Sands 2-16
Table 2-7. Description of the Spent Foundry Sands Analyzed for PCDDs, PCDFs, and
CoplanarPCBs 2-18
Table 2-8. Concentrations of PCDDs, PCDFs, and Coplanar PCBs and Homolog Totals
in the Spent Foundry Sands (n =1) 2-19
Table 2-9. Toxicity Equivalents (TEQs) of PCDDs, PCDFs, Coplanar PCBs, and Total
Dioxins in the Spent Foundry Sands 2-21
Table 2-10. Metal Concentrations in the TCLP Extracts from the Spent Foundry Sands 2-23
Table 2-11. Spent Foundry Sands TCLP Extracts Compared to Toxicity Characteristic
Regulatory Levels 2-24
Table 2-12. Metal Concentrations in the SPLP Extracts from the Spent Foundry Sands 2-25
Table 2-13. Metal Concentrations in Water Extracts from the Spent Foundry Sands 2-27
Table 2-14. Total and DTPA-Extractable Metal Concentrations in the Brass Green Sand
Blends 2-36
Table 4-1. Leaching Data for Silica-based Iron, Steel, and Aluminum SFSsOngL"1) 4-3
Table 4-2. Leachate Comparisons (mgL"1) 4-4
Table 4-3. Recommended Dermal Exposure Parameters for RME Residential Scenario 4-5
Table 4-4. Comparison of Water Dermal Absorbed Doses (DADs) to Health Benchmarks 4-5
Table 4-5. Inhalation Human Health Benchmarks 4-8
Table 4-6. Input Parameters for SCREENS 4-12
Table 4-7. SCREENS Output Summary 4-12
Table 4-8. Comparison to Screening Values: Inhalation Pathway 4-13
Table 4-9. Constituents Detected in at Least One Sample 4-15
Table 4-10. Residential Soil Screening Levels (mg kg"1)3 4-17
Table 4-11. Comparison to Dermal Soil Screening Levels 4-20
Table 4-12. Ecological Screening Criteria Used in the Analysis a 4-21
Table 4-13. Comparing SFS-manufactured Soil to Human and Ecological SSLs 4-22
Table 4-14: SFS Constituents Retained for Phase II Risk Modeling 4-24
Table 5-1. Tested Leachate Concentrations, Receptor Well Concentrations for the Home
Gardener Exposure Scenario, and Screening Levels (mgL"1) 5-5
Table 5-2. Human Exposure Pathways for SFS-Manufactured Soil in Home Gardens 5-13
Table 5-3. EPACMTP Arrival Times of Arsenic Plume at the Receptor Well 5-19
Table 5-4. Produce and Drinking Water Consumption Rate (CR), Body Weight, and
Exposure Duration Distributions for the Home Gardener 5-22
Risk Assessment of Spent Foundry Sands in Soil-Related Applications vii
-------
List of Abbreviations
Table 5-5. Summary of Exposure Parameters with Fixed Values Used in Probabilistic
Analysis 5-23
Table 5-6. Summary of Produce Consumption Rates (CR) 5-25
Table 5-8. Example 50th Percentile Child Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg"1 d"1) 5-27
Table 5-9. Example 90th Percentile Adult Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg"1 d"1) 5-28
Table 5-10. Example 90th Percentile Child Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg"1 d"1) 5-29
Table 5-11. 50th and 90th Percentile Ecological Exposure Model Outputs for SFS-
Manufactured Soil Constituents 5-33
Table 5-12. Human Health Benchmarks Used in Phase II Analysis 5-35
Table 5-13. 50th and 90th Percentile Adult Unitized Dose Ratios for SFS-Manufactured
Soil Constituents 5-37
Table 5-14. 50th and 90th Percentile Child Unitized Dose Ratios for SFS-Manufactured
Soil Constituents 5-38
Table 5-15. Eco-SSLs Used in Phase II Analysis (mg kg"1 soil) 5-40
Table 5-16. 50th and 90th Percentile Ecological Unitized Dose Ratios for SFS-
Manufactured Soil Constituents 5-40
Table 5-17. Modeled SFS-specific Screening Levels for the Home Garden Scenario 5-41
Table 5-18. Modeled SFS-specific Ecological Screening Levels for the Home Garden
Scenario (mg kg"1 SFS) 5-42
Table 5-19. Comparing SFS Constituent Concentrations to Modeled SFS-Specific
Screening Levels (mgkg"1 SFS) 5-42
Table 6-1. Comparison of PAH Concentrations in SFS to Screening Criteria (mg kg"1) 6-9
Table 6-2. Comparison of Total Dioxin TEQ Concentrations in SFS to Screening Criteria 6-10
Table 6-3. Comparison of Phenolic Concentrations in SFS to Screening Criteria 6-11
Table 6-4. Home Gardening 90th Percentile Modeled SFS-specific Screening Levels
for Arsenic 6-2
Table 6-5. Summary of Other SFS Metal Concentrations and Relevant Screening Criteria 6-21
Table 7-1. Comparing SFS Concentrations to Various Screening Values (mg kg"1 unless
otherwise noted) 7-4
Table 8-1. Quantity SFS Beneficially used, by Market (tons) 8-1
Table 8-2. Primary Environmental Benefits of Beneficial use of SFS, by Market 8-1
Risk Assessment of Spent Foundry Sands in Soil-Related Applications viii
-------
List of Abbreviations
Abbreviation
2,3,7,8-TCDD
3MRA
ADD
AGS
AMC
ARS
ASE
ASTM
ATSDR
AWC
BGS
BTEX
CASRN
CEC
CERCLIS
COC
CONUS
CPP
CR
CSEFH
CSF
CV
DAF
DHA
DMG
DTPA
DW
DWEL
Eco-SSL
EFH
El
EP
EPA
EPACMTP
ET
List of Abbreviations
Definition
2,3,7,8-tetrachlorodibenzo-/>-dioxin
Multipathway, Multimedia, Multireceptor Risk Assessment Modeling
System
average daily dose
aluminum green sand
antecedent moisture class
Agricultural Research Service
accelerated solvent extractor
American Society for Testing and Materials
Agency for Toxic Substances and Disease Registry
available water capacity
brass green sand
benzene, toluene, ethylbenzene, and xylenes
Chemical Abstract Service Registry Number
cation exchange capacity
Comprehensive Environmental Response, Compensation, and Liability
Information System
constituent of concern
contiguous United States
chemical properties processor
consumption rate
Child Specific Exposure Factors Handbook
cancer slope factor
coefficient of variation
dilution attenuation factor
dehydrogenase activity
dry matter growth
diethylenetriamine pentaacetic acid
dry weight
Drinking Water Equivalent Level
ecological soil screening level
Exposure Factors Handbook
erosivity index
extraction procedure
U.S. Environmental Protection Agency
EPA's Composite Model for Leachate Migration with Transformation
Products
Evapotranspiration
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
IX
-------
List of Abbreviations
Abbreviation
EXAFS
FC
foe
GC-MS
GIRAS
GIS
GSCM
HGDB
HH-SSL
HMTA
Hp
HQ
HS-SPME
Hx
ICP-AES
ICP-MS
IGS
IRIS
ISC3
ISCST3
IUPAC
IWEM
LADD
LDH
LOEL
LOQ
MAP
MCL
MDI
MDL
MRL
NAAQS
NAPL
NBS
NIST
NOEL
NOM
NPDWS
NRC
Definition
extended X-ray absorption fine structure spectroscopy
field capacity
fraction organic carbon
gas chromatography-mass spectrometry
Geographic Information Retrieval and Analysis System
geographic information system
Generic Soil Column Model
Hydrogeologic DataBase for Modeling
human health soil screening level
hexamethylenetetramine
hepta
hazard quotient
headspace solid-phase microextraction
hexa
inductively coupled plasma-atomic emission spectrometry
inductively coupled plasma-mass spectrometry
iron green sand
Integrated Risk Information System
Industrial Source Complex Model version 3
Industrial Source Complex-Short Term Model version 3
International Union of Pure and Applied Chemistry
Industrial Waste Management Evaluation Model
lifetime average daily dose
layered double hydroxide
lowest observable effects level
limit of quantitation
moisture adjustment factors
Maximum Contaminant Level
diphenylmethane-4,4-diisocyanate
method detection limit
Minimum Risk Level
National Ambient Air Quality Standard
nonaqueous phase liquid
steel phenolic urethane no-bake sand
National Institute of Standards and Technology
no observable effects level
natural organic matter
National Primary Drinking Water Standards
U.S. Nuclear Regulatory Commission
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
List of Abbreviations
Abbreviation
NSDWS
NWS
o
oc
ORCR
OSHA
OSU
OSWER
PAH
PCB
PCDD
PCDF
PDF
Pe
PET
PMio
ppm
PPRTV
PQL
PVC
RCRA
RDMG
REL
RfC
RfD
RSD
SAB
SAMSON
SCRAM
scs
Sdev
SFS
SMFC
SMS
SMWP
SOM
SPLP
SSL
STATSGO
Definition
National Secondary Drinking Water Standard
National Weather Service
octa
organic carbon
Office of Resource Conservation and Recovery
Occupational Safety and Health Administration
The Ohio State University
Office of Solid Waste and Emergency Response
polycyclic aromatic hydrocarbon
polychlorinated biphenyl
polychlorinated dibenzo-p-dioxin
polychlorinated dibenzofuran
probability distribution function
penta
potential evapotranspiration
particulate matter with a mean aerodynamic diameter of 10 microns or less
parts per million
Provisional Peer-Reviewed Toxicity Value
practical quantitation limit
polyvinyl chloride
Resource Conservation and Recovery Act
relative dry matter growth (relative to controls)
Reference Exposure Level
Reference Concentration
Reference Dose
relative standard deviations
Science Advisory Board
Solar and Meteorological Surface Observation Network
Support Center for Regulatory Air Models
Soil Conservation Service
standard deviation
spent foundry sand
Soil moisture field capacity
spent mushroom substrate
Soil moisture wilting point
soil organic matter
synthetic precipitation leaching procedure
Soil Screening Level
State Soil Geographic Database
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
XI
-------
List of Abbreviations
Abbreviation Definition
T Tetra
TCLP Toxicity Characteristic Leaching Procedure
TEF toxic equivalency factor
TEQ toxic equivalency value
UAC unitized air concentration
USDA U. S. Department of Agriculture
USLE Universal Soil Loss Equation
WHO World Health Organization
WMU waste management unit
WP wilting point
WW wet weight
Risk Assessment of Spent Foundry Sands in Soil-Related Applications xii
-------
Executive Summary
Executive Summary
Purpose: To provide states with a sound scientific basis from which to
evaluate the health risks to human and ecological receptors associated with
the beneficial use of silica-based spent foundry sand (SFS) from iron, steel,
and aluminum foundries in soil-related applications.
Within the scope and limitations of this evaluation, the following conclusions
were drawn:
• Metals found in SFS are present at concentrations similar to background
in U.S. and Canadian soils.
• The conclusions of this report apply to silica-based SFS from iron, steel,
and aluminum foundries.
• The evidence demonstrates that the evaluated uses of silica-based SFS
produced by iron, steel, and aluminum foundries (i.e., used in
manufactured soil, in soil-less potting media, and in road subbase) were
found to be protective of human health and ecological receptors.
Roughly 2.6 million tons of SFS is beneficially used each year outside of the foundries,
of which 14% is used in soil-related applications (USEPA, 2008c). In 2002, the U.S. Department
of Agriculture's Agricultural Research Service (USDA-ARS) implemented the Foundry Sand
Initiative under National Program 206 (Manure and Byproduct Utilization; renamed since to NP
214 - Agricultural and Industrial Byproducts) to address agricultural and horticultural uses of
SFS. A collaborative effort was initiated to evaluate the potential risks of using SFS in soil-
related applications and to encourage this beneficial use if found to be protective of human
health and the environment. USDA-ARS, The Ohio State University (OSU), and the U.S
Environmental Protection Agency (EPA) formed an expert team of agronomists, soil scientists,
and environmental health risk assessors to develop an SFS-specific risk assessment. The overall
goals for this document were to:
• Review the available information on SFS in soil-related applications
• Identify likely exposure pathways and receptors associated with various use scenarios
• Use a combination of screening and modeling methods to determine whether the
proposed unencapsulated uses of SFS are protective of human health and the environment
• Discuss the findings within the context of certain overarching concepts (e.g., the
complexities of soil chemistry) and provide conclusions.
Reviewing Available Information: SFS Characterization
Forty-three samples of spent molding and core sand from U.S. foundries were collected
and analyzed by USDA-ARS and OSU. Other materials, such as broken or unused cores, or floor
sweepings from core room operations, were not examined in this evaluation. The characteristics
of the samples taken are as follows:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
ES-1
-------
Executive Summary
• Metal cast: 4 aluminum sands, 31 iron sands, 6 steel sands, and 2 non-leaded brass sands
Only nonhazardous SFSs are within the scope of this evaluation. Sands from brass and
bronze foundries that use lead are frequently hazardous waste because they leach lead at
levels above the federal regulatory limit (see 40 CFR 261.24). Therefore, sands from
leaded brass and bronze foundries were not collected, and such sands were not evaluated
in this study.
• Mineral type: 41 silica sands and 2 olivine sands
• Binder type of molding sand: 36 green sands and 7 chemically bound sands.
USDA collected the initial 43 samples in June 2005. To test variation over time, USDA
trained foundry personnel in proper collection techniques, and most foundries collected and sent
USDA two additional sample sets, in September 2005 and July 2006.l USDA conducted total
constituent testing on all samples for elements (metals and metalloids), polycyclic aromatic
hydrocarbons (PAHs), and phenolics. Ten of the June 2005 samples were also analyzed for
dibenzodioxins, dibenzofurans, and dioxin-like polychlorinated biphenyls (PCBs).
OSU also conducted total constituent testing on the initial 43 samples for elements. The
test method that OSU used had a lower detection limit than the method used by USDA, and was
therefore able to more accurately estimate concentrations at the lower end of the range.
To characterize the leaching behavior of trace elements, USDA conducted leach tests on
SFS using the toxicity characteristic leaching procedure (TCLP), the synthetic precipitation
leaching procedure (SPLP), and the American Society for Testing and Materials (ASTM)
International method D 3897.2 The conditions simulated by SPLP (leaching from soil due to acid
rain) and the ASTM method (material's native leaching potential) are more relevant than TCLP
(highly acidic leaching in a municipal waste landfill) for evaluating the conditions considered in
this report. Therefore, TCLP leach data were only used in this evaluation if SPLP or ASTM
leach data were not available.
To assess plant uptake of trace metals, USDA grew spinach, radishes, and perennial
ryegrass in a 50% SFS mixture with added nutrients. Spinach and radish experienced typical
levels of elements. Ryegrass, on the other hand, was found to be iron deficient and contained
elevated but nontoxic concentrations of boron, manganese, and molybdenum.
USDA also assessed the potential of SFS to impact soil invertebrates. This was done in a
28-day experiment where earthworms were placed in blends of 10%, 30%, and 50% SFS. The
worms did not exhibit higher levels of any elements, except in the samples from the two non-
leaded brass foundries.
Data were identified from industry, academia, and the peer-reviewed literature. However,
based on the number, geographic distribution and types of sampled foundries and SFS, and the
breadth of aspects studied, as well as the types of analytical methods used and the level of
QA/QC built into the studies, the USDA and OSU datasets are considered the most complete and
1 38 foundries (88%) sent samples in September 2005, and 37 foundries (86%) sent samples in July 2006. 79% of
foundries sent samples on both dates.
2 TCLP (U.S. EPA SW-846, method 1311, U.S. EPA, 2007a)
SPLP (U.S. EPA SW-846, method 1312, U.S. EPA, 2007a)
ASTM (ASTM International, 2004)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-2
-------
Executive Summary
scientifically robust. The risk assessment therefore used the OSU totals dataset because it more
accurately represented the low end of concentration ranges, and the USDA leachate data.
The existing data on non-leaded brass sands and olivine sands demonstrated levels of
copper, lead, nickel, and zinc that were both potentially phytotoxic and much higher than the
other 39 SFSs, but insufficient samples existed to characterize constituent concentration
variability. Therefore, while descriptions of non-leaded brass sands and olivine sands are
retained for completeness, they are not evaluated in the risk assessment, and any risk assessment
findings apply only to silica-based SFS from iron, steel and aluminum foundries.
Identifying Likely Exposure Pathways/Receptors: Conceptual Model
The purpose of this analysis is to evaluate whether the use of silica-based SFSs from iron,
steel, and aluminum foundries will be protective of human and ecological receptors in the United
States if the SFSs are used in manufactured soils, soil-less potting media, or road subbase. This
evaluation defines "protective" as a reasonably maximally exposed individual incurring no more
than a 10~5 excess risk of cancer, or for noncancer effects, exposures to ensure that the effects
would not be expected over a lifetime, for both human and ecological receptors.
As discussed in Chapter 3, the quantitative evaluation focused on the use of SFS in
manufactured soils (comprised of 50% SFS, by weight), because potential exposure to human
and ecological receptors from constituents of concern was judged to be higher than potential
exposures in the other two uses. Therefore, if the potential for adverse effects to human and
ecological receptors from SFS-manufactured soils was found to be protective, then the other two
uses would also be protective.
The exposure scenarios that were judged to have the greatest potential for human and
ecological exposure from the use of SFS in manufactured soils included residents living near
commercial blending facilities,3 home gardeners that use SFS-manufactured soils, and ecological
receptors that come in contact with these home gardens. The conceptual models developed for
these scenarios describe potential exposures to adult and child receptors through three basic
pathways: (1) groundwaterpathway - the ingestion and dermal exposure to groundwater
contaminated by the leaching of SFS constituents; (2) ambient air pathway - the inhalation of
SFS emitted from soil blending operations; and (3) soil pathway - the incidental ingestion and
dermal exposure to SFS-manufactured soil, as well as ingestion of fruits and vegetables grown in
the soil. The conceptual models included exposures to ecological receptors through direct contact
with SFS-manufactured soil.
Screening and Modeling
Analytical data were available for 25 metals, 16 PAHs, 17 phenolics, and 20 dioxins and
dioxin-like compounds. In Phase I (screening), the SFS data and available screening criteria
(e.g., available health benchmarks, media-specific screening levels) and models were used to
determine which constituents, if any, required further evaluation. Phase II (risk modeling) used
constituent-, regional- and site-specific data to address the variability in home garden conditions
across the country.
3 Commercial soil blending facilities use construction equipment, such as a front-end loader, to combine large
volumes of the various mineral and organic components to manufacture soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-3
-------
Executive Summary
Each of the three pathways identified above was evaluated individually. In addition, the
soil pathway evaluation used screening levels that also addressed inhalation exposures. The
exposure scenarios and pathway evaluations were developed to produce conservative risk
estimates; that is, the methodology was designed to estimate risk from reasonable maximum
exposure, to ensure that the analysis included an ample margin of safety. This approach ensures
that the results of this analysis can be used to determine if soil-related uses of SFS are protective
of human health and the environment. The risk assessment provides decision makers with
information on the potential for adverse effects to the reasonably maximally exposed individuals
and ecological receptors that could come in contact with SFS.
Phase I Results
All PAHs, phenolics, and dioxin and dioxin-like compounds were screened out of all
three pathways, and therefore required no further evaluation. Inhalation screening eliminated all
SFS constituents (i.e., including the metals) from further evaluation; the inhalation pathway itself
therefore required no further evaluation. Dermal screening of soil and groundwater exposure
likewise found that all evaluated constituents were well below a level of concern, and dermal
exposure was also eliminated from further evaluation. However, based on groundwater ingestion
screening, soil multi-pathway exposure screening and ecological screening, 11 metals were
retained for further evaluation in the risk modeling phase. Table ES-1 lists the metals retained
for risk modeling.
Table ES-1: Phase I Results - SFS Constituents Requiring Further Evaluation
Human
Ecological
Groundwater Pathway
Antimony (Sb)
Arsenic (As)
Beryllium (Be)
Cadmium (Cd)
Lead (Pb)
Not evaluated
Inhalation
All constituents below a level of concern.
No need for further inhalation evaluation
Not evaluated
Soil/Produce
Arsenic (As)
Cobalt (Co)
Iron (Fe)
Antimony (Sb)
Chromium (Cr)
Copper (Cu)
Manganese (Mn)
Nickel (Ni)
Phase II Results
The SFS concentrations of all eleven modeled constituents fell below their respective
human and ecological modeled SFS-specific screening levels.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
ES-4
-------
Executive Summary
Table ES-2 summarizes the analytical and background soil information on metal
constituents in SFS.4 Human health SSLs and Eco-SSLs are provided. In addition, the table
provides the modeled screening values for the specific home gardener scenario developed in this
report, as well as modeled screening values based on median and high-end consumption by the
general public. As shown in this table, there is substantial evidence that the metal constituents
found in SFS are present at concentrations that are very similar to those found in native soils.
4 Table ES-2 lists only metals because all organics were screened out early in the analysis. Discussions and results
of the screening of organics can be found in Chapter 4.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-5
-------
Executive Summary
Table ES-2. Comparing SFS Concentrations to Various Screening Values (mg kg^dry weight, unless otherwise noted)
Elements
Alfekg-1)
As
B
Ba
Be
Cd
Co
Cr(III)
Cu
Fe (g kg'1)
Mn
Mo
Ni
Pb
Sb
Se
Tl
V
Zn
Silica-based Iron, Steel, and
Aluminum Sands"
Max
11.7
7.79
59.4
141
0.60
0.36
6.62
115
137
64.4
707
22.9
117
22.9
1.71
0.44
0.10
11.3
245
95%-ile
11.2
6.44
20.2
17.7
0.38
0.20
5.99
109
107
57.1
670
21.8
102
15.3
1.23
0.20
0.09
9.90
72.1
Median
5.56
1.05
10.0
5.00
0.15
0.05
0.88
4.93
6.22
4.26
54.5
0.50
3.46
3.74
0.17
0.20
0.04
2.88
5.00
SFS-
Manuf.
Soil
5.6
3.22
10.1
8.85
0.19
0.10
3.00
54.5
53.5
28.9
335
10.9
51.0
7.65
0.62
0.10
0.05
4.95
36.1
Human Screening Values
SSLd
77
6.7«
16,000
15,000
160
70
23
120,000
3,100
55
1,800
390
1,500
400
31
390
0.78
390
23,000
Modeled Consumption Rates0
Home
Gardener
—
8.0
-
—
—
-
22
—
—
160
-
-
—
—
-
-
—
—
--
Gen. Pop.
Median
—
30
-
—
—
-
58
—
—
230
-
-
—
—
-
-
—
—
--
Gen. Pop.
High
—
9.1
-
—
—
-
21
—
—
150
-
-
—
—
-
-
—
—
--
Eco Screening Values
Eco-
SSLse
ND
18
ND
330
21
0.36
13
34
49
ND
220
ND
38
56
0.27
0.52
ND
280
79
Modeled
(SFS-
specific)
—
40
-
—
—
-
-
510
159
-
1000
-
290
—
4.1
-
—
—
--
USDAf
—
-
-
—
—
-
-
—
200
-
-
-
200
—
-
-
—
—
300
U.S. and Canadian
Surface Soilsb
Max
87.3
18.0
ND
1800
4.0
5.2
143.4
5320
81.9
87.7
3,120
21.0
2,314
244.6
2.3
2.3
1.8
380
377
95%-ile
74.6
12.0
ND
840
2.3
0.6
17.6
70.0
30.1
42.6
1,630
2.16
37.5
38.0
1.39
1.0
0.7
119
103
Median
47.4
5.0
ND
526
1.3
0.2
7.1
27.0
12.7
19.2
490
0.82
13.8
19.2
0.60
0.3
0.5
55
56
— = No modeled value was generated because constituent was screened out of further study in an earlier stage of the evaluation. If a constituent screened out based on human
health SSL and had no Eco-SSL, the constituent was considered to have screened out for both human and eco.
ND = No Data.
a Source: Dayton et al. (2010).
b Source: Smith et al. (2005).
c See Chapter 5 for a detailed discussion of how the modeled values were generated.
d Concentrations of SFS constituents in manufactured soil (a 1:1 blend) were compared to an order-of-magnitude below the SSLs listed here, as discussed in Chapter 4,
Section 4.4.3. Values are from EPA Regional Screening Tables (http://www.epa.gov/reg3hwmd/risk/human/rb-concentration_table/Generic_Tables/index.htm'l. Unless
otherwise noted, all values are based on noncarcinogenic impacts.
e Concentrations of SFS constituents in manufactured soil (a 1:1 blend) were compared to the Eco-SSLs, as discussed in Chapter 4, Section 4.4.3.
f See Appendix C for an explanation of USDA Phytotoxicity Screening Values for copper, nickel, and zinc.
g Based on carcinogenic risk, set at the standard EPA Office of Resource Conservation and Recovery risk target level of IE-OS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
ES-6
-------
Executive Summary
Risk Characterization
Risk characterization summarizes the various lines of evidence presented earlier in the
evaluation and discusses them within the context of the conservative nature of screening risk
assessment and the complexities of soil chemistry. First, the context was set by reviewing the
high-level risk questions that the evaluation was designed to address, and by introducing
overarching concepts while reviewing constituent-specific information. Second, constituent-
specific information was reviewed and conclusions drawn.
When reviewing the various lines of evidence, it is important to keep in mind the key risk
assessment questions that this evaluation was designed to answer:
• Will the addition of SFSs to soil result in an increase in the constituent concentrations in
soil relative to background levels, and how should the results of the risk assessment be
interpreted across varied national soils?
• How do constituent forms found in the SFS matrix behave with respect to bioaccessibility
and bioavailability, and how does that affect potential risks?
• How will the behavior of individual constituents in manufactured soil, such as the soil-
plant barrier, impact the potential for exposure through the food chain pathway and,
ultimately, the potential for adverse human health and ecological effects?
• How do the risk assessment results compare to levels required to maintain nutritional
health in plants and animals?
When reviewing the various lines of evidence, there are also a number of other
overarching concepts to consider:
• Background Concentrations. Comparing the 95th percentile metal concentrations in
U.S. and Canadian soils to silica-based U.S. iron, steel, and aluminum SFSs reveals that
the concentrations of most trace metals in SFSs are below background concentrations in
U.S. and Canadian soils.
• Chemical Reactions in Soil. Metals reaching soils in elemental forms will oxidize
rapidly depending on the redox characteristics of the metal and the soil. Sorption is a
chemical process that buffers the partitioning of trace metals between solid and liquid
phases in soils and byproducts. Metal cations can sorb onto the metal oxides referred to
above, as well as onto soil organic matter.
• Soil-Plant Barrier. Soil chemical processes may limit the availability of metals for
uptake, while phytotoxicity limits the chances that contaminated plants will be consumed
(i.e., plant death acts as a barrier to contamination up the food chain).
• Interactions among Constituents. The presence (or absence) of some metals may affect
the toxicity of other metals. For example, copper-deficiency-stressed animals are more
sensitive to dietary zinc than animals fed with copper-adequate diets. Also, increased zinc
in forage diets strongly inhibits cadmium absorption and reduces liver and kidney
cadmium concentrations in cattle.
• SFS use as a manufactured soil component. The evaluation considered a high end use:
a 20 cm layer of manufactured soil containing 50% SFS (dry weight) in the blend. Blends
are much more likely to include 10% or less SFS (dry weight).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-7
-------
Executive Summary
Considering all of the above, and based on the evidence, most constituents were well
below SSLs and Eco-SSLs. Those that required further, more refined study were found to be
below levels of concern.
Conclusions
This assessment, driven by conservative assumptions and risk screening models, indicates
that the silica-based SFSs from iron, steel, and aluminum foundries evaluated in this report do
not pose risks of concern to human health or ecological receptors when used in manufactured
soils. Among other lines of evidence, the constituent concentrations in SFS-manufactured soils
are at or below relevant regulatory and health-based benchmarks for human and ecological
receptors. Because human and ecological exposure potential is lower for use in soil-less potting
media or road subbase than it is for use in manufactured soil, we similarly conclude that these
SFSs do not pose risks of concern when used in soil-less potting media, or road subbase.
Any conclusions drawn by this risk assessment should be understood within the
limitations and scope of the evaluation, including the following:
• Only silica-based SFS from iron, steel and aluminum foundries are evaluated. In contrast,
SFS from leaded brass and bronze foundries often qualify as RCRA hazardous waste.
Also, there weren't sufficient data to characterize SFS from non-leaded brass foundries
and SFS containing olivine sand, and therefore these SFSs are not evaluated in this risk
assessment.
• In addition to SFS, foundries can generate numerous other wastes (e.g., unused and
broken cores, core room sweepings, cupola slag, scrubber sludge, baghouse dust,
shotblast fines). This assessment, however, applies only to SFS as defined in the
assessment: molding and core sands that have been subjected to the metalcasting process
to such an extent that they can no longer be used to manufacture molds and cores. To the
extent that other foundry wastes are mixed with SFS, the conclusions drawn by this
assessment may not be applicable.
• Samples from 39 foundries (totals and pore water data from 39 samples, and leachate
data from 108 samples) were used to represent silica-based SFS from all iron, steel, and
aluminum foundries in the U.S. Because the foundries were not chosen randomly, there
is uncertainty regarding whether the data are statistically representative of SFS from all
iron, steel, and aluminum foundries. However, these foundries were specifically selected
to ensure that the full range of constituents and their concentrations were adequately
represented, and the analytical data from these samples are the best available for
characterizing SFS constituents.
• Analytical data were available for 25 metals, 16 PAHs, 17 phenolics, and 20 dioxins and
dioxin-like compounds. USDA analyzed for organic compounds that are major binder
components (i.e., phenolics) or might be generated during thermal degradation of
chemical binders and other organic additives (i.e., PAHs, dioxins, furans), because these
constituents present the greatest hazard if at elevated levels in the environment. Review
of the scientific literature for evidence of additional organic compounds present in SFS
indicated that they were well below levels of concern.
• Screening and modeling evaluated those constituents for which toxicity benchmarks
exist.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-8
-------
Executive Summary
• Evaluated beneficial uses include manufactured soil, soil-less growth media and road
subbase. The home garden using SFS-manufactured soil was modeled because it
demonstrated the greatest potential for exposure.
The beneficial use of spent foundry sand, when conducted in an environmentally sound
manner, can contribute significant environmental and economic benefits. These benefits can
include reduced energy use, water consumption, and greenhouse gas emissions. An EPA
analysis indicates current reuses in road base and manufactured soil result in energy savings of
43 billion BTUs per year, 7.8 million gallons of water, and prevention of more than 4,000 tons of
greenhouse gas emissions.
Based on the conclusions of the risk assessment conducted for the specific SFSs
applications as stated above, and the available environmental and economic benefits, the EPA
and USDA support the beneficial use of silica-based SFS specifically from iron, steel and
aluminum foundry operations when used in manufactured soils and soil-less potting media, and
roadway construction as subbase. Consistent with the assumptions, limitations, and scope of this
analysis, the beneficial uses of SFSs also provide significant opportunities to advance
Sustainable Materials Management (SMM) ( ).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-9
-------
Executive Summary
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications ES-10
-------
Chapter 1.0 Introduction
1. Introduction
Industrial and municipal byproducts were once traditionally viewed as wastes, but their
application to soils is now being practiced in the United States and many countries around the
world. A number of industrial byproducts have proven beneficial uses in agronomic settings,
including byproducts from coal combustion, fertilizer production, construction, and incineration
(Stout et al., 1988; Korcak, 1995; Wright et al., 1998; Clark et al., 1999), and paper
manufacturing (Beyer and Mueller, 1995; Phillips et al., 1997; Aitken et al., 1998; Simard et al.,
1998; Zibilske et al., 2000). Many of these byproducts can provide nutrients to crops or improve
the physical and chemical properties of soil. Because the beneficial use5 of these materials has
been shown to improve physical, chemical, and biological properties of soils, there currently
exists a demand for the approved use of these byproducts as low-cost soil amendments, as well
as for other uses (e.g., road construction). Of these byproducts, spent foundry sand (SFS) has
emerged as a material that may be currently underutilized in the production of manufactured
soils and other soil-related applications.
Foundries purchase virgin sand to create metalcasting molds and cores. The sand is
reused numerous times within the foundry itself. However, mechanical abrasion during the mold-
making process and sand reclamation, and exposure to high casting temperatures causes the sand
grains to eventually fracture. The fracturing changes the shape of the sand grains, rendering them
unsuitable for continued use in the foundry. The resulting residuals are generally managed as a
waste or beneficially used. A single foundry can generate numerous wastes, including spent
molding and core sands, unused and broken cores, core sand waste, core room sweepings, cupola
slag, scrubber sludge, baghouse dust, and shotblast fines. However, only spent molding and core
sands from ferrous and nonferrous foundries were considered in this assessment. That is, for the
purpose of this assessment, SFS will be used to indicate molding and core sands that have been
subjected to the metalcasting process to such an extent that they can no longer be used to
manufacture molds and cores. While not all molds contain cores (e.g., solid casting), molds that
do contain cores generally produce a commingled waste. Therefore, SFS should also be
considered a byproduct that contains only spent molding sand, or spent molding and core sand.
Core butts, which are pieces of core that did not break down to grain size after the casting
process, were not considered in this evaluation.
Approximately 2.6 million tons of the SFS produced annually are beneficially used
outside of the foundries, of which 14% is used in soil-related applications (USEPA, 2008c).
Spent foundry sand has been used as a substitute for virgin sand in certain markets. These
markets generally can be divided into three groups:
• Highway and Construction Uses - SFSs have been shown to perform well in bases and
subbases under roadways, paved surfaces and structures. In pavement surfaces, SFSs are
also used in hot mix asphalt and in portland cement concrete products.
5 The term "beneficial use," as defined in this document, is the reuse of an industrial material in a product that
provides a functional benefit; that may replace a product made from virgin raw materials, thus conserving natural
resources that would otherwise need to be obtained through practices such as extraction; and that meets relevant
product specifications and regulatory standards.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-1
-------
Chapter 1.0 Introduction
* Aggregate Substitutes - SFSs substitute for other fine aggregates in products that are
bound together in some manner. Such products include: portland cement, ready mix
concrete, pre-cast concrete, bricks, blocks and pavers, grouts and mortars, ceramic tiles
and other manufactured products where sand is a raw material.
• Manufactured Soils - Nurseries and landscaping companies are manufacturing soils by
blending SFSs with low-grade soils and organic materials.
Spent foundry sands are potentially useful in manufactured soils because of their
uniformity, consistency, and dark color in the case of green sands. The sands can be blended
with soils and/or organic amendments (e.g., peat, composted yard waste, manures, biosolids) to
develop manufactured soils suitable for horticultural, landscaping, and turfgrass applications
(Jing and Barnes, 1993; Naystrom et al., 2004; Lindsay and Logan, 2005). A high sand content
(as much as 50% by weight) is required in manufactured soils to reduce compaction and increase
water movement, especially in high foot traffic soils such as golf putting greens and athletic
fields (Swartz and Kardos, 1963; Brown and Duble, 1975; Davis, 1978; Taylor and Blake, 1979;
Baker, 1983). A laboratory study by McCoy (1998) demonstrated that progressive increases in
the sand content of silt loam and loam soils while maintaining a low organic matter content
greatly improved the quality of soil with respect to compaction properties and water movement.
In addition, SFSs have also been successfully used in non-agricultural applications, for example,
highway subbases, structural fills, flowable fills, cement, concrete, pipe bedding, and backfill
(Naik et al., 1994; Leidel et al., 1994; FIRST, 2004; Abichou et al., 2004; Guney et al., 2006;
Deng and Tikalsky, 2008). Spent foundry sands may also be useful as a low-cost reactive
medium to remove trace elements and organics from contaminated water (Lee et al., 2004a, b;
Lee and Benson, 2004).
While SFSs satisfy the engineering and other performance specifications for many of the
above-mentioned applications, their use has been constrained in many states, especially as an
ingredient in manufactured soils and for land application. The unencapsulated6 use of SFS is of
particular concern to many states because the application to land poses the highest potential for
human and ecological exposure to chemical constituents found in the material. To address this
concern for SFS and other byproducts, a number of states have established beneficial use
programs for industrial materials. With the increase in environmental, legislative, and economic
activities that are favorable to beneficial use of industrial byproducts, more states are beginning
to develop such beneficial use programs. States are generally receptive to beneficial use
proposals from industry that are backed by sound science, but frequently lack the necessary
resources to determine whether or not the proposed use could pose significant risks to human
health and the environment. Questions also persist among regulators and scientists as to whether
the levels of trace elements and organic compounds in industrial materials will cause adverse
effects to ecosystems or humans. Consequently, the availability of an evaluation based on sound
science would be enormously helpful to states that are just beginning to develop programs to
evaluate the beneficial use of SFS (Kauffmann et al., 1996), and for states with existing
6 Unencapsulated use is sometimes also referred to as unconsolidated or unbound use and means that the material is
not bound chemically or physically within a matrix such as cement or asphalt.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-2
-------
Chapter 1.0 Introduction
programs, such a risk assessment could serve as a confirmation of current methods or a template
to further refine and improve current methods used in evaluating beneficial use proposals.
Developed through collaboration between the U.S. Environmental Protection Agency
(EPA), the U.S. Department of Agriculture-Agricultural Research Service (USDA-ARS), and
The Ohio State University (OSU), this report characterizes the potential for adverse human
health and ecological effects associated with the beneficial use of SFS in soil-related
applications. By combining the results of current scientific research on SFS and metal and
organic behavior in soils with the results of risk screening modeling, this report is intended to
provide states with a sound scientific basis with which to evaluate the potential risks to human
health and the environment associated with the beneficial use of SFS in soil-related applications.
This chapter presents (1) the purpose, (2) the major features of the report, and (3) a
"roadmap" to this report that summarizes the major components of the SFS evaluation.
1.1 Purpose
In 2002, the USDA-ARS implemented the Foundry Sand Initiative under National
Program 206 (Manure and Byproduct Utilization) to address agricultural and horticultural uses of
SFS. Prior to the inception of this initiative, there was limited information on the use of SFS in
manufactured soils, although sands are commonly used as an ingredient in a variety of soil-
related applications. The USDA-ARS supports research to address the increasing national need
for manufactured soils, particularly for use in disturbed and degraded environments and
agricultural applications. A multiyear research project was conducted to characterize inorganic
and organic constituents of environmental concern in SFSs and to assess the potential mobility
and uptake of these constituents by environmental receptors. Research results were published as
peer-reviewed scientific articles, which are available in the public domain (Dungan 2006;
Dungan and Dees, 2006, 2007, and 2009; Dungan and Reeves, 2005 and 2007; Dungan et al.,
2006 and 2009 and Dayton et al., 2010). In an effort to address the potential risks of using SFS in
soil-related applications, the USDA-ARS and EPA formed an expert team of agronomists, soil
scientists, and environmental health risk assessors to develop a SFS-specific risk assessment. The
main purpose of this work was to determine whether or not SFSs pose unacceptable risks to
human health or the environment when used in manufactured soils. The risk management criteria
used in this evaluation stipulate that the estimated risks to human or ecological receptors exposed
to SFS chemical constituents in manufactured soils should not exceed a target cancer risk and
noncancer hazard as defined below:
• For carcinogenic (cancer-causing) constituents, the target cancer risk is defined as an
excess lifetime cancer risk above 1 chance in 100,000 (i.e., 10"5).
• For constituents that cause noncancer health effects, the target hazard level is defined as a
ratio of the estimated exposure level to a reference level—the hazard quotient (HQ)—of
1.
• For noncancer effects to ecological receptors (e.g., plants, animals, soil invertebrates), the
target hazard level is defined as the ratio of the predicted exposure level to a chosen
environmental quality criterion or allowable medium concentration.
Thus, the question to be answered by this evaluation may be stated as follows: is the use
of silica-based iron, steel, and aluminum SFSs in manufactured soils protective of human and
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-3
-------
Chapter 1.0 Introduction
ecological receptors in the United States where this material is used? This evaluation defines the
term "protective" in terms of Y excess risk of cancer (i.e., < 10~5) for human receptors and Z
hazard (i.e., < 1) for noncancer endpoints for both human and ecological receptors. The SFS
evaluation uses a lines-of-evidence approach to draw conclusions, taking advantage of a
significant body of research on SFS and constituent behavior in soils, as well as risk screening
modeling.
In pointing out that the SFS evaluation uses a lines-of-evidence approach, it is useful to
consider exactly what that means. As detailed in Chapter 2, the constituents of potential concern
in SFS include metals, metalloids, and a number of organics, including polycyclic aromatic
hydrocarbons (PAHs), phenolics, dibenzodioxins, dibenzofurans, and dioxin-like compounds.
With respect to the presence of metals and metalloids (hereafter simply referred to as metals), the
assessment considers a number of different issues that EPA has identified in the Metals
Framework for Risk Assessment (U.S. EPA, 2007b), including:
• Will the addition of SFS to soil result in an increase in the metal concentrations in soil
relative to background levels, and how should the results of the risk assessment be
interpreted across varied national soils?
• How do metal species found in the SFS matrix behave with respect to bioaccessibility
and bioavailability? What soil properties are most important to consider in evaluating the
metal behavior and toxicity (e.g., pH is often referred to as the master soil variable for
metals)?
• How will the behavior of individual metals in manufactured soil, such as the soil-plant
barrier, impact the potential for exposure through the food chain pathway and, ultimately,
the potential for adverse human health and ecological effects?
• How do the risk assessment results compare to levels required to maintain nutritional
health in plants and animals? Do issues of essentiality suggest that the predicted risks to
plants and animals overestimate the potential for adverse effects?
• How do the interactions among metals in the SFS matrix influence the mobility and
toxicity of metals? If used as a component of manufactured soils, would a decrease or
increase in toxicity be expected?
Each of these questions is important in assessing the potential risks posed by metal
constituents in SFS-manufactured soils, because the properties of this material may increase or
decrease the risk to human health and the environment. Therefore, the lines-of-evidence
approach taken in this risk assessment brings recent study information on SFS and metal
constituents—including both qualitative and quantitative information—to address these
questions and to ensure that the risk characterization presents a comprehensive view of the
potential for adverse effects.
1.2 Major Features of the SFS Evaluation
The problem formulation chapter (Chapter 3) and the analysis chapters (Chapters 4 and
5) provide a detailed description of the conceptual approach, as well as the models and data used
in considering the potential risks associated with SFS constituents in manufactured soil. The
following list of features provides a broad sense of the SFS evaluation:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-4
-------
Chapter 1.0 Introduction
The point of exposure is assumed to be the point of application. That is, the exposure
scenarios focus primarily on the potential risks associated with exposure at the point of
SFS application. Thus, the SFS risk assessment is based on conservative assumptions
regarding exposure (e.g., the drinking water well is immediately adjacent to the use
location).
The recent research conducted for this evaluation includes an analysis of the constituent
concentrations found in SFS, leaching potential, plant uptake, and toxicity to soil
invertebrates. Data include both constituent-specific information as well as studies on
SFS as a material (e.g., soil invertebrate toxicity). Taken together, this body of data
represents the best available characterization of SFS and its constituents.
The risk assessment draws upon a number of different sources of information in
developing conclusions regarding the potential risks to human health and the
environment. The information developed and presented in this report includes
- Qualitative (e.g., descriptions of how the soil-plant barrier renders certain exposure
pathways incomplete for certain SFS constituents)
- Semi-quantitative (e.g., comparisons of SFS constituent concentrations to
environmental quality criteria)
- Quantitative (e.g., quantitative estimates derived using risk assessment screening
models to evaluate the inhalation, groundwater ingestion, and plant ingestion
pathways).
A tiered risk assessment approach was used to identify constituents and exposure
pathways of concern; the information produced at each step was used to identify the
constituents to be included in the following step.
The EPA model SCREENS (U.S. EPA, 1995b) was used in screening-level modeling of
the inhalation pathway to develop conservative estimates of exposure concentrations for
comparison with EPA inhalation benchmarks.
EPA's Industrial Waste Management Evaluation Model (IWEM; U.S. EPA, 2002a,
2002b) was used in screening-level modeling of the groundwater ingestion pathway to
develop conservative estimates of groundwater exposure concentrations for use in
standard risk equations.
EPA's Composite Model for Leachate Migration with Transformation Products
(EPACMTP; U.S. EPA, 2003f, g, h; 1997a) was used in refined probabilistic
groundwater modeling of arsenic. Drinking water well exposure concentrations were
developed for use in standard risk equations.
The EPA model (with minor modifications) that is currently used to support EPA's 2004,
2005, and 2006 biosolids risk assessments under section 503 of the Clean Water Act
(U.S. EPA, 2002e) was used for selected constituents, screening-level probabilistic
modeling of the direct ingestion of soil and the ingestion of home-grown produce.
The risk characterization addresses the potential for adverse effects to both human and
ecological receptors for exposure scenarios involving direct contact with and use of
manufactured soils containing SFS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-5
-------
Chapter 1.0 Introduction
• The risk characterization uses recent research (largely conducted by USDA-ARS and
OSU) to interpret the screening-level estimates of risk, making full use of a wealth of
information describing and sometimes quantifying the behavior of chemical constituents
in soil, as well as exhaustive analytical data on constituent concentrations and leach tests
on SFS.
1.3 Roadmap to this Report
As shown in Figure 1-1, the SFS assessment framework is comprised of five key components:
(1) SFS Characterization; (2) Problem Formulation; (3) Analysis; (4) Risk Characterization; and
(5) Conclusions. Information gathered during the SFS characterization is used to support the risk
assessment, which is performed under the Problem Formulation, Analysis, and Risk
Characterization phases shown in Figure 1-1. The Analysis applied a phased approach where
Phase I identified SFS constituents and pathways of potential interest, and Phase II applied a
probabilistic screening approach to further evaluate those constituents and pathways that did not
pass the Phase I screen. As illustrated in Figure 1-1, the information collected during the SFS
Characterization (which included scientific research on the SFS constituents) was critically
important to the Risk Characterization; in conjunction with the risk modeling results, the
information on SFS and its constituents was synthesized to develop conclusions regarding the
potential health and ecological risks associated with soil-related SFS use. In summary, the
chapter organization is as follows:
SFS
Characterization
(Chapter 2)
Problem Formulation
(Chapters)
Risk Assessment
Analysis
(Chapters 4 and 5)
Risk
Characterization
(Chapter 6)
Evaluate all information
relevant to interpreting
screening risk modeling
results
Phase I. Identifying COCs
Identify constituents for further
evaluation
Phase II. Risk Modeling
Evaluate constituents identified
under Phase I
Figure 1-1. Framework for the SFS assessment.
Chapter 2—Background and Characteristics of Spent Foundry Sands. Summarizes
information on the sources and types of foundry sands, provides data on the physical and
chemical properties of U.S. iron, steel, and aluminum SFSs, and provides data on the
uptake of metals by plants and earthworms, and the impact of those metals on soil
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-6
-------
Chapter 1.0 Introduction
microorganisms. Chapter 2 also provides additional information, especially on metal
constituents, relevant to the potential exposure pathways of interest.
Chapter 3—Problem Formulation. Defines the scope of this risk assessment, presents
the conceptual models that illustrate the sources, exposure pathways, and receptors of
interest, and summarizes the analysis plan developed to characterize the potential for
adverse health and ecological effects associated with constituent releases from SFS in
manufactured soils.
Chapter 4—Analysis Phase I: Identification of COCs for Modeling. Describes the
rationale for selecting the constituents of concern (COCs) for the groundwater,
inhalation, and soil pathway modeling. This chapter presents the comparison of
constituent concentrations in SFS with screening criteria for groundwater, air, and soil
exposures, respectively. The screening results identified the COCs and exposure
pathways for probabilistic risk modeling.
Chapter 5—Analysis Phase II: Risk Modeling of COCs. Describes the probabilistic
screening and refined modeling of the groundwater and soil pathways for the home
gardener scenario. This chapter presents the methodology and inputs/outputs for each part
of the modeling and discusses the results of the model simulations.
Chapter 6—Risk Characterization. Presents the lines-of-evidence interpretation of the
potential for adverse health and ecological effects (1) for SFS as a material used in
manufactured soils, (2) by constituent category such as PAHs and dioxins, and (3) by
constituent for the majority of metals found in SFS. This chapter pulls together the
information and risk modeling results from the previous chapters, and incorporates
critical research on areas such as the soil-plant barrier that are essential to the
interpretation of the risk assessment results. In addition, this chapter discusses key
sources of uncertainty in the characterization of risk.
Chapter 7—Conclusions and Recommendations. Distills the findings from the risk
characterization into a concise summary to be used in interpreting the results of this risk
assessment for the purposes of decision making regarding the beneficial uses of SFS
addressed by the assessment.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-7
-------
Chapter 1.0 Introduction
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-8
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
2. Background and Characteristics of Spent Foundry Sand
The overall goals for this report are to (1) evaluate all available information on the
beneficial use of SFS in the various use scenarios addressed in this assessment, (2) identify likely
exposure pathways and receptors associated with those use scenarios, and (3) determine whether
the unencapsulated use of SFS in those beneficial scenarios have the potential to cause adverse
health or ecological effects. With these goals in mind, this chapter presents information on the
production, composition, characteristics, and uses of SFS.
2.1 Foundry Sand Characteristics
Sand is used by the foundry industry to create metalcasting molds and cores. The sand
has the ability to absorb and transmit heat because it allows gases generated during casting to
pass between the sand grains. The most commonly used sand is silica sand (silicon dioxide,
SiCh) because of its wide availability and relatively low cost. Several other sands are used for
specialty casting because of the specific properties related to limited expansion upon heating
(e.g., chromite, olivine, zircon, and staurolite). While thermal expansion is an important physical
property that must be considered before selecting a sand, other important physical properties are
grain shape, grain fineness, permeability, and density. Specifically:
• Sand grain shapes can be classified as round, subangular, angular, and compound. Round
sand is superior for green sand systems (see discussion on green sands in Chapter 2.2.1,
below), while subangular sand with obtuse angles is the most common type of silica sand.
Angular sands have grains with edges that form acute angles, and compound sands have
grains that are fused together; both angular and compound sands are poor sands for
making castings.
• Grain fineness is based on the average sand-grain size. Steel castings typically use very
coarse sand, while nonferrous castings (e.g., aluminum, brass, bronze) use finer sand.
• Permeability is a measure of how fast gases will pass through the mold. If the gases do
not freely pass through the sand, then the resulting pressure buildup may crack the mold.
On the other hand, if the gases pass too quickly, then the molten metal may penetrate the
voids, causing a very rough casting.
• Higher sand density is desirable because high-density sands will absorb heat faster and
result in fewer surface defects. A smaller coefficient of thermal expansion is also
preferred. High-quality silica sand has about a 1.8% thermal expansion from ambient
temperature up through casting temperatures of 1,540-1,590°C (2,800-2,900°F). This is
an important consideration when trying to hold dimensional tolerances.
2.2 Molding and Core Sands
2.2.1 Green Sands
Green sand is the most widely used in the molding process. The main components of
green sand systems are sand, sodium and/or calcium bentonite clay, and carbonaceous additives
(e.g., bituminous coal, gilsonite, cellulose). Green sands are named not because of their color,
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-1
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
but because the sand mixture contains water and provides "green strength." Green strength is the
ability of an incompletely cured material to be handled without distortion. Green sands contain
about 85-99% sand and up to 10% clay and 5% carbonaceous material. Sodium and calcium
bentonite are hydrous alumina silicates, which provide cohesion and plasticity in the green state
(i.e., wet) and when dried. Sodium bentonite (also called western bentonite) can swell to 10-20
times its original volume when thoroughly wetted and has a burnout temperature of about
1,290°C (2,350°F). Calcium bentonite (also called southern bentonite) is a non-swelling clay.
Calcium bentonite's burnout temperature of 1,100°C (1,950°F) results in a sand that is less
durable than sodium bentonite.
Bituminous coal (called seacoal by the foundry industry) and gilsonite partially combust
in the presence of the molten metal, leading to off-gassing of vapors. Release of the organic
vapors from within the mold is necessary to prevent the mold from splitting and causing casting
defects. Cellulosic additives (such as wood flour, corn flour, cotton hulls, rice hulls, walnut
shells, and pecan shells) absorb the moisture, prevent expansion defects, and can improve the
flowability of the sands. The individual sand grains are coated with clay and water through the
use of a mulling process.
2.2.2 Chemically Bonded Sands
In addition to clay or other inorganic binders, individual sand grains can also be held
together using a variety of organic resins. These resins are used to create molds and cores. Cores
are used to create a hollow cavity within a metal casting and are exclusively made using resin-
coated sand prepared by a number of different processes. Some of the most commonly used
resins/processes are the phenolic urethane coldbox7 and no-bake; furan no-bake and warmbox;
novolac; resole; and sodium silicate.8
Phenolic Urethane
All phenolic urethanes are three-part systems consisting of a phenolic resin,
polyisocyanate, and a tertiary amine catalyst (Gardziella et al., 2000). The phenolic resin is a
phenol-formaldehyde polymer and is adjusted to a specific viscosity with a complex mixture of
high-boiling aromatic hydrocarbons. The polyisocyanate used is diphenylmethane-4,4-
diisocyanate (MDI) and is similarly diluted with solvents. MDI is produced from aniline and
formaldehyde. Additives of a proprietary nature are often added to coldbox formulations to
increase moisture resistance, bench life, and core box release. The urethane is formed when the
isocyanate group reacts with a hydroxyl group in the phenolic resin (all urethanes share a
common functional group, i.e., R-NHC=OO-R). Amine catalysts are used in both coldbox and
no-bake core and mold making to accelerate the polyurethane reaction. The tertiary amine
catalysts—dimethylethylamine and triethylamine—are used in coldbox systems.
7 "Coldbox" is a term used to describe any binder process that uses a gas or vaporized catalyst to cure the resin while
at ambient temperature.
8 In addition to these resins, a new class of sand binder was created by General Motors and is known as GMBOND.
This protein-based binder is made from high strength collagens with an additive to promote thermal breakdown of
the binder coating. The minimum protein content of the binder is 99.5% and it contains trace quantities of iron
oxide, methyl paraben, propyl paraben, benzalkonium chloride, and sodium benzoate. Unlike the thermoset
polymers of many binder systems, this protein-based binder system forms a biopolymer crystalline structure.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-2
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Furan
In the furan (i.e., heterocyclic organic compound, but not related to dibenzofurans) no-
bake process, polymerization occurs when the liquid resin is exposed to an acid catalyst at
ambient temperature. While the major component of furan resins is furfuryl alcohol, other
additives such as phenol, formaldehyde, urea, 2-furancarboxaldehyde (furfural), and
2,5-bis(hydroxymethyl)furan are often used to improve specific resin properties (Gandini and
Belgacem, 1997). The acid catalyst is a combination of acid (phosphoric acid-based or sulfonic
acid-based), methanol, and water. Optimum binder concentrations vary from approximately 0.8-
1.5% of the sand mixture by weight before metalcasting. The furan warmbox process uses the
same equipment and procedures as the no-bake process, except that heat is applied (130-180°C)
to aid in resin curing.
Novolac
Novolac oligomers are thermoplastic, brittle, and do not cross-polymerize without the
help of a cross-linking agent. The oligomers are produced under reflux at 100°C with a molar
ratio of formaldehyde to phenol <1 and the addition of an acid catalyst (e.g., sulfuric acid, oxalic
acid). Cross-polymerization or curing of the oligomers occurs when they are heated in the
presence of hexamethylenetetramine (HMTA), which decomposes to formaldehyde and
ammonia. The shell process is used to produce free-flowing, storable sand that is coated with a
novolac-HMTA film (1.6-3.8% based on sand weight before metalcasting), which is then cured
on hot pattern plates or in heated coreboxes (180-350°C) to form hollow and solid cores
(Gardziella et al., 2000). To reduce brittleness, 1-2% iron oxide is often added to the resin.
Resole
Phenolic resoles are prepared by a reaction of excess formaldehyde with phenol and the
addition of a base catalyst (e.g., sodium hydroxide, potassium hydroxide) at temperatures up to
100°C (Gardziella et al., 2000). Curing occurs when the phenolic resoles react with an acid at
ambient temperature (no-bake process) or heating to 180-250°C (hotbox process), or from a
reaction with an aliphatic ester (ester no-bake process).
Sodium Silicate
Sodium silicate (Na2O- SiCh) is an inorganic system that can be cured using an organic
ester or during gassing with carbon dioxide (CCh) (Owusu, 1982; Gardziella et al., 2000). In the
ester-cured system, the ester is hydrolyzed by alkaline sodium silicate. The acid produced during
this reaction then reacts with the sodium silicate to form a gel, which bonds the sand grains.
Some typical organic esters used are glycerol diacetate, ethylene glycol diacetate, and glycerol
triacetate (Winkler and Bol'shakov, 2000).
2.3 Reclamation and Disposal
Many foundries have invested in sand reclamation systems that can recover up to 90% of
the sand used in the casting process (Stevenson, 1996; Zanetti and Fiore, 2002). Used molding
and core sands can be reclaimed through mechanical and/or thermal treatment. During
mechanical reclamation, the sand is crushed to grain size, then dry abrasion is used to separate
the binder from the sand grains. Thermal reclamation is a process where all organic binders and
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-3
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
carbonaceous additives are burned off after the sand is pre-crushed. This is a more expensive
process than mechanical attrition because it requires high-energy inputs to heat the sand to 500-
800°C. Reclaimed sand can be reused a number of times in the casting process; however,
because heat and mechanical abrasion eventually render the sand unsuitable for continued use in
the foundry, the resulting sand must be managed as a waste or beneficially used outside the
foundry. Much of the SFS sent to landfills is used as daily cover, but it is not uncommon for
foundries to dispose of their SFS in monofills at the foundry.
2.4 Collection and Analysis of U.S. SFSs
An examination of the peer-reviewed literature on metals and organics in SFS revealed
that many peer-reviewed reports on this topic have been published over the last two decades.
Because there was great interest in using SFS in geotechnical applications, prior to its use in
manufactured soils, the majority of the research addressed the leaching potential of various
constituents (Riediker et al., 2000; Ji et al., 2001; Kendall, 2003; Lee and Benson, 2006; Deng
and Tikalsky, 2008). The most comprehensive data sets on metals and organics in SFS have been
generated by the USD A. The USDA data sets contain information on total and teachable metals
(Dungan, 2008; Dungan and Dees, 2009; Dayton et al., 2010), PAHs and phenolics (Dungan,
2006), and dioxins (Dungan et al., 2009). A database was also created by The Pennsylvania State
University (Penn State), where industry data on different foundry waste materials were compiled
(Tikalsky et al., 2004). This database contains information on total and teachable concentrations
of various constituents in foundry byproducts, many of which were not suitable for beneficial use
in soil-related applications. While the Penn State database was not used in this risk evaluation as
a result of inconsistent analytical data among the foundry byproducts, a comparison of the
database with the USDA data set revealed that total and teachable concentrations of organic and
inorganic constituents in molding sands were very similar. USDA analyzed for organic
compounds that are major binder components (i.e., phenolics) or might be generated during the
thermal degradation of chemical binders and other organic additives (i.e., PAHs, dioxins, furans),
because these constituents present the greatest hazard if at elevated levels in the environment.
Evidence of additional organic compounds present in SFS found them at concentrations well
below levels of concern. Therefore, additional organic compounds, beyond those analyzed by the
USDA, were not considered in this assessment.
2.4.1 Spent Foundry Sand Collection
In June 2005, September 2005, and July 2006, 43 SFSs (36 green and 7 chemically
bonded molding sands) were collected from ferrous and nonferrous foundries located in 12 states
(Alabama, Georgia, Iowa, Indiana, Michigan, North Carolina, Ohio, Pennsylvania, South
Carolina, Tennessee, Virginia, and Wisconsin). A description of the SFSs can be found in
Table 2-1. The June 2005 samples were collected as described by Dungan (2006), while the
remaining sets were collected by foundry personnel after receiving training on sample collection.
Briefly, a clean section of polyvinyl chloride (PVC) pipe was used as a sampling device to
collect four samples from each SFS pile. The samples were transferred into 500-mL glass jars
with Teflon-lined polypropylene closures and immediately shipped to the laboratory in
Styrofoam coolers with ice packs. Upon receipt, the samples were stored at 4°C for no longer
than 2 weeks until processed. All SFSs were passed through a 0.5-mm sieve to remove any core
butts before being analyzed.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-4
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-1. Description of the U.S. Spent Foundry Sands
Sand
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
State
PA
PA
PA
PA
PA
PA
PA
OH
OH
OH
OH
IN
OH
OH
IN
OH
OH
IN
WI
OH
IN
MI
MI
WI
WI
MI
OH
TN
WI
WI
TN
TN
AL
AL
VA
GA
SC
IA
IA
NC
IN
IN
WI
Sampling Dates
6/05
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
9/05
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
7/06
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Metal Poured
Iron
Iron
Iron
Aluminum
Iron
Steel
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Aluminum
Iron
Iron
Iron
Iron
Iron
Aluminum
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Iron
Steel
Iron
Iron
Iron
No lead brass
No lead brass
Iron
Iron
Iron
Steel
Steel
Iron
Steel
Iron
Steel
Molding Sand
Green sand
Green sand3
Green sand
Green sand
Green sand
PU no-bake3
Green sand
Green sand
Green sand
Green sand
Green sand
ShelP
Green sand
Green sand
Green sand
Green sand
Green sand
Green sand
Green sand
Green sand
PU no-bake
Green sand
Green sand
Green sand
Green sand
Green sand
Green sand
Green sand
PU no-bake
Green sand
Green sand
Green sand
PU no-bake
Green sand
Green sand
Green sand
Green sand
Phenolic ester-cured
Green sand
Green sand
PU no-bake
Green sand
Green sand
Core Binder System and Process
PUb coldbox, PU no-bake, shell, core oil
Shell
Shell, furan warmbox
Shell
PU no-bake, shell, sodium silicate
PU no-bake
PU no-bake
PU coldbox, PU hotbox
PU coldbox, PU hotbox
PU coldbox, PU hotbox
PU coldbox, PU no-bake, shell
Shell
PU coldbox, PU no-bake, shell
PU no-bake, shell, core oil
PU coldbox, shell
PU coldbox, PU hotbox
PU coldbox, PU hotbox
PU coldbox, PU hotbox, shell
PU coldbox
Shell
PU coldbox, PU no-bake, furan warmbox
PU no-bake, shell
PU coldbox, shell
Shell
PU coldbox
None
PU no-bake, shell
None
PU no-bake
PU coldbox, shell
Shell, resin/CO2
PU coldbox
PU no-bake
PU no-bake
PU coldbox
PU coldbox, shell
PU coldbox, shell
PU coldbox, shell, resin/CO2
PU coldbox, shell, resin/CO2
PU coldbox, shell
PU no-bake
PU coldbox
PU no-bake, shell, core oil, resin/CO2
3 Olivine sand utilized
1 PU = phenolic urethane ° Shell process associated with use of novolac resin
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-5
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
2.4.2 PAHs and Phenolics9
An accelerated solvent extractor (ASE 200, Dionex, Sunnyvale, CA) was used to extract
the PAHs and phenolics for analysis by gas chromatography-mass spectrometry (GC-MS).
Twenty grams of SFS was placed into the center of a 33-mL stainless steel extraction cell, which
was then packed at each end with clean Ottawa sand (20-30 mesh, U.S. Silica Corp., Ottawa, IL)
to fill the void. If the SFS was moist, anhydrous Na2SO4 was mixed with the sand prior to the
addition to the cells. The conditions of the ASE were as follows: solvent, dichloromethane/
acetone (1:1); static extraction for 5 min at a pressure of 14 MPa (2,000 psi) and an oven
temperature of 100°C; flush volume, 60% of the cell volume; N2 purge, 1 MPa (150 psi) for 60 s.
All extracts were collected in 40-mL vials. Immediately after the extraction, the extracts were
evaporated to near dryness under N2 and then reconstituted with 2 mL of dichloromethane. The
method detection limit (MDL) for this data set was calculated by multiplying the standard
deviation of replicate standards (n = 6) by the Student's lvalue at the 99% confidence interval.
Calculating the MDL at the 99% confidence interval allows for the possibility that 1% of the
samples analyzed, which have a true concentration at the MDL, will be false positives.
2.4.3 Dioxins and Dioxin-like Compounds
The SFSs were processed and analyzed for polychlorinated dibenzo-p-dioxins (PCDDs),
polychlorinated dibenzofurans (PCDFs), and coplanar polychlorinated biphenyls (PCBs) by EPA
Method 1613 (tetra- through octa-chlorinated dioxins and furans by isotope dilution
HRGC/HRMS, 1994B) modified to include the coplanar PCBs (IUPAC nos. 77, 126, and 169).
Toxic equivalency values (TEQs) were calculated by summing the products of each congener
concentration and its World Health Organization (WHO) 2005 toxic equivalency factor (TEF)
(Van den Berg et al., 2006).
2.4.4 Trace Elements
USDA-ARS Data Set
The SFSs were digested according to EPA method 3050B. The digests were filtered
through Whatman no. 40 paper layered with Whatman 2V fluted filters (Florham Park, NJ). The
filtrate was diluted to 100 mL with 0.1 MHC1 and analyzed by inductively coupled plasma-
atomic emission spectrometry (ICP-AES). Blanks and standard reference material 2709 (San
Joaquin Soil, National Institute of Standards and Technology [NIST], Gaithersburg, MD) were
run regularly to ensure quality control. The limit of quantitation (LOQ) was calculated as 10 or
30 times the standard deviation of digestion blank values (n = 20) and was expressed as mass of
element per sample dry weight.
Ohio State University Data Set
Elemental concentrations were determined by EPA method 3051A (U.S. EPA, 2007d); a
microwave-assisted aqua regia digestion followed by ICP-AES analysis and inductively coupled
plasma-mass spectrometry (ICP-MS) for elements below detection by ICP-AES. ICP-AES and
ICP-MS analyses for total elemental analysis were carried out according to EPA methods 60IOC
and 6020A, respectively. Quality control operations included analysis of laboratory control
See Section 2.5.3 for a discussion of the selection process for organics.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-6
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
samples (CRM 059-050; RTC Corporation, Laramie, WY) with each microwave tray, pre-
digestion spikes, initial calibration verification, initial calibration blank, continuing calibration
verification for every 10 samples, continuing calibration blank for every 10 samples, and low
LOQ verification for every 20 samples. All checks were within the quality control limits set in
EPA, ILM04.0b (U.S. EPA, 1999a).
2.4.5 Leach Tests
The toxicity characteristic leaching procedure (TCLP) and the synthetic precipitation
leaching procedure (SPLP) were conducted according to EPA methods 1311 and 1312,
respectively. The water leach test was conducted according to American Society for Testing and
Materials (ASTM International) method D 3897 (ASTM International, 2004). All leaching
procedures were slightly modified as described by Dungan and Dees (2009). The extracts were
analyzed by ICP-AES. Also, Dayton et al. (2010) estimated pore water elemental content on the
SFS by equilibrating SFS in a 1:1 SFS:deionized water saturated paste for 24 hours. Extracts
were analyzed by ICP-AES. The LOQ was calculated as 10 times the standard deviation of
matrix blanks (n = 10) and was expressed as mass of element per volume of leaching solution.
2.5 Constituents and Properties of Spent Foundry Sand
2.5.1 Properties Important to Soil Quality and Function
Manufactured soils, such as horticultural potting soils or those made for landscaping,
generally contain some low-grade native soil. Soils made for such purposes are created by
blending organic and mineral components, such as SFS. For SFSs to be considered for beneficial
use as a soil amendment or a component of a soil blend, they must have soil-like qualities, make
a contribution to soil quality/fertility, or provide a functional benefit (e.g., acid neutralization,
contaminant sorption/binding). SFSs tend to have low fertility, but they often have soil-like
qualities that make them attractive as components in a soil blend. Soil quality has been defined
as "the capacity of a soil to function, within ecosystem and land-use boundaries, to sustain
biological productivity, maintain environmental quality and promote plant and animal health"
(Doran and Parkin, 1996). A manufactured soil suitable for plant growth should have desirable
chemical (e.g., pH, salinity) and physical (e.g., drainage, texture, water holding capacity)
properties. Components used in a manufactured soil are chosen to provide suitable levels of these
properties. An added advantage of manufactured soils is that component ratios can be adjusted so
a soil blend can be "tailored" to specific uses. For example, in horticultural applications, soils
used for market pack containers need to be light and well drained, while soils used for
landscaping or container mixes for trees and shrubs need to be heavier and have a good water-
holding capacity. To be beneficial, a manufactured soil also must not cause toxicity to plants and
biota.
Properties important to soil quality and function were measured in 43 ferrous and
nonferrous SFSs to characterize the sands as potential components in manufactured soil blends.
Soil Organic Carbon
Soil organic carbon (OC) typically comprises 0.5-3% by weight of mineral soils (Brady
and Weil, 2007), but its importance to soil chemistry and function is greater than these numbers
suggest. Soil OC contributes to soil quality in many ways. It increases water-holding capacity
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-7
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
and infiltration. It also improves soil structure by promoting soil aggregate formation and serves
as a major reservoir for plant nutrients and carbon and energy for soil microorganisms (Brady
and Weil, 2007). Soil organic matter has a large surface area (800-900 m2 g"1) and is rich in
reactive functional groups (e.g., carboxyl or phenolic) (McBride, 1994; Bohn et. al., 2001;
Sparks, 2003). The ionization of these groups, as mediated by pH, imparts a high pH-dependent
cation exchange capacity (CEC, 150-300 cmolc kg"1) to soil organic matter (Bohn et. al., 2001;
Adriano, 2001; Sparks, 2003). Binding of nutrient cations to the exchange sites reduces leaching
and provides nutrient storage for plant nutrition. Nutrients in equilibrium with the soil solution
are readily resupplied to the solution as plants feed. A more stable form of metal complexation
with soil organic matter is through chelation of cationic micronutrients with soil organic matter.
The OC content of the 43 SFSs, measured using dry combustion after acid pretreatment
to remove inorganic carbonates, ranged from 0.29-2.99%, with a mean of 1.71%. The SFS OC
includes OC additions made to the molding sands (i.e., seacoal, polymers) and is within the
typical range for native soils.
Soil Texture
Soil texture is determined by the proportionate content of different sized soil particles.
Particle size distribution determines the soil textural class. Knowing a soil's particle size
distribution or textural class provides insight into important aspects of the soils behavior (e.g.,
water retention, infiltration, bulk density).
Many horticultural manufactured soil blends are composed of high levels of coarse
materials (e.g., bark, rice hulls, perlite). These soil blend components are light weight and freely
drain, but finer fractions also are needed to increase the water holding capacity and provide plant
nutrient storage. Clay-size particles or clay minerals are a highly reactive component of soil
characterized by having a particle size <2.0 jim and a large surface area.10
A small but important component of many foundry sands is their clay content. Although
we refer to SFS as sand, the addition of clay, seacoal, and other carbonaceous additives
contribute finer particles that can affect the soil textural class and properties of SFS. The particle
size distribution for the 43 sands was determined using the hydrometer method (Gee and Bauder,
1986) and is summarized in Table 2-2 (a more complete breakdown is provided in Appendix B,
Table B-25). Sand (0.05-2 mm) was the dominant size fraction, ranging from 76.6-100% with a
mean of 91%, while silt size particles (2-50 jim) ranged from 0-16.9%, with a mean of 3.43%,
and clay size particles ranged from 0-11.3%, with a mean of 5.54%. Using the USDA Soil
Texture Calculator (USDA, 1993), the SFS textural class was calculated based on the particle
size distribution. The SFS bulk density was calculated using the Saxton equation (Saxton et al.,
1986). Soil texture, in general, ranges from sand (coarse) to clay (fine). Not surprisingly, the
10 The reactions between clay minerals are primarily attributed to their cationic exchange capacity (CEC) or ligand
exchange (specific adsorption) reactions that occur on non-crystalline or amorphous metal oxide clays, typically
of iron or aluminum. The permanent, negatively charged portion of the soil CEC is associated with isomorphic-
substituted 2:1 clay minerals, such as smectite and montmorillonite. These clay minerals have a large surface area
and high CEC. Montmorillonite, for example, has a surface area of 600-800 m2 g"1 and a CEC of 80-150 cmolc
kg"1. The pH-dependent CEC sites are associated primarily with non-crystalline metal oxide clays. These
amorphous metal oxides also have a large surface area. For example, iron and aluminum oxides have a specific
surface area of 70-250 and 100-220 m2 g-1, respectively (Bohn et. al., 2001; Adriano, 2001; Sparks, 2003).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-8
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
texture of the SFSs ranged from sand to sandy loam and the bulk density ranged from 1.57-1.66
g cm"3, with a mean of 1.64 g cm"3.
Table 2-2. Particle Size Distribution, USDA Textural Class, and Bulk Density for 43 SFSs
Minimum
Maximum
Mean
Sand (0.05-2mm)
%
76.6
100
91
Silt (2-50 jam)
%
0
16.9
3.43
Clay (<2 jam)
%
0
11.3
5.54
Bulk Density
gem3
1.57
1.66
1.64
The hydrous metal oxides of aluminum and iron were measured using an acid ammonium
oxalate extraction (McKeague and Day, 1996). The aluminum oxide content ranged from 0.072-
2.43 g Al kg"1, with a median of 0.386 g Al kg"1, while the iron oxide content ranged from 0.213-
32.1 g Fe kg"1, with a median of 1.39 g kg"1. These values are within the typical range for natural
soils (Brady and Weil, 2007). The clay/silt component of SFS suggests that they could increase
the water-holding capacity of coarse horticultural soil blends, but is not so high as to inhibit
drainage. The higher bulk density (see Table 2-2) compared to typical mineral soils (1.25 g cm"3,
Brady and Weil, 2007) suggests that SFS alone may be heavy, which could inhibit root
penetration. Due to relatively high concentrations of bentonite clays in foundry sands, the use of
SFS alone as a potting medium is likely to inhibit root penetration, as they exhibit high rupture
strength under dry conditions (de Koff et al., 2008). However, the addition of SFS to potting or
landscape media may be beneficial where shrubs or trees are planted and a heavier mix is
advantageous.
pH
Soil pH is often called the "master variable." It has the potential to modify metal/nutrient
solubility/availability in several ways. It controls dissolution/precipitation and therefore
influences the speciation of minerals. It regulates the ionization of pH-dependent cation
exchange sites on organic matter and metal oxide clay minerals. The ionization of pH-dependent
functional groups on soil organic matter also affects stable organic complex formation (McBride,
1994; Adriano, 2001; Sparks, 2003).
The pH of the 43 SFSs ranged from 6.67-10.2, with a mean of 8.76. In some instances,
the pH of the SFSs was higher than a typical productive soil. Certainly, the pH will moderate
upon blending SFS with other components. There would only be a concern if the pH of the final
blended soil remained high, as high pH can reduce plant nutrient availability. In addition, the
potential for the formation of unstable aluminum species due to high pH is apparent in the pore
water soluble aluminum (Appendix B, Table B-26), which ranged from 0.1-1,847 mg L"1, with
a median of 1.79. High pH can also induce plant deficiencies of metal cation micronutrients,
including iron, manganese, copper, and zinc. Iron chlorosis is the visual symptom of iron plant
deficiency induced at soil pH >8.5. Blending SFS with organic materials (e.g., compost,
biosolids, manure) and/or soil will buffer the soil pH. SFS will likely be combined with organics,
soil, and other materials to make topsoil. The pH buffer capacity of the organic and/or soil
materials is much greater than SFS. Therefore, the final pH of the manufactured soil will be
closer to the pH of the organic and/or soil materials than the original SFS pH. That is, the final
pH of the manufactured soil will be more relevant than the original pH of the SFS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-9
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
2.5.2 Metals and Metalloids
The sand and other materials used to create metalcasting molds contain natural levels of
metals and metalloids (which will collectively be called metals), but metals may also be
transferred to the molding sands during the casting process (Dungan et al., 2006). Abundant
industry data are available characterizing the leaching of metals in SFS and other foundry
wastes. That is, much of the data are not total numbers, but were derived using leaching
procedures, such as the TCLP (U.S. EPA SW-846, method 1311, U.S. EPA, 2007a). TCLP
concentrations are used to assess risk of metals in landfill leachates, but have limited relevance to
risk assessment for surface soil. The following metals (i.e., aluminum, antimony, arsenic,
barium, beryllium, boron, cadmium, chromium, cobalt, copper, iron, lead, magnesium,
manganese, molybdenum, nickel, selenium, silver, thallium, vanadium, zinc) were specifically
targeted for testing in the 43 U.S. SFSs because they are potential contaminants of ground and
surface waters and are a toxicity threat to plants, animals, and humans if present at elevated
concentrations.
In a study conducted by Dungan and Dees (2009), a totals analysis was conducted for 19
metals in the 43 SFSs listed in Table 2-1. The total metal concentrations in the SFSs, as
determined by EPA method 3050B (SW-846), are summarized in Table 2-3. Of the 19 metals
analyzed for total concentrations, four (antimony, boron, cadmium, and silver) were not detected
in any of the SFSs above the LOQ. The LOQ for antimony, boron, cadmium, and silver were 4.5,
19.2, 5.9, and 17.6 mg kg"1, respectively. The remaining metals (aluminum, arsenic, barium,
beryllium, chromium, cobalt, copper, iron, lead, magnesium, manganese, molybdenum, nickel,
vanadium, and zinc) were detected above the LOQ in some, but not all, of the SFSs.
In the June 2005 set of SFS samples, sand #2 (green sand from an iron foundry)
contained the highest total concentrations of beryllium, cobalt, iron, magnesium, manganese, and
nickel at 3.1; 95; 44,320; 51,574; 671; and 2,328 mg kg'1, respectively. For the remainder of the
sands, beryllium, cobalt, magnesium, and manganese were generally below the LOQ of 1.2,
0.84, 720, and 45 mg kg"1, respectively. Sand #6 contained the second-highest concentration of
nickel at 1,022 mg kg"1. It is likely that the nickel in sands #2 and #6 came from the olivine sand
that these foundries use, which typically contains about 2,000 mg Ni kg"1 (Dungan and Dees,
2009). The mineral olivine is a magnesium iron silicate and contains naturally elevated
concentrations of nickel, cobalt, and chromium. Although silica sand is the most abundantly used
sand, olivine sands are used by some foundries because they have a lower thermal expansion
coefficient, and therefore hold tighter dimensional tolerances. Olivine sands also produce a better
cast surface than silica sands.11 Sand #39 (green sand from a steel foundry) contained nickel at
107 mg kg"1, which was elevated due to the metal alloy, not because they use olivine sands.
Sands #2 and #6 also contained elevated concentrations of chromium at 57 and 149 mg
kg"1, respectively. In sand #22 (green sand from an iron foundry), the molybdenum concentration
was 9.6 mg kg"1. In all of the other SFSs, chromium was generally well below 50 mg kg"1 and
molybdenum was less than the LOQ of 4.4 mg kg"1.
Arsenic was detected in all 43 SFSs at concentrations above the LOQ of 0.03 mg kg"1, but
no higher than 7.79 mg kg"1. The arsenic results (and chromium results discussed above) are
similar to those obtained by Lee and Benson (2006), who analyzed 12 green sands from gray-
iron foundries and found respective ranges of 0.002-2.9 and 1.5-66.4 mg kg"1.
11 Characterization of sands #2 and #6 are included for completeness; however, they were not evaluated as part of
the risk assessment because they contain olivine sand.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-10
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-3. Total Metal Concentrations in the Spent Foundry Sands as Determined by EPA Method 3050B
Metal
Ag°
Al
As
Bc
Ba
Be
Cdc
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
Sbc
V
Zn
Collected June 2005% 43 samples
(mgkg1)
Min
<17.6
<311
0.04
<19.2
<8.7
<1.2
<5.9
<0.84
<1.0
<23.1
<352
<720
<45.0
<4.4
<1.2
<7.7
<4.5
<7.4
<33.4
Max
10,048
4.8
151
3.1
95.3
149
3,318
44,320
51,574
671
9.6
2,328
25.7
9.1
1,640
Meanb
8.8
1,853
1.0
9.6
23.3
0.8
3.0
3.7
11.6
97.1
5976
2,804
96.0
2.4
85.7
5.1
2.3
3.8
60.1
No. of
Detects
0
37
43
0
30
5
0
7
40
9
42
11
18
2
40
4
0
1
5
Collected September 2005, 38 samples
(mgkg1)
Min
<17.6
<311
0.13
<19.2
<8.7
<1.2
<5.9
0.84
<1.0
<23.1
727
<720
<45
<4.4
<1.2
<7.7
<4.5
<7.4
<33.4
Max
6,940
5.1
72.5
3.5
9.1
196
14,360
60,020
26,994
920
19.8
139
28.9
19.3
1,732
Mean
8.8
1,771
1.7
9.6
19.2
0.72
3.0
0.77
12.
772
6,262
1,313
91.8
2.9
10.9
5.8
2.3
4.1
91.1
No. of
Detects
0
33
37
0
28
3
0
5
37
6
38
13
16
3
34
5
0
1
4
Collected July 2006, 37 samples
(mg kg-1)
Min
<17.6
<311
0.07
<19.2
<8.7
2.47
<5.9
0.84
<1.0
<23.1
<352
<720
<45
<4.4
<1.2
<7.7
<4.5
<7.4
<33.4
Max
6,189
4.9
149
2.5
9.1
132
4,668
45,120
16,566
845
54.6
189
212
9.7
2,829
Mean
8.8
1,656
1.0
9.6
25.3
0.65
3.0
0.88
8.8
148
4,867
1,285
75.9
3.6
12.2
13.6
2.3
3.9
102
No. of
Detects
0
33
37
0
27
1
0
4
33
8
36
4
15
1
31
10
0
1
3
< means less than the LOQ.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the LOQ.
0 All concentrations recorded below the LOQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-11
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
The highest concentrations of copper and zinc at 3,318 and 1,640 mg kg"1, respectively,
were found in sand #34, which is a green sand from a non-leaded brass foundry. This is of little
surprise, as brass is an alloy of copper and zinc. The lead concentration in sand #34 was only 19
mg kg"1, which is relatively low due to the fact that it was a non-leaded brass foundry. In
contrast, sand #33 is a chemically bonded molding sand from the same brass foundry, but it
contained considerably less copper and zinc at 70 and 44 mg kg"1, respectively, and lead was
<7.7mgkg"1.
Table 2-3 also shows total element data from samples collected in September 2005 and
July 2006 from a subset of the same 43 foundries. Overall, the data show that there is little
change in the element concentrations in sands collected from specific foundries over time.
Except for sand #6, only the non-leaded brass foundry sands showed a large temporal variation.
The nickel concentration in sand #6 decreased from 1,022 to 111 mg kg"1 by the third sampling
event, while copper in sand #34 increased to 14,200 mg kg"1 by the second sampling event, but
was lower at 4,670 mg kg"1 by the third sampling event. In sand #33 (from the same foundry as
sand #34), the copper increased to 14,360 mg kg"1 by the second sampling event, but was down
to 38.5 mg kg"1 by the third sampling event. Although sands #33 and #34 are from a non-leaded
brass foundry, lead in sand #34 increased from 19 to 212 mg kg"1 by the third sampling event.
Detection limits for some SFS constituents in the USDA dataset are higher than those
required for risk assessment (e.g., the detection limit for antimony (4.5 mg kg"1) is higher than
the human screening level (3.1 mg kg"1), and the detection limit for cadmium (5.9 mg kg"1) is
higher than the ecological screening level (0.36 mg kg"1)). For this reason, and for comparative
purposes, Dayton et al. (2010) analyzed the 43 SFSs from the June 2005 sampling event using an
analytical method able to reach lower detection limits (i.e., EPA method 3051 A), and the data
are presented in Table 2-4. Because of the lower detection limits, total elemental data generated
Dayton et al. (2010) were used for analysis in the risk assessment.
The existing data on non-leaded brass sands and olivine sands demonstrated levels of
copper, lead, nickel, and zinc that were both potentially phytotoxic and much higher than the
other 39 SFSs, but insufficient samples existed to characterize constituent concentration
variability in non-leaded brass and olivine sands. Therefore, while descriptions of non-leaded
brass sands and olivine sands (i.e., sands #2, #6, #33, and #34) are retained for completeness,
they are not evaluated in the risk assessment.
Table 2-4. Metal Concentrations in 39 of 43 Spent Foundry Sands (June 2005 Samples)
as Determined by EPA Method 3051Aa
Metal
Al
As
B
Ba
Be
Ca
Cd
Co
Units
gkg-1
mg kg'1
mg kg'1
mg kg'1
mg kg-1
gkg'1
mg kg'1
mg kg-1
Minimum
0.19
0.13
<20.0
<10.0
<0.1
0.09
O.04
<0.5
Maximum
11.7
7.79
59.4
141
0.60
44.1
0.36
6.62
Mean
5.14
1.70
11.5
8.81
0.17
1.89
0.07
1.26
Median
5.56
1.05
10.0
5.00
0.15
1.89
0.051
0.88
95%-ile
11.2
6.44
20.2
17.7
0.38
3.23
0.20
5.99
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-12
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Metal
Cr
Cu
Fe
K
Mg
Mn
Mo
Na
Ni
P
Pb
S
Sb
Se
Tl
V
Zn
Units
mg kg-1
mg kg'1
gkg'1
mg kg-1
gkg-1
mg kg'1
mg kg-1
gkg-1
mg kg'1
mg kg-1
mg kg'1
gkg-1
mg kg-1
mg kg'1
mg kg'1
mg kg-1
mg kg'1
Minimum
<0.5
<0.5
1.28
<50.0
0.05
5.56
<1.0
O.02
1.11
5.41
<1.0
0.05
0.04
O.4
0.04
<1.0
<10.0
Maximum
115
137
64.4
1,780
3.20
707
22.9
1.93
117
96.6
22.9
2.04
1.71
0.44
0.096
11.3
245
Mean
17.6
21.2
9.20
388
1.26
112
2.98
0.93
15.2
51.2
4.38
0.62
0.30
0.21
0.04
3.44
20.0
Median
4.93
6.22
4.26
328
1.28
54.5
0.50
1.02
3.46
50.9
3.74
0.59
0.17
0.20
0.04
2.88
5.00
95%-ile
109
107
57.1
1300
3.02
670
21.8
1.85
102
85.9
15.3
1.64
1.23
0.20
0.089
9.90
72.1
Source: Dayton et al. (2010)
a Brass green sands and olivine sands (i.e., sands #2,
calculations; calculations based on setting samples
#6, #33, and #34 from Table 2-1) were omitted from
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
other mold components (Dungan and Reeves, 2005; 2007). While every effort was made to
target the widest range of organic constituents that are of concern from an environmental and
human health standpoint, it is possible that additional organics were present in the SFSs and not
addressed in this risk evaluation. However, evidence of additional organics found them at
concentrations well below levels of concern.
In early studies conducted by Gwin et al. (1976), Scott et al. (1976, 1977), and Palmer
et al. (1985), some of the most abundant organics emitted from green sand molds were BTEX,
phenolics, and PAHs such as acenaphthalene, benzo[a]pyrene, naphthalene, phenanthrene, and
pyrene. These organic compounds are a potential threat to the environment and human health
(Alberg et al., 2002; Bostrom et al., 2002; Rana and Verma, 2005; Baird et al., 2007). In green
sand molds, volatile organics are generated during the thermal decomposition of carbonaceous
additives such as coal, gilsonite, lignite, and cellulose (Dungan and Reeves, 2007; Wang et al.,
2007). During the pyrolysis of a green sand at temperatures up to 1,000°C, Dungan and Reeves
(2007) tentatively identified substituted benzenes (e.g., BTEX), phenolics, and PAHs
(Appendix B, Figure B-l and Table B-27). When novolac, phenolic urethane, and furan resins
were pyrolyzed at temperatures up to 1,000°C, similar thermal decomposition products were
identified (Lytle et al., 1998a,b; Helper and Sobera, 1999; Sobera and Helper, 2003; Dungan and
Reeves, 2005).
In a study conducted by Dungan (2006), all samples from the 43 foundries listed in Table
2-3 were analyzed for 15 PAHs and 17 phenolics that are identified as priority pollutants by
EPA. Summary concentration information of the PAHs and phenolics in the SFSs are shown in
Tables 2-5 and 2-6, respectively. Although no published reports are available on BTEX
compounds in SFSs, a preliminary scan of the SFSs using headspace solid-phase microextraction
(HS-SPME) was conducted. The benzene; toluene; ethylbenzene; o- and w-xylene; and/>-xylene
concentrations ranged from below the MDL to maximum values of 50.9, 79.2, 32.9, 72.0, and
41.9 |j,g kg"1, respectively, for the June 2005 samples. In the September 2005 samples, the
maximum concentrations were 1,670; 164; 14.5; 16.4; and 16.8 |j,g kg"1, respectively (R.S.
Dungan, unpublished data).
The majority of the PAHs that were present at concentrations above the MDLs were
2-ring and 3-ringPAHs (i.e., acenaphthene, acenaphthylene, anthracene, fluorene, naphthalene,
and phenanthrene). For most of the SFSs, naphthalene was at the highest concentrations,
followed by phenanthrene. Three SFSs in particular (sands #6, #33, and #41) had the highest
concentrations of naphthalene, which ranged from 28-48 mg kg"1. These sands were from
foundries that used both phenolic urethane molding and core sands (i.e., not green sands). The
4-ring PAHs, benz[a]anthracene and chrysene, were at concentrations slightly above the MDL
only in SFSs #9, #34, and #40 (and fluoranthene in SFSs 8 and 27). The respective MDLs for
benz[a]anthracene, chrysene, and fluoranthene were 0.10, 0.08, and 0.06 mg kg"1. The following
5-ring and 6-ring PAHs were all below the MDLs in every SFS:
• Benzo[b]fluoranthene
• Benzo[k]fluoranthene
• Benzo[g,h,i]perylene
• Benzo[a]pyrene
• Dibenz [a, h] anthracene
• Indeno[l,2,3-cd]pyrene.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-14
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-5. Concentrations of the PAHs in Spent Foundry Sands
Compound
Acenaphthene
Acenaphthylene
Anthracene
Benz[a]anthracene
Benzo [b]fluoranthene°
Benzo [k]fluoranthene°
Benzo [g,h,i]perylene°
Benzo [a]pyrene°
Chrysene
Dibenz [a,h] anthracene0
Fluoranthene
Fluorene
Indeno [1,2,3 -cd]pyrene°
Naphthalene
Phenanthrene
Pyrene
Collected June 2005, 43 Samples3
(mg kg-1)
Min
0.04
0.03
O.03
O.10
0.12
O.13
O.14
0.20
0.08
O.16
0.06
0.04
O.14
O.03
0.03
0.03
Max
11.7
0.29
0.95
0.31
0.30
0.50
2.58
48.1
2.2
0.53
Meanb
0.39
0.06
0.32
0.06
0.06
0.07
0.07
0.10
0.05
0.08
0.05
0.31
0.07
3.67
0.62
0.14
No. of
Detects
12
20
34
o
6
0
0
0
0
3
0
2
39
0
40
41
23
Collected September 2005, 38 Samples
(mg kg-1)
Min
0.04
0.03
O.03
O.10
0.12
O.13
O.14
0.20
0.08
O.16
0.06
0.04
O.14
O.03
0.03
0.03
Max
0.18
0.32
0.99
0.20
0.11
0.17
1.03
1.19
14.6
1.91
0.86
Mean
0.04
0.06
0.41
0.06
0.06
0.07
0.07
0.10
0.04
0.08
0.07
0.34
0.07
1.46
0.73
0.17
No. of
Detects
10
13
34
o
6
0
0
0
0
1
1
5
32
0
35
37
24
Collected July 2006, 37 Samples
(mg kg-1)
Min
0.04
0.03
O.03
O.10
0.12
O.13
O.14
0.20
0.08
O.16
0.06
0.04
O.14
O.03
0.03
0.03
Max
0.40
0.33
0.69
0.15
0.33
1.05
42.2
1.86
0.73
Mean
0.05
0.05
0.19
0.06
0.06
0.07
0.07
0.10
0.04
0.08
0.05
0.23
0.07
2.01
0.49
0.11
No. of
Detects
8
13
31
2
0
0
0
0
0
0
6
30
0
34
35
33
< means less than the MDL.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the MDL.
0 All concentrations recorded below the MDL.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-15
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-6. Concentrations of Phenolics in Spent Foundry Sands
Compound
2-sec-Butyl-4,6-
dinitrophenol0
4-Chloro-3-methylphenol
2-Chlorophenolc
2,4-DichlorophenoP
2,6-DichlorophenoP
2,4-Dimethylphenol
2,4-Dinitrophenol
2-Methylphenol
3- and 4-Methylphenol
2-Methyl-4,6-
dinitrophenoP
2-NitrophenoP
4-NitrophenoP
PentachlorophenoP
Phenol
2,3,4,6-TetrachlorophenoP
2,4,6-TrichlorophenoP
2,4,5-TrichlorophenoP
Collected June 2005, 43 Samples3
(mg kg-1)
Min
0.21
0.18
O.ll
O.13
0.06
0.08
0.24
O.21
O.08
0.16
0.09
0.44
O.24
O.07
0.09
0.12
0.14
Max
0.82
12.3
0.86
14.9
6.11
186
Meanb
0.11
0.11
0.06
0.07
0.03
1.13
0.14
2.19
0.99
0.08
0.05
0.22
0.12
11.2
0.05
0.06
0.07
No. of
Detects
0
2
0
0
0
27
1
32
30
0
0
0
0
39
0
0
0
Collected September 2005, 38 Samples
(mg kg -1)
Min
0.21
0.18
O.ll
O.13
0.06
0.08
0.24
O.21
O.08
0.16
0.09
0.44
O.24
O.07
0.09
0.12
0.14
Max
0.45
7.45
9.90
3.98
50.0
Mean
0.11
0.10
0.06
0.07
0.03
0.72
0.12
1.29
0.58
0.08
0.05
0.22
0.12
4.41
0.05
0.06
0.07
No. of
Detects
0
1
0
0
0
24
0
27
33
0
0
0
0
35
0
0
0
Collected July 2006, 37 Samples
(mg kg -1)
Min
0.21
0.18
O.ll
O.13
0.06
0.08
0.24
O.21
O.08
0.16
0.09
0.44
O.24
O.07
0.09
0.12
0.14
Max
10.9
10.5
4.70
28.5
Mean
0.11
0.09
0.06
0.07
0.03
1.12
0.12
1.85
0.9
0.08
0.05
0.22
0.12
4.78
0.05
0.06
0.07
No. of
Detects
0
0
0
0
0
25
0
24
27
0
0
0
0
30
0
0
0
< means less than the MDL.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the MDL.
0 All concentrations recorded below the MDL.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-16
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Sand #12 (iron foundry that used novolac molds and cores) was the only sand where all
of the PAHs were below the MDLs. A summary of PAH data from the two additional sampling
events—that is, September 2005 and July 2006 can also be found in Table 2-5; the results are
markedly similar to those found in the first sampling event. One exception is sand #6, where the
naphthalene concentration during the first sampling event was 48.1 mg kg"1, but by the second
and third sampling event, it decreased to 8.3 and 0.16 mg kg"1, respectively. The other exception
is sand #5, in which the naphthalene concentration increased from 0.41 to 42.2 mg kg"1 by the
third sampling event. It is possible that the sand storage practices at the foundries account for
these differences.
Anthracene, fluorene, naphthalene, and phenanthrene were the most prevalent PAHs,
detected in >79% of the SFSs (Dungan, 2006). No discernible trend between the PAH
concentration and the type of molding sand, core binder, or metal poured was apparent. It is
likely that other variables, such as casting and core size and sand handling and storage, play a
role in the amount of organics found in the SFSs. Except for the naphthalene concentrations in
SFSs #6, #33, and #41, the results obtained by Dungan (2006) were similar to those obtained by
Lee and Benson (2006), who found that naphthalene (0.02-4.6 mg kg"1), phenanthrene (0.08-
0.9 mg kg"1), and 2-methylnaphthalene (0.004-9.8 mg kg"1) were generally present at higher
concentrations than the other PAHs. PAH-specific data for individual samples are found in
Appendix B, Tables B-4, B-5, and B-6
In a study conducted by Ji et al. (2001), naphthalene, 1- and 2- methylnaphthalene, and
phenanthrene were also at the highest concentrations in waste green sands from iron, steel, and
aluminum foundries. When compared to chemically bonded sands, the PAH concentrations were
higher in the green sands. Naphthalene accounted for about 30% of the PAHs found in all of the
SFSs.
Of the 17 phenolics analyzed, 11 were at concentrations less than the MDL in all 43 SFSs
in the June 2005 sampling event. Phenolics that were quantitatively detected in the majority of
the SFSs were phenol, 2-methylphenol, 3- and 4-methylphenol, and 2,4-dimethylphenol. In
general, phenol was found at the highest concentration, followed by 2-methylphenol and then 3-
and 4-methylphenol and 2,4-dimethylphenol. Phenol was present in samples from 39 of 43
foundries at concentrations ranging from 0.12-186 mg kg"1. Sand #6, from a steel foundry that
used both phenolic urethane no-bake molds and cores, contained the highest concentration of
phenol. In contrast, sand #29 was from a steel foundry that used the same mold and core binders,
but it contained substantially less phenol at 0.36 mg kg"1. The highest concentrations of 2-
methylphenol, 3- and 4-methylphenol, and 2,4-dimethylphenol were 14.9 mg kg"1 (sand #34), 6.1
mg kg"1 (sand #20), and 12.3 mg kg"1 (sand #20), respectively. Of the remaining phenolics, only
2,4-dinitrophenol and 4-chloro-3-methylphenol were found at concentrations that slightly
exceeded the MDL of 0.24 and 0.18 mg kg"1, respectively, in sands #6, #38, and #41. Phenolic
data from the two additional sampling events can also be found in Table 2-6. Constituent-
specific data for individual samples are found in Appendix B, Tables B-7, B-8, and B-9.
PCDDs, PCDFs, and PCBs are ubiquitous environmental contaminants. They are
nonpolar, lipophilic, persistent in the environment, and bioaccumulate in the food chain. Unlike
PCBs, PCDDs and PCDFs were never intentionally manufactured, but are largely released into
the environment during combustion processes. Ten representative spent sands from iron,
aluminum, and steel foundries, shown in Table 2-7, were analyzed for PCDD/PCDFs and PCBs
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-17
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
(Dungan et al., 2009). The concentrations of the PCDD/PCDFs and PCBs in the SFSs, expressed
as ng kg"1, are presented in Table 2-8. Except for 1,2,3,7,8,9-HxCDF, the tetra (T), penta (Pe),
hexa (Hx), hepta (Hp) and octa (O) congeners of PCDD and PCDF were found above the MDLs,
but not in all SFSs. Concentrations of the PCDD congeners ranged from <0.01-44.8 ng kg"1,
with 1,2,3,4,6,7,8,9-OCDD being found at the highest concentration in all of the SFSs. Although
the OCDD concentrations were the greatest, based on the TEF, OCDD is considered to be less
toxic than 2,3,7,8-TCDD by four orders of magnitude. 2,3,7,8-TCDD, with concentrations
ranging from <0.01-0.14 ng kg"1, was detected in only 50% of the SFSs.
Table 2-7. Description of the Spent Foundry Sands Analyzed for
PCDDs, PCDFs, and Coplanar PCBs
Sand
4
8
12
14
16
20
28
29
39
43
Metal Poured
Aluminum
Iron
Iron
Aluminum
Iron
Aluminum
Iron
Steel
Steel
Steel
Molding Sand
Green sand
Green sand
Shell
Green sand
Green sand
Green sand
Green sand
PU no-bake
Green sand
Green sand
Core Binder System and Process
Shell3
PUb coldbox, PU hotbox
Shell
PU no-bake, shell, core oil
PU coldbox, PU hotbox
Shell
None
PU no-bake
PU coldbox, shell, resin/CO2
PU no-bake, shell, core oil, resin/CCh
a Shell process associated with the use of novolac resin
b PU = phenolic urethane
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-18
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-8. Concentrations of PCDDs, PCDFs, and Coplanar PCBs and Homolog Totals in the Spent Foundry Sands (« =1)
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
TEFa
1
1
0.1
0.1
0.1
0.01
0.0003
Spent Foundry Sand (ng kg -1)
4
0.02
0.03
0.02
0.05
0.03
0.38
27.8
8
0.03
0.13
0.09
0.60
0.35
5.29
44.8
12
0.01
0.02
O.02
0.02
0.02
0.42
2.89
14
0.02
0.03
0.01
0.02
0.03
0.15
1.60
16
0.05
0.04
0.02
0.05
0.06
0.60
8.76
20
0.02
0.07
O.02
0.18
0.13
0.74
5.89
28
0.03
0.03
O.04
0.04
0.04
0.21
2.95
29
0.02
0.15
0.16
0.21
0.15
1.24
3.01
39
0.14
0.72
0.58
0.81
0.66
5.00
12.5
43
0.07
0.24
0.21
0.33
0.23
1.62
2.42
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,6,7,8,9-OCDF
0.1
0.03
0.3
0.1
0.1
0.1
0.1
0.1
0.01
0.0003
0.03
0.03
0.04
0.06
0.04
0.04
O.02
0.17
0.03
0.12
0.46
0.19
0.29
0.25
0.18
0.22
O.03
1.01
0.11
1.51
0.03
0.01
0.01
0.01
0.01
0.01
O.02
0.13
0.02
0.48
0.03
0.01
0.01
O.01
0.01
0.01
O.01
0.02
0.02
0.09
0.16
0.07
0.08
0.10
0.04
0.02
O.03
0.11
0.03
0.16
0.09
0.13
0.20
0.18
0.15
0.24
O.02
0.48
0.06
0.36
0.01
0.02
0.04
O.04
0.03
0.03
O.03
0.14
0.17
0.16
0.13
0.15
0.21
0.18
0.15
0.17
O.02
0.73
0.06
0.26
1.69
1.50
2.61
2.32
2.30
2.34
O.04
9.93
0.50
3.10
0.45
0.46
0.72
0.63
0.56
0.55
O.02
1.72
0.10
0.26
PCB-77
PCB-126
PCB-169
0.0001
0.1
0.03
0.30
0.12
0.02
47.4
1.22
0.09
0.43
0.02
0.01
2.03
0.06
0.02
7.14
0.24
0.03
2.13
0.72
0.06
0.53
0.01
0.02
0.81
0.22
0.05
4.35
1.99
0.68
1.21
0.38
0.12
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-19
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Congener
Sum TCDD
Sum PeCDD
Sum HxCDD
Sum HpCDD
TEFa
Spent Foundry Sand (ng kg -1)
4
0.33
0.33
0.42
0.76
8
1.41
1.37
5.01
10.3
12
0.01
0.00
0.07
0.63
14
0.22
0.17
0.23
0.35
16
0.58
0.42
0.90
1.48
20
2.80
1.51
2.24
1.52
28
0.24
0.83
0.42
0.44
29
9.78
8.39
8.12
2.71
39
21.8
20.7
22.7
10.2
43
9.58
9.70
9.64
3.54
Sum TCDF
Sum PeCDF
SumHxCDF
Sum HpCDF
0.66
0.55
0.46
0.28
5.10
2.75
2.22
2.07
0.33
0.15
0.14
0.36
0.50
0.16
0.10
0.06
1.59
0.57
0.45
0.25
5.32
2.89
1.52
0.78
0.08
0.21
0.37
0.32
6.06
3.25
1.89
0.94
53.0
32.8
22.1
12.1
16.8
9.31
5.55
2.10
< means less than the MDL.
Source: Dungan et al. (2009).
a Values assigned by WHO (Van den Berg et al., 2006).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-20
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-9 shows the PCDD, PCDF, PCB, and total dioxin concentrations corrected for
their TEFs and expressed as TEQs. However, because PCB-81 and mono-ort/zo-substituted
PCBs were not measured, the PCB contribution to the total TEQ concentration is not known.
Total dioxin concentrations ranged from 0.01-3.13 ng TEQ kg"1, with an average concentration
of 0.58 ng TEQ kg"1. The highest total dioxin concentration of 3.13 ng TEQ kg"1 was found in
sand #39, (a green sand from a steel foundry). This concentration is about 100 times lower than
the 300 ng TEQ kg"1 limit considered by EPA for biosolids (U.S. EPA, 2002e). In sand #39,
23%, 25%, and 22% of the TEQ was attributed to 1,2,3,7,8-PeCDD, 2,3,4,7,8-PeCDF, and
HxCDFs, respectively. Only 5% of the TEQ could be attributed 2,3,7,8-TCDD, the most toxic
dioxin congener. Other SFSs with higher TEQs were sands #8 and #43 (green sands from iron
and steel foundries), at 0.68 and 0.91 ng TEQ kg"1, respectively. In sand #8, 49%, 32%, and 19%
of the TEQ was attributed to PCDDs, PCDFs, and PCBs, respectively. In sand #43, 44%, 51%,
and 5% of the TEQ was attributed to PCDDs, PCDFs, and PCBs, respectively. In the remaining
SFSs, PCDDs and PCDFs accounted for 76 to 94% of the total TEQ.
Table 2-9. Toxicity Equivalents (TEQs) of PCDDs, PCDFs, Coplanar PCBs,
and Total Dioxins in the Spent Foundry Sands
PCDDs
PCDFs
PCBs
Total3
Spent Foundry Sand (ng TEQ kg'1)
4
0.05
0.03
0.01
0.10
8
0.33
0.22
0.13
0.68
12
0.02
0.01
0.00
0.04
14
0.03
0.01
0.01
0.05
16
0.02
0.06
0.03
0.11
20
0.13
0.14
0.07
0.34
28
o.ooa
0.01
0.00 a
0.01
29
0.23
0.14
0.02
0.40
39
1.12
1.80
0.22
3.13
43
0.40
0.47
0.04
0.91
a Sufficiently low that it rounds to zero.
b Sum of the PCDDs, PCDFs, and PCBs; does not include mono-or//zo-substituted PCBs.
2.5.4 Constituent Leaching Potential
The amount of any constituent that might be mobilized (leached) from a waste or material
depends on the constituent of concern, the matrix of the waste or material, and the environmental
conditions under which the waste or material is managed. It is important to have information
about the potential for the constituents to leach because leached constituents could be transported
to groundwater. Laboratory leaching tests are often used to determine the potential for a given
waste material to contaminate groundwater. Over the past two decades, a number of studies have
characterized the leaching potential of chemical constituents from SFSs and their impact on the
environment (Ham et al., 1981, 1986, 1993; Stanforth et al., 1988; Krueger et al., 1989; Regan et
al, 1994; Riediker et al., 2000; Lee and Benson, 2006). Many of these studies used the extraction
procedure (EP) toxicity test (U.S. EPA, SW-846 method 1310B), which was later replaced by the
TCLP. The TCLP was designed to determine the teachability of 25 organic compounds, 8 trace
elements, and 6 pesticides regulated under the Resource Conservation and Recovery Act of 1976
(RCRA).
The main drawback of the TCLP and EP for gathering data to assess SFS soil-related
applications is that they simulate leaching in an environment very different from that found in
such beneficial use scenarios. For example, the TCLP uses organic acids to simulate the
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-21
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
conditions found within municipal
solid waste landfills. A buffered
solution of acetic acid is used as the
extraction fluid in the TCLP, and the
pH of the extraction fluid is 4.93 ±
0.05 (or 2.88 ± 0.05 for highly
alkaline wastes). SFS would be used
in various surficial applications and
would not be exposed to water
containing high concentrations of
acetic acid or water with such a low
pH; thus, TCLP test conditions have
limited relevance to determining the
acceptability of soil-related beneficial
uses of SFS.
The TCLP
The TCLP estimates leachate concentrations, which are used by
EPA to determine whether a solid waste exhibits the hazardous
characteristic of toxicity (Kendall, 2003). A waste exhibits the
toxicity characteristic under RCRA if any one of the
constituents in the TCLP leachate exceeds its RCRA Toxicity
Characteristic regulatory limit. Conversely, if leachate estimates
do not exceed the regulatory limits, the waste is not considered
to exhibit the characteristic of toxicity and thus, is not a
hazardous waste under RCRA. The test was designed to
determine the mobility of both inorganic and organic analytes
present in liquids, solids, and multiphasic wastes in landfills.
The Toxicity Characteristic regulatory levels are 100 times the
National Primary Drinking Water Standards (NPDWSs). This
factor was established by EPA because it is assumed that
constituents in the leachate will be diluted and attenuated as
they seep from an unlined landfill.
Nevertheless, TCLP is often
used because (1) it is commercially available and (2) some state beneficial use determination
processes require that SFSs be tested using EPA-approved methods for the analysis of solid
wastes. The concentrations of 10 elements in TCLP extracts from SFSs are summarized in Table
2-10 (Dungan and Dees, 2009). Similar TCLP results were obtained for samples that were
collected from the same foundries at later dates (also in Table 2-10). Element-specific data for
each sample are detailed in Appendix B.
Dungan and Dees (2006) used the TCLP to assess the teachability of other elements that
are not regulated under RCRA Subtitle C, including antimony, beryllium, copper, nickel, and
zinc. In the vast majority of cases, these elements were not detected. A few exceptions did occur
where copper, nickel, and zinc were detected in the TCLP extracts. During the first sampling
event, both copper and zinc at 3.5 and 37.6 mg L"1, respectively, were at the highest levels in the
extract from sand #34 (i.e., non-leaded brass green sand), which also contained the highest total
copper and zinc concentrations (see Table 2-3). The TCLP extract from sand #2 (which had the
highest total nickel concentration at 2,328 mg kg"1) contained 0.94 mg Ni L"1. However, the
TCLP extract from sand #39 contained the highest concentration of nickel at 1.5 mg L"1,
although its total nickel concentration was about 22 times lower than that of sand #2. These data
appear to support the premise that the total element concentrations should not be used to predict
the amount of the element that is likely to leach from the SFS.
To our knowledge, published data do not exist that link the trace element concentrations
in TCLP leachates and their relationship to an industrial landfill or beneficial use field results.
Ham et al. (1986) found no relationship between the trace element concentrations in laboratory
leach extracts and those found in the unsaturated zone, saturated zone, and groundwater at
ferrous foundry landfills. As discussed above, the environmental conditions that the TCLP
simulates are unlike the conditions in which SFS would be beneficially used in soil-related
applications. Therefore, the most appropriate use of TCLP analytical data is to test whether SFSs
are hazardous waste under RCRA Subtitle C. As illustrated in Table 2-11, based on existing
data, SFSs do not exhibit the Toxicity Characteristic.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-22
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-10. Metal Concentrations in the TCLP Extracts from the Spent Foundry Sands
Element
Agc
As
Ba
Be
Cd
Crb
Cu
Ni
Pb
Sbc
Zn
Collected June 2005. 43 Samples3
(Concentration, mg I/1)
Min
0.04
0.001
O.86
0.01
0.01
O.46
0.10
0.14
O.05
0.02
0.41
Max
2.40
1.13
0.043
0.065
3.52
1.50
0.098
37.6
Meanb
0.020
0.058
0.446
0.007
0.007
0.230
0.193
0.163
0.027
0.010
1.16
No. of
Detects
0
24
1
3
3
0
8
9
1
0
3
Collected September 2005. 38 Samples
(Concentration, mg L"1)
Min
0.04
0.001
O.86
0.01
0.01
O.46
0.10
0.14
O.05
0.02
0.41
Max
0.019
43.9
0.298
40.3
Mean
0.020
0.003
0.430
0.005
0.005
0.230
1.23
0.092
0.025
0.010
1.47
No. of
Detects
0
25
0
0
0
0
6
6
0
0
4
Collected July 2006. 37 Samples
(Concentration, mg L"1)
Min
0.04
0.001
O.86
0.01
0.01
O.46
0.10
0.14
O.05
0.02
0.41
Max
0.017
0.064
5.39
1.71
1.13
42.5
Mean
0.020
0.003
0.430
0.005
0.007
0.230
0.194
0.128
0.055
0.010
1.49
No. of
Detects
0
23
0
0
1
0
1
4
1
0
4
< means less than the LOQ.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the
0 All concentrations recorded below the LOQ.
LOQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-23
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-11. Spent Foundry Sands TCLP Extracts Compared to
Toxicity Characteristic Regulatory Levels
Element
Ag3
As
Ba
Be
Cd
Cr3
Cu
Ni
Pb
Sba
Zn
All Sampling Events, 118 Samples
(Concentration, mg I/1)
Min
0.04
O.001
<0.86
0.01
O.01
O.46
0.10
O.14
O.05
0.02
O.41
Max
2.40
1.13
0.04
0.06
43.9
1.71
1.13
42.5
Mean
0.02
0.44
0.01
0.01
0.53
0.13
0.03
1.36
Toxicity
Characteristic
Regulatory
Level
5.0
5.0
100.0
1.0
5.0
5.0
< means less than the LOQ.
a All levels recorded below LOQ.
An alternative leaching procedure, the SPLP (SW-846 method 1312) was designed to
simulate the leaching of trace elements and organics from wastes or contaminated soils due to
acidic rainfall. Because the environmental conditions being mimicked or approximated by the
SPLP are more similar to some beneficial use situations than those approximated by the TCLP,
SPLP provides a more realistic estimate of trace element and organic mobility under field
conditions during precipitation events.12 Summary SPLP extract data from the 43 SFSs are
presented in Table 2-12. In every extract, antimony, beryllium, cadmium, chromium, lead,
nickel, and silver were below their respective LOQ. Arsenic, barium, copper, and zinc were
detected in some of the SPLP extracts. SPLP extracts of SFSs from the second and third
sampling events demonstrate similar results (also in Table 2-12). Compared to the TCLP
leaching results, which is run at a pH of 4.93 buffered by acetic acid, fewer trace elements were
found to be above the LOQ in the SPLP extract, which has an initial pH of 4.2. This can be
explained by the fact that the strong mineral acids used to make the SPLP extracting solution
provide little buffering capacity. After the extraction, the pH in the SPLP extracts was higher
(pH range of 4.8-9.9) than in the TCLP extracts (pH range of 4.6-5.7). Some elements tend to be
less soluble at the higher pH range found in the SPLP extracts.
12 The SPLP may not be used to assess the Toxicity Characteristic of a solid waste.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-24
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-12. Metal Concentrations in the SPLP Extracts from the Spent Foundry Sands
Element
Ag°
As
Ba
Bec
Cdc
Cr0
Cu
Ni
Pbb
Sbc
Zn
Collected June 2005, 43 Samples3
(Concentration, mg I/1)
Min
<0.08
O.001
0.23
<0.02
<0.01
0.01
O.21
O.05
0.08
O.04
O.18
Max
0.098
0.612
0.546
0.238
3.05
Meanb
0.040
0.006
0.161
0.010
0.005
0.005
0.115
0.030
0.040
0.020
0.165
No. of
Detects
0
25
9
0
0
0
1
1
0
0
2
Collected September 2005, 38 Samples
(Concentration, mg I/1)
Min
O.08
O.001
0.23
O.02
O.01
0.01
O.21
O.05
0.08
O.04
O.18
Max
0.024
0.371
0.748
0.089
1.62
Mean
0.040
0.008
0.129
0.010
0.005
0.005
0.122
0.028
0.040
0.020
0.130
No. of
Detects
0
24
3
0
0
0
1
o
6
0
0
1
Collected July 2006, 37 Samples
(Concentration, mg L"1)
Min
O.08
O.001
0.23
O.02
O.01
0.01
O.21
O.05
0.08
O.04
O.18
Max
0.017
0.634
1.66
0.070
0.284
3.95
Mean
0.040
0.004
0.154
0.010
0.005
0.005
0.147
0.026
0.047
0.020
0.194
No. of
Detects
0
28
5
0
0
0
1
1
1
0
1
< means less than the LOQ.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the LOQ.
0 All concentrations recorded below the LOQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-25
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
The TCLP and SPLP represent standard tests that are widely used by the EPA and other
regulatory agencies to evaluate the potential for constituent release into the subsurface. With few
exceptions,13 the aggressive conditions of the TCLP are thought to provide a very conservative
screen for leach potential. The scenario that the TCLP mimics, however, is not representative of
SFS use in manufactured soil because the level of acidity will overestimate constituent release.
In addition, the organic component of manufactured soils (e.g., composts, peat moss, pine bark,
biosolids) would likely sorb elements released from the molding sand (Basta et al., 2005;
Kumpiene et al., 2008). The SPLP conditions that mimic acid rain are more relevant than TCLP
for evaluating the conditions considered in this report.
Dungan and Dees (2009) also performed a shake extraction procedure using deionized
water, known as ASTM D 3987 (ASTM International, 2004), on the 43 SFSs at a liquid-to-solid
ratio of 1:20 (see Table 2-13). A comparison of the ASTM and TCLP results reveals that fewer
elements were above the LOQ in the water extracts; also, concentrations were generally lower in
the water extracts than concentrations from the TCLP. As discussed above, these results indicate
that pH is a factor affecting the leaching of elements from the SFSs. As with the non-buffered
SPLP extracting solution, the water used for the ASTM procedure is non-buffered. The pH of the
extracts from the ASTM procedure ranged from 4.7 to as high as 9.9, which explains why the
results are similar to those from the SPLP. In the water extracts from all SFSs, the concentrations
of silver, barium, beryllium, cadmium, lead, and antimony were below their respective LOQ.
The only water extracts that contained copper and zinc at concentrations that were one to two
orders of magnitude higher than the LOQ were from sands #33 and #34. The copper and zinc
concentrations in the extract from sand #33 were 1.1 and 1.0 mg L"1, while in sand #34, they
were 0.3 and 1.3 mg L"1, respectively. With respect to arsenic in the water extracts, 21 of 43
sands were below the LOQ. The water extract from sand #5 (green sand from an iron foundry
with 0.65 mg arsenic kg"1) had the highest concentration of arsenic at 0.018 mg L"1. Sand #27
(another green sand from an iron foundry), however, with the highest total concentration of
arsenic at 3.0 mg kg"1, leached <0.003 mg arsenic L"1. In a study by Lee and Benson (2006),
arsenic in water extracts from 12 green sands ranged from 0.003 to 0.008 mg L"1. Water extract
data from the second and third sampling events can also be found in Table 2-13. As with the
TCLP and SPLP results, the ASTM extract data from the subsequent sample sets were very
similar to data from the first set.
For most elements, pore water concentrations (Appendix B, Table B-26) were low, and
for many sands were below detection limits. However, plant nutrients are evident in SFS pore
water. The 39 SFSs (brass and olivine sands were omitted) have median soluble concentrations
of the macro nutrients calcium, magnesium, potassium, phosphorus, and sulfur of 32.5, 13.5,
27.3, 0.39, and 125 mg kg"1, respectively, and median concentrations of the soluble micro
nutrients boron, iron, manganese, zinc, copper, and molybdenum of 0.53, 1.14, 0.09, 0.05, 0.01,
and 0.11 mg kg"1, respectively. Only pore water aluminum is occasionally elevated, ranging from
<0.2-1,847 mg Al kg"1, with a median of 3.89 mg Al kg"1. However, despite this large range,
33.3% of SFS pore waters were below the aluminum detection limit of 0.2 mg kg"1. Not all
aluminum species are phytotoxic, and it is unlikely that the soluble aluminum found in the raw
SFS will remain stable in solution for long once blended with other soil components (Kinraide,
1991).
13 Recent research indicates that the TCLP may not provide an adequately conservative test for arsenic in mature
landfills characterized by alkaline pH, low redox potential, biological activity, long retention time, and organic
composition of mature landfills (e.g., Ghosh et al., 2004).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-26
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-13. Metal Concentrations in Water Extracts from the Spent Foundry Sands
Element
Ag°
As
Bac
Bec
Cdc
Cf=
Cu
Ni
Pbc
Sbc
Zn
Collected June 2005. 43 Samples3
(Concentration, mg I/1)
Min
0.05
0.003
O.24
0.01
0.01
O.02
0.07
0.05
O.ll
0.04
0.22
Max
0.018
1.06
0.046
1.34
Meanb
0.030
0.005
0.120
0.005
0.005
0.010
0.070
0.026
0.055
0.020
0.159
No. of
Detects
0
23
0
0
0
0
2
1
0
0
2
Collected September 2005. 38 Samples
(Concentration, mg I/1)
Min
0.05
0.003
O.24
0.01
0.01
O.02
0.07
0.05
O.ll
0.04
0.22
Max
0.024
0.218
Mean
0.030
0.008
0.120
0.005
0.005
0.010
0.045
0.026
0.055
0.020
0.110
No. of
Detects
0
24
0
0
0
0
2
0
0
0
0
Collected July 2006. 37 Samples
(Concentration, mg L"1)
Min
0.05
0.003
O.24
0.01
0.01
O.02
0.07
0.05
O.ll
0.04
0.22
Max
0.017
0.080
1.57
Mean
0.030
0.005
0.120
0.005
0.005
0.010
0.041
0.026
0.055
0.020
0.150
No. of
Detects
0
24
0
0
0
0
1
0
0
0
1
< means less than the LOQ.
a Source: Dungan (2008) and Dungan and Dees (2009).
b Mean calculated with all non-detects set at one half the LOQ.
0 All concentrations recorded below the LOQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-27
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
2.5.5 Plant Uptake of Trace Metals from Spent Foundry Sands
To date, only a few studies on the growth of plants in SFSs have been reported. In a
laboratory study conducted by Dungan and Dees (2007), high purity silica sand was mixed with
50% SFS (dry weight). Spinach (Spinacia oleracea, cv. Bloomsdale), radish (Raphanus sativus,
cv. Cherry Belle), and perennial ryegrass (Loliumperenne, cv. Pizzazz) were grown with added
soluble fertilizers (i.e., Hoagland's solution) to assess the phytoavailability of aluminum, barium,
beryllium, boron, cadmium, chromium, cobalt, copper, iron, lead, magnesium, manganese,
molybdenum, nickel, vanadium, and zinc. The SFSs used in this study were from two aluminum,
two iron, and two steel foundries. Plastic pots were used and filled with 1,500 g of the foundry
sand blend. There were four replicates of each treatment, plus a control. The sand blends were
adjusted to pH 6 with a dilute solution of FbSO/t, because the pH of foundry sands tends to be
slightly alkaline. After germination, the spinach and radish seedlings were thinned to three plants
per pot. The ryegrass was planted with 1 g of seed per pot. The pots were watered with 150 mL
of full-strength Hoagland's solution, alternating with the same volume of deionized water.
Plastic saucers were used at the bottom of each pot so that the applied volume of deionized water
and nutrient solution was allowed to be taken up. The pots were kept in a growth chamber at 20
± 2°C, 50% humidity, and under a light-dark cycle of 16 hours light and 8 hours darkness.
Radish globes and leaves were harvested at 27 days, and the spinach leaves with stems were
harvested at 39 days. The perennial ryegrass was harvested three times, at 27, 57 and 87 days, by
collecting all of the top growth when it reached a height of about 15 cm. After harvest, all plant
parts were thoroughly rinsed with deionized water and then dried to constant weight at 65°C. The
plant samples were digested to determine total metals following the method of Kukier et al.
(2004).
Although there were differences in the amounts of trace metals accumulated by the
various plant species, excessive amounts of trace metals (i.e., above the amount necessary for
proper plant nutrition and health) were not taken up, regardless of the SFS treatment (see
Appendix B, Tables B-20, B-21, and B-22). For the spinach and radish, boron, copper, iron,
manganese, and zinc were found to be within or close to the sufficiency range for agronomic
crops. In the ryegrass cuttings at 27, 57, and 87 days, copper and zinc were within sufficiency
ranges, but plants were iron deficient and contained elevated nontoxic concentrations of boron,
manganese, and molybdenum.
To evaluate the transmission of nutrients and trace metals from SFS into plant tissue,
Romaine lettuce (Lactuca sativa, cv. Parris Island Cos) was grown in 100% of a subset of 10
SFSs and a silica sand (play sand) control. Prior to planting, the SFS pH was reduced to a target
pH of 7.5 ± 0.5 using 3 applications of a 2% acetic acid solution, with wetting and drying cycles
between applications. Pots were prepared with 1 kg of pH-adjusted SFS or silica sand, the top
1.3 cm of which was amended with vermiculite to facilitate germination. To ensure nutrient
sufficiency, each pot was amended with Miracle-Gro® (15% N + 30% P2O5 + 15% K2O) to
supply nitrogen, phosphorus, and potassium at 200, 230, and 190 mg kg"1, respectively, in a split
application. An additional 100 mg N kg"1 was added as NFUNCb. Twenty lettuce seeds were
planted per pot. Three replicates of each SFS and the silica sand control were grown in a
completely randomized design. Plants were grown in a controlled environment growth chamber
with 18 hours of light per day, light temperatures of 20°C, and dark temperatures of 18.5°C. Pots
were thinned to four lettuce plants per pot (if more than four plants were present) at 14 days.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-28
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Lettuce was harvested after 40 days, rinsed in deionized water, and dried at 70°C for 48 hours
and crushed by hand. The dried material was weighed to determine dry matter growth (DMG).
Dry lettuce tissue (0.25 g) was predigested for 4 hours in 10 mL of aqua regia. Predigested
samples were digested at 140°C for 4 hours, or until clear. Filtered (0.45 |j,m) solutions were
analyzed by ICP-AES.
By growing lettuce in 100% sand and not allowing the pots to drain, there was no
opportunity for dilution of either the plant nutrients or other trace metals. However, the poor
physical properties of the sand, due to high bulk density, made germination difficult.
Germination ranged from 23.3-100% with a mean of 67%. The only pots that had full
germination were the silica sand control pots. However, for lettuce grown in SFS, DMG relative
to that in the control pots (RDMG) ranged from 9.6-226%, with a mean of 110%. The SFS with
low RDMG was also low for germination, so there were fewer plants. Generally, despite a slow
start, lettuce grown in the SFS performed well. The average plant tissue concentration of the
plant macronutrients nitrogen, phosphorus, potassium, and sulfur were all within the nutrient
sufficiency levels, although calcium and magnesium were slightly low. For the micronutrients
boron, copper, iron, manganese, molybdenum, and zinc, the tissue concentrations were all
adequate. Arsenic tissue concentrations were below 1 mg kg"1, except in the control sand, where
they were 1.43 mg kg"1, which is within the typical range for arsenic in plant tissue. Similarly,
other trace metals found in SFS tissue were within or below the levels typically found in plant
tissue.
In a greenhouse study conducted by Hindman et al. (2008), SFSs from two iron foundries
and one aluminum foundry were blended with either yard trimmings compost, spent mushroom
substrate (SMS), orbiosolids compost, and a silt loam soil at a dry weight ratio of 6.5:1.5:2.0
(SFS: compost: subsoil). All manufactured soils were characterized as sandy loams. Each of the
manufactured soils was initially amended with inorganic fertilizer and seeded with annual
ryegrass, which was harvested seven times. The grass cuttings were analyzed for aluminum,
boron, calcium, cadmium, copper, iron, potassium, magnesium, manganese, molybdenum,
sodium, nickel, phosphorus, lead, sulfur, and zinc. The ryegrass yields in the manufactured soils
exceeded the growth in natural topsoil, which was likely the result of the more available
nitrogen. Among the manufactured soils, the SMS plus biosolids compost showed larger yields
than blends containing yard compost. There was no evidence of trace metal deficiencies or
toxicities in ryegrass on the manufactured soils. Ryegrass tissue analyses indicated that most
tissue trace metal concentrations were lower or the same as the control and that most tissue
nutrient concentrations fell within the sufficiency range.
2.5.6 Potential to Impact Soil Biota
Microorganisms
Bacteria are the most numerous organisms in soils, and are important because they are
involved in essential processes, such as cycling of nutrients, biodegradation of organic
pollutants, formation of humus, and the stabilization of soil structure. Inputs of toxic elements
can alter the biological activity of soil microorganisms, sometimes causing a severe ecosystem
disturbance. Affected soils often exhibit decreased microbial diversity, microbial biomass and
enzyme activities, and lower respiration rates per unit biomass. An increasing body of evidence
suggests that microorganisms are more sensitive to heavy metal pollution than the faunal or
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-29
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
floral community growing on the same soil (Giller et al., 1998). However, a summary of the
effects of trace elements on soil microorganisms from laboratory and field studies shows
enormous differences between studies as to which element concentrations are toxic (Baath,
1989). In particular, the addition of metal salts during short-term (acute toxicity) laboratory
experiments is a poor predictor of long-term (chronic toxicity) effects on microbial biomass and
activity (Renella et al., 2002). Further complications arise as pollution in field studies generally
involves multiple elements, while laboratory studies often focus on a single element, making it
difficult to draw conclusions regarding the toxicity of element combinations. Perhaps because of
these difficulties, no advanced risk assessment schemes or regulatory policies have dealt with
impacts on microorganisms in environmental risk assessments (Giller et al., 1999). Despite these
obstacles, soil microorganisms are being examined as indicators of adverse effects of trace
element pollution, which could potentially be used to define critical trace element loadings for
soil protection (Chapman, 1999). Some measures used to investigate the response of soil
microorganisms to trace element pollution are enzyme activity, microbial biomass, respiration
rate, carbon mineralization, nitrogen fixation, and fatty acid composition (Rother et al. 1983;
Ibekwe et al., 1995; Aceves et al., 1999; Lee et al., 2002; Broos et al., 2004; Zhang et al., 2006;
Vasquez-Murrieta et al., 2006).
Leguminous plants are important in maintaining soil fertility because they contain within
their root nodules symbiotic bacteria capable of fixing atmospheric nitrogen. Within soils, free-
living associative and asymbiotic nitrogen-fixing microorganisms also play an important role,
but generally fix less nitrogen (Stevenson, 1982). To date, many laboratory and field studies
have investigated the impacts of trace elements on legumes and nitrogen-fixing bacteria (Rother
et al., 1983; McGrath et al., 1988; Giller et al., 1986; Ibekwe et al., 1995, 1997; Smith, 1997;
Lakzian et al., 2002; Broos et al., 2004, 2005). In an early experiment, Rother et al. (1983)
reported only minor decreases in nitrogenase activity, plant size, and nodulation of white clover
(Trifolium repens) growing on mine spoils containing up to 216 mg Cd kg"1; 30,000 mg Pb kg"1;
and 20,000 mg Zn kg"1. Rhizobia from other legume species have not been found to be inhibited
by soil element concentrations below those which cause significant phytotoxicity (Heckman et
al., 1986; Kinkle et al., 1987; Angle and Chaney, 1991; Angle et al., 1988; El-Aziz et al., 1991).
Although no specific studies have been conducted to assess the impacts of trace elements
in SFSs on rhizobia, the results from the above-mentioned studies do not implicate SFS as
having possible adverse effects on soil microbes, except for brass or other spent sands where
trace element concentrations are up to a few orders of magnitude higher than element
concentrations in native background soils. With the exception of a few SFSs where the
concentrations of copper, nickel, and/or zinc are strongly elevated, minimal impacts on rhizobia
can be expected to occur in SFS-amended soils. Due to the naturally low trace element
concentrations in most ferrous and aluminum foundry sands (see Table 2-3), manufactured soils
and agricultural soils amended with these SFSs will not reach element levels required to cause
adverse effects on soil microbes. Furthermore, compared to the results obtained by Broos et al.
(2005), all of the SFSs from iron, steel, and aluminum foundries contained cadmium at <5.9 mg
kg"1 and zinc no higher than 352 mg kg"1 (Appendix B, Table B-24).
Dehydrogenases are intracellular enzymes involved in microbial respiratory metabolism
(von Mersi and Schinner, 1991). The dehydrogenase activity (DHA) assay is a sensitive
technique that has been used to assess microbial activities in soil amended with organic residues,
composted municipal solid wastes, and biosolids (Obbard et al., 1994; Albiach et al., 2000;
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-30
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Garcia-Gil et al., 2000; Yang et al., 2003). In a study conducted by Dungan et al. (2006), the
DHA of a sandy loam soil amended with green sands or chemically bonded sands at 10, 30, and
50% (dry weight) was determined. The green sands were obtained from iron, aluminum, and
brass foundries, and the chemically bonded sands were made with phenol-formaldehyde or
furfuryl alcohol based resins. Overall, the addition of these sands resulted in a decrease in the
DHA that lasted throughout the 12-week experimental period (see Figures 2-1 and 2-2). This
effect was largely determined to be a result of blending the sand into the soil, which
subsequently reduced the total microbial population in the sample, and thus, resulted in
decreased DHA. When plain silica sand with very low trace element levels was added to the soil
at the same application rates, there was a decrease in the DHA as the blending ratio increased,
which also lasted throughout the 12-week period. A brass green sand that contained high
concentrations of copper, lead, and zinc at 8,496; 943; and 4,596 mg kg"1, respectively, severely
impacted the DHA. By week 12, no DHA was detected in the 30% and 50% treatments. In
contrast, the DHA in soil amended with an aluminum green sand was 2.1 times higher (all
blending ratios), on average, at week 4, and 1.4 times greater (30% and 50% treatments only)
than the controls by week 12. In core sand-amended soil, the DHA results were similar to soils
amended with aluminum and iron green sands. Increased activity in some treatments may be a
result of the soil microorganisms utilizing the core resins as a carbon source.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-31
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
^.uu
150 -
100 -
50 -
0
200
!c 150 -
CN
1
73 100 -
'ra
LJ-
I —
1 5°-
n -
a
^ 10% a
^^ 30% h "
^H 50%
a
b
b
1
r
a
r
a
a
c
"
c
T
1 d
3
r
a
d
b
p-
=> a
f b
1
b
b
r
a
a
T
1 a
I
b
b
a
c
r
a
a
T
!a
b
a
r
a
a
c
1 a
^
1
hd
ZUU
150 -
100 -
50 -
I
aa
11
C
a
a
T
1 a
Id
T
II
a
T
1 a
a a
a
d
' '
a
r
a
a
b
aa
cd
Figure 2-1. Dehydrogenase activities at (a) week 4, (b) week 8, and
(c) week 12 in Sassafras sandy loam soil amended with 10%, 30%, and
50% (dry wt.) spent green sand from iron, aluminum, or brass foundries.
Error bars represent the standard deviation of triplicate samples. Treatments with letter a were
significantly different (p <0.05) from the soil only control, while those with a letter b, c, or d were
significantly different (p <0.05) from the respective silica sand treated soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-32
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
^.uu
150 -
100 -
50 -
g
200 -
•^ 150 -
CN
1
73 100 -
LJ_
1
1
Z
— 50 -
n -
a
^m 10%
^ 30% a
^m so% a b
c
b b
1
a
a
I
a
a
ac
r
a
d
n
)
aaa
1 bTd
d c
T
a
d
T
b
a b
r
a
a
T
a
|
b
a
c
a
c
1
c b
T
a
X
b
a
'•a
a
r
a
d
•ra
d
1
zuu
150 -
100 -
50 -
n -
I
a
aa
be
a
T
a
r
:
[
a
a
a
T
. .
3
t
a
3
a
T
C
i
;a
T
Figure 2-2. Dehydrogenase activities at (a) week 4, (b) week 8, and
(c) week 12 in Sassafras sandy loam soil amended with 10%, 30%, and 50%
(dry wt.) fresh core sand made with either phenol-formaldehyde,
phenolic urethane, or furfuryl alcohol based resins.
Treatments with letter a were significantly different (p <0.05) from the soil only control, while
those with a letter b, c, or d were significantly different (p <0.05) from the respective silica sand
treated soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-33
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Earthworms
Earthworms play a beneficial role in the development of soil structure and fertility by
enhancing the decomposition and mixing of organic matter and improving water infiltration and
aeration (Lee, 1985). Earthworm activities are important in native grassland and woodland soils,
as well as agricultural soils; however, earthworms have difficulty performing essential functions
when they are exposed to harmful soil concentrations of trace elements (Edwards and Bohlen,
1996). Earthworms bioaccumulate some trace elements in their tissues as a result of oral (i.e.,
ingestion of large quantities of soil) and dermal routes of exposure (Helmke et al., 1979; Morgan
and Morgan, 1999). As a result, earthworms living in some contaminated soils present a
significant element-transfer risk to animals whose diet consists largely of earthworms (e.g.,
shrews, moles, badgers). If earthworms do survive in element-contaminated soils, it is more
pertinent to examine the element risk to the earthworm-consuming animals than to assess the
toxicity to the earthworms themselves (Chaney and Ryan, 1993; Brown et al., 2002). The
accumulation of cadmium, lead, and zinc in moles has been shown to reflect the bioavailability
of these elements to earthworms (Ma, 1987). In acidic sandy soils, cadmium accumulated in the
earthworms to a considerable extent, and critical concentrations of cadmium toxicity in moles
can be exceeded even when the soil cadmium concentration is relatively low. Earthworms and
moles also accumulated much more lead from the contaminated acidic sandy soils than from
soils that have been limed (Ma, 1987), demonstrating the importance of soil pH on element
bioavailability to earthworms.
Many earthworm studies have been conducted to determine the effects of trace elements
on survival, growth, cocoon production, litter breakdown, and the bioaccumulation of elements
(Anderson, 1979; Hartenstein et al., 1980; Beyer et al., 1982, 1987; Ma, 1982, 1984; Khalil et al.,
1996; Spurgeon and Hopkin, 1996; Morgan and Morgan, 1988, 1999; Posthuma et al., 1997;
Conder and Lanno, 2000; Dai et al., 2004). A potential shortcoming of some of these studies is
that they examined the effect of added metal salts (Ma, 1982, 1984; Khalil et al., 1996; Posthuma
et al., 1997; Conder and Lanno, 2000), rather than contaminated field soils nearer equilibrium.
When metal salts are added to soils (i.e., metal-spiking studies), they become more acidic with
increasing metal rate as protons are displaced. Trace elements applied as salts are generally more
bioavailable than those from mineralized or environmentally contaminated soils (Basta et al.,
2005). When Ma (1984) corrected the acidity of copper salt amended soils, the high earthworm
toxicity observed at low pH was reversed. Due to long-term soil-ageing processes, trace element
availability generally decreases with time (Ford et al., 1997; Trivedi and Axe, 2000; Lock and
Janssen, 2001). However, depending on the element and pH of the system, aging will not
necessarily result in decreased element bioavailability (Lock and Janssen, 2003).
There is a relatively large amount of data on the concentration of trace elements in
earthworms from biosolid-amended soils, smelter-contaminated soils, and mine spoils. In most
reports, earthworms were not found to bioconcentrate lead and zinc, but earthworms have been
found to bioconcentrate cadmium (Pietz et al., 1984; Beyer and Stafford, 1993). Cadmium
concentrations in earthworms are generally greater than soil concentrations, while lead
concentrations in earthworms are generally similar to or lower than soil concentrations. Beyer et
al. (1990) examined the ratio of chromium in earthworms to that in soil of dredged material
deposit sites and found no evidence of chromium accumulation. Helmke et al. (1979) found that
chromium measured in earthworms was related to residual soil contamination. Many of these
studies generally report the element concentrations in earthworms after the internal soil has been
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 2-34
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
removed (i.e., purged or depurated). However, there is little evidence to suggest that earthworm
consumers can avoid ingestion of the internal soil. From a risk perspective, it may be more
appropriate to consider the element-transfer risk posed by earthworms that have not been purged
(Brown et al., 2002) as approximately 35% of the unpurged earthworm dry weight is soil (Beyer
and Stafford, 1993).
Dungan and Dees (2006) conducted a 28-day experiment with Eisenia fetida (a red worm
adapted to manure or composts) to assess the bioavailability of trace elements in iron, aluminum,
steel, and brass SFSs. The soil blends contained 10%, 30%, and 50% SFS on a dry-weight basis.
After 28 days, the number of viable adult earthworms across all treatments and blending ratios
was not significantly different from the control, except in blends containing 30% and 50% SFS
from a brass foundry (see Figure 2-3). The high earthworm mortality in the brass sand blends
correlated well with the high total and diethylenetriamine pentaacetic acid (DTPA)-extractable
concentrations of copper, lead, and zinc (see Table 2-14). The DTPA procedure is widely used
to determine plant available micronutrients in soils (Lindsay and Norvell, 1978) and has also
been used to assess the accumulation of trace elements by earthworms (Dai et al., 2004). Trace
element concentrations in the tissues of purged earthworms from iron, aluminum, and steel SFS
blends did not exceed those in the control. The copper and zinc concentrations in worm tissue
from the 10% brass blend were about 10 and 2 times higher than the control, respectively.
Because of the high copper, lead, and zinc concentrations (i.e., above those found in background
soils) in many brass molding sands, they should not be considered for beneficial use in
manufactured soils or other unencapsulated uses.
(D
LU
1 H-
12 -
10 -
8 -
6 -
4-
2 -
n
^m 10%
I 1 30%
rzzi 50%
I
a
a
1
J*
Figure 2-3. Adult earthworm survival after 28 days in the SFS blends.
Treatments marked with a letter are significantly different than the control (p <0.05, Holm-Sidak
method). Error bars represent the standard deviation of four replicates (eight replicates in the case
of the control). AGS = aluminum green sand; IGS = iron green sand; NBS = steel phenolic
urethane no-bake sand; BGS = brass green sand.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-35
-------
Chapter 2.0 Background and Characteristics of Spent Foundry Sand
Table 2-14. Total and DTPA-Extractable Metal Concentrations
in the Brass Green Sand Blends
Blending
Ratio
10%
30%
50%
Copper
Total3
812.0
2198.7
3713.3
DTPA
(l:5)a
154.8
494.7
884.5
%b
19.1
22.5
23.8
Lead
Total
87.2
243.4
386.2
DTPA
(1:5)
31.8
135.2
216.7
%
36.4
55.5
56.1
Zinc
Total
438.4
1186.4
1975.3
DTPA
(1:5)
72.7
194.7
320.0
%
16.6
16.4
16.2
a mg kg'1
b Percent of total metal that was DTPA extractable.
PAHs are common xenobiotic compounds in soils and are persistent because of their low
mobility and resistance to degradation. Because PAHs are hydrophobic in nature, they tend to
associate with soil organic matter and mineral fractions (Semple et al., 2003). The lipophilic
nature of PAHs can result in the bioaccumulation of these chemicals by soil biota, such as
earthworms (Krauss et al., 2000; Tang et al., 2002; Jager et al., 2003). As with trace element
contaminants, the bioaccumulation of PAHs and other persistent lipophilic compounds (e.g.,
PCBs) by earthworms presents a potential risk to earthworm-consuming animals. However, as
the soil-PAH contact time increases, there is a corresponding decrease in the extractability of the
PAHs in the soil, and their bioavailability to earthworms also decreases with time (Kelsey and
Alexander, 1997; Johnson et al., 2002). Johnson et al. (2002) found that tissue concentrations of
pyrene and benz[a]anthracene in earthworms declined by 58% and 43%, respectively, after
spiked soils were incubated for 240 days. In general, the extractability (via chemical extraction
procedures) and bioavailability of xenobiotics in soils, composts, and biosolids has been found to
decline substantially within months after application (Hatzinger and Alexander, 1995; Wang et
al., 1995; Puglisi et al., 2007). This process is known as "aging" and results from the slow
diffusion of xenobiotics to microsites or adsorption deeper into lipophilic soil organic matter
particles (Alexander, 1995). Even low molecular weight xenobiotics can become aged and less
bioavailable over time in soils (Frink and Bugbee, 1989; Guo et al., 2003). PAHs and phenolics
are present in SFSs below background soil concentrations (Dungan, 2006), and because of the
aging process, it is likely that these compounds will present a minimal risk to earthworms and
higher organisms. Thus, as long as SFSs are managed appropriately, the concentrations of most
organic compounds of concern will remain low and sensible land application of byproducts will
result in minimal risk to animals, humans, and the environment from organics (Kester et al.,
2005; Overcash et al., 2005).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
2-36
-------
Chapter 3.0 Problem Formulation
3. Problem Formulation
The overall goals for this assessment are to (1) evaluate all available information on
behavior of SFS in soils; (2) identify likely exposure pathways and receptors associated with
various unencapsulated SFS use scenarios; and (3) determine whether the proposed SFS uses
have the potential to cause adverse health or ecological effects (defined in this assessment as
above 10~5 risk for cancer, and an HQ of 1 for noncancer and ecological effects). With these
goals in mind, this chapter presents
• A description of the overall scope of the risk assessment, including the types, relevant
characteristics, and proposed uses of SFS
• Conceptual models illustrating the environmental behavior and potential exposure
pathways relevant to constituent releases from SFS in three soil-related applications
• The analysis plan developed to identify COCs and screen for potential risks associated
with SFS use in manufactured soils, soil-less media, and road subbase.
3.1 Scope of the SFS Risk Screening
Chapter 2 presented the body of data used in this analysis. This is the most rigorous and
consistent body of data available characterizing SFS and its constituents to date. The scope of
this screening risk assessment focuses on specific "unencapsulated" uses of SFS. Unencapsulated
uses present the highest potential for release of a material and its constituents because the
material is not chemically or physically bound. Below is a summary of the types of SFS,
constituents in SFS, and beneficial uses that are included in the scope of this analysis, as well as
other information about the scope.
3.1.1 Types of SFSs
As described in Chapter 2, there are many different types of SFS. The assessment
categorized SFSs according to three characteristics: the type of metal cast (e.g., aluminum, iron,
brass), the mineral type of the virgin sand (e.g., silica, olivine), and the type of binder used (e.g.,
clay, chemical binders). Samples from 43 U.S. foundries were collected by USDA-ARS and
industry, and analyzed by USDA-ARS. The characteristics of these samples are as follows:
• Metal cast type: 4 aluminum, 31 iron, 6 steel, and 2 non-leaded brass sands14
• Mineral type: 41 silica sands and 2 olivine sands
• Binder type of molding sand: 36 green sands and 7 chemically bound sands.
After a thorough review of the analytical data, described in Chapter 2, it was determined
that the remainder of this evaluation would focus on silica-based SFSs from iron, steel, and
aluminum foundries. Therefore, non-leaded brass sands and olivine sands would not be included
in this analysis. One of the two non-leaded brass sand samples had high levels of copper and zinc
14 Sands from brass and bronze foundries that use lead are frequently hazardous waste because they leach lead at
levels above the federal regulatory limit (see 40 CFR 261.24). Only nonhazardous SFSs are included in the scope
of this evaluation. Therefore, sands from leaded brass and bronze foundries were not collected, and such sands
were not evaluated in this study.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-1
-------
Chapter 3.0 Problem Formulation
(3,318 and 1,640 mg kg"1, respectively). Additionally, both olivine sand samples had high levels
of nickel (2,328 and 1,022 mg kg"1). As discussed in Chapter 2, Section 2.5.2, the nickel in
those sand samples did not come from the foundry operations; rather, the mineral olivine is a
magnesium iron silicate and contains naturally elevated concentrations of nickel, cobalt, and
chromium. It is important to note that the olivine sands were not included in this assessment
because there was limited constituent-specific data on those sand types. Further assessment of
olivine sands from aluminum, iron, and steel foundries could be performed to determine the risk
associated with the use of these sands in unencapsulated applications, and their exclusion from
this assessment should not be interpreted to mean that olivine sands could not be considered or
approved for such uses, where analytical data indicate they are acceptable.
3.1.2 SFS Characteristics
Both the chemical and physical characteristics of SFS are relevant to effects associated
with their use. The sand, clay, and silt content of the SFS affect the potential for particle
emissions and leaching. Smaller particle sizes (i.e., higher silt content and lower sand content)
result in greater potential for particle emissions (because the individual particles are more readily
released into the air) and for leaching (because a greater surface area of each particle is exposed
to the precipitation and groundwater that leaches the constituents from the particle). As shown in
Table 2-2, the silt content of SFS ranges from 0-16.9%, whereas the sand content ranges from
76.6-100%. The particle size information was used in the inhalation pathway screening
assessment to calculate emission rates for SFS.
As discussed in Chapter 2, Section 2.5.1, leaching potential is affected by pH, especially
for metals. For most metals, higher leaching occurs at the extreme ends of the acid/alkaline
spectrum and lower leaching occurs when the leachate is neutral. However, other variables, such
as redox potential, can significantly alter the leaching behavior of some metals (e.g., arsenic).
Agricultural and horticultural uses of SFS generally require that the soil remain near neutral pH
to promote healthy plant growth. Of the various types of leaching data presented in Chapter 2
(i.e., TCLP, SPLP, ASTM D3987, and pore water), this evaluation primarily used SPLP and
ASTM data. SPLP simulates leaching due to acid rain, and is run at an unbuffered pH of 4.2.
ASTM method estimates leaching at the material's natural pH, which for SFS ranged from 6.67-
10.2. These tests were performed on each SFS sample to empirically estimate the leaching
potential. Leaching data are described in Chapter 2, Section 2.5.4, and presented in Tables 2-12
and 2-13. These data were used in this assessment to evaluate the groundwater and produce
consumption pathways. In addition, TCLP data, estimated under very acidic conditions, were
used when neither SPLP nor ASTM data were available (see Chapter 4, section 4.2.1). Finally,
pore water data were used in refined ecological exposure modeling (see Chapter 5, Section
5.3.8).
The total concentrations of constituents were important inputs into both the screening
process and the predictive risk modeling. Used initially to identify constituents for evaluation,
total concentrations were also used to assess the inhalation pathway, the groundwater ingestion
pathway, and the soil pathways (i.e., the ingestion of soil and home grown produce and dermal
contact with soil). In addition, total concentrations were used in evaluating the potential for
adverse effects to ecological receptors. Total concentration data for metals used in this evaluation
are described in Chapter 2, Section 2.5.2, and presented in Table 2-4, and total concentrations
of organics used in this evaluation are described in Chapter 2, Section 2.5.3, and presented in
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-2
-------
Chapter 3.0 Problem Formulation
Tables 2-5, 2-6, and 2-8. With the exception of arsenic exposure through incidental soil
ingestion, the conservative assumption made in this screening risk assessment is that 100% of the
total concentration of each constituent is biologically available to the receptors. This is a
conservative assumption because, as discussed in Chapter 2, Section 2.5.6, metals exist in soils
in solid phases, not as the more bioavailable soluble salts, and the metals become less
bioavailable over time as soil organic components age. When assessing exposures to arsenic in
soil, U.S. EPA (2012b) recommends applying a default relative bioavailability (RBA) value of
60% when a site-specific value in unavailable. This assessment used the recommended default
value.
3.1.3 Beneficial Uses of SFS
In general, SFS can be used as an effective replacement for virgin sand in many
geotechnical and agricultural applications. This evaluation focused on the following potential
unencapsulated beneficial uses of SFS:
• Roadway construction as subbase
• Soil-less potting media for horticultural purposes
• Mineral component of manufactured soils.
Road subbase, soil-less potting media, and manufactured soils are discussed in greater
detail below in Section 3.1.4.
3.1.4 Conceptual Models
The information on the SFS characteristics and constituents presented in Chapter 2 was
used to develop the conceptual models. The conceptual models describe the sources, exposure
pathways, and receptors associated with SFS use in roadway construction, blending operations
that produce manufactured soils and soil-less potting media, and use of manufactured soils in
home gardens.
Figure 3-1 shows the conceptual model for SFS used as road subbase. Road subbase is a
layer of material required in some roadway applications to change the physical characteristics of
the land area on which the roadway is to be built so that the pavement is capable of withstanding
the stress of vehicle traffic and seasonal changes (e.g., freeze/thaw cycles). The subbase is placed
directly onto the subgrade and is covered by the base course, which is the layer in the roadway
beneath the pavement. Subbase thickness varies depending on road type, site requirements, and
material used, but sand subbase thickness typically ranges from 10-25 cm (i.e., 4-9 inches, U.S.
ACE, 1984). Pre-use storage and processing would vary by proposed use, but would likely
involve at least some storage in open areas. Rainfall on stored SFS piles or not yet covered
subbase could potentially leach constituents that could migrate through the subsurface and
contaminate an underlying groundwater aquifer. While possible, constituent releases into surface
waterbodies are not likely to be significant because standard road construction practices include
engineering controls to prevent significant runoff/erosion15. During loading and unloading
15 Runoff controls are a legal requirement under the National Pollutant Discharge Elimination System (NPDES) that
is part of the Clean Water Act. Most states have been authorized to implement the NPDES storm water program
(http://cfpub.epa.gov/npdes/stormwater/authorizationstatus.cfm), although some areas (e.g., tribal lands) remain
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-3
-------
Chapter 3.0 Problem Formulation
operations at roadway construction sites, nearby residents could be exposed via the inhalation of
particulate emissions and/or the incidental ingestion of soil following particle deposition;
terrestrial receptors (e.g., small mammals, soil invertebrates) could be exposed to chemical
constituents in SFS through direct and indirect exposure pathways.
SFS Source
Exposure Pathways
Receptors
Roadway Subbase
r
Temporary
storage pile
1
Particulate/
Volatile
Emissions
>
Leaching
Dispersion
^
Deposition
Air
Inhalation
qnji
Ingestion/
Dermal contact
Complete pathway
Incomplete pathway
-> The scenario assumes that engineering controls would
be used to significantly reduce the particulate and
volatile emissions from the temporary storage pile.
-> The scenario assumes that engineering controls would
prevent significant runoff/erosion from releasing
constituents into surface waters.
Figure 3-1. Conceptual model: the use of SFS in roadway subbase.
Given their inherent properties and low cost, SFS could potentially be of value as
feedstock for the blending of soil-less potting media and manufactured soil. Soil-less potting
media are generally used by nurseries as temporary growth media while individual plants await
sale, whereas manufactured soils more closely mimic native soils, and can be used on a much
larger scale as a long-term replacement for degraded native soils. Soil-less potting media and
manufactured soil could be mixed at the site of application (e.g., manufactured soil blended at a
construction site to landscape degraded topsoil), or mixed at a nursery, landscaping company, or
commercial soil-blending operation (hereafter referred to collectively as blending sites). SFS
used in these horticultural or agricultural applications is not encapsulated, and piles of SFS
feedstock may be uncovered for short periods of time. Figure 3-2 shows the conceptual model
for residents near a blending site. This scenario assumes that SFS would be temporarily stored on
site near other media components, along with piles of various blended soil and soil-less potting
media.
If uncovered, rainfall on stored SFS and blended piles could potentially leach
constituents; if the piles are stored on a pervious surface, these constituents could potentially
under the direction of EPA. The NPDES regulations establish best management practices (BMPs) for any source
of sediment, from sites or operations (e.g., construction, agricultural, or industrial), that might impact surface
waters. Many of the BMPs applicable to the control of runoff are similarly used to control fugitive dust emissions
as required under the Clean Air Act.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
3-4
-------
Chapter 3.0 Problem Formulation
migrate through the subsurface and contaminate an underlying aquifer. In addition, rainfall and
windblown erosion could result in some portion of the SFS running off and possibly reaching
nearby surface waters, assuming that the blending site did not include any sort of runoff
collection system. Storage and blending processes at commercial soil-blending facilities could
potentially be conducted on a much larger scale relative to storage and blending soil-less potting
media, and cover a wide range of manufactured soil "recipes." During storage, and particularly
during the blending process, chemical constituents could volatilize or be released via particulate
emissions. Nearby residents could be exposed through the groundwater pathways or the
inhalation of ambient air. Terrestrial receptors could be exposed to chemical constituents in SFS
through direct and indirect exposure pathways.
SFS Source
Exposure Pathways
Receptors
Complete pathway
Incomplete pathway
—> The scenario assumes that deposition would result in
insignificant exposures for the soil pathways when
compared to the home gardener scenario (Figure 3-3).
—> The scenario assumes that engineering controls would
prevent significant runoff/erosion from releasing
constituents into surface waters.
Figure 3-2. Conceptual model: the blending site.
Figure 3-3 shows the conceptual model for the use of SFS-manufactured soil (i.e.,
blended soils containing SFS) in home gardens. Although SFS-manufactured soil could be used
in corporate and residential landscaping (e.g., resurfacing construction sites), the home gardener
could potentially receive a much higher exposure to SFS constituents under the following
assumptions
• The home gardener incorporates a significant amount of SFS-manufactured soil into the
home garden
• The home gardener frequently works in the garden, thereby increasing the opportunities
of dermal contact and incidental ingestion of the SFS-manufactured soil, and
• A significant portion of produce consumed by the home gardener would be taken from
the garden consisting of SFS-manufactured soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
3-5
-------
Chapter 3.0 Problem Formulation
Because the SFS-manufactured soil is unencapsulated, direct exposures (e.g., ingestion,
dermal contact) could occur, and constituents could leach from the home garden following
rainfall events and/or irrigation. Additionally, terrestrial receptors could be exposed to chemical
constituents in SFS through direct and indirect exposure pathways.
SFS Source
Exposure Pathways
Receptors
Complete pathway
Incomplete pathway
-> The scenario assumes that manufactured soil is used
soon after delivery, so constituent releases from the
temporary storage pile are insignificant.
—> The scenario assumes that the home gardener would
impose controls to prevent significant runoff/erosion of
manufactured soil from the garden.
Figure 3-3. Conceptual model: the use of SFS-manufactured soils in home gardens.
The three conceptual models shown above were used in developing the Analytical Plan
discussed in Section 3.3.
3.1.5 Assumptions Behind the Risk Screening
The development of these conceptual models included assumptions that influenced the
selection of which exposure pathways to evaluate. These assumptions include the following:
• Acute and short-term worker exposures during application would be addressed by
existing standards developed by the Occupational Safety and Health Administration
(OSHA), and therefore potential worker exposures were not evaluated.
• For the temporary storage and use of SFS, indirect exposure pathways (e.g., air emissions
to soil deposition to soil-to-plant uptake to ingestion) would be unlikely to produce
significant exposures because
- there would likely be engineered controls to prevent the loss of valued commodities,
such as SFS feedstocks or blended soils,
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
3-6
-------
Chapter 3.0 Problem Formulation
- few chemical constituents have been shown to biomagnify in terrestrial food webs,16
- the time to reach steady state with respect to plant and animal concentrations would
be insufficient, so bioaccumulation would be limited, and
- releases during roadway construction using SFS would be temporary and intermittent
and, as a result, the potential for exposure would be very limited.
• The potential for exposure via direct contact (e.g., human incidental soil ingestion,
ecological exposures) is greater in the home garden scenario than the temporary storage
and use at blending facilities because air emissions and deposition from blending
facilities or other temporary storage piles are unlikely to result in residential soil
concentrations greater than those found in home gardens.
• The potential to contaminate groundwater is greater in the home garden scenario than the
other scenarios because (1) the SFS would remain in the garden indefinitely, (2) the SFS
is incorporated into the soil rather than sitting on top of the soil, (3) the garden presents a
much larger footprint (approximately 405 m2) than the temporary storage pile (assumed
to be 150 m2 in size), and (4) the soil underlying a garden would likely have a higher
hydraulic conductivity than a compacted soil or concrete pad used for the temporary
storage of SFS.
• Because SFS and manufactured soils have economic value17, blending sites would
process the SFS as rapidly as possible to generate revenue. This means that (1) the
temporary storage pile would remain in place for a relatively short period of time before
soil blending, and (2) the storage pile would likely be managed to protect the material's
value and workability (e.g., use of a temporary cover to prevent loss due to runoff, and
prevent the pile from becoming saturated with water).
• Commercial blending facilities demonstrate the greatest potential for nearby residential
inhalation exposures, because they tend to work with larger volumes of feedstock and
product (thereby emitting greater volumes of particulates) and conduct operations
throughout the year.
• The economics of purchasing, transporting, and applying SFS-manufactured soil would
make its large-scale agronomic application untenable - farmers could not afford it.18
Other potential agronomic uses for SFS (e.g., to improve soil texture) involve application
rates that would result in SFS concentrations lower than the assumed 1:1 blend (i.e., the
soil is 50% SFS, by weight) in SFS-manufactured soil.
In addition to these overarching assumptions, the risk assessment was predicated on a
number of conservative assumptions intended to ensure that the results could be used to support
management decisions with a high degree of confidence. That is, the assessment was
intentionally designed not to underestimate the potential risks to human health and the
environment.
16 With the exception of certain persistent organic pollutants, such as dioxins and PCBs, we are not aware of any
studies demonstrating biomagnification for multiple trophic levels (e.g., from terrestrial soil invertebrates up
through top predators).
17 In 2007 manufactured soil sold for approximately $21.50 yd"3 (cost of product and delivery), which would be
about $22,800 A'1 for a 20 cm-deep layer (Kurtz Bros., Inc. 2007).
18 See previous footnote.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-7
-------
Chapter 3.0 Problem Formulation
• The exposure scenarios focus on sensitive populations with respect to behaviors that tend
to increase exposures. For example, the home gardener scenario represents adults and
children that will have a relatively high level of direct contact (e.g., incidental soil
ingestion) and indirect contact (e.g., ingestion of home grown produce) when compared
to other populations.
• For carcinogenic (i.e., cancer-causing) constituents, the target cancer risk was defined as
an excess lifetime cancer risk of 1 chance in 100,000 (i.e., IE-OS).
• For constituents that cause noncancer health effects, the target hazard level was defined
as a ratio of predicted intake levels to safe intake levels—the HQ—of 1.
• The Phase II modeling (explained further in Section 3.2.2, below) used the upper end of
the exposure concentration distribution (i.e., groundwater screening modeling used the
90th percentile receptor well concentration, and refined surface and groundwater
modeling used the 90th percentile of the exposure distribution) rather than a central
tendency measure.
• Exposure assumptions used in the risk modeling were designed to overestimate, rather
than underestimate, potential exposures. For example, the exposure estimates from
ingestion of home-grown produce assumed that the receptor consumes a very large
amount of produce because the total produce diet is the sum of multiple produce
categories (e.g., root vegetables, leafy greens). This implies that (1) all of these categories
can be grown in the 0.1 acre garden in the same season, (2) all of these categories are
consumed at relatively high rates, and (3) all these categories are consumed year round.
• For effects to ecological receptors (e.g., plants, animals, soil invertebrates), conservative
environmental quality criteria (i.e. Eco-SSLs - see section 4.4.3 for more on the
conservative nature of these screening levels) were used to define the target levels.
• The home garden was accessible to all residents, including children at all times; and
• The addition of SFS-manufactured soil (containing SFS at 50% of the soil dry weight) to
the home garden essentially replaced the existing top 20-cm layer of local soil.
3.2 Analysis Plan
The analysis plan presents the overall approach used to (1) identify which, if any, SFS
constituents have the potential to cause adverse health and environmental effects, and (2) model
those constituent in the scenarios described in Section 3.1 associated with the greatest potential
for exposure to SFS constituents.
Of the exposure scenarios described in Section 3.1, it was judged that the home garden
scenario involved the greatest potential for exposure to SFS constituents. If risks from the use of
SFS-manufactured soil in home gardens was below levels of concern for human health and
ecological receptors, then risks from the other uses of SFS addressed by this assessment (i.e.,
soil-less potting media and road subbase) would also be below levels of concern. The exposure
pathways evaluated included in the home garden scenario are (1) the ingestion of and dermal
exposure to groundwater contaminated by SFS constituents leaching from SFS-manufactured
soil in a home garden; (2) the inhalation of SFS emitted from soil-blending operations; and (3)
the incidental ingestion and dermal exposure to SFS-manufactured soil, as well as ingestion of
fruits and vegetables grown in SFS-manufactured soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-8
-------
Chapter 3.0 Problem Formulation
As illustrated in Figure 3-4, the analysis plan involved a two-phase process for (1)
identifying the COCs using a lines-of-evidence approach for the groundwater, inhalation, and
soil pathways; and (2) conducting probabilistic risk modeling of any COCs identified for further
study. Information gathered in Phase I, as well as the risk modeling results, represent lines of
evidence. The risk characterization, presented in Chapter 6, integrates these lines of evidence
with the substantial body of scientific research on SFSs presented in Chapter 2 to develop a
complete picture of the potential for adverse effects to both human and ecological receptors.
3.2.1 Analysis Phase I: Identifying Constituents of Concern
As illustrated in Figure 3-4, Phase I of the analysis was designed to identify the universe
of SFS constituents needing more refined study; the COCs. This initial step included a review
and synthesis of a wide variety of information on the types of SFS, production processes,
properties of constituents in SFS (e.g., total constituent concentrations, leach test data),
toxicological studies, and relevant soil science on the uptake and accumulation of chemicals
(particularly metals) in plants and animals. Under Phase I, SFS constituents that met relevant
pathway-specific screening criteria would need no further evaluation. SFS constituents that did
not meet relevant pathway-specific screening criteria, however, would be evaluated further under
Phase II.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-9
-------
Chapter 3.0 Problem Formulation
Constituents in SFS
!
Groundwater Inhalation Soj| p g
Pathways Pathway
O
O
o
O)
c
"c
fli
^n
fli
t/>
(0
.c
Q_
.—
(/)
t
M—
nsidered
O
o
"5
i
NO /Leac
/ avail;
r n
hate^W
ta \ _2
able? / t
\ X a
TYES 0
~U
T ^
/\ 8
/ Above / "o
^ / drinking water \ z
~/ or dermal //
^xcriterion?^^
\, ,/
///
15 //
c
1
NO /He
•^ ( bench
/^ avail;
x
r
alth ^x
mark y
able? .^1
\v /
(YES
NO / Fail /
•^ / SCREENS y
^x modeling? ^^
^x ^/
N. ,/
1 y
YES
^1^ N° r Above ^N
[^ "/detection limit?/
1
t A.
^ / \
"~ / /
a) NO y Soil ^\
•^ / available? /
c / /
0 / /
I YYES
/\
/ A \
/ Above /
, NO / SSL/10, /
V y ^/DermalSSL, or/
^~~^ /EcoSSL?/
YES / /
/,/
YES
Analysis Phase II:
Risk Modeling
1
Probabilistic
groundwater
screening
model (IWEM)/
refined model
(EPACMTP)
|
Groundwater Pathways:
screening level; Compare
(90%-ile) to SFS concentr
Air Pathway: All SFS con
Probabilistic air
pathway
screening
model
Probabilistic soil
pathways
screening model
Compared IWEM modeled well concentrations (90%-ile) to lowest
d EPACMTP risk-based, groundwater protective, soil concentrations
ations.
stituents were eliminated from further consideration after Phase 1.
Soil Pathways: Compared risk-based soil concentrations (90%-ile) to SFS concentrations.
Figure 3-4. Analysis Plan for the risk assessment of SFS uses in soil-related applications.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
3-10
-------
Chapter 3.0 Problem Formulation
Analysis Phase I: Groundwater Pathway
In the home gardening scenario, the only source of drinking water for the home was a
well located directly downgradient from the garden. As shown in Figure 3-4, a two-step process
was used to identify which SFS constituents, if any, would require further evaluation for the
groundwater pathway.
• Step 1: Leachability of constituents. Leachability was evaluated based on the
availability of leachate data obtained using appropriate test methods (i.e., SPLP or
ASTM). If a constituent was not detected in any samples, the constituent was removed
from further evaluation.
• Step 2: Comparison to Drinking water or Dermal criteria. SFS leachate data were
compared directly (i.e. undiluted) to the EPA water quality criteria, including Regional
Tapwater Screening Levels, Maximum Contaminant Levels (MCLs), and National
Secondary Drinking Water Standards (NSDWS). Water dermal exposure was evaluated
by comparing dermal absorbed doses to dermal benchmarks (i.e., oral benchmarks that
were adjusted using EPA gastrointestinal absorption factors). If a constituent
concentration was at or below the various drinking water criteria and the dermal absorbed
dose was at or below the dermal benchmark, the constituent was removed from further
evaluation.
COCs that were not removed through this initial two-step screen would be modeled under
Phase II of the analysis. A detailed description of the groundwater pathway analysis, including
inputs and results, is found in Chapter 4, Section 4.2.
Analysis Phase I: Inhalation Pathway
In the inhalation pathway, a resident living immediately downwind of a soil-blending
operation (either at the use site, or a commercial blending operation) was exposed to fugitive
dust released via windblown emissions from a storage pile, as well as emissions that occur as the
result of loading/unloading operations. As shown in Figure 3-4, a two-step process was used to
identify which SFS constituents, if any, would require further evaluation for the inhalation
pathway.
Step 1: Availability of health benchmarks. The
availability of inhalation benchmarks was
determined based on the Office of Solid Waste and
Emergency Response (OSWER) toxicity value
hierarchy (USEPA, 2003a). Because benchmarks
are required for the quantitative evaluation of
health effects, those constituents lacking inhalation
benchmarks were removed from further inhalation
evaluation.
Step 2: SCREENS Modeling. SCREENS was
used to estimate constituent-specific air
concentrations associated with loading/unloading
activities and windblown emissions. These modeled air concentrations were used to
calculate the allowable concentration for each constituent in SFS based on potential risk
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-11
OSWER Benchmark Hierarchy
EPA's Integrated Risk Information
System (IRIS; U.S. EPA, 2012)
EPA's Provisional Peer-Reviewed
Toxicity Values (PPRTVs)
Other toxicity values (e.g., California
Environmental Protection Agency
[CalEPA] chronic inhalation
Reference Exposure Levels [RELs]
and cancer potency factors [CalEPA,
2005]; Agency for Toxic Substances
and Disease Registry [ATSDR]
Minimum Risk Levels [MRLs]).
-------
Chapter 3.0 Problem Formulation
via the inhalation pathway. The allowable concentration of each constituent in SFS—the
health-based screening level for SFS—was intended to be protective of human health for
the inhalation pathway. If a constituent concentration was at or below the allowable
concentration in SFS, the constituent was removed from further evaluation.
COCs not removed based on available inhalation benchmarks and the SCREENS
simulation would be modeled under Phase II of the analysis. A detailed description of the Phase I
inhalation pathway analysis, including inputs and results, is found in Chapter 4, Section 4.3.
Analysis Phase I: Soil Pathway
In the home gardening scenario described in Section 3.1.4 and illustrated in Figure 3-3,
home gardeners (adults or children) could be exposed via two direct pathways and five indirect
pathways. Direct pathways included incidental ingestion and dermal contact with soil, and
indirect pathways included ingestion of exposed fruits (e.g., strawberries), protected fruits (e.g.,
oranges), exposed vegetables (e.g., lettuce), protected vegetables (e.g., corn), and root vegetables
(e.g., carrots). The home garden was assumed to supply a significant fraction of the home
gardener's produce diet. As shown in Figure 3-4, a three step process was used to identify SFS
constituents that may pose risk above levels of concern for the soil pathways.
• Step 1: Samples above detection limit. As discussed in Chapter 2, numerous SFS
samples were collected and analyzed. Analytes not identified in any sample were not
evaluated further.
• Step 2: Availability of Soil Screening Levels. EPA's Soil Screening Levels (SSLs) for
soil ingestion were available for a large number of SFS constituents. Constituents with
soil ingestion SSLs have EPA-approved ingestion benchmarks; therefore, those
constituents lacking SSLs, and lacking health benchmarks with which to derive SSLs,
were not evaluated further.
• Step 3: Soil SSL Comparison. For manufactured soils, concentrations of SFS
constituents remaining after Step 2 were compared to human and ecological SSLs. The
human health SSL was divided by a factor of 10 to account for Home Gardener indirect
exposure pathways (i.e., ingestion of home-grown produce) not already accounted for in
the SSL. If the constituent concentration was at or below the Adjusted SSL, Dermal-SSL,
and Eco-SSL, then the constituent was not evaluated further.
Detected COCs not removed based on soil screening levels would be modeled under
Phase II of the analysis. A more detailed description of the Phase I soil pathway analysis,
including inputs and results, is found in Chapter 4, Section 4.4.
3.2.2 Analysis Phase II: Risk Modeling
A national-scale evaluation needs to account for variability in conditions across the
country. The Phase II evaluation of SFS constituents used probabilistic modeling to account for
national-scale variability. Specifically, Phase II used a Monte Carlo approach to probabilistically
model site-specific conditions across the country. Monte Carlo simulation techniques are useful
when there is substantial variability in the data and probability distributions19 can be developed
19 A probability distribution for a parameter describes both the range of possible values and the likelihood of where
in the possible range any single value will be.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-12
-------
Chapter 3.0 Problem Formulation
for most or all of the input parameters. The Monte Carlo approach essentially performs a series
of many site-specific evaluations of randomly selected locations, using input parameter values
appropriate for each location. Taken together, the results of these many individual evaluations
comprise a distribution of results from across the country. This approach is particularly
appropriate for a risk analysis of soil-blending operations and home gardens that can be located
across a wide range of environmental conditions.
The Phase II modeling methodology for each pathway is briefly described below.
Chapter 5 includes additional introductory information on probabilistic modeling in general, as
well as more detailed descriptions of how it was applied to evaluate the home garden scenario.
Groundwater Pathway: EPA's Industrial Waste Management Model V2.0 (IWEM) and
EPA's Composite Model for Leachate Migration with Transformation Products (EPACMTP)
were used to evaluate risks from exposure to groundwater. Both models have undergone external
peer review, including the EPACMTP model having been subjected to peer review by the
Science Advisory Board (SAB). Modeling performed with each of these models is described
below.
Screening Modeling
IWEM provides a flexible basis for considering the potential leaching from SFS in
manufactured soils. Detailed information on this model can be found in the IWEM User's Guide
(U.S. EPA, 2002a) and Technical Background Document (U.S. EPA, 2002b).20 Some modeling
input parameter values (e.g., distance from the garden to the drinking water well) were chosen to
be conservative (i.e., protective of human health). When data were available, values for other
input parameters (e.g., depth to the water table) were chosen from distributions representing
variable conditions across the country. The remaining parameters used default values provided in
the IWEM User's Guide (U.S. EPA, 2002a).
Probabilistic modeling calculated groundwater concentrations at a hypothetical receptor
well located from 1 to 200 m from the edge of the garden. Using the 95th percentile SFS leachate
concentration for each of the COCs,21 the model estimated groundwater concentrations at the
receptor well. The model ran each leachate concentration 10,000 times, varying site conditions
based on user inputs. The 90th percentile groundwater well concentration for each constituent
was selected from the output distributions. Each constituent-specific concentration was then
compared to the lowest of the health benchmarks collected during Phase I (e.g., drinking water
MCLs). If the 90th percentile concentration estimate was at or below the benchmark, the leachate
concentration was considered protective.
If the 90th percentile concentration estimate from the IWEM model was above the
benchmark, more refined probabilistic groundwater modeling was performed using EPACMTP
and source model leachate concentrations.
Refined Modeling
Consistent with other EPA national-scale groundwater modeling assessments,
probabilistic groundwater modeling was performed using EPACMTP (U.S. EPA, 2003f,g,h;
20 Supporting documentation for IWEM, IWAIR, and EPACMTP can be found
http://www.epa.gov/waste/nonhaz/industrial/tools/index.htm
21 This analysis used the higher of the 95th percentile leachate concentrations found by either SPLP or the ASTM
leachate methods.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-13
-------
Chapter 3.0 Problem Formulation
1997a). As described in Section 5.3, the refined groundwater modeling was performed
concurrently with the probabilistic modeling of the soil pathways by using the source model
outputs (i.e., garden leachate fluxes and annual average leachate infiltration rates) as EPACMTP
model inputs. Coupling the groundwater and surface pathways in this way both addressed
environmental variability (e.g., local meteorological patterns, soil types) and ensured that the
groundwater pathway and surface pathway exposure estimates were based on the same
environmental conditions. Refined groundwater modeling placed the drinking water receptor
well 1 m from the edge of the garden in the centerline of the plume.
The probabilistic simulation produced distributions of risk for the adult and child
receptors, which reflect the variability in environmental setting. As described in Chapter 5,
these distributions were subsequently used to estimate protective target SFS concentrations based
on EPA's risk management criteria (e.g., HQ of 1). These target SFS concentrations represent
conservative estimates which, if the SFS were a component of manufactured soil, would result in
exposures (and risk) via groundwater pathway below the risk management criteria. A SFS
constituent concentration at or below the target concentration would be considered protective.
Please note that although the groundwater and soil pathways were evaluated concurrently,
separate target SFS concentrations were developed for each pathway based on analyses discussed
in Section 5.3.5 and Appendix J that indicate that these exposures will not occur within the
same timeframe.
A more detailed description of the Phase II groundwater pathway analysis is found in
Chapter 5, Sections 5.2 and 5.3.
Inhalation Pathway: The Phase I analysis found that no constituents required further
evaluation, and therefore no Phase II inhalation modeling took place. However, for
completeness, a description of the Phase II inhalation modeling methodology is included below.
EPA's Industrial Waste Air Model (IWAIR) would have been used to evaluate risks from
inhalation. IWAIR was developed to assist facility managers and regulatory agency staff in
evaluating inhalation risks for workers and residents in the vicinity of a management unit.
Detailed information on this model can be found in the IWAIR User's Guide (U.S. EPA, 2002c)
and Technical Background Document (U.S. EPA, 2002d). With a limited amount of blending
site-specific information (e.g., pile surface area and height, and constituent-specific emission
rates), IWAIR can estimate whether temporary storage piles of SFS and SFS-manufactured soils
might pose an unacceptable inhalation risk to human health. IWAIR default dispersion factors
address variability in environmental settings across the country. These dispersion factors were
developed based on dispersion modeling with the EPA's Industrial Source Complex - Short
Term (ISCST3). Modeling was performed for many separate scenarios designed to cover a broad
range of unit characteristics, including a range of storage pile surface areas and heights, 6
receptor distances from the unit and 60 meteorological stations, chosen to represent the different
climatic and geographical regions of the contiguous 48 states, Hawaii, Puerto Rico, and parts of
Alaska. The model would have been run thousands of times based on user inputs. The 90th
percentile air concentration for each constituent would be compared to human health benchmarks
identified under Phase I. If the 90th percentile concentration estimate was at or below the
benchmark value, the SFS concentration would be considered protective.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-14
-------
Chapter 3.0 Problem Formulation
Soil Pathway: The home-gardener scenario assumed that a raised garden received a
single "addition" of SFS-manufactured soil containing 50% SFS by weight,22 to a depth of 20 cm
(a typical tilling depth). Based on this composition, it was further assumed that the basic
properties and characteristics of the manufactured soil were similar to those of natural soil in the
area.
The risk modeling framework currently used by EPA to support the Part 503 biosolids
program was modified and adopted to evaluate soil pathway risks. This framework represents
variability in soil and meteorological conditions in areas that produce SFS, as well as variability
in consumption rates for fresh fruits and vegetables that are home grown. This risk modeling
framework was adapted to capture variability in environmental settings within the context of
"economic feasibility areas" for the use of SFS, defined as areas within 50 km of the foundry.23
Locations within these areas were selected at random; no locations outside of the economic
feasibility areas were included in the Monte Carlo simulations. The assumed application site and
rates were also modified from the Biosolids framework to reflect home gardening practices
rather than farming practices.
The probabilistic simulation produced distributions of risk/hazard for the adult and child
receptors, as well as for plants, soil invertebrates and small mammals, which reflect the
variability in conditions within the economic feasibility areas. As described in Chapter 5, these
distributions (and the groundwater pathway distributions discussed above) were developed using
an initial "unitized" soil concentration of 1 part per million (ppm) for each constituent. Based on
the model's linearity with respect to constituent concentration, the 90th percentile of each
constituent-specific unitized risk estimate was scaled to estimate a protective SFS-specific
screening level based on EPA's risk management criteria (e.g., HQ of 1). These SFS-specific
screening levels represent conservative estimates of the selected SFS constituent concentrations
which, if the SFS were used in manufactured soil, would be protective of human health and the
environment. An SFS constituent concentration at or below the target SFS screening level would
be considered protective.
A more detailed description of the Phase II soil pathway analysis is found in Chapter 5,
Section 5.3.
22 This is a conservative blend, as most manufactured soil blends would contain 5-10% SFS by weight. See Chapter
2 for more details on soil blend recipes.
23 SFS use areas are based on the ZIP codes of the membership of the American Foundry Society as of November
2007. Since we did not know a foundry's exact location within its ZIP Code area, we extended the ZIP Code
boundary out 50 km to establish the economic feasibility areas.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-15
-------
Chapter 3.0 Problem Formulation
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 3-16
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4. Analysis Phase I: Identification of COCs for Modeling
Chapter 3 described the three beneficial use scenarios for SFS under consideration in
this assessment, and presented conceptual models for the exposure pathways and receptors for
using SFS in roadway subbase, in blending operations, and in home gardening. As shown by the
conceptual models, the home gardener scenario includes all of the exposure pathways in the
other two scenarios. However, the blending scenario represents the highest potential inhalation
exposure among any of the three scenarios. Therefore, the assessment used the home gardening
scenario and the blending scenario to represent the exposure pathways that are most likely to
present an unacceptable risk to human health and the environment. By focusing attention on the
exposure pathways associated with manufactured soils that are potentially of greatest concern,
the assessment could confidently identify the COCs (Analysis Phase I) and model only those
COCs that might pose unacceptable risks to human health and the environment (Analysis Phase
II). This chapter describes the process used to select COCs for further modeling evaluation and,
by default, determine whether the exposure pathways are of concern.
4.1 Purpose
The primary purpose of the first phase of the analysis was to identify COCs for additional
analysis in the risk modeling phase. If all constituents screened out for a particular exposure
pathway, the potential risks for that pathway would no longer need to be evaluated using
probabilistic risk models. Because this phase was designed to perform a screening function, a
very conservative approach was used to ensure that an ample margin of safety was applied
before eliminating a constituent from further consideration. For example, leachate concentrations
were compared directly with EPA screening criteria for the protection of drinking water; this
assumes that there would be no attenuation or dilution of the leachate and no degradation of
organic compounds as they move through the subsurface to the drinking water well. Importantly,
the following pathway-specific high-end concentrations provided the basis for the various Phase
I analyses performed as described in this section:
• Groundwater pathway: 95th percentile leachate concentrations;
• Inhalation pathway: 95th percentile SFS constituent concentrations;
• Soil pathway: Manufactured soil concentrations (ConcMs) reflecting a soil/SFS mixture
that contained SFS with 95th percentile constituent concentrations.
As seen in the conceptual models for SFS-manufactured soils (see Figures 3-2 and 3-3),
there are three basic media-specific exposure pathways to be evaluated: (1) groundwater
pathway - the ingestion of, and dermal contact with, groundwater contaminated by the leaching
of SFS constituents; (2) ambient air pathway - the inhalation of SFS emitted from soil blending
operations; and (3) soil pathway - dermal contact with, and incidental ingestion of soil, as well as
ingestion of fruits and vegetables grown in the SFS-manufactured soil. Although some
constituents, such as manganese elicit similar toxicological responses (e.g., neurotoxicity) via
different exposure pathways, neither the screening nor the modeling stages of the analysis
considered cumulative exposures across these three pathways. Rather, the exposure scenarios
and pathway evaluations were developed and parameterized to produce conservative risk
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-1
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
estimates. The risk assessment is therefore an efficient approach to providing decision makers
with information on the potential for adverse effects to the most highly exposed individuals and
ecological receptors that could come in contact with SFS constituents.
4.2 Groundwater Exposure
Given the use of SFS-manufactured soil in a home garden, leaching to groundwater is a
potential pathway of concern. Under this pathway, residents could be exposed to SFS
constituents through the ingestion of contaminated drinking water or through dermal contact
while bathing. Thus, this section: (1) examines the potential for SFS to leach constituents of
potential concern; (2) evaluates drinking water ingestion exposure by comparing leachate data to
regulatory levels and screening criteria developed to protect water use; and (3) evaluates water
dermal exposure by comparing dermal absorbed doses to oral benchmarks adjusted using EPA
gastrointestinal absorption factors. If a constituent concentration exceeded one of the drinking
water criteria or if a dermal absorbed dose exceeded the adjusted oral benchmark, the constituent
was flagged for further evaluation under Phase II.
4.2.1 Leachate Data
The first step in the groundwater analysis was to examine the teachability of SFS
constituents. As discussed in Chapter 2, Dungan and Dees (2009) used the TCLP, SPLP and
ASTM methods to estimate the leaching potential of metals from ferrous and aluminum foundry
SFSs. The TCLP method, however, was designed to predict leaching potential under conditions
very different from SFS use in manufactured soil or other soil-related applications (see Chapter
2, Section 2.5.4 for a more detailed discussion of the relevance of TCLP data to SFS soil-related
applications). Therefore, the conditions reproduced by TCLP are not relevant to the SFS uses
evaluated in this assessment.
The SPLP method was designed to mimic leaching from soil due to acid rain conditions,
and the ASTM method tests leaching potential at a material's "natural" pH. The conditions
reproduced by the SPLP and ASTM methods are more relevant than TCLP for characterizing
SFS leaching potential under the conditions evaluated in this report. This part of the evaluation
therefore only used SPLP or ASTM leach data.
Table 4-1 presents a summary of the SPLP and ASTM leachate data for the 39 silica-
based iron, steel, and aluminum SFSs.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-2
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-1. Leaching Data for Silica-based Iron, Steel, and Aluminum SFSs (mg I/1)
Metal
Ag
As
Ba
Be
Cd
Cr(III)
Cu
Ni
Pb
Sb
Zn
SPLP
Maximum
0.08
0.098
0.63
0.02
0.01
O.01
0.21
0.24
O.08
0.04
0.18
95%-ile
N/A
0.017
0.37
N/A
N/A
N/A
N/A
0.025
N/A
N/A
N/A
Median
N/A
0.004
0.12
N/A
N/A
N/A
N/A
0.025
N/A
N/A
N/A
ASTM
Maximum
0.05
0.024
O.24
0.01
0.01
O.02
1.1
0.046
O.ll
0.04
0.22
95%-ile
N/A
0.018
N/A
N/A
N/A
N/A
0.04
0.025
N/A
N/A
N/A
Median
N/A
0.005
N/A
N/A
N/A
N/A
0.04
0.025
N/A
N/A
N/A
Data from Dungan (2008) and Dungan and Dees (2009), all three sampling events of 39 silica-based iron, steel,
and aluminum SFSs.
4.2.2 Selection of Constituents
Because leachate data for only 11 constituents (i.e., antimony, arsenic, barium, beryllium,
cadmium, chromium, copper, lead, nickel, silver, and zinc) are available from Dungan and Dees
(2009), these were the constituents of potential concern that were evaluated. A limitation of this
data set is that for some constituents, the analytical detection limits were higher than the
screening levels (or regulatory levels) to which they were being compared. In addition, this
leachate analysis did not include mercury and selenium. Therefore, mercury and selenium were
not evaluated quantitatively. However, the leaching potential of mercury and selenium from
SFSs is discussed below.
4.2.3 Comparisons to Screening Levels and Regulatory Levels
To evaluate drinking water ingestion exposures, several risk levels were available for
comparison to SFS leachate data. EPA's Superfund program developed Tapwater Screening
Levels to be protective at 1E-06 cancer level24 and an HQ of 1 for noncancer risk levels. EPA
has also developed National Drinking Water Regulations. These include primary standards such
as Maximum Contaminant Limits (MCLs), as well as secondary standards. Table 4-2 provides
the comparison of SFS leachate concentrations to all three screening and regulatory levels.
24 This cancer risk target is an order of magnitude lower than the risk target level that the EPA Office of Resource
Conservation and Recovery typically uses in risk assessments. As mentioned elsewhere in this report, this
evaluation used a risk target of 1E-05 for cancer.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-3
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-2. Leachate Comparisons (mg I/1)
Metal
Ag
As
Ba
Be
Cd
Cr(III)
Cu
Ni
Pb
Sb
Zn
SFS 95%-ile a
SPLP
0.08
0.017
0.37
0.02
0.01
O.01
0.21
0.025
O.08
0.04
0.18
ASTM
0.05
0.018
O.24
0.01
0.01
O.02
0.040
0.025
O.ll
0.04
0.22
Screening and Regulatory Levels
Tapwater SLb
0.094
0.000526
3.8
0.025
0.0092
22
0.8
0.39
N/A
0.0078
6.0
MCLC
N/A
0.01
2.0
0.004
0.005
O.lf
1.3
N/A
0.015
0.006
N/A
NSDWSd
0.1
N/A
N/A
N/A
N/A
N/A
1.0
N/A
N/A
N/A
5.0
a Data from Table 4-1
b Tapwater Screening Levels can be found at http://www.epa.gov/reg3hwmd/risk/human/rb-
concentration table/Generic Tables/index.htm
0 MCLs are primary drinking water standards that can be found at
http ://water. epa. gov/drink/contaminants/index. cfm#Primary
d NSDWSs can be found at http://water.epa.gov/drink/contaminants/index.cfm#Secondary
e To be consistent with other ORCR risk assessments, the listed Tapwater SL for arsenic
represents the Regional Tapwater SL converted to a 10~5 risk level
f Based on total Cr
To examine the potential for groundwater dermal exposure, the evaluation performed a
screening level dermal assessment based on guidance provided in EPA's Risk Assessment
Guidance for Superfund Volume I: Human Health Evaluation Manual (Part E, Supplemental
Guidance for Dermal Risk Assessment) (U.S. EPA; 2004). The assessment evaluated the SFS
COCs identified in Section 4.2.2 using a three step process:
1. Identify COCs for quantitative analysis: Constituents for quantitative analysis were
identified using the RAGs Part E Screening Tables, which flag chemicals where the
dermal pathway has been estimated to contribute more than 10% of the oral pathway,
using conservative residential exposure criteria. The screening tables reflect the
comparison of two main household daily uses of water: as a source for drinking and for
showering or bathing. This step determined that beryllium, cadmium, chromium (III), and
zinc should be quantitatively evaluated for dermal exposure.25
2. Calculate dermal absorbed dose (DAD): Adult and child-specific DADs were calculated
for beryllium, cadmium, chromium(III), and zinc using the reasonable maximum
exposure (RME) scenario for residential settings as defined in U.S. EPA (2004). For the
home garden use of SFS-manufactured soil scenario, the evaluation assumed that the
adult and child showered or bathed with groundwater concentrations equivalent to
25 Lead was not included in U.S. EPA (2004) and sufficient data were not available to quantitatively assess dermal exposures for
this constituent. However, the U.S. EPA notes that cutaneous absorption is generally not a significant route of exposure for
inorganic lead (http://www.epa.gov/superfund/lead/ahnfaq.htmtfdermal).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-4
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
leachate concentrations. As discussed in Section 4.2.1, leachate data were available from
both the SPLP and ASTM leachate methods. In this assessment, the higher of the two
leachate values were used to calculate the DADs. These calculations were performed
using the Inorganic Chemicals in Water spreadsheet available from U.S. EPA (2004),
and using exposure parameter values shown in Table 4-3.
Table 4-3. Recommended Dermal Exposure Parameters
for RME Residential Scenario
Exposure Parameters
Event Frequency (events d"1)
Exposure Frequency (days yr1)
Averaging Time (d)
Event Duration (h event"1)
Exposure Duration (yr)
Skin Surface Area (cm2)
Body Weight (kg)
Showering/ Bathing
1
350
25,550
Adult
0.58
30
18,000
70
Child
1.0
6
6,600
15
3. Compare DAD to dermal criterion: The resulting DAD estimates were then used to
calculate constituent-specific HQs. Methods for estimating dermal risk are based on
absorbed dose - the fraction of administered dose that is absorbed into the body.
However, oral benchmarks such as RfDs and Slope Factors are typically based on
administered dose. Use of oral benchmarks to estimate dermal risk required the
adjustment of oral benchmarks using gastrointestinal absorption factors (ABSoi). In
accordance with U.S. EPA (2004), the oral reference dose (RfD) for noncarcinogens was
multiplied by the constituent-specific ABSoi to estimate a reference dose based on
absorbed dose (RfDABs). The DAD estimates were then divided by the RfDABss to
calculate the constituent-specific hazard quotients. As seen in Table 4-4, the dermal
hazard quotients were all below a level of concern (i.e., HQ = 1).
Table 4-4. Comparison of Water Dermal Absorbed Doses (DADs) to Health Benchmarks
Constituent
Be
Cd
Cr(III)
Zn
SFS 95%-ile
Concentration
SPLP
(mg I/1)
O.02
<0.01
0.01
0.18
ASTM
(mg L-1)
O.01
O.01
0.02
0.22
Benchmark
Oral RfD
(mgkg-'d-1)
2.0E-03
5.0E-04b
1.5
0.3
RfDABs"
(mgkg-'d-1)
1.4E-05
1.3E-05
2.0E-02
0.3
DAD
Adult DAD
(mg kg-'d-1 )
1.2E-06
6.2E-07
1.2E-06
8.1E-06
Child DAD
(mg kg-'d-1)
7.2E-07
3.6E-07
7.2E-07
4.8E-06
Dermal Hazard
HQ
Adult
8.6E-02
4.8E-02
6.0E-05
2.7E-05
HQ
Child
5.1E-02
2.8E-02
3.6E-05
1.6E-05
a U.S. EPA (2004) presents gastrointestinal absorption efficiencies for beryllium (0.7%), cadmium (2.5%), and
chromium (III) (1.3%), and recommends an efficiency of 100% for zinc in the absence of a reported value.
b Oral RfD (water)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-5
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.2.4 Results
Only samples of arsenic had detectable leachate levels that exceeded the screening or
regulatory levels for drinking water. That is, using the SPLP and ASTM leachate methods,
several SFSs exceeded the MCL for arsenic (0.01 mg L"1). In addition, the detection limit for
arsenic (0.001 mg L"1) was above the Tapwater Screening Level (0.00045 mg L"1).
Analyses for the remaining constituents showed no samples that exceeded the screening
or regulatory levels for drinking water. However, while all leachate samples of antimony,
beryllium, cadmium, and lead were below their respective detection limits, the detection limits
were higher than their respective MCLs. The detection limit for antimony also exceeded its
Tapwater Screening Level.
Results from the water dermal screening assessment indicated that none of the
constituents needed to be further evaluated for groundwater dermal exposure. As seen in
Table 4-4, the dermal hazard quotients were all below a level of concern (i.e., HQ = 1).
With respect to mercury and selenium leachate concentrations, they are also not expected
to exceed their regulatory levels based on the following considerations. In a study conducted by
Fahnline and Regan (1995), the maximum concentrations of mercury and selenium in TCLP
extracts from 50 spent foundry molding sands (from foundries of unknown type) were <0.10 mg
L"1 and <0.83 mg L"1, respectively. These TCLP data are being used here because no SPLP or
ASTM data are available. Also, the TCLP method is likely more aggressive than either the SPLP
or ASTM method when testing SFS (see Chapter 2, Section 2.5.4, for TCLP, SPLP and ASTM
leaching results), such that actual leachate concentrations are unlikely to be greater than those
listed in Fahnline and Regan (1995). Also, with respect to selenium, even if one assumes
complete leaching of all selenium in the 39 SFSs considered (see Appendix B), no sand would
exceed the regulatory level of 1.0 mg L"1.
Therefore, as a result of the high detection limits for some constituents, and the
exceedances of arsenic described above, the following constituents were retained for Phase II
risk modeling (see Chapter 5):
• Antimony
• Arsenic
• Beryllium
• Cadmium
• Lead.
All remaining constituents were screened out from the groundwater pathway and were
not retained for Phase II modeling.
4.3 Inhalation Exposure
As discussed earlier, SFS can replace mined sand as a mineral component of
manufactured soil. It is probable that during storage and mixing, some components of the SFS
(e.g., clays) will be emitted into the air and migrate offsite as fugitive dust. Therefore, as shown
in the blending site conceptual model (Figure 3-2), nearby residents could be exposed to SFS
constituents through the inhalation of this fugitive dust. Manufactured soils can be blended at the
site where they will be used, or at a separate commercial blending facility. Residents living near
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-6
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
a commercial blending facility would likely be exposed to fugitive dust for longer durations
(potentially years) than those living near a site where the soil was blended once and then applied
to the land. Activities at a soil-blending facility are also likely to result in higher emission rates
and higher potential exposure frequencies than would be expected from gardening activities.
This assessment therefore evaluated residential inhalation exposures to fugitive emissions from a
soil-blending facility.
4.3.1 Scenario
In this scenario SFS is loaded and unloaded from a storage pile at an active soil blending
facility. Soil blending involves using construction equipment, such as a front-end loader, to
combine large volumes of the various mineral and organic components. The blending site was
assumed to blend SFS-manufactured soil year-round. Some of the information used to develop
the exposure scenario was based on the only commercial soil blender that currently uses SFS in
soil-blending operations (Bailey, 2007); specifically,
• The amount of SFS managed
• The size of the SFS storage pile
• The distance from the site to the nearest residence.
Within the soil-blending industry this facility is considered quite large. Use of
information from this facility (e.g., size of the SFS storage pile) is therefore considered a
conservative assumption.
4.3.2 Selection of Constituents of Potential Concern
Constituents were chosen to undergo screening based on the availability of human health
benchmarks for inhalation. Because benchmarks are required for the quantitative evaluation of
health effects, those without benchmarks were not evaluated here. Cancer and noncancer
benchmarks were chosen based on the Office of Solid Waste and Emergency Response
(OSWER) toxicity value hierarchy.26 Table 4-5 provides the health benchmarks used to calculate
the screening criteria for inhalation. The benchmarks in Table 4-5 are based on chronic exposure,
24 h d"1, 365 d yr"1. All 14 of the SFS constituents with inhalation exposure benchmarks (listed in
Table 4-5) were screened.
26 The hierarchy is listed in the 2003 OSWER Directive 9285.7-53. This directive can be found at
http://www.epa.gov/oswer/riskassessment/pdf/hhmemo.pdf.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-7
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-5. Inhalation Human Health Benchmarks
Constituent
Concentration
(mg m 3)
Non-cancer target organ/ toxicological
endpoint
Carcinogenic
Arsenic a
Benz[a]anthracene h
Benzo[b]fluoranthene h
Benzofkjfluoranthene h
Benzo[a]pyrene h
Beryllium a
Cadmium3
Chrysene h
Dibenz[a,h]anthracene h
Indeno[l,2,3-cd]pyrene h
Naphthalene °
Pentachlorophenol
2,3,7,8-TCDD d-f
2,4,6-Trichlorophenol a
2E-06
2E-04
2E-04
2E-04
2E-05
4E-06
6E-06
1.1E-05
2E-05
2E-04
3E-03
5E-03
1E-09
3E-02
--
--
--
--
--
--
--
--
--
--
--
--
--
--
Noncarcinogenic
Aluminum b
Barium g
Boron g
Cobalt d
Manganese °
2-Methylphenol e
3- and 4-Methylphenol e
Nickel6
Phenol e
Selenium e
5E-03
5E-04
2E-02
1E-04
5E-05
6E-01
6E-01
5E-05
2E-01
2E-02
Neurological
Fetotoxicity
Respiratory system
Respiratory system
Impaired neurobehavioral function
nervous system
nervous system
Respiratory system, hematologic system
Liver, cardiovascular system, kidney, nervous
system
Liver, cardiovascular system, nervous
system
a Source: IRIS - Air concentration that would elicit a carcinogenic risk estimate of 1E-05 (U.S. EPA, 2012a)
b Source: PPRTVs - RfC for chronic inhalation exposure (U.S. EPA, 2006)
c Source: IRIS - RfC (U.S. EPA, 2012a)
d Source: ATSDR - MRL (ATSDR, 2007)
e Source: CalEPA - REL (CalEPA, 2005)
f 2,3,7,8-TCDD is used as the benchmark for the toxicity equivalent of all dioxins, furans, and dioxin-like
PCBs
« Source: Health Effects Summary Table (HEAST, U.S. EPA, 1997b)
h Source: CalEPA - Inhalation Unit Risk (CalEPA, 2009) used in the methodology for generating Regional
Screening Levels (the User's Guide is available at http://www.epa.gov/reg3hwmd/risk/human/rb-
concentration table/index.htm) to estimate an air concentration that would elicit a carcinogenic risk
estimate of 1E-05
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-8
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.3.3 Deterministic Modeling
To perform a screening assessment for the inhalation pathway, it was necessary to
determine whether residents living near the soil-blending site could be exposed via inhalation at
levels above the benchmarks in Table 4-5. In general, air exposure for a particular constituent
would be the concentration of that constituent in the fugitive dust multiplied by the concentration
of fugitive dust in the air:
Exposure = [X\* [FD] x 10"6
Where:
Exposure = Exposure to the constituent (mg m"3)
[X] = Concentration of the constituent in fugitive dust (mg kg"1)
[FD] = Concentration of fugitive dust in the air (mg m"3)
10~6 = Conversion factor from mg to kg (kg mg"1).
The SCREENS model (U.S. EPA, 1995b) was used to estimate the concentration of
fugitive dust in the air near a soil-blending site.27 SCREENS (a screening version of ISC3) is a
single source Gaussian plume model that provides maximum ground-level concentrations for
point, area, flare, and volume sources. It was developed to provide an easy-to-use method of
obtaining pollutant concentration estimates based on Screening Procedures for Estimating the
Air Quality Impact of Stationary Sources (U.S. EPA, 1992b). SCREENS outputs were used in
conjunction with the health benchmarks in Table 4-5 to calculate screening levels for each
constituent, as follows:
[FD]
Where:
SL = Screening level (mg constituent kg"1 fugitive dust)
[HB] = Health benchmark (mg m"3)
[FD] = Concentration of fugitive dust in the air (mg m"3)
106 = Conversion factor from mg to kg (mg kg"1).
The inhalation pathway was evaluated by comparing the calculated screening level for
each constituent to the 95th percentile concentration of the constituent in SFS. If the 95th
percentile concentrations are less than the screening level concentrations, it is reasonable to
assume that the inhalation pathway, when taken in isolation, does not pose risks requiring further
analysis and modeling, for the following reasons:
• The health benchmarks used to calculate the screening level are based on the worst-case
exposure duration and frequency of 24 h d"1, 365 d yr"1
• The health benchmarks are protective of the general population and sensitive
subpopulations
• The SCREENS model was implemented based on guidance provided in Section 4. 1 .2 of
the Workbook of Screening Techniques for Assessing Impacts of Toxic Air Pollutants
27 SCREENS is publicly available at http://www.epa. gov/scramOO I/dispersion screening.htm.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-9
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
(U.S. EPA, 1992a) for continuous fugitive/windblown dust emissions. Modeling options
were selected to examine the full range of meteorological conditions and wind directions
to ensure that the highest maximum concentrations were identified. Outputs from this
model are short-term, maximum 1-hour air concentrations. These short-term
concentrations were then combined with chronic health benchmarks to develop
conservative screening levels. Input parameters for the model (described in the following
subsections), including emission factors, were selected to increase potential exposure,
and
• The 95th percentile concentrations of constituents in SFS were used instead of median
concentrations.
4.3.3.1 Emission Factors
To model the concentration of the SFS in the air, it was necessary to estimate the
emission rate for the SFS managed in the soil-blending scenario. Two emission factors were
calculated and converted into emission rates: one for loading and unloading the sand onto and off
of the storage pile, and the other for windblown emissions. The loading/unloading emission
factor was based on AP-42 (Compilation of Air Pollutant Emission Factors) Section 13.2.4
"Aggregate Handling and Storage Piles" (U.S. EPA, 1995a):
Where:
E = Emission factor (kg Mg"1)
k = Particle size multiplier (dimensionless)
U = Mean wind speed (m s"1)
M = Material moisture content (%).
Information from U.S. EPA (1995a) was used to determine the values for k and U. For k,
0.35 was chosen based on an aerodynamic particle size of <10 |j,m (i.e., clay- and silt-sized
fractions). AP-42, Section 13.2.4, reports a range of wind speeds for calculating parti culate
emissions by batch or continuous drop operations as 0.6-6.7 (m s"1), and 5.4 m s"1 was selected
to serve as the high-end wind speed to be consistent with wind conditions used to calculate
windblown particulate emissions from a storage pile. The material moisture content of 3% was
based on Table 1 in Foundry Sand Facts for Civil Engineers (FIRST, 2004), assuming that the
foundry sand contains some clay-sized particles. The calculated emission factor for
loading/unloading was 1.02E-03 kgMg"1.
Approximately 86,450 tons (78,410 Mg) per year of SFS is used at the active soil-
blending site described in this assessment (Bailey, 2007). Based on the mass of sand managed
per year, the area of the storage pile (150 m2), and the assumption that the sand is being
loaded/unloaded 4 h d"1, 260 d yr"1, the calculated emission factor (1.02E-03 kg Mg"1) was
converted to an emission rate of 1.42E-04 g s"1 m"2.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-10
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
The windblown emission factor was calculated using the equation for "Continuous
Fugitive/Windblown Dust Emissions" (U.S. EPA, 1992a):
Where:
E = Emission factor (kg d"1 ha"1)
s = Material silt content (%)
p = Number of days per year with more than 25 mm of precipitation (dimensionless)
w = Percent of time wind speed exceeds 5.4 m s"1 (%).
The material silt content of 12% was based on particle size analysis of the 39 samples of
silica-based SFS from iron, steel and aluminum foundries (see Chapter 2 Section 2.5.1, and
Appendix B Table B-25). The default values in U.S. EPA (1992a) of 0 for p and 20% for w
were used in calculating this emission factor. The result (31.5 kg d"1 ha"1) was converted to g s"1
m"2, with a final emission rate of 3.64E-05 g s"1 m"2.
4.3.3.2 Other Input Parameters for SCREENS
In addition to the emission rates, SCREENS also required the following input parameters:
• Source Type: An area source was chosen because the emissions would be coming off of
a storage pile and not from a smokestack or other point source
• Length, Width, and Height of Storage Pile: 15m, 10m, and 4 m were chosen based on
an aerial photograph of the only currently operating facility that uses foundry sand in soil
blending operations (Bailey, 2007). Within the soil-blending industry this facility is
considered quite large.
• Receptor Height: 0 m was chosen to be protective of a child or infant receptor close to
the ground
• Urban or Rural: Rural was chosen because it is more conservative than the urban option
and based on the location of the blending operation in the aerial photograph referenced
above
• Search for Maximum Direction: A positive response was chosen as a conservative
assumption so that the maximum air concentration would be located.
SCREENS requires the user to specify the modeling area, defined as the region between
two distances from the source, within which to estimate maximum concentrations. For this study,
the modeling area was defined as the region from 0 to 1,000 m from the source to ensure that the
maximum concentration of airborne SFS would be included in the range. SCREENS gives the
user the option to specify "discrete" distances, which are specific distances from the source at
which to identify maximum concentrations. Because the distance to the nearest resident was
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-11
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
estimated to be 500 m, based on the aerial photograph referenced above, SCREENS calculated
the concentration 500 m away.28 Table 4-6 summarizes the SCREENS input parameters used.
Table 4-6. Input Parameters for SCREENS
Parameter Description
Source type
Emission rate (g s"1 m"2)
Height of storage pile (m)
Length of storage pile (m)
Width of storage pile (m)
Receptor height (m)
Urban or rural
Search for maximum direction
Choice of meteorology
Automated distance array
Minimum distance (m)
Maximum distance (m)
Use discrete distances
Distance (m)
Scenario
Loading and Unloading
Area
1.42E-04
4
15
10
0
Rural
Yes
Full
Yes
0
1,000
Yes
500
Windblown Erosion
Area
3.64E-05
4
15
10
0
Rural
Yes
Full
Yes
0
1,000
Yes
500
4.3.3.3 SCREENS Outputs
Using the inputs listed in Table 4-6, SCREENS estimated the concentration of SFS in the
air at ground level under both the loading/unloading and windblown erosion scenarios. Table 4-7
shows both outputs from SCREENS at a distance of 500 m. In addition, the estimated
concentrations for these two scenarios were summed to provide a total concentration that a
receptor might be exposed to. This calculated total concentration was 49.7 jig m"3.
Table 4-7. SCREENS Output Summary
Parameter Description
Concentration at 500 m (ug m"3)
Scenario
Loading and
Unloading
39.6
Windblown
Erosion
10.2
All Scenarios
(Sum Total)
49.7
28 While the assumption of a 500 m distance to the nearest residence is based on empirical evidence, it may not be a
conservative assumption. However, a preliminary analysis found that reducing the distance to 100 m would not
change the Phase I results: all modeled constituents would pass the screen, and therefore no constituents would
require Phase II evaluation.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-12
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.3.4 Results
Neither the loading and unloading scenario nor the windblown erosion scenario estimated
levels of particulates higher than the Primary National Ambient Air Quality Standard (NAAQS)
for coarse inhalable parti culates29 (150 jig m"3). The combined concentration of both scenarios
also fell below the NAAQS for coarse inhalable particulates. However, even when the particulate
levels do not exceed their primary air standard, it is still possible that one or more constituents in
the fugitive dust could exceed chemical-specific, health-based target levels (see Table 4-5).
As described above, conservative screening concentrations were calculated for each of
the constituents in Table 4-5 by dividing the health benchmarks by the total SFS air
concentration listed in Table 4-7. Exposure was assumed to be at the total concentration 24 h d"1,
365 d yr"1. Table 4-8 shows the actual 95th percentile concentrations of constituents in SFS and
the calculated conservative screening concentrations for the inhalation pathway.
Table 4-8. Comparison to Screening Values: Inhalation Pathway
SFS Constituent a'b
SFS 95%-ile
(mgkg1)
Calculated Screening
Concentration (mg kg -1)
Carcinogens
Arsenic
Benz[a]anthracene
Benzo [b]fluoranthene
Benzo [kjfluoranthene
Benzo [a]pyrene
Beryllium
Cadmium
Chrysene
Dibenz[a,h]anthracene
Indeno [1,2,3 -cd]pyrene
Naphthalene
Pentachlorophenol
2,3,7,8-TCDDTEQd
2,4,6-Trichlorophenol
6.44
0.13
0.06 c
0.07 c
0.10C
0.38
0.20
0.04
0.08
0.07 c
3.45
0.12
3.13E-6
0.06
40.2
4,020
4,020
4,020
402
80.4
121
221
402
4,020
60,300
100,500
0.0201
603,000
Noncarcinogens
Aluminum
Barium
11,200
17.7
100,500
10,060
29 A standard for particulate matter with a mean aerodynamic diameter of 10 microns or less (PMio)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-13
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
SFS Constituent a'b
Boron
Cobalt
Manganese
2-Methylphenol
3- and 4-Methylphenol
Nickel
Phenol
Selenium
SFS 95%-ile
(mg kg-1)
20.2
5.99
670
8.74
3.41
102
20.2
0.20
Calculated Screening
Concentration (mg kg -1)
402,000
2,010
1,005
Capped
Capped
1,005
Capped
402,000
a Constituent data from silica-based iron, steel, and aluminum sands (Dayton et al., 2010)
b PAH and phenolic data from Dungan (2006,2008)
c Not detected in any samples. Value represents one half the detection limit
d Due to the small size of the dataset for dioxins and dioxin-like compounds, the maximum value for
2,3,7,8 TCDD TEQ was used, rather than the 95th percentile.
Capped = Screening modeling estimates indicated risks below levels of concern at concentrations above
1E06 mg kg"1 (i.e., SFS could be comprised entirely of this constituent and still not cause risk).
None of the constituent concentrations in SFS exceeded their respective screening levels.
Therefore, no SFS constituents required further evaluation and Phase II risk modeling for the
inhalation pathway was not performed.
4.4 Soil Pathways Exposure
When SFS-manufactured soil is used in a home garden, potential exposure pathways
include incidental ingestion of soils, dermal contact with soils, and the ingestion of produce
grown in the home garden. The three-step process used to identify COCs for the soil pathways
included the following:
1. Remove SFS constituents that were not detected in any samples
2. Remove SFS constituents with no human health benchmarks
3. Remove SFS constituents by comparing the constituent concentrations to (a) adjusted
SSLs for the ingestion pathways (use of adjusted SSLs is discussed in Section 4.4.3), (b)
DermalSSLs for soil dermal exposure, and (c) Eco-SSLs.
Although Dungan and Dees (2009) examined total metals, data from Dayton et al. (2010)
were used because their analytical methods had lower detection limits. Data from Dungan and
Dees (2009) were used to screen PAHs and phenolics, and data from Dungan et al. (2009) were
used to screen dioxins and dioxin-like compounds.
It is also important to note that different categories of semi-volatiles were handled
differently. Specifically, PAHs were each dealt with individually, while dioxins and dioxin-like
compounds were dealt with in terms of their toxic equivalence values (TEQs - which estimate
toxicity relative to 2,3,7,8-TCDD). Evaluation of dioxins and dioxin-like compounds in terms of
their TEQ is an accepted approach that the Agency often uses. Therefore, from this point forward
all dioxin-like compounds will be represented by an aggregated toxicity equivalent, or 2,3,7,8-
TCDD TEQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-14
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.4.1 Remove SFS Constituents that are Nondetects
Although SFS samples were analyzed for numerous constituents of potential concern, not
all analytes were necessarily detected in the samples. Therefore, constituents of potential concern
that were not identified in any sample were not retained for further evaluation. Table 4-9 lists all
constituents of potential concern, identifying those that were not detected in any sample.
As shown in Table 4-9, all metals were detected in at least one sample, and were
therefore retained for further screening. Of the PAHs, benzo[b]fluoranthene,
benzo[k]fluoranthene, benzo[g,h,i]perylene, benzo[a]pyrene, and indeno[l,2,3-cd]pyrene were
not detected in any of the samples and were dropped from further study. Most phenolics also
were not detected in any of the samples and were also dropped from further study. Only 4-
chloro-3-dinitrophenol, 2,4-dichlorophenol, 2,4-dinitrophenol, 2-methylphenol, 3- and 4-
methylphenol, and phenol were detected in at least one sample, and were therefore retained for
further screening. Finally, 1,2,3,7,8,9-HCDF was not detected in any of the samples, and was
therefore dropped from further study; all other dioxins and dioxin-like compounds were retained
for further screening.
Table 4-9. Constituents Detected in at Least One Sample
Constituent
Al
As
B
Ba
Be
Ca
Cd
Co
Cr
Cu
Fe
K
Mg
Mn
Mo
Na
Ni
P
Pb
S
Detect
^=Yes
x=No
^
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
Constituent
Sb
Se
Tl
V
Zn
Acenaphthene
Acenaphthylene
Anthracene
Benz[a]anthracene
Benzo [b]fluoranthene
Benzo [k]fluoranthene
Benzo [g,h,i]perylene
Benzo [a]pyrene
2-sec-Butyl-4,6-dinitrophenol
Chrysene
4-Chloro-3 -methylphenol
2-Chlorophenol
Dibenz[a,h]anthracene
2,4-Dichlorophenol
2,6-Dichlorophenol
Detect
^=Yes
x=No
•/
S
S
S
V
V
•/
S
S
X
X
X
X
X
^
s
X
•/
X
X
Constituent
2,4-Dimethylphenol
2,4-Dinitrophenol
Fluoranthene
Fluorene
Indeno [ 1 ,2,3 -cd]pyrene
2-Methylphenol
3- and 4-Methylphenol
2-Methyl-4,6-dinitrophenol
Naphthalene
2-Nitrophenol
4-Nitrophenol
Pentachlorophenol
Phenanthrene
Phenol
Pyrene
2,3,7,8-TCDD TEQ
2,3,4,6-Tetrachlorophenol
2,4,6-Trichlorophenol
2,4,5-Trichlorophenol
Detect
^=Yes
x=No
^
X
^
S
X
^
•/
X
S
X
X
X
•/
S
S
S*
X
X
X
a All dioxin-like compounds except for 1,2,3,7,8,9-HxCDF were detected.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-15
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.4.2 Remove SFS Constituents without Benchmarks
Health benchmarks are required to quantify potential health risks, and the screening
criteria developed by EPA require an EPA-approved health benchmark. The SSLs developed by
EPA (U.S. EPA, 1996) to be protective of the soil ingestion pathway are based on EPA-approved
health benchmarks, as well as conservative exposure assumptions. Table 4-10 lists SSLs for
constituents of potential concern or indicates that no benchmark exists for generating SSLs.30
Of the constituents of potential concern remaining after the first step, there were no
health benchmarks for calcium, magnesium, phosphorus, potassium, sodium, and sulfur, all six
of which are also essential plant nutrients. Therefore, these constituents were removed from
further quantitative evaluation. Eighteen metals, 9 PAHs, 20 dioxins and dioxin-like compounds,
and 5 phenolics remained after the first two steps in the screening process for soil pathways.
30 SSLs are not national cleanup standards, nor do they define "unacceptable" levels of contaminants in soil. They
were designed as tools for the Superfund program to quickly identify sites that no longer need federal attention.
Because of this, soil concentrations above SSLs do not in and of themselves denote a problem, only that further
study may be warranted. More information on SSLs can be found at http://rais.ornl.gov/calc start.shtml.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-16
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-10. Residential Soil Screening Levels (mg kg"1)
-l\a
Analyte
Aluminum
Arsenic
Antimony
Barium
Beryllium
Boron
Cadmium
Calcium
Chromium(III)
Cobalt
Copper
Iron
Leadc
Magnesium
Manganese
Molybdenum
Nickel
Phosphorus
Potassium
Selenium
Sodium
Sulfur
Thallium
Vanadium
Zinc
Acenaphthene
Acenaphthylene
Anthracene
Benz [a] anthracene
Chrysene
4-Chloro-3 -methylphenol
Dibenz[a,h]anthracene
2,4-Dimethylphenol
Fluoranthene
Fluorene
2-Methylphenol
3- and 4-Methylphenol
Naphthalene
Phenanthrene
Phenol
Pyrene
2,3,7,8 TCDD (TEQ)
Carcinogenic SSL b
Pathways included in the
Screening Level
Ingestion
X
Dermal
X
Inhalation
X
X
X
Resi-
dential
SSL
N/A
6.7E+00
N/A
N/A
1.6E+03
N/A
2.1E+03
No Benchmark
X
N/A
4.2E+02
N/A
N/A
N/A
No Benchmark
X
N/A
N/A
1.5E+04
No Benchmark
No Benchmark
N/A
No Benchmark
No Benchmark
N/A
N/A
N/A
N/A
No Benchmark
X
X
X
X
X
X
X
X
X
X
N/A
1.5E-01
1.5E+01
N/A
1.5E-02
N/A
N/A
N/A
N/A
N/A
3.8E+00
No Benchmark
X
X
X
N/A
N/A
4.9E-06
Noncarcinogenic SSL
Pathways included in the
Screening Level
Ingestion
X
X
X
X
X
X
X
Dermal
X
X
Inhalation
X
X
X
X
X
X
Resi-
dential
SSL
7.7E+04
3.4E+01
3.1E+01
1.5E+04
1.6E+02
1.6E+04
7.0E+01
No Benchmark
X
X
X
X
X
1.2E+05
2.3E+01
3.1E+03
5.5E+04
4.0E+02
No Benchmark
X
X
X
X
X
1.8E+03
3.9E+02
1.5E+03
No Benchmark
No Benchmark
X
X
3.9E+02
No Benchmark
No Benchmark
X
X
X
X
X
X
7.8E-01
3.9E+02
2.3E+04
3.5E+03
No Benchmark
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
1.7E+04
N/A
N/A
6.2E+03
N/A
1.2E+03
2.3E+03
2.3E+03
3.1E+03
3.1E+03
1.3E+02
No Benchmark
X
X
X
X
X
X
X
X
1.8E+04
1.7E+03
5.1E-05
N/A = Not Available a SOURCE: EPA (2009)
b Cancer values are based on 10~5 risk level
0 The health benchmark for lead was being revised while this evaluation was conducted.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-17
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
4.4.3 Remove SFS Constituents by Comparing to SSLs and Eco-SSLs
The home gardener scenario represents a sensitive population because of the assumption
that individuals live near their garden and grow food using SFS-manufactured soils. These
individuals are potentially more exposed to SFS constituents than the general population. As
shown in the conceptual model (see Figure 3-3), the exposure pathways of concern include the
incidental ingestion of soil, dermal contact with soil, and the ingestion of exposed fruits (e.g.,
strawberries), protected fruits (e.g., oranges), exposed vegetables (e.g., lettuce), protected
vegetables (e.g., corn), and root vegetables (e.g., carrots).
Comparing the soil concentrations to EPA's Residential SSLs is a common technique to
identify COCs for exposure via soil ingestion, dermal exposure to soil, or inhalation of fugitive
dust in residential (as opposed to industrial) exposure scenarios (U.S. EPA, 2002c). Residential
SSLs are also available, on a constituent-specific basis, which address cumulative exposures
from two or more of the above-referenced exposure pathways. Table 4-10 lists the exposure
pathways addressed by the Residential SSLs for the remaining SFS constituents. Residential
SSLs are screening values for soil, regardless of the source of the contamination; in addition, the
Residential SSLs do not consider exposure via ingestion of produce grown on the soil. Therefore,
knowing that Residential SSLs are conservative screening levels for soil ingestion (and in some
instances dermal and inhalation exposures), the Residential SSLs were divided by a factor of 10
to account for indirect exposure associated with the ingestion of produce grown in SFS-
manufactured soil. Work by U.S. EPA (1993) on biosolids strongly suggests that the soil
ingestion pathway is the dominant exposure pathway when compared to the ingestion of plant or
animal products grown on amended soil. Based on EPA's insights on biosolids-amended soil, the
adjustment factor of 10 was used to provide a reasonably conservative adjustment to the
Residential SSLs. Thus, this screening step was only satisfied if the blended soil concentration
(ConcMs) was below the Adjusted SSL (i.e., an order of magnitude below the respective
Residential SSL). If the ConcMs for a constituent was below the Adjusted SSL, the constituent
was removed from further evaluation of the soil pathways. Constituent concentrations in SFS-
manufactured soil were calculated as follows:
ConcMs = Concps x FracMSps
Where:
ConcMs = Concentration of the constituent in SFS-manufactured soil (mg kg"1)
Concrs = 95th percentile constituent concentration in SFS (mg kg"1)
FracMSrs = SFS fraction of manufactured soil (dimensionless). Under this assessment, set to
0.5 representing 50% SFS.
This equation assumes that the SFS is the sole source of the constituent in the
manufactured soil (i.e., background concentrations are not considered).31
As discussed above and listed in Table 4-10, many of the Residential SSLs used in the
assessment address dermal exposure. However, to further evaluate direct dermal contact with
31 Failure to be screened out by this very conservative approach does not imply that the constituent presents a risk,
but rather that for the purposes of this assessment, the constituent was included in a more refined evaluation
discussed in Chapter 5.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-18
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
SFS-manufactured soils, a screening assessment compared SFS-manufactured soil
concentrations to dermal soil screening levels (dermal SSLs). For this conservative assessment,
the SFS-manufactured soil concentrations were calculated based on the 95th percentile SFS
concentrations shown in Table 4-11. The SFS-manufactured soil concentrations were then
compared to U.S. EPA's standardized risk-based dermal SSLs to determine if the pathway
should be further evaluated.
Dermal SSLs were obtained from the U.S. EPA's Mid Atlantic Risk Assessment website
(U.S. EPA 2009). This website provides tables of screening levels for various exposure
scenarios, including a residential soil scenario. The residential soil scenario table presents both
the dermal screening levels and the toxicity values used in the derivation of these levels. Those
COCs for which both noncancer (i.e., RfD) and cancer oral benchmarks (i.e., cancer slope factor,
or CSF) were available, two dermal SSLs were provided, one for each endpoint. The noncancer
SSL is based on a hazard quotient of 1 and the carcinogenic SSL is based on a cancer risk of 1E-
05.32 For those COCs with both noncancer and cancer risk-based SSLs, the SFS-manufactured
soil concentration was compared to the lower of the two SSLs. The calculation of dermal SSLs
also requires the input of a dermal absorption fraction from soils (ABS) and a gastrointestinal
absorption factor (ABSoi). The ABS factors are included in the soil dermal calculations to
account for uncertainty due to different soil types and other variable conditions. The ABSoi
values are used to adjust the oral benchmarks which are usually based on administered dose and
include GI absorption. Table 4-11 presents the dermal SSLs, the associated benchmarks, and
ABS values. With the exception of cadmium, an ABSoi factor of 1 (i.e., 100%) is applied for all
of the COCs shown in this table. The ABSoi value applied for cadmium was 0.025 or 2.5%, as
recommended by U.S. EPA (2004).
32 The carcinogenic SSL presented in the screening level table was based on a cancer risk of 1E-06. For the current
assessment, the carcinogenic SSLs were adjusted to reflect the established allowable cancer risk level of 1E-05.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 4-19
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-11. Comparison to Dermal Soil Screening Levels
Constituent
Arsenic (As)
Cadmium (Cd)
Acenaphthene
Anthracene
Benz [a] anthracene
Chrysene
Dibenz[a,h]anthracene
2,4-Dimethylphenol
Fluoranthene
Fluorene
2-Methylphenol°
3- and 4-Methylphenold
Naphthalene
Phenol
Pyrene
2,3,7,8-TCDD TEQ e
SFS
95%-ile e
(mgkg1)
6.44
0.20
0.26
0.87
0.13
0.04
0.08
4.38
0.18
0.71
8.74
3.41
3.45
20.2
0.47
3.13E-06
Manuf.
Soil Cone.
(mgkg1)
3.22
0.10
0.13
0.44
0.07
0.02
0.04
2.19
0.09
0.36
4.37
1.71
1.73
10.1
0.24
1.57E-06
Human Health
Benchmark
RfD (mg kg -1 d'1) or
CSF (per mg kg ' d'1)
1.5E+00 (CSF)
l.OE-3 (RfD)f
6.0E-02 (RfD)
3.0E-01 (RfD)
7.3E-01 (CSF)
7.3E-03 (CSF)
7.3E+00 (CSF)
2.0E-02 (RfD)
4.0E-02 (RfD)
4.0E-02 (RfD)
5.0E-02 (RfD)
5.0E-02 (RfD)
2.0E-02 (RfD)
3.0E-01 (RfD)
3.0E-02 (RfD)
1.3E+05 (CSF)
Cited
Refa
I
I
I
I
E
E
E
I
I
I
I
I
I
I
I
CalEPA
ABS
(unitless)
0.03
0.001
0.13
0.13
0.13
0.13
0.13
0.1
0.13
0.13
0.1
0.1
0.13
0.1
0.13
0.03
Dermal
SSLb
(mg kg-1)
51
730
13,000
67,000
5.7
570
0.57
5,800
8,900
8,900
15,000
15,000
4,500
87,000
6,700
5.80E-04
I=IRIS; E = (EPA/ORD) Environmental Criteria and Assessment Office
a Reference: Cited in U.S. EPA Mid Atlantic Risk Assessment Generic Tables for Residential Soil Scenario.
b Dermal SSLs based on oral cancer slope factors (CSFs) reflect a cancer risk of IE-OS; noncancer SSLs based on RiDs reflect
a hazard quotient of 1.
c Synonym: o-Cresol.
d RfD and Dermal SSL for 3-Methylphenol (m-Cresol) applied; IRIS reports RfD for 4-methylphenol (p-Cresol) withdrawn.
e Maximum concentration applied instead of 95th percentile due to small sample size.
f Oral RfD (food)
The ecological risk screening focused on receptors that are in direct contact with the SFS-
manufactured soil, and the potential for food web exposures specific to the garden. To screen
SFS constituents for potential ecological impacts, constituent concentrations in SFS-
manufactured soil (ConcMs) were compared to the Eco-SSLs for plants, soil invertebrates, or
mammals,33 whichever was lowest. Table 4-12 shows the ecological screening criteria used in
this assessment. Constituents with ConcMs levels below their respective Eco-SSL passed the
screen, and therefore were removed from further evaluation.
33 Like their human toxicity counterparts, Eco-SSLs are very conservative screening values. Eco-SSLs were
designed to overestimate potential impacts to ecological receptors. For example, the most bioavailable forms of a
constituent are chosen to estimate exposure.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-20
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-12. Ecological Screening Criteria Used in the Analysis a
Constituent
As
Ba
Be
Cd
Co
Cr(III)
Cu
Mn
Ni
Pb
Sb
Se
V
Zn
Low Molecular Weight
PAHsb- Total
High Molecular Weight
PAHs c - Total
Eco-SSL for
Terrestrial Plants
(mg kg -1 soil)
18
—
—
32
13
—
70
220
38
120
—
0.52
—
160
—
—
Eco-SSL for
Soil Invertebrates
(mg kg -1 soil)
—
330
40
140
—
—
80
450
280
1,700
78
4.1
—
120
29
18
Eco-SSL for
Mammals
(mg kg -1 soil)
46
2000
21
0.36
230
34
49
4,000
130
56
0.27
0.63
280
79
100
1.1
a Eco-SSLs are available at http://www.epa.gov/ecotox/ecossl/
b PAHs composed of fewer than four condensed aromatic ring structures (EPA, 2007e)
0 PAHs composed of four or more condensed aromatic ring structures (EPA, 2007e)
Table 4-13 compares the constituent concentrations in SFS-manufactured soil (ConcMs)
to human and ecological SSLs.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-21
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Table 4-13. Comparing SFS-manufactured Soil to Human and Ecological
(mg kg"1 unless otherwise noted)
SSLs
Constituent
Alfekg-1)
As
B
Ba
Be
Cd
Co
Cr (III)
Cu
FeCgkg-1)
Mn
Mo
Ni
Pb
Sb
Se
Tl
V
Zn
Low Molecular Weight
PAHs a - Total
Acenaphthene
Acenaphthylene
Anthracene
Fluorene
Naphthalene
Phenanthrene
High Molecular Weight
PAHs a - Total
Benz[a]anthracene
Chrysene
Dibenz [a,h] anthracene
Fluoranthene
Pyrene
SFS
95%-ile
11.2
6.44
20.2
17.7
0.38
0.20
5.99
109
107
57.1
670
21.8
102
15.3
1.23
0.20
0.09
9.90
72.1
7.59
0.34
0.20
0.88
0.73
3.89
1.56
0.95
0.14
0.04
0.08
0.21
0.48
ConcMs
5.60
3.22
10.1
8.85
0.19
0.10
3.00
54.5
53.5
28.9
335
10.9
51.0
7.65
0.62
0.10
0.05
4.95
36.1
3.79
0.17
0.10
0.44
0.37
1.94
0.78
0.48
0.07
0.02
0.04
0.10
0.24
Adjusted
SSL
7.7
0.67
1,600
1,500
16
7.0
2.3
1.2E+04
310
5.5
1,800
39
150
40
3.1
39
0.078
39
2,300
350
1,700
230
3.8
0.15
1.5
0.15
230
170
Dermal
SSL
51
730
1.3E+04
6.7E+04
8,900
4,500
5.7
570
0.57
8,900
6,700
Passes the Human
Health Screen?
K= Yes)
•/
No
^
•/
•/
^
No
•/
•/
No
,/
•/
•/
^
,/
•/
•/
^
,/
^
•/
•/
^
^
^
•/
^
^
Eco-
SSL
18
330
21
0.36
13
34
49
220
38
56
0.27
0.52
280
79
29
1.1
Passes the Eco
Screen?
K= Yes)
^
•/
^
^
^
No
No
No
No
,/
No
,/
,/
^
•/
•/
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-22
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
Constituent
4-Chloro-3-methylphenol
2,4-Dimethylphenol
2-Methylphenol
3- and 4-Methylphenol
Phenol
2,3,7,8-TCDD TEQ b
SFS
95%-ile
0.09
5.60
8.76
3.59
22.1
3.1E-06
Conors
0.05
2.80
4.38
1.79
11.1
1.6E-06
Adjusted
SSL
620
120
310
310
1,800
4.9E-06
Dermal
SSL
5,800
1.5E+04
1.5E+04
8.7E+04
5.8E-04
Passes the Human
Health Screen?
K= Yes)
^
•/
•/
^
^
•/
Eco-
SSL
Passes the Eco
Screen?
K= Yes)
a Low Molecular Weight PAHs are composed of fewer than four condensed aromatic ring structures, and High
Molecular Weight PAHs are composed of four or more condensed aromatic ring structures (EPA, 2007e).
b Maximum concentration applied instead of 95th percentile due to small sample size
4.4.4 Results
The 95th percentile SFS-manufactured soil concentrations of many of the SFS constituents
were below their respective Adjusted SSL, dermal SSL and ecological SSL, and therefore required
no further evaluation. For example, the SFS-manufactured soil concentrations for all of the
phenolics, PAHs, dioxins, and dioxin-like compounds were below the screening criteria. In
addition, all constituents with dermal SSLs were below the screening criteria, suggesting that these
constituents do not require further evaluation for this pathway. However, the SFS-manufactured soil
concentrations of three metals—arsenic, cobalt, and iron—were above the Adjusted SSL for multi-
pathway exposures. Also, the SFS-manufactured soil concentrations for five metals - antimony,
trivalent chromium, copper, manganese and nickel - were above the Eco-SSL. Based on these
findings and constituent-specific information, the following decisions were made:
• Arsenic was retained for further study in Phase II.
• Due to their potential for phytotoxicity, both manganese and nickel were retained for further
study in Phase II.
• The SFS-manufactured soil concentrations for antimony, trivalent chromium, and copper
were similar to, but above their Eco-SSL's for small insectivorous mammals. Therefore,
antimony, chromium (III) and copper were retained for further study in Phase II.
• The SFS-manufactured soil concentrations of cobalt and iron were above their respective
Adjusted SSLs. Therefore, cobalt and iron were retained for further study in Phase II.
4.5 Analysis Phase I Results
At the beginning of this evaluation, there were three major media-specific exposure
pathways under consideration: (1) groundwater pathway - the ingestion and dermal exposure to
groundwater contaminated by the leaching of SFS constituents; (2) ambient air pathway- the
inhalation of SFS emitted from soil-blending operations; and (3) soil pathway - the incidental
ingestion and dermal exposure to soil, as well as ingestion of fruits and vegetables grown in SFS-
manufactured soil. Because all evaluated SFS constituents were removed from further consideration
by the inhalation screening, the inhalation pathway itself will not be further evaluated. Under the
soil and groundwater dermal screening assessment, all evaluated SFS constituents were well below
a level of concern, and dermal exposure likewise will not be further evaluated. However, based on
other groundwater and soil evaluation criteria (e.g., Adjusted SSL screen for multi-pathway
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-23
-------
Chapter 4.0 Analysis Phase I. Identification of COCs for Modeling
exposures), eleven metals were retained for further evaluation in the risk modeling phase. Table 4-
14 lists the metals retained for Phase II risk modeling.
Table 4-14: SFS Constituents Retained for Phase II Risk Modeling
Human Risk Modeling
Antimony (groundwater)
Arsenic (groundwater and soil/produce)
Beryllium (groundwater)
Cadmium (groundwater)
Cobalt (soil/produce)
Iron (soil/produce)
Lead (groundwater)
Ecological
Risk Modeling
Antimony
Chromium III
Copper
Manganese
Nickel
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
4-24
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
5. Analysis Phase II: Risk Modeling of COCs
Based on the screening evaluations described in Chapter 4, five metals (antimony,
arsenic, beryllium, cadmium, and lead) were retained for probabilistic modeling of the
groundwater pathway, while eight metals (antimony, arsenic, chromium III, cobalt, copper, iron,
manganese, and nickel) were retained for probabilistic modeling of the soil pathways. Arsenic,
cobalt, and iron were evaluated for human exposures through the soil/produce ingestion pathway,
but only arsenic was evaluated under the groundwater pathway. Manganese and nickel in SFS
were modeled in the home gardening scenario because of their potential for phytotoxicity.
Finally, concentrations of antimony, trivalent chromium, and copper were retained for further
study due to the potential to impact small insectivorous mammals as described in Chapter 4.
Probabilistic modeling was conducted to address the variability in conditions across the
country. This was done by using metal-, regional- and site-specific data to conduct probabilistic
analyses of the remaining constituents of potential concern and exposure pathways.
This chapter is organized as follows:
• Section 5.1 provides an overview of Phase II probabilistic modeling
• Section 5.2 explains the screening probabilistic modeling of exposure via groundwater
ingestion
• Section 5.3 describes the more refined probabilistic modeling of exposures via soil,
produce consumption, and groundwater ingestion, the results of the modeling, and the
derivation of screening levels for the modeled constituents of potential concern in SFS.
5.1 Overview of Phase II Probabilistic Modeling
Figure 5-1 is a simple depiction of how the Monte Carlo probabilistic approach was
implemented for the SFS evaluation. It shows how the distributions for input parameters were
sampled and used to produce the probability distribution and cumulative distribution function
from which specific percentiles (e.g., 90th percentile) can be identified. The example parameters
A, B, and C each have their own distributions, which may represent variability or uncertainty or
both. For each model run, a single value was sampled from each input distribution regardless of
the type of variation (i.e., variability or uncertainty). For each modeling scenario (e.g., adult or
child), the simulation produced the probability distribution of risk results, as shown in Figure 5-1
(i.e., the distribution of risk across exposed individuals across all sites represented in the
analysis). Lastly, the cumulative distribution function was created, and the specific percentiles
(e.g., 90th percentile) were selected and used to characterize risks.
Home garden location was the primary determinant for selecting parameter values that
describe the environmental setting. A geographic information system (GIS) sampling procedure
was used that correlated location, climate station, and soil type, thus ensuring that feasible
combinations were modeled. The rest of the regional data (i.e., long-term climate data and daily
meteorological data) were held constant for all sampled locations within a given climate region.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-1
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Figure 5-1. How the Monte Carlo approach addresses uncertainty.
5.2 Screening Probabilistic Modeling of the Groundwater Ingestion
Pathway
Five constituents (antimony, arsenic, beryllium, cadmium, and lead) were retained for
Phase II evaluation. EPA's IWEM probabilistic groundwater screening model was used to
evaluate the home garden scenario groundwater pathway. IWEM has undergone extensive peer
review, and provides a flexible scenario for considering the potential leaching from the use of
SFS in manufactured soils. Detailed information on this model can be found in the IWEM User's
Guide (U.S. EPA, 2002a) and Technical Background Document (U.S. EPA, 2002b).34
As a conservative assumption, the 95th percentile SFS leachate concentration for each of
the five constituents was used with site-descriptive parameter values. The model ran each
constituent 10,000 times for 10,000 years assuming a constant leachate profile from a single
application of SFS-manufactured soil, varying site conditions based on original inputs. Figure 5-
2 illustrates a conceptual cross-section of the subsurface modeled in the SFS evaluation. After all
runs were completed, the estimated well-water concentration representing the 90th percentile
(i.e., higher than 90 percent of the other estimates) was compared to the lowest Phase I screening
level (i.e., Tapwater Screening Level, MCL, or National Secondary Drinking Water Standard -
see Chapter 4 Section 4.2.3 for more information on these screening levels). If the constituent's
34 Supporting documentation for IWEM and EPACMTP can be found at http://www.epa.gov/osw/nonhaz/
industrial/tools/
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-2
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
90 percentile well-water concentration estimate was at or below the screening level, then the
constituent was not retained for further evaluation.
L£ACHATE CONCENTRATION
UNSATURATED
ZONE
LEACHATE
SATURATED
ZONE LEACHATE
LAND SURFACE
WATER TABLE
Figure 5-2. Conceptual Cross-Section View of the Modeled Subsurface
5.2.1 Groundwater Model Inputs
Some modeling input parameter values (e.g., distance from the garden to the drinking
water well) were chosen to be conservative and to maximize drinking water estimates. Values for
some other input parameters (e.g., depth to aquifer) were chosen from distributions representing
variable conditions around the country. For the remaining parameters, the default values
provided in the IWEM User's Guide (U.S. EPA, 2002a) were used. The model used the
following parameters to define the use scenario:
• A 405 m2 (i.e., 0.1 acres) land application unit (i.e., unconsolidated application to land)
was operated for 40 years.35 An area of 0.1 acres was selected to be conservatively
representative of a garden suitable for SFS-manufactured soil use and that is of sufficient
size to feed a home gardening family for a year.36
• To test the effect of distance from the garden to the drinking water well, separate sets of
10,000 runs were performed for each of the following distances: 1 m, 15 m, 30 m, and
50m.
35 An operating life of 40 years for the land application unit is consistent with the default operating life applied in
EPACMTP and in the Multi-media, Multi-pathway, Multi-receptor Risk Analysis (3MRA) modeling system for
land application (U.S. EPA, 2003d, g).
36 A 0.1 acre garden is more than sufficient to support the home gardener scenario that includes an adult and child
receptor. The North Carolina State University, Department of Horticultural Science, reports that a garden of 25 ft
x 40 ft (approximately 0.02 acres) will produce most of the vegetables needed by 2 people for one year
(http://www.ces.ncsu.edu/depts/hort/hil/ag-06.htmlX Additional references also report garden sizes much smaller
than the modeled 0.1 acres. For example, The National Gardening Association reported in 2009 that only 6% of
U.S. gardens were larger than 2,000 ft2 (0.05 acres) (http://www.gardenresearch.com/files/2009-Impact-of-
Gardening-in-America-White-Paper.pdf).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-3
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
• The following subsurface parameters (e.g., groundwater pH, depth to water table) were
set to model defaults:37
- Depth to water table: 5.18m (IWEM default for a shallow aquifer)
- Aquifer thickness: 10.1 m
- Hydraulic conductivity: 1.89E+03 m yr"1
- Regional hydraulic gradient: 0.0057
- Groundwater pH: 7
- Chemical-specific decay rate: 0 for metals
- Soil-water partition coefficient: selected from isotherms generated by the
MINTEQA2 geochemical speciation model38
• Each model run included a randomly selected well-screen depth, constrained to occur
within the aquifer (i.e., between 5.18 m and 15.28 m below the ground surface).
• Other unsaturated zone parameters were varied for each run based on a nationwide
distribution of three soil types: sandy loam (15.4%), silt loam (56.6%), or silty clay loam
(28%)
• To represent conditions across the country, three climates were modeled: a representative
dry climate (Phoenix, AZ), a moderate climate (Indianapolis, IN), and a wet climate
(Seattle, WA)
• For arsenic, the higher of the 95th percentile leachate concentrations determined by either
the SPLP or ASTM leachate methods (0.018 mg L"1) was modeled. Antimony, beryllium,
cadmium, and lead were not detected in any samples, and were therefore modeled at one
half their detection limits in accordance with U.S. EPA (1991b). Thus, their modeled
leachate values were 0.02, 0.01, 0.005, and 0.055 mg L"1, respectively.
Effect of well distance on drinking water concentration: As illustrated in Figure 5-2,
some horizontal distance is required for the constituent plume to mix to the bottom of the
aquifer. The horizontal distance required for a constituent to mix to the bottom of the aquifer
depends on constituent-specific characteristics (e.g., soil-water partitioning), and therefore the
distance will vary by constituent. Constituent concentrations within the groundwater plume will
be highest directly under and near the garden. Concentrations will decrease as the plume travels
horizontally, the constituent mass diluting into an ever larger volume of groundwater.
The random selection of well-screen depth (see bullet 4, above) will, for some model
runs, result in the contaminant plume "missing" the well. For instance, if the screen depth
illustrated in Figure 5-2 had been chosen to be 15 m (i.e., near the bottom of the aquifer) rather
than within the contaminant plume, the plume would have moved above the screen and produced
a zero well concentration. Existence of these zero concentrations in the output distribution would
skew percentile calculations lower (i.e., a lower value can be above 90% of the other values
when some of the other values are zero).
To test the interplay between constituent dilution and the effects of zero concentrations
on output percentiles, a complete set of 10,000 model runs was completed for each constituent in
37 See U.S. EPA (2002b), section 4.2.3.1 for details on how these defaults were chosen for IWEM.
38 See U.S. EPA (2002b) section 4.3.4.3.2 for details on how MINTEQA2 was used to produce the isotherms
sampled for partition coefficients.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-4
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
each climate at well distances of 1 m, 15 m, 30 m and 50 m, respectively. If IWEM returned a
receptor well concentration of "0" (i.e., below l.OE-20 mg L"1) at all four distances, then no
further modeling of that constituent was performed in that climate. If IWEM returned non-zero
values that demonstrated a peak and dilution with greater distance, then no further modeling of
that constituent was performed in that climate. If no peak concentration was demonstrated (i.e.
modeling at greater distances elicited higher receptor well concentrations), IWEM was run again
and re-evaluated at 75 m, 100 m, 150 m, and 200 m, or until a peak was demonstrated.
5.2.2 Groundwater Model Outputs
Table 5-1 lists the groundwater modeling results at the 90th percentile. As shown in the
table, the exposure estimates for arsenic in the Wet and Central Tendency climates were above
the screening level, and below the screening level for the Dry climate. The exposure estimates
for beryllium, cadmium, lead, and antimony were consistently lower than the screening levels in
all three climates.
Table 5-1. Tested Leachate Concentrations, Receptor Well Concentrations for the Home
Gardener Exposure Scenario, and Screening Levels (mg L'1)
Constituent
Tested
Leachate
Cone.
90th Percentile Modeled Exposure Level a
1m
15m
30m
50m
75m
100m
Lowest
Screening
Level b
Wet Climate
As
Be
Cd
Pb
Sb
0.018
0.01
0.005
0.055
0.02
4.9E-03
1.7E-09
2.3E-03
5.9E-03
5.9E-03
3.4E-03
3.8E-08
1.5E-03
3.00-03
4.5E-03
2.5E-03
7.2E-07
1.1E-03
1.7E-03
3.2E-03
1.8E-03
1.5E-06
7.3E-04
1.1E-03
2.4E-03
NM
1.1E-06
NM
NM
NM
NM
NM
NM
NM
NM
4.5E-04C
4.0E-03
5.0E-03
1.5E-02
6.0E-03
Moderate Climate
As
Be
Cd
Pb
Sb
0.018
0.01
0.005
0.055
0.02
5.2E-04
0
2.6E-04
2.0E-03
1.1E-03
9.6-04
6.9E-14
4.3E-04
2.1E-03
1.8E-03
8.9E-04
8.2E-13
3.7E-04
l.OE-03
1.7E-03
6.8E-04
2.9E-12
2.6E-04
5.3E-04
1.3E-03
NM
4.1E-12
NM
NM
NM
NM
3.4E-12
NM
NM
NM
4.5E-04C
4.0E-03
5.0E-03
1.5E-02
6.0E-03
Dry Climate
As
Be
Cd
Pb
Sb
0.018
0.01
0.005
0.055
0.02
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
4.5E-04C
4.0E-03
5.0E-03
1.5E-02
6.0E-03
a The model reports a"0" level if the 90th percentile modeled well concentration is lower than l.OE-20 mgL"1.
Unless otherwise noted, MCLs were the lowest screening level.
0 For arsenic, the Tapwater Screening Level was the lowest screening level. The arsenic Tapwater Screening
Level used in this evaluation is based on a 10~5 risk level
NM = Not Modeled
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-5
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
5.2.3 Results
The well distance demonstrating peak concentration varied by constituent and climate,
but in no case was further than 75 m. In modeling a wet climate, antimony, arsenic, cadmium,
and lead demonstrated peak concentrations at a well distance of 1 m, and were therefore not
modeled beyond 50 m. Beryllium demonstrated a peak receptor well concentration at a well
distance of 50 m.
In modeling the constituents in a moderate climate, antimony, arsenic, cadmium, and lead
demonstrated peak concentrations at a well distance of 15 m, and were therefore not modeled
beyond 50 m. IWEM estimated a receptor well concentration of zero for beryllium at a 1 m well
distance, but ultimately peaked at a distance of 75 m. In modeling the constituents in a dry
climate, IWEM estimated receptor well concentrations of zero for all constituents across the first
four distances, and therefore no further modeling performed.
The screening probabilistic modeling for groundwater ingestion found that estimated
exposures for antimony, beryllium, cadmium, and lead were below drinking water screening
levels in all climates and at all well distances. Therefore, no further evaluation of exposure to
those constituents via groundwater ingestion was necessary. Estimated exposures for arsenic
were consistently above the drinking water screening level in the Wet and Moderate climates,
and consistently below the screening level in the Dry climate. Arsenic was therefore retained for
more refined study.
5.3 Refined Probabilistic Modeling of the Soil/Produce and Groundwater
Ingestion Pathways
As described in Chapter 4, four constituents of potential concern required further
evaluation of the soil/produce ingestion pathway: arsenic, lead, manganese, and nickel. In
addition, as described in Section 5.2, arsenic was retained for refined evaluation of the
groundwater pathway. As part of this evaluation, probabilistic modeling of these constituents
was performed to derive risk-based modeled screening levels for comparison to SFS constituent
concentrations. If the SFS concentrations were below these conservative SFS-specific screening
levels, then the beneficial use of SFS as a component of manufactured soil would be considered
protective of human health and the environment. The following provides an overview of the
process used to derive the modeled screening levels.
Risk distributions were developed using an initial soil concentration of 1 ppm for each
constituent; this initial concentration is referred to as a "unitized" concentration in the sense that
it does not represent an actual concentration in SFS or soil; rather, it represents an arbitrarily
chosen concentration that is used to estimate risk per "unit" of constituent in soil. Consistent with
previous EPA risk assessments and based on the model's linearity with respect to constituent
concentration, the 90th percentile of the unitized risk estimates was scaled to estimate protective
target SFS constituent concentrations based on EPA's risk management criteria (e.g., hazard
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-6
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
quotient of I).39'40 These SFS-specific concentrations (i.e., concentrations in SFS, rather than
concentrations in soil) are conservative estimates of the selected SFS constituents that would be
protective of human health and the environment if the SFS were used in manufactured soil. The
following summarizes the individual steps taken to develop the target SFS concentrations from
the unitized risk distributions.
Step 1. Estimate Environmental Releases
Using an initial soil concentration of 1 ppm, the source model was run to simulate the
release of constituents to surrounding media from a home garden assumed to receive a single
"addition" of SFS-manufactured soil to a depth of 20 cm (a typical tilling depth). As discussed in
Sections 5.3.3 and 5.3.4, release mechanisms simulated by the model include losses due to
leaching, volatile and particle releases to the air, and horizontal movement of pollutants (i.e.,
runoff and erosion from the garden). The model generates time-series estimates for these
releases, as well as estimates for surficial and root zone soil concentrations. For arsenic (i.e., the
only SFS constituent requiring refined groundwater modeling), leachate fluxes (g m^yr"1)
estimated by the source model were used by the groundwater fate and transport model to
estimate arsenic concentrations at the drinking water receptor well.
Step 2. Calculate Unitized Ratios
Calculating risk from the source modeling outputs involved fate and transport modeling
(Section 5.3.5, groundwater modeling, and Section 5.3.6, food chain modeling), human
exposure and health effects modeling (Sections 5.3.7 and 5.3.8), and ecological exposure and
health effects modeling (Section 5.3.9). The probabilistic simulation generated distributions of
unitized risks for adult and child home gardeners, as well as for ecological receptors, that reflect
the variability in conditions within the economic feasibility areas.
Step 3. Calculate SFS Screening Level
Using 90th percentile unitized risk estimates, and EPA's risk management criteria (e.g.,
HQ of 1), screening levels were calculated for each constituent. As shown in Section 5.3.11, the
calculation of SFS screening levels also allows for the adjustment of levels based on the fraction
of SFS in manufactured soil. The resulting soil concentrations represent conservative estimates
of SFS constituent concentrations considered protective of human health and the environment.
The remainder of this chapter is organized as follows:
• Section 5.3.1 provides an overview of the risk modeling framework implemented to
perform probabilistic modeling.
• Section 5.3.2 describes the exposure scenario, including conservative screening
assumptions, developed for application of SFS in home gardens.
39 Similar unitized approaches have been applied under previous U.S. EPA risk assessments. For example, the
unitized approach was applied in the Risk-Based Mass Loading Limits for Solvents in Disposed Wipes and
Laundry Sludges Managed in Municipal Landfills. This risk assessment and the unitized approach have been
extensively reviewed, and the final rule based on this risk assessment, Solvent-Contaminated Wipes, was
published July 31, 2013 (U.S. EPA, 2013a)
40 Appendix J describes the analysis that was performed to confirm that the unitized calculation method was
appropriate for the groundwater modeling of arsenic.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-7
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
• Section 5.3.3 describes the receptors (both human and ecological) and the exposure
pathways by which receptors could potentially be exposed to SFS constituents.
• Sections 5.3.4 through 5.3.10 describe the models, inputs, and outputs used in the
probabilistic screening of health and ecological risk associated with SFS use in home
gardens.
• Section 5.3.11 describes how the human and ecological modeling results were used to
calculate SFS-specific screening levels.
• Section 5.3.12 compares the SFS constituent concentrations to the lowest human health-
based SFS-specific screening levels, as well as ecological SFS screening levels.
5.3.1 Modeling Framework Overview
Unitized risk distributions were developed for this analysis using a risk modeling
framework currently used by EPA to support the Part 503 biosolids program. The risk modeling
framework integrates a variety of models and input datasets facilitating site-based and national-
level exposure and risk assessments. The SFS assessment modified and adopted the system to
evaluate soil/produce and groundwater ingestion risks associated with the use of SFS in
manufactured soils.
Under this assessment, we used a Monte Carlo approach that essentially loops over
randomly selected locations within the area of economic feasibility, selecting input parameter
values that correspond to each particular location. Within the looping structure, a series of
modules are executed in a specific order. The modeling process can be summarized as follows:
• The source models estimate pollutant releases to the environment
• The environmental fate and transport models estimate concentrations in environmental
media (e.g., soil, groundwater, ambient air) and in dietary items (e.g., fruits and
vegetables)
• The exposure models estimate the pollutant levels to which receptors are exposed
• The human risk model estimates the chemical-specific human health risk, and the
ecological effects model estimates chemical-specific hazard quotients.
The major functionality of the models implemented in this risk analysis is described in
Sections 5.3.4 through 5.3.10.
As illustrated in Figure 5-3, the looping structure is comprised of four nested loops:
Chemical; RunID; Human Receptor; and Ecological Receptor. The outmost loop is the chemical
loop, which allows a Monte Carlo simulation to be performed on a constituent-specific basis.
The next loop is the RunID loop, which controls the number of iterations performed in a given
simulation and is used as the primary index to input datasets, including site location. As shown in
Figure 5-3, the source, media, and food modules are executed for each Monte Carlo iteration.
Outputs from the source model are used as inputs to the downstream groundwater, media and
food modules to estimate concentrations that receptors can potentially be exposed to.
Within the Monte Carlo loop, the next loop in the probabilistic analysis cycles through
the different types of receptors. The model considers both adult and child receptors and various
ecological receptors. The receptor type determines the exposure factors used. Receptor type and
exposure factors were not specific to location; as a result, any receptor (human or ecological)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-8
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
could be present at any location with any applicable exposure parameter values. Receptor-
specific exposure factors for humans include exposure duration, the receptor's age when
exposure begins, dietary consumption rates, and individual body weight. A set of adult and child
exposure parameters was chosen for each iteration. Exposure parameters were not correlated
with each other or with geographic locations. Ecological exposure parameters included the
receptor-specific health benchmarks. More detailed descriptions of human and ecological
exposure modeling are found in Sections 5.3.7 and 5.3.8, respectively.
Sou reel D = Home Garden
Chemical (CAS) Loop
-*• RunID Loop (Monte Carlo iterations)
Call Source Module: calculate emission rates; soil concentrations and losses
due to leaching, runoff, and erosion
Call Media Module: calculate groundwater and air concentrations
Call Food Module: calculate concentrations for food items
*• Human Receptor Loop (adult, child)
Select pathways and exposure data based on human receptor type
For Adult Receptor
Calculate intake over exposure duration
For Child Receptor
r Cohorts Loop (ages child through age cohorts)
Calculate cohort intake
Next Cohort
Calculate intake over exposure duration
Call Human Risk Module: calculate risk based on human health benchmarks
1— Next Human Receptor
> Ecological Receptor Loop
Select pathways and ecological exposure data based on ecological receptor type
Call Ecological Exposure Module and calculate ratios of media concentrations to
ecological concentration benchmarks
1— Next Ecological Receptor
L- Next RunID
Next Chemical
Figure 5-3. Basic Monte Carlo looping structure for the home garden.
The Monte Carlo simulation represents a set of individual model realizations, with each
realization defined in terms of a unique set of values for the input parameters required by the
model. The approach is implemented by creating input files prior to the assessment that include
data that are randomly selected based on the regional setting and scenario selected for each
iteration. Chemical-specific data are generally constant across all iterations and are not correlated
with other input parameters. The SFS-manufactured soil concentration was also held constant
under this assessment to allow the calculation of the unitized risk estimates. The input of the
fixed initial soil concentration of 1 ppm wet weight (i.e., unit concentration) into this linear
system allowed for the development of unitized risk estimates that reflect national variability.
The unitized approach was ideal for the SFS analysis since it provided the flexibility to generate
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-9
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
distributions of unitized risk estimates that could be scaled to calculate screening concentrations
using a variety of recipes for SFS-manufactured soils.
Under the SFS analysis, 7,500 Monte Carlos iterations were executed. To ensure the
stability of the model results and determine the appropriate number of Monte Carlo simulations,
the model was run for 4 different sets of iterations: 1,000; 3,000; 5,000; and 7,500 iterations.
Tolerance criteria were established at 5%; that is, the model would be considered to be stable if
the mean, variance, and the 50th and 90th percentile results did not change by more than 5%.
Based on previous experience, the model was expected to converge in less than 5,000 iterations.
The results of the stability test are shown in Figure 5-4. The table shown in the figure presents
the absolute percent changes between samples. As demonstrated by this figure, the model is
stable before 5,000 iterations for the mean, variance, and at the 50th and 90th percentiles, and
extending the simulation to 10,000 iterations was considered unnecessary.
03
I
Eard Quotient Estimates (uni
° ° S
m ro m
• ' ift
?H psziuun.
Arsenic
Unitized Hazard Quotient Estimates
Total Inaestion: Child of Home Gardener
J
J
/
/
— i— IWO
---3000
- sooo
-.-.'.'
: W 70 80 90 100
Percent! le
value
SOtn Dercentile
75tn percentiie
90th percentile
95th percentile
94th percentile
mean
variance
Percent Difference
1000JOOO
6.S
6.5
2.1
0.4
0.7
2.5
2.7
3000 SOOO
0.6
O.fi
l.i,
1.3
4.)
0.03
4.0
5W07HK
0.5
0.5
0.9
1,1
^$
0.6
2.6
Figure 5-4. Model stability.
5.3.2 Exposure Scenario—Use of SFS in Home Gardens
The modeled use of SFS in home gardens assumed that a portion of a residential yard is
used for home gardening: either the yard itself is tilled or raised beds are constructed. A single
application of 20 cm (approximately 8 inches) of SFS-manufactured soil is spread in the
residential construction area as topsoil, or a single application of 20 cm of SFS-manufactured
soil is used in the construction of raised gardening beds. SFS is generated across the United
States; therefore, the evaluation used a regional approach to capture the variability across site
conditions. The modeling framework used regional climate and soil data to estimate constituent-
specific releases and to predict their fate and transport in the environment. For example, the
source model used soil data and daily precipitation data to estimate events such as runoff,
erosion, and leaching.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-10
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
The SFS was assumed to be used within 50 km of the foundry (EPA, 2008q).41 This
approach thereby focused the evaluation on climate and soil conditions relevant to where SFSs
might reasonably be used as a component of manufactured soil. Figure 5-5 shows the areas
included in the assessment.
| | Met Station Region
^ Foundry Sands Economic Feasibility Area
Figure 5-5. Meteorological regions and SFS use areas.
The scenario consists of the following elements:
• Regional data for 41 climate regions. Climate regions were shaped such that climate data
from a single location would represent any location within the region, taking into account
geographic boundaries, such as mountains, and other parameters that differentiate
meteorological conditions (e.g., temperature and wind speed) as described in
Appendix D.
• Locations of foundries in the United States, in the form of ZIP Code boundaries extended
50km.
• Using a geographic information system (GIS), a soil layer was overlaid with the
meteorological regions to identify location-specific soil texture and characterize soil
parameters as described in Appendix E.
5.3.3 Potential Release Pathways and Receptors
Chapter 3 described the conceptual models that define the sources, releases, exposure
pathways, and receptors relevant to the use of SFS in manufactured soil. The potential exposure
pathways not fully modeled previously—incidental soil ingestion and ingestion of fruits and
vegetables grown in SFS-manufactured soil—were modeled in this phase of the evaluation. In
addition, the groundwater pathway was further evaluated for arsenic. Figure 5-6 diagrams the
41 SFS use areas are based on the ZIP codes of the membership of the American Foundry Society as of November
2007. Since a foundry's exact location within its ZIP Code area was not provided, the ZIP code boundary was
extended by 50km.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-11
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
portions of the conceptual model (described in the problem formulation) that were addressed by
this national-scale modeling. The diagram shows how the data flow from the source models,
which are used to estimate releases to the environment, to the environmental fate and transport
models, which are used to estimate concentrations in the soil, leachate, groundwater, eroded soil
and air, to the exposure models, which are used to estimate concentrations in the food chain and
resulting exposures to human and ecological receptors.
Source
Release, Fate & Transport
Exposure Pathways
Receptors
Figure 5-6. Conceptual model for modeling the home gardener.
As shown in Figure 5-6, the human and ecological receptors identified in the conceptual
model could be exposed through various pathways. To estimate screening SFS concentrations,
human and ecological receptors that would be subject to reasonable maximum exposures were
identified. The potentially exposed human receptors are assumed to be members of a family that
live and grow food in a garden on property where manufactured soil contains SFS. These
individuals would be more highly exposed to SFS than the general population. In addition, the
percentage of the gardening receptor's diet that consists of home-grown produce is assumed to
be higher than the percentage for the general population. Throughout the modeling, exposure
assumptions were designed to be conservative; that is, they were likely to overestimate, rather
than underestimate potential exposures.
The exposure pathways considered for adult and child receptors are summarized in
Table 5-2. Although these pathways were evaluated concurrently within the modeling
framework, analyses were performed as discussed in Section 5.3.5 and Appendix J that indicated
that the maximum groundwater and soil/produce pathway exposures would not occur within the
same period of time. As a result, separate SFS screening levels were developed for the
groundwater and the soil/produce pathways.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-12
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-2. Human Exposure Pathways for SFS-Manufactured Soil in Home Gardens
Receptor
Resident Adult
Gardener
Resident Child
Ingestion
of
Ground-
water
^
^
Ingestion
of Soil
^
^
Ingestion of
Exposed
Fruits
(e.g., apples)
/
/
Ingestion of
Protected
Fruits
(e.g., oranges)
^
^
Ingestion of
Exposed
Vegetables
(e.g., lettuce)
^
^
Ingestion of
Protected
Vegetables
(e.g., corn)
^
^
Ingestion of
Root
Vegetables
(e.g., carrots)
^
^
5.3.4 Source Modeling
This section provides an overview of the source model and modeling approach, and
identifies model inputs and outputs.
5.3.4.1 Conceptual Source Model
The source model used in this assessment was the land application unit model developed
for ORCR as part of the 3MRA modeling system (U.S. EPA, 2003 c). The land application unit
model was developed to estimate annual average surface soil constituent concentrations and
constituent mass release rates to the air, downslope land, and groundwater. The model simulates
the vertical movement of pollutants within the agricultural land (releases through leaching to
groundwater), volatile and particle releases to the air, and horizontal movement of pollutants
(runoff and erosion from the agricultural land across any buffer area to a nearby waterbody). The
model considers losses from the agricultural land due to hydrolysis and biodegradation, as well
as leaching, volatilization, and particle emissions due to tilling (mixing) operations and wind
erosion.
The model has been extensively peer reviewed and has been used to support several risk
assessments conducted for EPA's ORCR and Office of Water. Although the source model was
initially developed to assess hazardous wastes, it has been used to support regulatory risk
assessments, including the 2003 and 2013 biosolids exposure and hazard assessments. Under
these national assessments, biosolids were assumed to be applied to agricultural fields used to
grow crops or used as pastureland. Under the SFS assessment, the crop modeling scenario was
adopted and modified to assess human and ecological impacts associated with the application of
SFS-manufactured soil in residential gardens. The following highlight areas where the current
screening approach deviated from the biosolids methodology:
• A "soil replacement" assumption was applied instead of the "soil amendment"
assumption in biosolids. The soil replacement scenario definition represents a reasonably
conservative description regarding the use of SFS in manufactured soil.
• In the biosolids analyses, farm areas are varied stochastically by sampling from a
distribution using data from Hoppe et al. (2001) that spans a range from 45 - 73 hectares
(i.e., Ill to 180 acres). Because residential gardens are significantly smaller, the modeled
application area was reduced to better reflect actual gardening practices. The garden was
modeled as a 405 m2 (i.e., 0.1 acres) area consistent with the IWEM modeling discussed
in Section 5.2.1.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-13
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
• Choices of garden location, meteorological data, and soil data were constrained to fall
within the SFS economic feasibility areas. A discussion of this approach is provided in
Chapter 3, Problem Formulation.
• The current screening level ecological assessment focused on direct contact with the soil.
To evaluate potential ecological risks associated with the SFS constituents, EPA's Eco-
SSLs for soil invertebrates, terrestrial plants, and small insectivorous mammals were
identified and compared to predicted maximum predicted soil concentrations.
Under the soil replacement scenario, it was assumed that SFS-manufactured soil is
applied one time, evenly across an area used for home gardening to a depth of 20 cm. Based on
the assumed composition of SFS-manufactured soil, it was also assumed that the properties and
characteristics of the SFS-manufactured soil would mimic those of natural soil in the area. That
is, the SFS-manufactured soil used in the garden will be similar to the local native soil (which is
a function of the garden location).
Consistent with the 2013 biosolids exposure and hazard assessments, the source model
was coupled with EPACMTP to evaluate impacts to the groundwater pathway. The leachate
fluxes (g m"2 yr"1) and infiltration water fluxes (m d"1) estimated by the source model were
subsequently used as input to EPACMTP to estimate arsenic concentrations at the receptor well.
5.3.4.2 Source Model Inputs
The source model requires numerous input parameters, including location-specific
parameters, constituent-specific parameters, and parameters that describe the garden's
dimensions and operating practices. The following identifies key inputs and describes the
approach used in characterizing the parameters; additional details on the source model mass-
balance governing equations and parameter inputs are provided in Appendix G, Home Garden
Source Model, and Appendix F, Chemical Data:
• Constituent Concentrations. Constituent concentrations were fixed to a unit
concentration of 1 mg kg"1. In applying a unitized concentration approach, the resulting
constituent-specific hazard estimates were used to estimate concentrations in SFS-
manufactured soil that could be applied without exceeding the hazard criterion adopted
for this analysis. The criterion for this analysis was a Unitized Dose Ratio (UDR) of 1 for
cancer and noncancer effects42 at the 90th percentile of the hazard probability
distribution.43 A detailed discussion of the UDR is found in Section 5.3.9.1.
• Chemical properties. The model requires the input of several parameters, such as
diffusivity in air and water. The chemical-specific properties used in this assessment are
presented in Appendix F. The primary data source for these parameters is the Superfund
Chemical Data Matrix (SCDM; U.S. EPA, 2008b), because it is peer reviewed and
contains all of the constituents evaluated. Other sources include the Hazardous
42 In this evaluation, UDR refers to the generic ratio of estimated exposure divided by health benchmark, regardless
of the type of adverse effect (i.e., cancer or noncancer) the benchmark is based on.
43 EPA's Guidance for Risk Characterization (U.S. EPA, 1995c) defines the risk criterion for the hazard-based
calculation to be protective of 90% of hypothetically exposed individuals, stating that "For the Agency's purposes,
high end risk descriptors are plausible estimates of the individual risk for those persons at the upper end of the risk
distribution," or conceptually, individuals with "exposure above about the 90th percentile of the population
distribution."
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-14
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Substances Data Bank (HSDB) (NLM, 2006) and the Merck Index (Budavari, 1989).
Distributions for soil water partition coefficients (Kd) were derived from U.S. EPA
(2005a).
• Area of the garden. As discussed in Section 5.2.1, the size of the garden was assumed to
be 405 m2 (i.e., 0.1 acres). An area of 0.1 acres was selected to be conservatively
representative of a typical residential garden that is of sufficient size to feed a home
gardening family for a year.
• Characteristics of the SFS-manufactured soil (e.g., percent solids, bulk density,
fraction organic carbon). Properties and characteristics of the SFS-manufactured soil
were assumed to mimic those of natural soil in the area. Because soil characteristics vary
spatially, it was necessary to assign gardens to specific locations. With the added
consideration of economic feasibility areas, the approach applied in making these
assignments was consistent with the approach used in the biosolids assessments.
Considering the joint probability of occurrence, gardens were assigned to one of 41
climate regions. Using a geographic information system (GIS), a soil layer was overlaid
with the climatic regions to identify the predominant soil texture for the top 20 cm of soil.
Specific soil parameters, such as bulk density and fraction of organic carbon, were
characterized based on the selected soil type. The percent solid of the SFS-manufactured
soil was calculated based on soil moisture at field capacity and soil bulk density.
• Climate conditions at the garden site. Gardens were assigned to one of the 41 climate
regions. As discussed in Appendix D, a representative meteorological station and data set
was selected for each climate. This data set was assumed to be representative of the
conditions throughout the entire region.
• Tilling depth. The soil mixing depth for the garden was set to a default value of 20 cm to
reflect tilling conditions. This value is consistent with the recommended default value for
tilled soil in EPA's Human Health Risk Assessment Protocol (U.S. EPA, 2005b).
5.3.4.3 Source Model Outputs
The outputs of the source model include the following:
• Annual average constituent concentration in the surface of the garden soil
• Annual average constituent concentration in the root zone of the garden soil
• Annual emission of volatile constituents from the surface of the garden soil
• Annual emission of constituents sorbed to particles from the surface of the garden soil
due to tilling and wind erosion
• Daily concentrations and mass of soil eroded from the garden soil
• Daily concentrations and volume of runoff from the garden (used in calculating the load
to the buffer)
• Daily concentrations and volume of runoff from the buffer area
• Annual infiltration rate of water from the garden
• Annual leachate flux of constituents from the garden.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-15
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
5.3.5 Fate and Transport: Refined Groundwater Modeling
Refined probabilistic groundwater modeling used the EPACMTP (U.S. EPA, 2003f, g, h;
1997a). Consistent with other EPA national-scale assessments, EPACMTP inputs included
distributions of leachate fluxes and infiltration rates from the home garden source model, rather
than the single, 95th percentile leachate concentration used in screening probabilistic modeling.
Coupling the source and groundwater modeling in this way captures national variability in
conditions through the use of location-specific climate and soil distributions, as well as
constituent-specific input parameters (e.g., soil Kd distributions) to estimate constituent-specific
releases and to probabilistically predict their fate and transport in the environment.
EPACMTP accounts for advection, hydrodynamic dispersion, equilibrium linear or
nonlinear sorption, and transformation processes via chemical hydrolysis. In this analysis,
sorption of arsenic being leached from SFS-manufactured soil into the unsaturated and saturated
zones was modeled using soil-water partitioning coefficients (Kd values) selected from nonlinear
sorption isotherms generated from the equilibrium geochemical speciation model MINTEQA2
(U.S. EPA, 1991). As discussed in Appendix J, maintaining linearity with respect to sorption was
critical to supporting the appropriateness of applying the unitized approach to estimate SFS
Screening Levels. Kd selection was therefore monitored during the EPACMTP simulations,
ensuring that the assumption of linearity was valid.
The groundwater concentrations are used in estimating drinking water exposures as
shown in the equations presented in Appendix H.
5.3.5.1 Conceptual Groundwater Model
The groundwater pathway was modeled to estimate receptor well concentrations that
result from a predicted release of arsenic from SFS-manufactured soil used in a home garden.
The release of a constituent occurs by leachate, containing the constituent, percolating through
the soils into the subsurface as a result of precipitation water infiltrating through the SFS-
manufactured soil. The released constituent is transported via aqueous-phase migration through
the unsaturated zone (the soil layer beneath the garden and above the aquifer) to the underlying
saturated zone (i.e., groundwater), and then downgradient in the groundwater to a hypothetical
residential drinking water well (the "receptor well") located near the home garden.
Receptor Well Location
One of the key inputs for EPACMTP is the receptor well location. EPACMTP estimates
the exposure concentration at the intake point of a hypothetical residential drinking water well
located at a specified distance from the downgradient edge of the source area and at a specified
depth below the water table. For this analysis, modeling simulated groundwater impacts to a well
assumed to be placed in the centerline of the plume at a fixed distance of 1 m from the edge of
the garden. The depth of the well was varied uniformly throughout the aquifer thickness, to a
maximum of 10 m, whichever was less. That is, the well depth was never allowed to exceed 10
m below the water table. This limitation for well depth, used in several previous EPA analyses, is
applied primarily for two reasons: (1) to be representative of typical residential well scenarios
where wells are generally shallow because of the higher cost of drilling a deeper well and (2) to
produce a conservative estimate of risk (because the infiltration rate is generally lower than the
groundwater seepage velocity, groundwater plumes tend to be relatively shallow).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-16
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Key Assumptions
The groundwater modeling approach included the following key assumptions. More
comprehensive documentation of the EPACMTP model and associated assumptions are available
in the EPACMTP Technical Background Document (U.S. EPA, 2003f).
• The model assumes that the vertical migration is 1-D and that transverse dispersion is
negligible in the unsaturated zone.
• The model assumes linear and non-linear equilibrium sorption and homogeneous aquifer
conditions. However, as discussed in Appendix J, linearity has been demonstrated for
the SFS arsenic leachate concentrations modeled in this analysis.
• The model assumes that receptors use the uppermost aquifer, rather than a deeper aquifer,
as their drinking water source. This assumption could overestimate risks in cases in
which the uppermost aquifer is not used.
• The model assumes that long-term average conditions are sufficient and that shorter
frequency fluctuations (e.g., in rainfall/infiltration) are insignificant in estimating
long-term risk.
Preferential flow in karst aquifers or in fractures was not considered, although such
conditions are known to exist over broad areas. Preferential flow can allow contamination to
migrate faster and at a higher concentration than in a standard porous medium. However, the
contamination typically does not spread over such a broad area. As a result, the modeling may
under- or overestimate the concentrations in groundwater.
5.3.5.2 Groundwater Model Inputs
EPACMTP requires a number of input parameters. Provided below is a summary of the
key types of EPACMTP inputs and how they were parameterized in the SFS evaluation.
The leachate fluxes (g m"2 • yr) estimated by the home garden source model were used as
inputs to EPACMTP to estimate arsenic concentrations at the receptor well. All leachate fluxes
from the source model were applied uniformly over the footprint of the garden, immediately
below the garden.
To model the unsaturated zone, EPACMTP requires inputs for the following soil-related
hydrological parameters: saturated hydraulic conductivity, van Genuchten soil moisture
parameters, residual and saturated water contents, percent organic matter, and soil bulk density.
Values for these parameters vary, and EPACMTP includes distributions of appropriate values
organized by soil texture. EPACMTP requires a site-specific soil texture be input in order to
determine which soil-related hydrologic parameter distributions will supply the unsaturated zone
model input parameters. A pre-sampled distribution of saturated hydraulic conductivity (a
particularly important variable) was shared by the home garden source model and the
unsaturated zone model.
Similarly, the hydrogeological setting assigned to each garden was used to select
appropriate aquifer conditions from EPACMTP's Hydrogeologic DataBase (HGDB). Given an
aquifer code setting for a garden site, a correlated sample of key aquifer model input parameters
(hydraulic conductivity, hydraulic gradient, depth to the water table, and saturated thickness) was
selected from a population of samples taken from similar hydrogeological settings. Details of the
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-17
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
data used to parameterize the unsaturated zone and the development and use of the HGDB are
given in the EPACMTP Parameters/Data BackgroundDocument (U.S. EPA, 2003b).
Recharge is water percolating through the soil to the aquifer outside the footprint of the
garden. EPACMTP selects a recharge rate using a meteorological station assignment (based on
the geographic location and topography of a garden setting) and by the garden's associated soil
texture. Using the soil texture and station assignment, a recharge rate is selected from a database
of the Hydrologic Evaluation of Landfill Performance (HELP) model-derived recharge rates for
climate stations across the country and for various soil textures. Further details about how these
rates are determined and other options for determining recharge rates outside of the EPACMTP
model can be found in the EPACMTP Parameters/Data Background Document (U.S. EPA,
2003b). A few required inputs are based upon established empirical distributions and are
described in the EPACMTP Parameters/Data Background Document (U.S. EPA, 2003b).
5.3.5.3 Groundwater Model Outputs
EPACMTP's outputs (i.e. predictions of the contaminant concentrations arriving at a
downgradient receptor location) are time-dependent; they can vary over time. The model can
calculate both the peak concentration arriving at the well and maximum time-averaged
concentrations. The SFS Evaluation used maximum time-averaged concentrations (based on the
exposure duration for each receptor type) to develop human risk estimates.
In some cases, it may take a long time for the plume to reach the receptor well, and the
maximum groundwater exposure may not occur until a very long time after the application. This
time delay may be on the order of thousands of years. If the model predicts that the maximum
exposure will not have occurred after 10,000 years, the actual receptor concentration at 10,000
years will be used in the risk calculations.
An analysis was performed to evaluate anticipated arrival times to determine if the
exposure through the soil/produce pathway would overlap with exposure through the
groundwater pathway. To determine the approximate timeframe when the peak groundwater
exposure might occur, estimates were made of the time at which the contaminant plume would
arrive at the receptor well and the time when the contaminant plume would finish passing the
well. Arrival of peak concentrations would only occur somewhere within this time period. These
estimates were based upon two additional outputs from the unsaturated zone transport
simulation: 1) first arrival time of leachate at the water table and 2) cessation time of leachate
arrival at the water table. Retardation effects were used to account for horizontal travel to the
receptor well. The results of this analysis are summarized in Table 5-3.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-18
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-3. EPACMTP Arrival Times of Arsenic Plume at the Receptor Well
Percentile
90%
80%
70%
60%
50%
40%
30%
20%
10%
Arrival Time Zone (year)
Beginning
29
61
100
150
201
203
207
229
398
End
200
200
202
220
272
345
457
663
1,112
Based on the analysis (see Appendix J for more details), it is unlikely that peak
soil/produce pathway exposures and peak groundwater exposures will occur within the same
timeframe. For example, the earliest estimated timeframe for groundwater arrival of arsenic from
the garden spanned from 29 to almost 400 years following the application of the SFS. It is
therefore likely that the peak well concentrations will not occur until well past the timeframe for
peak soil/produce pathway exposures, and perhaps even past the timeframe of residency (i.e.,
exposure duration of the gardeners who originally applied the SFS-manufactured soil).
Therefore, separate screening levels were developed for the groundwater and soil/produce
pathways.
5.3.6 Fate and Transport: Produce Modeling
The food chain model calculates constituent concentrations in food items using soil
concentrations and emissions predicted by the source model and using air concentrations and
deposition rates from the dispersion model. Constituents pass from contaminated soil and air
through the food chain to the gardening family. For example, constituents in air may be
deposited on plants growing in the garden. Simultaneously, these plants may take up constituents
from the soil and accumulate constituents from both routes in the fruits and vegetables consumed
by the receptors.
This section presents the methodology used to calculate constituent concentrations in the
aboveground and belowground produce grown in the residential garden.
5.3.6.1 Conceptual Produce Model
The human food chain model is designed to predict the accumulation of a constituent in
the edible parts of food crops eaten by the human receptor. Edible crops include exposed and
protected fruits, exposed and protected vegetables, and root vegetables. The term "exposed"
refers to the fact that the edible portion of the produce is exposed to the atmosphere. The term
"protected" refers to the fact that the edible portion of the produce is shielded from the
atmosphere (i.e., not impacted by air-to-plant transfer and particle deposition). Examples of the
categories include tomatoes (exposed vegetable), corn (protected vegetable), apples (exposed
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-19
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
fruit), oranges (protected fruit), and potatoes (root vegetables). The equations used to calculate
the food chain concentrations of constituents are presented in Appendix H.
5.3.6.2 Produce Model Inputs
The inputs to the food chain model are vegetation-specific properties, soil and air
concentrations, deposition rates, and other chemical-specific properties, such as bio-uptake
factors. Estimation of soil concentrations is discussed in Section 5.3.4. The following identifies
the additional input parameters that are needed to calculate constituent concentrations in
aboveground and belowground (i.e., root vegetables) produce:
• Aboveground produce. Concentrations in aboveground produce consider impacts due to
air-to-plant transfer, root uptake, and particle deposition. Exposed fruits and vegetables
are susceptible to contamination through all three mechanisms, while protected
vegetation is assumed to be impacted only through root uptake. The vegetation-specific
parameters used in calculating these impacts are presented in Appendix H. The air-to-
plant and root uptake factors for each constituent are identified in Appendix F.
• Belowground produce. Concentrations in belowground produce consider impacts due to
root uptake, which is calculated for metals using chemical-specific soil-to-plant
bioconcentration factors. These chemical-specific factors are presented in Appendix F.
• Conversion to Wet Weight (WW). The implemented equations predict aboveground
and belowground concentrations on a dry weight basis. The model must convert these
values to a wet-weight basis for use in the downstream exposure model, which applies
wet-weight consumption rates. As shown in Appendix H, this conversion is made using
plant-specific moisture adjustment factors (MAFs) (i.e., percent moisture). These factors,
which vary by vegetation type, are identified in Appendix H.
5.3.6.3 Produce Model Outputs
The food chain model outputs constituent-specific concentrations in exposed and
protected fruits, exposed and protected vegetables, and root vegetables. These concentrations
serve as input to the exposure model, where they are combined with human consumption rates
and other exposure factors to calculate an individual's ingested dose.
5.3.7 Human Exposure Modeling
The predicted constituent concentrations in soil, drinking water, and food chain items are
used to estimate human exposures. This section describes the human exposure modeling that was
performed to estimate exposure based on the potential dose ingested. Appendix H presents the
equations used to calculate dose for each pathway and for total ingestion.
5.3.7.1 Human Exposure Conceptual Model
Exposure through the ingestion route was estimated by multiplying the concentration of
the constituent in the soil, drinking water, or food item by the consumption rate of the individual.
This is the average daily dose (ADD) for an individual. Calculation of a lifetime average daily
dose (LADD) for constituents with cancer endpoints also considers the individual's exposure
duration, averaging across an assumed lifetime (70 yr), and exposure frequency (350 d yr"1).
Appendix H presents the equations used to calculate ADD and LADD.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-20
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Exposure modeling relies heavily on default assumptions concerning population activity
patterns, mobility, dietary habits, body weights, and other factors. The following highlights the
key assumptions that were applied in estimating the level of constituents that the hypothetical
home gardener and child were exposed to via ingestion of soil and homegrown aboveground and
belowground produce.
• Both the adult and child members of the family were exposed to constituents through the
application of SFS-manufactured soil to their own home garden. The exposure period for
the receptors was constrained to begin at the time of application of the soil to the garden.
• The adult was 20 years old when exposure began, and the child was 1 year of age when
exposure began. Application of these start ages maintains the conservative nature of this
screening assessment. Infant exposure (i.e., 0 to 1 year of age) via the breastmilk pathway
was not evaluated under this modeling scenario given that none of the metals included in
the probabilistic modeling phase have been identified in current studies as being of
significant concern via the breastmilk pathway.
• Receptors both lived and worked at the exposure location. This assumption may
overestimate exposure, because individuals may live at the exposure location, but
commute to work (or school or daycare) outside of the study area, or commute to areas
within the study area where SFS-manufactured soil had not been used.
• In the case of incidental soil ingestion, the EPA's default relative bioavailability (KB A)
value of 60% (U.S. EPA, 2012b) was used to adjust the distribution of arsenic
concentration in soil for the exposure modeling. All other constituents were assumed to
be 100%bioavailable.
5.3.7.2 Human Exposure Model Inputs
The inputs to the exposure model are human exposure factors and soil, drinking water,
and food concentrations. Estimation of soil, drinking water, and food item concentrations is
discussed in Sections 5.3.4, 5.3.5 and 5.3.6, respectively. The key human exposure factors used
as inputs to the analysis include the following:
• Averaging time for carcinogens
• Exposure duration
• Exposure frequency
• Ingestion rate for soil
• Ingestion rate for drinking water
• Consumption rates for exposed vegetables, protected vegetables, exposed fruit, protected
fruit, root vegetables
• Fraction food preparation loss for exposed vegetables, protected vegetables, exposed
fruit, protected fruit, root vegetables.
These exposure factors were used to calculate the dose for the soil and produce ingestion
pathways. The primary data source of human exposure model inputs used in this analysis was
EPA's Exposure Factors Handbook (EFH; U.S. EPA, 2011) and Child-Specific Exposure
Factors Handbook (CSEFH; U.S. EPA, 2008a). These references summarize data on human
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-21
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
behaviors and characteristics related to human exposure from relevant key studies and provide
recommendations and associated confidence estimates on the values of the exposure factors.
These data were carefully reviewed and evaluated for quality before being included in the EFH
and CSEFH. EPA's evaluation criteria included peer review, reproducibility, pertinence to the
United States, currency, adequacy of the data collection period, validity of the approach,
representativeness of the population, characterization of the variability, lack of bias in study
design, and measurement error (U.S. EPA, 2011). Table 5-4 characterizes the distributions of
consumption rates for produce items and drinking water, as well as the distributions of body
weights and exposure durations used in this analysis. Table 5-5 identifies the exposure
parameters, including soil ingestion, that were fixed at constant values in this analysis.
Table 5-4. Produce and Drinking Water Consumption Rate (CR), Body Weight, and
Exposure Duration Distributions for the Home Gardener
Age
Distribution
Type
Mean
(or Shape)3
Std Dev
(or Scale)3
Minimum
Maximum
Reference 3
Exposed Fruit (g [WW] kg * body weight d'1)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Gamma
Lognormal
Lognormal
Lognormal
1.43E+00
2.78E+00
1.54E+00
1.57E+00
1.58E+00
5.12E+00
2.44E+00
2.3E+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
1.60E+01
3.60E+01
1.80E+01
1.29E+01
U.S. EPA (2011);
Table 13-58
Exposed Vegetables (g [WW] kg * body weight d"1)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Gamma
Lognormal
Gamma
Weibull
9.70E-01
1.64E+00
9.10E-01
1.57E+00
2.62E+00
3.95E+00
1.19E+00
1.76E+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
2.10E+01
2.70E+01
1.10E+01
1.03E+01
U.S. EPA (2011);
Table 1 13-60
Protected Fruit (g [WW] kg * body weight d'1)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Gamma
Gamma
Gamma
Lognormal
7.37E-01
7.37E-01
7.36E-01
6.63E+00
1.59E+01
8.15E+00
3.56E+00
1.57E+01
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
4.50E+01
2.60E+01
3.80E+01
4.73E+01
U.S. EPA (2011);
Table 13-59
Protected Vegetables (g [WW] kg * body weight d'1)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Lognormal
Lognormal
Lognormal
Lognormal
1.88E+00
1.07E+00
7.70E-01
1.01E+00
1.98E+00
1.04E+00
6.90E-01
1.19E+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
1.60E+01
8.00E+00
6.00E+00
6.49E+00
U.S. EPA (2011);
Table 13-61
Root Vegetables (g [WW] kg-1 body weight d'1)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Lognormal
Weibull
Weibull
Weibull
2.31E+00
6.80E-01
8.40E-01
1.15E+00
6.05E+00
1.06E+00
9.10E-01
1.32E+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
4.10E+01
1.50E+01
9.00E+00
7.47E+00
U.S. EPA (2011);
Table 13-62
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-22
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Age
Distribution
Type
Mean
(or Shape)3
Std Dev
(or Scale)3
Minimum
Maximum
Reference 3
Drinking Water Ingestion (mL kg"1 body weight d"1'
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Weibull
Weibull
Weibull
Weibull
1.15E+00
1.14E+00
1.08E+00
1.16E+00
2.56E+01
1.75E+01
1.14E+01
1.66E+01
2.23E-03
2.23E-03
2.23E-03
l.OOE-02
1.86E+02
1.86E+02
1.86E+02
1.26E+02
U.S. EPA (2008a);
Table 3-19
U.S. EPA (2011)
Table 3-38
Body Weight (kg)
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
Lognormal
Lognormal
Lognormal
Lognormal
1.55E+01
3.07E+01
5.82E+01
7.12E+01
2.05E+00
5.96E+00
1.02E+01
1.33E+01
4.00E+00
6.00E+00
1.30E+01
1.50E+01
5.00E+01
2.00E+02
3.00E+02
3.00E+02
U.S. EPA (2011);
Table 8-3
Exposure Duration (yr)b
Child (1-19 yrs)
Adult (20-69 yrs)
Weibull
Weibull
1.32E+00
1.34E+00
7.06E+00
1.74E+01
l.OOE+00
l.OOE+00
3.80E+01
5.00E+01b
U.S. EPA (2011);
Table 16-109
a Shape and scale are presented for Gamma and Weibull distributions.
b Exposure duration was capped at 50 years so it would never exceed the 70-year lifetime assumption
implicit in the averaging time used, given the starting age of 20 years.
Table 5-5. Summary of Exposure Parameters with Fixed Values
Used in Probabilistic Analysis
Parameter
Averaging time for carcinogens
Exposure frequency
Fraction food preparation loss: exposed fruit
Fraction food preparation loss: exposed vegetables
Fraction food preparation loss: protected fruit
Fraction food preparation loss: protected vegetables
Fraction food preparation loss: root vegetables
Fraction contaminated: drinking water
Fraction contaminated: soil
Ingestion rate: soil (adult)
Ingestion rate: soil (child 1, child 2, child 3)
Units
yr
dyr1
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
Fraction
mgd"1
mgd'1
Constant
Values
7.00E+01
3.50E+02
2.10E-01
1.61E-01
2.90E-01
1.30E-01
5.30E-02
l.OOE+00
l.OOE+00
5.00E+01
l.OOE+02
Reference
U.S. EPA(1991a)
U.S. EPA(1991a)
U.S. EPA (20 11);
Table 13-69
U.S. EPA Policy
U.S. EPA Policy
U.S. EPA (1997c);
Table 5-1
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-23
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
The conservative nature of the distributions used to estimate home gardener adult and
child consumption rates could result in overly conservative consumption rates of home-grown
produce. Two additional sets of runs were therefore added for comparison: one using point
estimates of 50th percentile annual produce consumption rates for the general population,
multiplied by 50% to account for crop growth periods and climate limitations to crop harvest
periods (reducing the effective consumption rate to home-grown produce); and a set of runs
using the 90th percentile annual produce consumption rates for the general population, similarly
multiplied by 50%. All other distributions and constant values were the same. Thus, the three
sets of runs are as follows:
• Set 1: Home gardener, modeled distributions of consumption rates (for home gardeners)
—the produce consumption rates specific to home-grown produce;
• Set 2: General population, 50th percentile (for the general population) consumption rates
—the median produce consumption rates for the general population were multiplied by
0.5 to derive a value specific to home-grown produce; and
• Set 3: General population, 90th percentile (for the general population) consumption rates
—the high produce consumption rates for the general population were multiplied by 0.5
to derive a value specific to home-grown produce.
Table 5-6 identifies the 90th percentile home gardener produce consumption rates, and
the general population median and high produce consumption rates that were used in the
additional runs. Evaluation of the groundwater pathway did not require the development of
different drinking water consumption rate datasets for each population type; it was assumed that
the general population receptor and the home gardener receptors would ingest drinking water at
consistent rates.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-24
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-6. Summary of Produce Consumption Rates (CR)
(g [WW] produce kg"1 body weight d"1)
Age
Home Gardener
Estimates
90%-ile
General Population Estimates a
Median
High
Exposed Fruit
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
5.41
6.98
3.41
5.00
1.95
1.10
0.44*
0.32*
10.62
3.15
1.45
1.06
Exposed Vegetables
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
6.43
3.22
2.35
6.01
0.32
0.30
0.27
0.45
2.48
1.70
1.25
1.63
Protected Fruit
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
13.00
6.92
7.44
15.00
2.70
0.17
1.80
0.93
7.19
4.05
2.70
2.09
Protected Vegetables
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
3.05
2.14
1.85
3.55
0.63*
0.39*
0.23*
0.27*
1.93
1.30
0.75
0.85
Root Vegetables
Child 1-5 yrs
Child 6-1 lyrs
Child 12-19 yrs
Adult (20-69 yrs)
5.72
3.83
2.26
3.11
0.72
0.50
0.41
0.35
3.01
2.10
1.50
1.29
SOURCE: Values derived from EPA's Exposure Factors Handbook (U.S. EPA, 2011).
a The listed general population values are the general population consumption rates listed in
U.S. EPA (2011) multiplied by 0.5 to derive a value specific to home-grown produce.
* Based on mean values.
5.3.7.3 Human Exposure Model Outputs
The outputs from the exposure model are receptor- and pathway-specific ADDs for
constituents with noncancer endpoints, and LADDs for constituents with cancer endpoints. As
discussed in Section 5.3.1, each model run generated an ADD/LADD for each of the exposure
pathways (i.e., separate ADDs/LADDs for exposure from ingestion of soil, exposed fruits,
exposed vegetables, etc). Each model run also combined the pathway-specific ADDs/LADDs
into a "Total Ingestion" ADD/LADD.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-25
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Running the model probabilistically generated distributions of exposure values for each
pathway, as well as a distribution of Total Ingestion values. Table 5-7 thru 5-10 list pathway-
specific and Total Ingestion values taken from example runs that generated the 50th and 90th
percentile Total Ingestion values.
Table 5-7. Example 50th Percentile Adult Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg'1 d'1)
Constituent
Pathway
Home Gardener
RunID
ADD/
LADD
General Population
Median Consumption
Rates
RunID
ADD/
LADD
High Consumption
Rates
RunID
ADD/
LADD
Cancer
As
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Groundwater
4772
2.0E-08
1.1E-07
9.2E-08
5.8E-08
7.5E-08
2.5E-08
3.7E-07
8883
3.6E-08
LIE-OS
3.7E-08
8.1E-09
9.1E-09
2.3E-08
1.2E-07
7041
4.8E-08
3.6E-08
1.4E-07
3.8E-08
3.3E-08
8.7E-08
3.8E-07
PI
Noncancer
Co
Fe
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
569
959
3.6E-07
8.0E-07
2.6E-06
6.2E-07
8.4E-07
2.7E-06
7.9E-06
5.7E-07
1.3E-07
4.5E-07
5.3E-08
2.7E-07
l.OE-07
1.6E-06
5410
1301
3.4E-07
3.1E-07
5.9E-07
2.2E-07
2.5E-07
7.8E-07
2.5E-06
2.3E-07
4.1E-08
1.1E-07
3.0E-08
3.4E-08
1.5E-07
5.9E-07
509
7952
2.3E-07
9.8E-07
2.2E-06
l.OE-06
8.4E-07
3.0E-06
8.2E-06
3.4E-07
1.2E-07
3.7E-07
1.3E-07
l.OE-07
5.0E-07
1.6E-06
PI = Pathway incomplete (constituent does not reach receptor well during simulation)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-26
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-8. Example 50th Percentile Child Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg'1 d"1)
Constituent
Pathway
Home Gardener
RunID
ADD/
LADD
General Population
Median Consumption
Rates
RunID
ADD/
LADD
High Consumption
Rates
RunID
ADD/
LADD
Cancer
As
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Groundwater
5114
1.7E-07
6.9E-08
4.9E-08
1.9E-07
3.2E-08
2.9E-08
5.4E-07
5208
1.8E-07
1.6E-08
1.6E-08
1.5E-08
3.5E-08
3.0E-08
2.9E-07
2701
1.7E-07
5.0E-08
1.3E-07
7.9E-08
1.8E-07
1.3E-07
7.4E-07
PI
Noncancer
Co
Fe
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
495
7672
5.4E-06
6.8E-07
7.5E-06
4.8E-07
1.7E-06
8.3E-07
1.7E-05
3.7E-06
1.4E-07
4.1E-07
3.6E-07
3.8E-07
5.6E-07
5.5E-06
3059
6883
3.5E-06
6.9E-07
4.0E-07
6.3E-07
1.5E-06
1.6E-06
8.2E-06
3.0E-06
9.3E-08
8.5E-08
7.8E-08
2.5E-07
3.0E-07
3.8E-06
9733
2508
4.3E-06
1.9E-06
2.9E-06
3.1E-06
7.6E-06
6.0E-06
2.6E-05
2.8E-06
3.1E-07
6.4E-07
4.9E-07
1.2E-06
1.3E-06
6.8E-06
PI = Pathway incomplete (constituent does not reach receptor well during simulation)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-27
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-9. Example 90th Percentile Adult Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg'1 d"1)
Constituent
Pathway
Home Gardener
RunID
ADD/
LADD
General Population
Median Consumption
Rates
RunID
ADD/
LADD
High Consumption
Rates
RunID
ADD/
LADD
Cancer
As
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Groundwater
7831
9716
1.6E-07
7.7E-08
7.2E-07
1.3E-07
8.7E-08
6.7E-07
1.8E-06
2.1E-07
1770
8.5E-08
5.2E-08
1.7E-07
3.8E-08
4.2E-08
1.1E-07
5.0E-07
3447
6.6E-08
1.7E-07
6.5E-07
1.8E-07
1.5E-07
4.1E-07
1.6E-06
Same as Home Gardener
Noncancer
Co
Fe
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
5661
9766
9.4E-08
4.8E-07
1.9E-08
1.5E-05
1.1E-06
9.4E-07
1.8E-05
4.4E-07
1.4E-07
2.3E-06
2.0E-07
1.1E-07
7.8E-09
3.2E-06
5260
5677
6.8E-07
3.2E-07
6.1E-07
2.3E-07
2.6E-07
8.2E-07
2.9E-06
5.7E-07
4.7E-08
1.4E-07
3.4E-08
5.7E-08
1.7E-07
l.OE-06
9534
4181
4.9E-07
l.OE-06
2.3E-06
1.1E-06
9.5E-07
3.1E-06
8.9E-06
5.5E-07
1.4E-07
4.9E-07
1.5E-07
1.8E-07
6.0E-07
2.1E-06
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-28
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-10. Example 90th Percentile Child Unitized Doses for SFS-Manufactured Soil
Constituents—Total Ingestion Pathway (mg kg'1 d"1)
Constituent
Pathway
Home Gardener
RunID
ADD/
LADD
General Population
Median
Consumption Rates
RunID
ADD/
LADD
High Consumption
Rates
RunID
ADD/
LADD)
Cancer
As
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Groundwater
4734
4302
3.2E-07
l.OE-07
1.6E-07
3.1E-07
1.8E-07
6.5E-08
1.1E-06
2.5E-07
2116
2.2E-07
3.7E-08
6.1E-08
3.0E-08
6.6E-08
7.8E-08
5.0E-07
1692
1.4E-07
l.OE-07
2.8E-07
2.2E-07
2.5E-07
2.8E-07
1.3E-06
Same as Home Gardener
Noncancer
Co
Fe
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
5049
3020
6.8E-06
4.6E-06
3.3E-06
6.7E-07
2.9E-06
1.2E-05
3.1E-05
4.0E-06
3.1E-07
4.0E-06
6.1E-08
5.4E-07
6.0E-07
9.6E-06
8674
4792
6.8E-06
6.7E-07
4.3E-07
5.1E-07
1.4E-06
1.6E-06
1.1E-05
6.0E-06
l.OE-07
8.5E-08
9.6E-08
2.3E-07
3.3E-07
6.8E-06
4005
7537
6.2E-06
2.3E-06
3.4E-06
3.7E-06
9.0E-06
7.1E-06
3.2E-05
6.1E-06
3.3E-07
6.9E-07
5.3E-07
1.3E-06
1.4E-06
l.OE-05
It is important to note that the pathway-specific values listed in Tables 5-7 thru 5-10 are
those which, when totaled, result in the 50th (or 90th) percentile Total Ingestion ADD/LADD.
Each pathway-specific value is not necessarily the 50th (or 90th) percentile value for that
individual pathway. For example, in the distribution of child Total Ingestion LADDs for arsenic
based on home gardener ingestion rates, the 50th percentile value (i.e., the Total Ingestion LADD
at the exact center of the distribution) was generated in model run 5114 (see Table 5-8). This
Total Ingestion LADD includes an LADD of 6.9E-08 mg kg"1 d"1 from ingestion of protected
vegetables. However, in the distribution of child LADDs for arsenic specific to ingestion of
protected produce, the 50th percentile LADD of 1.1E-08 mg kg"1 d"1 was generated in model run
8883. Pathway-specific 50th and 90th percentile ADDs/LADDs for adult and child receptors
(including the probabilistic runs that generated them) are listed in Appendix K, Tables K-l
through K-4. Example Total Ingestion 50th and 90th percentile ADDs/LADDs for adult and child
receptors, including their respective pathway-specific contributions and the probabilistic runs
that generated them, are listed in Appendix K, Tables K-5 through K-8.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-29
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
These ADDs/LADDs are used as input to the human health effects model, as discussed in
Section 5.3.10
5.3.8 Ecological Exposure Modeling
The following sections describe the ecological exposure modeling. Section 5.3.8.1
provides an overview of the conceptual model, including the basic approach and assumptions.
Section 5.3.8.2 discusses the input parameters and values used in this risk analysis. Section
5.3.8.3 discusses the model outputs.
5.3.8.1 Ecological Conceptual Exposure Model
As described in Section 5.3.3, ecological receptors could be exposed to SFS constituents
via direct contact with soil. Depending on the receptor (i.e., plants, soil invertebrates, or small
mammals), ecological exposure was estimated by adjusting the concentration of the constituent
in soil to reflect the phyto-available fraction or the receptor's home range.
Exposure modeling relies heavily on default assumptions concerning population activity
patterns, mobility, dietary habits, body weights, and other factors. For example, Phase I
screening assumed that 100% of SFS-bound metals were bioavailable to ecological receptors.
This assumes that SFS-bound metals are equally available to biological systems as soluble metal
salts added to soils in laboratory studies. Phase I screening also assumed that animals received
100% of their diet from the home garden; they do not forage or feed beyond the boundaries of
the garden. Both of these assumptions are upper bound estimates that are reasonable for a
screening analysis.
One function of refined probabilistic modeling is to replace upper bound estimates with
more realistic conservative inputs. The key assumptions that were applied in refined ecological
exposure modeling include:
• Plants were grown in the home garden, and therefore 100% of the soil they were exposed
to was SFS-manufactured soil. However, soil concentrations were adjusted to reflect the
soluble, and therefore phyto-available, fraction of SFS constituents (see Section 5.3.8.2
for a more detailed discussion of this assumption).
• Soil invertebrates spend their entire lives in home garden soils.
• As a highly exposed species, the short-tailed shrew was the surrogate species used to
derive the Eco-SSL for mammals, and evaluated for potential adverse impacts.
Constituent soil concentrations were adjusted to reflect the fraction of shrew diet to come
from the garden (see Section 5.3.8.2 for a more detailed discussion of this assumption).
5.3.8.2 Ecological Exposure Model Inputs
The inputs to the ecological exposure model are soil concentrations and ecological
exposure factors. Estimation of soil concentrations is discussed in Section 5.3.4. The key
ecological exposure factors used as inputs to the analysis include the following factors.
Plant Toxicity
Manganese and nickel were retained for further study in Phase II due to the potential for
phyto-toxicity. Because the toxicity of metals is dependent on the soluble soil fraction, the risk
posed to terrestrial plants will be directly related to the amount of metal that can desorb from
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-30
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
SFS particles and become available in the soluble fraction. In her review of plant responses to
metal toxicity, Reichman (2002) noted that:
The total metal concentration of a soil includes all fractions of a metal, from the
readily available to the highly unavailable. Other soil factors, such aspH, organic
matter, clay andredox conditions, determine the proportion of total metal which is
in the soil solution. Hence, while total metal provides the maximum pool of metal
in the soil, other factors have a greater importance in determining how much of
this soil pool will be available to plants (Wolt, 1994). In addition, researchers have
found that while total metal correlates with bioavailable soil pools of metal, it is
inadequate by itself to reflect bioavailability (Lexmond, 1980; Sauve et al, 1996;
McBride etal, 1997; Sauve etal, 1997; Peijnenburg etal, 2000).
Lacking empirical data on the soluble fraction of metals in SFS-amended soil, this
evaluation used SFS sample-specific pore water concentrations as a surrogate to develop
estimates of the soluble (and therefore bioavailable) fraction in soil. This approach defines the
constituent-specific bioavailable fractions as the ratio of SFS sample-specific pore water
concentrations to corresponding total concentrations (see Appendix B Tables B-26 and B-19).
The empirical distributions of the "pore water/total" ratios establishes a reasonable range for the
bioavailable fraction. The 95th percentile of the ratio range (i.e., an estimate of the bioavailable
fraction that is higher than 95 percent of other estimates) was used as a reasonably conservative
estimate of the bioavailable fraction. Therefore, the maximum soil concentrations for manganese
and nickel would be adjusted by a fraction of 0.10 and 0.07, respectively. In effect, this
adjustment estimates that the majority of manganese and nickel is in a solid form unavailable for
plant uptake. That is, only a fraction of the metals found in SFS-amended soil behaves similarly
to the metals added in spiked soil studies (e.g., soluble metal salts).
Dietary Exposure to Mammals
Antimony, chromium, and copper were retained for further study in Phase II due to the
potential for toxicity to small insectivorous mammals (based on studies for the short tailed
shrew). The area of the home garden (i.e. 405 m2) may be substantially less than the home range
for the shrew. In developing the ecological risk assessment methodology for 3MRA, EPA
determined that it was reasonable to prorate exposures based on a comparison between the
"habitat" (i.e., the area in which the material is managed - the home garden in the SFS
evaluation), and the median home range for the animal so that dietary exposure was not grossly
overestimated. This methodology was reviewed and approved by EPA's Science Advisory Board
in 2003, as a reasonable method to account for the spatial heterogeneity in animals' use of
feeding and foraging areas.44 The same method is used in this risk assessment to avoid the
unrealistic and overly conservative assumption that 100% of the shrew diet comes from the home
garden.
Information on home ranges of species was reviewed for northern, southern, Adirondack,
Sherman's, and Elliot's short-tailed shrews (ADCNR, 2008; FFWCC, 2013; Getz and McGuire,
2008; KBS, 2014; MNHP, 2014; Saunders, 1988; U.S. EPA, 1993 and 2002; VDGIF, 2014). The
short-tailed shrew diet consists primarily of insects, earthworms, slugs, and snails, while plants,
44 The SAB review report is available at
http://vosemite.epa.gov/sab/sabproduct.nsf/95eac6037dbee075852573a00075f732/99390efbfc255ae885256fFe005
79745/$FILE/SAB-05-003 unsigned.pdf
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-31
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
fungi, millipedes, centipedes, arachnids, and small mammals also are consumed (U.S. EPA,
1993b). The literature on short-tailed shrews noted that these animals can be found in a wide
variety of habitats, although areas with litter/grass cover (e.g., forest, wetlands) and high
moisture levels are clearly preferred (Miller and Getz, 1977; van Zyll de Jong, 1983). A variety
of factors that influence the home range and habitat preference for short-tailed shrews were
identified; for example, the availability of prey, season, and reproductive status were shown to
influence movement and home ranges for short-tailed shrews in east-central Illinois (Getz and
McGuire, 2008). Figure 5-7 presents the median home range values identified in that review,
ranging from 0.06 to 6.2 acres with a median (of the medians) of 2.4 acres (9700 m2), and a 10th
percentile value of 0.7 acres (2800 m2). The variability in results shown in Figure 5-7 suggests
that the species, as well as the geographical location, has a significant influence on the home
range and movement (a surrogate for foraging behavior) for the short-tailed shrew.
Short-tailed Shrew
Median Home Ranges
0)
b
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
...Mill
Figure 5-7. Analysis of Home Range Sizes for the Short Tailed Shrew.
Comparing the home garden area of 0.1 acres (405 m2) to the 10th percentile value for
home ranges shown in Figure 5-7, 0.7 acres (2800 m2) attributes roughly 15% of the short-tailed
shrew diet to the home garden. As a consequence, a fraction of 0.15 was assumed for all three
COCs to reflect the percentage of diet likely to come from the home garden.
5.3.8.3 Ecological Exposure Model Outputs
The outputs from the ecological exposure model are distributions of predicted receptor-
and constituent-specific soil concentrations adjusted to reflect bioavailability and mammal home
range. Table 5-11 lists the 50th and 90th modeled soil concentrations, adjustment factors, and
adjusted soil concentrations.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-32
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-11. 50th and 90th Percentile Ecological Exposure Model Outputs for SFS-
Manufactured Soil Constituents
Constituent
Terrestrial Plants
Modeled
Soil Cone.
(mg kg-1)
Adjustment
Factor
(unitless)
Adjusted
Soil Cone.
(mg kg-1)
Soil Invertebrates
Modeled
soil cone.
(mg kg-1)
Adjustment
Factor
(unitless)
Adjusted
Soil Cone.
(mg kg-1)
Mammals
Modeled
soil cone.
(mg kg-1)
Adjustment
Factor
(unitless)
Adjusted
Soil Cone.
(mg kg-1)
50th percentile
Cr (III)
Cu
Mn
Ni
Sb
—
0.90
0.93
0.92
—
—
NA
0.10
0.07
—
—
0.90
0.093
0.064
—
—
0.90
0.93
0.92
0.82
—
NA
NA
NA
NA
—
0.90
0.93
0.92
0.82
0.94
0.90
0.93
0.92
0.82
0.15
0.15
NA
NA
0.15
0.14
0.13
0.93
0.92
0.12
90th percentile
Cr(III)
Cu
Mn
Ni
Sb
—
0.97
0.97
0.97
—
—
NA
0.10
0.07
—
—
0.97
0.097
0.068
—
—
0.97
0.97
0.97
0.96
—
NA
NA
NA
NA
—
0.97
0.97
0.97
0.96
0.98
0.97
0.97
0.97
0.96
0.15
0.15
NA
NA
0.15
0.15
0.15
0.97
0.97
0.14
The adjusted soil concentrations are used as input to the ecological effects model
described in Section 5.3.10.
5.3.9 Human Health Effects Modeling
This section presents the human health benchmarks and the modeling approach used to
estimate potential health hazards. Section 5.3.9.1 provides an overview of the conceptual model,
including the basic approach and assumptions. Section 5.3.9.2 discusses the input parameters
and values used in this hazard analysis. Section 5.3.9.3 discusses the model outputs. The hazard
equations used in the human health effects modeling are presented in Appendix H.
5.3.9.1 Human Health Effects Conceptual Model
Human health effects modeling was performed to estimate cancer and noncancer health
impacts due to ingestion of soil and home-grown produce. A chemical constituent's ability to
cause an adverse health effect depends on the toxicity of the particular constituent, the route of
exposure, the duration and intensity of exposure, and the resulting dose that an individual
receives. The human health benchmarks used in this assessment were compared to the ADD for
noncarcinogens or the LADD for carcinogens. For constituents with noncancer endpoints, the
health benchmark was the RfD. For constituents with cancer endpoints, the health benchmark
was the dose that yields a cancer risk level of 10~5 (1 in 100,000) over a lifetime (calculated as
10"5/oral cancer slope factor [CSF]). The ratio of the ADD or LADD to the health benchmark
(shown below) is referred to as a Unitized Dose Ratio (UDR) and was used to establish a
threshold of concern for a specific health effect. The level of concern established by EPA for this
analysis is a UDR of 1.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-33
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
1T^ ADDorLADD
Umtized Dose Ratio =
Health Benchmark
where
Umtized Dose Ratio = Comparison of exposure dose to benchmark dose (unitless)
For noncarcinogens:
ADD = Average daily dose (mg kg"1 d"1)
Health Benchmark = RfD (mg kg'1 d'1).
For carcinogens:
LADD = Lifetime average daily dose (mg kg"1 d"1)
Health Benchmark = Cancer risk level of 10~5/oral CSF (mg kg"1 d"1).
Although some constituents such as manganese elicit similar toxicological responses
(e.g., neurotoxicity) via different exposure pathways, the modeling stages of the analysis did not
consider cumulative exposures or impacts. The exposure scenarios and pathway evaluations were
developed and parameterized to produce conservative risk estimates; that is, the methodology
was designed to overestimate the actual risk to ensure that an ample margin of safety was built
into the analysis.
5.3.9.2 Human Health Model Inputs
Inputs to the human health effects model include estimates of toxicity (the human health
benchmarks) and exposure doses. The estimation of exposure dose is discussed in Section 5.3.7.
The human health benchmarks used as input to the model are discussed below.
Human health benchmarks for chronic exposures were used in this analysis to
characterize the potential cancer and noncancer hazards associated with the use of SFS-
manufactured soil in residential gardens. Oral CSFs and RfDs were used to estimate the cancer
and noncancer hazards from oral exposures, respectively.
The CSF is an upper-bound estimate (approximating a 95% confidence limit) of the
increased human cancer risk from a lifetime of exposure to an agent. This estimate is usually
expressed in units of proportion (of a population) affected per mg of agent per kg body weight
per day (per (mg kg"1 d"1)). Unlike RfDs, CSFs relate levels of exposure to a probability of
developing cancer.
The RfD is the primary benchmark used to evaluate noncarcinogenic hazards posed by
environmental exposures to chemical constituents. The RfD is an estimate of a daily exposure to
the human population (including sensitive subgroups) that is likely to be without appreciable risk
of deleterious noncancer effects during a lifetime (U.S. EPA, 2012a). However, an average
lifetime exposure above the RfD does not imply that an adverse health effect would necessarily
occur.
The chronic human health benchmarks used in the Phase II analyses are summarized in
Table 5-12. This table provides the constituent's name, Chemical Abstract Service Registry
Number (CASRN), RfD (in units of mg kg"1 d"1) and oral CSF (CSFo) [per (mg kg"1 d"1)], if
available. Health benchmarks for arsenic are from EPA's Integrated Risk Information System
(IRIS, U.S. EPA, 2012a), which is EPA's official electronic repository of chronic human health
benchmarks and represents EPA-wide consensus on critical human health effects associated with
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-34
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
exposure to chemical constituents released into the environment (U.S. EPA, 2012a). Benchmarks
in IRIS have been extensively reviewed, and each file contains descriptive and quantitative
information on potential health effects associated with the benchmark and other studies evaluated
during its derivation.
The health benchmarks for cobalt and iron are Provisional Peer-Reviewed Toxicity
Values (PPRTVs). The second tier of human health toxicity values in the OSWER toxicity value
hierarchy (USEPA, 2003a), PPRTVs are derived when such values are not available in IRIS.
PPRTVs are derived after a review of the relevant scientific literature using the methods, data
sources and guidance for value derivation used by the EPA IRIS Program. All PPRTVs receive
internal review by EPA scientists and external peer review by independent scientific experts.
PPRTVs differ in part from IRIS values in that PPRTVs do not receive the multi-program
consensus review provided for IRIS values. This is because IRIS values are generally intended to
be used in all EPA programs, while PPRTVs are developed specifically for the Superfund and
RCRA programs.
Table 5-12. Human Health Benchmarks Used in Phase II Analysis
Constituent
Asa
Cob
Feb
CASRN
7440382
7440484
7439896
RfD
(mgkg'd1)
0.0003
0.0003
0.7
CSF
(per mg kg * d'1)
1.5
—
—
a SOURCE: IRIS (U.S. EPA, 2012a)
b SOURCE: PPRTV(U.S. EPA, 2014)
5.3.9.3 Health Model Outputs
The human health effect model generated a distribution of Unitized Dose Ratio estimates
(UDRs) for adult and child receptors and each exposure pathway, as well as aggregates for the
soil exposure pathways (titled "Total Ingestion" reflecting exposures through incidental soil and
ingestion of produce). Analyses discussed in Section 5.3.5.3 and Appendix J indicate that
exposures via groundwater will not occur within the same timeframe as exposures via soil
pathways. Consequently, UDRs for soil and groundwater pathways were not combined. Rather,
the individual, pathway-specific UDRs were used to develop separate pathway-specific SFS
screening levels.
As discussed in Section 5.3.9.1, the UDRs represent a ratio of the ADD (or LADD) and
the health benchmarks listed in Table 5-12. Any UDR less than one equates to estimates below
the health benchmark. As discussed in Section 5.3.7.2, three separate sets of model runs were
performed: the first set produced home gardener exposure estimates using consumption rates
based on distributions from the EFH (U.S. EPA, 2011) and CSEFH (U.S. EPA, 2008a); sets 2
and 3 produced exposure estimates based on constant values for general population median and
high-end annual consumption rates assuming that no more than 50% of the produce consumed
was grown on the home garden. The 50th and 90th percentile UDRs from the two sets of general
population runs were then compared to the 50th and 90th percentile UDRs from the set of home
gardener runs. Table 5-13 lists the 50th and 90th percentile Total Ingestion ADD/LADDs, health
benchmarks, and UDRs for each adult receptor (home gardener, general population median
consumption rate, and general population high consumption rate). Table 5-14 lists parallel
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-35
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
information for the child receptor. Detailed 50th and 90th percentile values for adult and child
receptors for all pathways are listed in Appendix K. Tables 5-13 and 5-14 also present
information on arsenic exposure for the groundwater pathway. However, in the case of the 50th
percentile groundwater UDR a value of "PI" is reported indicating that the constituent did not
reach the receptor well during the simulation.
The UDRs in Tables 5-13 and 5-14 were used to estimate SFS-specific screening
concentrations, as discussed in Section 5.3.11.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-36
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-13. 50th and 90th Percentile Adult Unitized Dose Ratios for SFS-Manufactured Soil Constituents
Constituent
Health
Benchmark a
Pathway
Home Gardener
RunID
ADD/
LADD
(mg kg -1
BW d-1)
Unitized
Dose
Ratio
(unitless)
General Population
Median Consumption Rates
RunID
ADD/
LADD
(mg kg -1
BW d-1)
Unitized
Dose
Ratio
(unitless)
High Consumption Rates
RunID
ADD/
LADD
(mg kg-1
BW d-1)
Unitized
Dose
Ratio
(unitless)
50th Percentile
Cancer
As
6.67E-06
(CSF based)
Soil/Produce
Groundwater
4772
3.7E-07
0.056
8883
1.2E-07
0.019
7041
3.8E-07
0.057
PI
Noncancer
Co
Fe
0.0003 (RfD)
0.7 (RfD)
Soil/Produce
Soil/Produce
569
959
7.9E-06
1.6E-06
0.026
2.2E-6
5410
1301
2.5E-06
5.9E-07
0.0083
8.5E-7
509
7952
8.2E-06
1.6E-06
0.027
2.2E-6
90th Percentile
Cancer
As
6.67E-06
(CSF based)
Soil/Produce
Groundwater
7831
9716
1.8E-06
2.1E-07
0.28
0.031
1770
5.0E-07
0.074
3447
1.6E-06
0.24
Same as Gardener
Noncancer
Co
Fe
0.0003 (RfD)
0.7 (RfD)
Soil/Produce
Soil/Produce
5661
9766
1.8E-05
3.2E-06
0.058
4.6E-06
5260
5677
2.9E-06
l.OE-06
0.0097
1.4E-06
9534
4181
8.9E-06
2.1E-06
0.030
3.0E-6
aHealth Benchmark = RfD (mg kg'1 d'1) for noncancer risk and 10-5/oral CSF (per mg kg'1 d'1) for cancer risk.
PI = Pathway incomplete (constituent does not reach receptor well during simulation)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-37
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-14. 50th and 90th Percentile Child Unitized Dose Ratios for SFS-Manufactured Soil Constituents
Constituent
Health
Benchmark"
Pathway
Home Gardener
RunID
ADD/
LADD
(mg kg -1
BW d-1)
Unitized
Dose
Ratio
(unitless)
General Population
50%-ile Consumption Rate
RunID
ADD/
LADD
(mg kg -1
BW d-1)
Unitized
Dose
Ratio
(unitless)
90%-ile Consumption Rate
RunID
ADD/
LADD
(mg kg -1
BW d-1)
Unitized
Dose
Ratio
(unitless)
50th Percentile
Cancer
As
6.67E-06
(CSF based)
Soil/Produce
Groundwater
5114
5.4E-07
0.081
5208
2.9E-07
0.044
2701
7.4E-07
0.11
PI
Noncancer
Co
Fe
0.0003 (RfD)
0.7 (RfD)
Soil/Produce
Soil/Produce
495
7672
1.7E-05
5.5E-06
0.055
7.9E-06
3059
6883
8.2E-06
3.8E-06
0.027
5.4E-6
9733
2508
2.6E-05
6.8E-06
0.086
9.7E-06
90th Percentile
Cancer
As
6.67E-06
(CSF based)
Soil/Produce
Groundwater
4734
4302
1.1E-06
2.5E-07
0.17
0.037
2116
5.0E-07
0.075
1692
1.3E-06
0.19
Same as Gardener
Noncancer
Co
Fe
0.0003 (RfD)
0.7 (RfD)
Soil/Produce
Soil/Produce
5049
3020
3.1E-05
9.6E-06
0.10
1.4E-5
8674
4792
1.1E-05
6.8E-06
0.038
9.7E-06
4005
7537
3.2E-05
l.OE-05
0.11
1.5E-05
aHealth Benchmark = RfD (mg kg"1 d"1) for noncancer risk and 10~5/oral CSF (per mg kg"1 d"1) for cancer risk.
PI = Pathway incomplete (constituent does not reach receptor well during simulation)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-38
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
In all cases, UDRs based on home gardener consumption rates were higher than estimates
based on the general population median consumption rates, for both the adult and child. On the
other hand, at the 50th percentile of all model runs - as summarized in Tables 5-13 and 5-14 -
UDRs based on home gardener consumption rates were often lower than UDRs based on general
population high consumption rates, for both the adult and child. This is likely because home
gardener consumption rates varied with each model run (i.e. the consumption rate probability
distributions in Table 5-4 were sampled for each run, generating run-specific consumption rates)
and reflect consumption rates from across the entire range, whereas the general population
consumption rates were constrained at the high end of the range.
At the 90th percentile of all model runs, home gardener UDRs were almost always higher
than general population high consumption rate UDRs for both adult and child. For arsenic, the
home gardener child UDR was slightly lower than the general population high consumption rate
child UDR (i.e., 0.17 and 0.19, respectively).
5.3.10 Ecological Effects Modeling
Based on the conceptual model used for SFS in manufactured soil identified in Chapter
3 and depicted in Figure 5-6, this assessment evaluated the potential for adverse impacts to
plants, animals and soil invertebrates from the use of SFS in manufactured soil.
5.3.10.1 Conceptual Ecological Effects Model
This screening ecological assessment evaluated only direct contact with soil. Ecological
risk was expressed in terms of risk ratios. Risk ratios were calculated as the ratio of the
maximum soil concentration to the relevant SSL. For example, the risk ratio for soil invertebrates
was calculated as the ratio of the soil concentration to the soil invertebrate SSL.
5.3.10.2 Ecological Effects Model Inputs
The inputs to the ecological effects model for direct contact are surficial soil
concentrations and ecological health benchmarks. Estimation of soil concentrations is discussed
in Section 5.3.4. Table 5-15 presents EPA's Ecological SSLs (Eco-SSLs)45 that were used, with
maximum soil concentrations, to calculate the constituent-specific HQs for terrestrial plants and
soil invertebrates.
45 Developed by EPA's Superfund program, Eco-SSLs are concentrations of contaminants in soil that are protective
of ecological receptors that commonly come into contact with soil or ingest biota that live in or on soil. These
values can be used to identify those contaminants of potential concern in soils requiring further evaluation in a
baseline ecological risk assessment. Although these very conservative screening levels were developed
specifically to be used during the Superfund ecological risk assessment process, EPA envisions that any federal,
state, tribal, or private environmental assessment can use these values to screen soil contaminants to determine if
additional ecological study is warranted (U.S. EPA, 2005c).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 5-39
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-15. Eco-SSLs Used in Phase II Analysis (mg kg'1 soil)
Constituent
Cr(III)
Cu
Mn
Ni
Sb
CASRN
16065831
7440508
7439965
7440020
7440360
Terrestrial
Plants
—
70
220
38
—
Soil
Invertebrates
—
80
450
280
78
Mammals
34
49
4000
130
0.27
5.3.10.3 Ecological Effects Model Outputs
The ecological effects model generates distributions of constituent-specific Unit Dose
Ratios. As discussed in Section 5.3.10.1, these values represent a ratio of the modeled exposure
value and the ecological health benchmarks listed in Table 5-15. Any UDR less than one equates
to exposure estimate below the benchmark. As listed in Table 5-16 and discussed in Section
5.3.11, values representing the 50th and 90th percentiles of these UDR distributions were used to
estimate risk-based SFS-specific ecological screening concentrations.
Table 5-16. 50th and 90th Percentile Ecological Unitized Dose Ratios for SFS-Manufactured
Soil Constituents
Constituent
Terrestrial Plants
Adjusted
Soil Cone.
(mg kg-1)
Eco-SSL
(mg kg-1)
UDR
(unitless)
Soil Invertebrates
Adjusted
Soil Cone.
(mg kg-1)
Eco-SSL
(mg kg-1)
UDR
(unitless)
Mammals
Adjusted
Soil Cone.
(mg kg-1)
Eco-SSL
(mg kg-1)
UDR
(unitless)
50th percentile
Cr (III)
Cu
Mn
Ni
Sb
—
0.90
0.093
0.064
—
—
70
220
38
—
—
0.013
0.00042
0.0017
—
—
0.90
0.93
0.92
0.82
—
80
450
280
78
—
0.011
0.0021
0.0033
0.010
0.14
0.13
0.93
0.92
0.12
34
49
4000
130
0.27
0.0041
0.0027
0.00023
0.0071
0.45
90th percentile
Cr(III)
Cu
Mn
Ni
Sb
—
0.97
0.097
0.068
—
—
70
220
38
—
—
0.014
.00044
0.0018
—
—
0.97
0.97
0.97
0.96
—
80
450
280
78
—
0.012
0.0022
0.0035
0.012
0.15
0.15
0.97
0.97
0.14
34
49
4000
130
0.27
0.0043
0.0030
0.00024
0.0075
0.53
5.3.11 Calculating Modeled SFS-Specific Screening Levels
Health model outputs compare health benchmarks to exposure estimates assuming a
starting constituent concentration in SFS-manufactured soil of 1 mg constituent in one kilogram
of soil on a wet weight basis. The home garden conceptual model assumes a soil recipe that
includes 50% SFS. Therefore, SFS-manufactured soil UDRs listed in Tables 5-13 and 5-14 were
converted to modeled SFS-specific screening concentrations using the following equation:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-40
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Screening ConcSFS =
Unit Dose Ratio
xl.l
where:
Screening Cone
= Concentration of the constituent in SFS unlikely to cause adverse effect
(mg kg"1 SFS dry weight)
Unit Dose Ratio = exposure dose to health benchmark (unitless, based on a starting soil
concentration in mg kg"1 wet weight)
2 = 1/SFS fraction of manufactured soil (unitless).
1.1 = Factor for converting from wet weight to dry weight reflecting average
modeled solids content of 90 percent (10 percent moisture) (unitless).46
Table 5-17 lists the SFS screening values protective of human health. These values
represent the concentration of the constituent that could be found in SFS and not exceed the
health benchmark.
Table 5-17. Modeled SFS-specific Screening Levels for the Home Garden Scenario
(mgkg-'SFS)
Constituent
Adult
Home
Gardener
General Population
Median
Consumption
Rates
High
Consumption
Rates
Child
Home
Gardener
General Population
Median
Consumption
Rates
High
Consumption
Rates
Soil/Produce Pathway
As
Co
FeCgkg-1)
8.0
38
480
30
230
Capped
9.1
74
730
13
22
160
30
58
230
12
21
150
Groundwater Pathway
As
71
59
Capped = Calculated SFS-specific screening level would allow SFS to be 100% Fe, so value capped.
Table 5-18 lists the SFS screening values protective of ecological receptors. Appendix L
presents the 50th and 90th percentile values and their corresponding soil concentrations. These
values represent constituent concentrations that could be found in SFS and not exceed the
ecological health benchmark.
46 As required by the source model, chemical-specific concentrations are input on a wet weight basis as a mass
concentration. Noting that the SFS concentrations are similarly mass concentration-based, except that they are
expressed on a dry weight basis, it is necessary to account for the modeled solids content.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-41
-------
Chapter 5.0 Analysis Phase II. Risk Modeling ofCOCs
Table 5-18. Modeled SFS-specific Ecological Screening Levels for the Home Garden
Scenario (mg kgJ SFS)
Constituent
Cr(III)
Cu
Mn
Ni
Sb
Terrestrial Plants
50%-ile
—
170
5200
1300
—
90%-ile
—
160
5000
1200
—
Soil Invertebrates
50%-ile
—
200
1,100
670
210
90%-ile
—
180
1,000
630
179
Mammals
50%-ile
530
800
9,500
310
4.8
90%-ile
510
740
9,000
290
4.1
5.3.12 Results: Comparing Screening Values to SFS Constituent Concentrations
Table 5-19 compares SFS constituent concentrations to the lowest human health-based
SFS-specific screening values, as well as the ecological SFS-specific screening values, derived in
Section 5.3.11. For each constituent, the human health-based value is the lower of the adult or
child screening values. Likewise, the listed ecological health-based value is the lowest of the
plant, soil invertebrates, or mammal screening values.
Table 5-19. Comparing SFS Constituent Concentrations to Modeled SFS-Specific
Screening Levels (mg kg"1 SFS)
Constituent
As
Co
Cr
Cu
FeCgkg-1)
Mn
Ni
Sb
SFS 95%-ile
Concentration
6.44
5.99
109
107
57.1
670
102
1.23
Modeled SFS-Specific Screening Levels
Home
Gardener
8.0
22
--
--
160
--
--
--
General Population
Median
Consumption
Rates
30
58
--
--
230
--
--
--
High
Consumption
Rates
9.1
21
--
--
150
--
--
--
Ecological
--
--
510
160
--
1,000
290
4.1
- - = Constituent was screened out in Phase I and did not require modeling for this receptor.
The SFS concentrations of all eight modeled constituents fell below their respective
human and ecological modeled SFS-specific screening levels.
Modeling results are specific to the assumptions used in the modeling, and should be
understood within the context of the complexity of the environmental conditions they represent.
Chapter 6 discusses the various lines of evidence described in the report, including the modeling
results presented above and the information provided by them, as well as the uncertainties in and
limitations of the analysis.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
5-42
-------
Chapter 6.0 Risk Characterization
6. Risk Characterization
Chapter 2 introduced the current state of research on the origins, characteristics, and
behavior of SFS in soil. Chapters 3, 4, and 5 detailed the screening and modeling steps
undertaken to assess the potential for human and ecological health impacts from soil-related uses
of SFS: the results from quantitative evaluation of SFS-manufactured soil in home gardens
would also apply to SFS use in soil-less potting media and use in road subbase. The results of
these various efforts represent lines of evidence.
EPA's Risk Characterization Handbook (U.S. EPA, 2000) states that a risk
characterization "integrates information from the preceding components... and synthesizes an
overall conclusion about risk that is complete, informative, and useful for decision makers." This
chapter provides the risk characterization for the evaluation. This chapter first discusses
overarching concepts, such as the conservative nature of the risk screen used and the
complexities of soil science. This information is then integrated with the results of the risk
evaluation to provide a summary of the potential for human health and environmental impacts.
As discussed in Chapter 2 of this report, generating industries, consumers, and
regulatory agencies need to be confident that the scientific basis for making beneficial use
decisions on SFS provides a high degree of certainty that potential risks to human health and the
environment have been thoroughly evaluated. To address this need, the human health risk
analysis was specifically designed to focus on the upper end of the distribution of risk to
individuals that could potentially be exposed to SFS constituents because they (1) live near soil
manufacturing facilities that include SFS among their soil recipes; (2) live near roadway
construction projects that use unencapsulated SFS as a subbase for roads; or (3) use
manufactured soil containing SFS in home gardens. In the Guidance for Risk Characterization
developed by EPA's Science Policy Council (U.S. EPA, 1995c), EPA defined the high end of the
risk distribution as being at or above the 90th percentile risk or hazard estimate generated during
the Monte Carlo simulation.
Similarly, the ecological risk analysis focused on receptors that are in direct contact with
the soil media and the potential for food web exposures specific to the area of use. This is
particularly conservative because small perturbations and stresses to a field that represents a
small fraction of the landscape may not be significant from either an ecological or societal
perspective. Therefore, the portion of this report that addresses the potential for adverse effects to
ecological receptors is also conservative and should be considered as a high-end approach
analogous to the human health risk analysis.
With the conservative nature of the analysis in mind, Section 6.1 provides an overview of
the risk characterization by describing how a lines-of-evidence approach has been used to
organize the information on modeling and scientific research.
6.1 Overview of the Risk Characterization
The goal of this evaluation was to determine whether SFS used in certain soil-related
applications will be protective of human health and the environment. This assessment defines
"protective" in terms of specific cancer risk (not to exceed an incremental risk of 10"5, or 1 in
100,000) and noncancer risk for human and ecological receptors (not to exceed a threshold dose
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-1
-------
Chapter 6.0 Risk Characterization
or concentration). There are many ways to present information relevant to these goals, all of
which would satisfy the requirements of a risk characterization. However, given the complexity
of risk-related issues surrounding the use of SFS (e.g., the relevance of comparing background
metal content of soil to SFS metal content), as well as the need to integrate the screening
modeling results with research on SFSs, soil chemistry, and toxicity, the most effective way to
create transparency in this section was to begin at a high level by laying out a series of risk
assessment questions, and then work through the analysis, ultimately presenting this information
at the constituent level of detail. As discussed later in this chapter, the use of available scientific
research on SFS and SFS constituent behavior and toxicology is critical to the interpretation of
the screening modeling results. All models are simplifications of reality, and although they are
extremely useful tools for predictive risk assessment, the modeling results should be considered
in conjunction with the science of chemical behavior in the environment as it relates to exposure
and, ultimately, risk. The remainder of this chapter is organized as follows:
• Section 6.2, Key Risk Assessment Questions. This section presents key risk assessment
questions that pertain to certain soil-related beneficial uses of SFS. These questions are
presented and discussed at a level that is intended to be accessible to risk managers, and
they provide the context for the entire risk characterization. These questions may be
tracked through all of the subsequent sections of the risk characterization.
• Section 6.3, Overarching Concepts. This evaluation is unique in that it deals with the
beneficial use of a material and needs to address several technical issues. Because these
issues are important to the interpretation of the risk modeling results and affect more than
one SFS constituent, this section describes these concepts as a prelude to the more
detailed elements of the risk characterization that follow.
• Section 6.4, SFS Product Risks. This section reviews the qualitative and semi-
quantitative information on SFS as a material that may be beneficially used. It is
important to understand what is known and what issues should be considered when
interpreting the scientific research and screening-level modeling results.
• Section 6.5, PAHs, Dioxins, Furans, and Dioxin-like PCBs in SFS. PAHs, dioxins,
furans, and dioxin-like PCBs constitute major groups of chemical constituents that have
been quantified above detection limits in SFS. In some cases, these constituents have
been addressed in risk assessments of other materials, such as dioxins in biosolids. The
results of these risk assessments are clearly relevant to the interpretation of information
specific to SFS; however, differences in exposure scenarios, modeling assumptions, the
constituent-specific matrix, and other determinants of risk should be carefully considered
when comparing the results of a risk assessment of those other materials to the SFS risk
assessment. Therefore, this section will consider both the interpretation of other risk
assessments, as well as the information and screening results developed in this report.
• Section 6.6, Phenolics in SFS. Although most phenolics were below detection limits,
some have been found above detection limits in SFS (e.g., phenol, 2,4- dimethylphenol,
2-methylphenol). These compounds were evaluated as part of this risk assessment. This
section presents the risk assessment modeling results and discusses the potential for
adverse effects on human health and the environment associated with phenolics above
detection limits in SFS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-2
-------
Chapter 6.0 Risk Characterization
• Section 6.7, Metals and Metalloids in SFS. Because of their persistence and potential
toxicity in the environment, metals represent a critical group of chemical constituents
found in SFS. A wide range of metals have been found above detection limits, and given
the complexity of metals' behavior in soil systems and critical science-policy issues (e.g.,
background soil levels), this section presents a detailed lines-of-evidence determination
for each metal constituent of concern. In addition to presenting the modeling results for
various exposure pathways and scenarios, this section integrates scientific research on
metals' behavior and toxicity and discusses whether this information (1) indicates that the
results are conservative, and (2) suggests that an exposure pathway could not be
completed at levels of concern because of natural obstacles, such as the soil-plant barrier.
• Section 6.8, Uncertainty Characterization. This section presents and discusses the data
gaps and major sources of uncertainty in this risk assessment, focusing again on the
overall goal to ensure that soil-related applications of SFS will not pose risks to human
health and ecological receptors above levels of concern. Therefore, this section does not
provide detailed information on modeling; that aspect of the risk assessment was
designed to be conservative, and the bias inherent in data inputs and scenario assumptions
is in the direction of overestimating risk. This section discusses the uncertainties from a
decision-maker's perspective; that is, it examines whether or not the uncertainties in this
risk assessment either (1) support or discourage the use of SFS in soil-related activities,
or (2) require additional research to improve the quality of the information.
6.2 Key Risk Assessment Questions
To ensure that this report provides a high level of confidence, it is important to articulate
the key risk assessment questions that this analysis was designed to address:
• Will the addition of SFSs to soil result in an increase in the constituent concentrations in
soil relative to background levels, and how should the results of the risk assessment be
interpreted across varied national soils?
• How do constituent forms found in the SFS matrix behave with respect to bioaccessibility
and bioavailability, and how does that affect potential risks?
• How will the behaviors of individual constituents in SFS-manufactured soil, such as the
soil-plant barrier, impact the potential for exposure through the food chain pathway and,
ultimately, the potential for adverse human health and ecological effects?
• How do the risk assessment results compare to levels required to maintain nutritional
health in plants and animals? Do issues of essentiality suggest that the predicted risks to
plants and animals overestimate the potential for adverse effects?
6.3 Overarching Concepts
6.3.1 Background Concentrations
The components used to create metalcasting molds are not anthropogenically derived, but
are obtained from the natural environment. Sands are either mined from terrestrial or aquatic
(e.g., lakebeds) environments, while phyllosilicate clays (bentonites) are mined from terrestrial
environments. A typical green sand contains as much as 90% sand, 5-10% clay, 5%
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-3
-------
Chapter 6.0 Risk Characterization
carbonaceous material (e.g., seacoal, cellulose), and 2-5% water by weight. These mold
components, like soils, contain a variety of trace metals at concentrations found in native soils.
Soils themselves contain metals because they are composed of weathered rock and
minerals (e.g., sand, clay) and decomposed plant and animal debris. However, metal levels in
some soils can be elevated through human activities and even natural processes (Adriano, 2001;
He et al., 2005). Good examples of natural element mineralization of soil are found in
California's central valley, where soils are enriched with selenium due to a high-selenium parent
material (Dungan and Frankenberger, 1999); or in northern California, where soils contain nickel
levels as high as 1,000-2,000 mg kg"1, because the parent material is serpentine, a mineral with
high natural levels of nickel. As discussed in Appendices A and C, risks to plants and grazing
livestock from most trace metals in soil are low. Serpentine soils with high nickel concentrations
(as much as 50 times greater than other background soils) are rarely phytotoxic if the pH does
not fall below 6 (Kukier and Chaney, 2004). Even at these extreme soil nickel concentrations,
natural flora and fauna thrive without detriment.
Comparing metal concentrations in background soils and silica-based U.S. iron, steel, and
aluminum SFSs (see Table 7-1) reveals that the concentrations of most metals and metalloids in
SFS fall below those in most background U.S. and Canadian soils. However, the 95th percentile
concentrations of arsenic, chromium, copper, manganese, molybdenum, and nickel in SFS
exceed the median soil background concentrations for these metals. This does not, however, by
itself mean that SFS should not be used as a soil amendment or component in a manufactured
soil, as other lines of evidence (e.g., comparison to human and eco screening values) may
mitigate concern. Based on the total metal data for silica-based iron, steel, and aluminum SFSs
reported here, applications of most SFSs to average U.S. soils will not cause significant increases
in the total soil metal concentrations.
6.3.2 Chemical Reactions in Soil
Soils contain metals at concentrations dependent on the parent material from which the
soil is derived (Kabata-Pendias, 2001). Metals may also reach soils as components of fertilizers,
manures, byproducts, and aerosols, and hence may exist in varied chemical forms. If metals
reach soils in elemental forms, they will oxidize rapidly depending on the redox characteristics
of the metal and the soil. For example, silver, gold, and even copper are found in a metallic form
in some reducing soils, but copper and silver are usually oxidized in aerobic soils over time.
Some are oxidized rapidly, but a few persist for long periods depending on the particle size of the
metal that reached the soil (smaller particles have higher surface area and react more rapidly) or
the redox status of the soil. Flooded soils (e.g., peat soils) may provide a reducing soil
environment, which will allow metallic or metal sulfide particles to persist for long periods.
The soluble cation and oxyanion forms of trace metals in aerobic soils are potentially
more mobile, and thus potentially more bioavailable than the elemental forms of the trace metals,
so a risk assessment for the aerobic soil forms is appropriate. In a normal aerobic soil, most
metals are present as hydrated or complexed cations or anions controlled by their chemistry in
equilibrium with the ions bound to the soil surfaces or precipitated as minerals in the soil
(Langmuir et al., 2005), such as Zn2+, Cu2+, Ni2+, Pb2+, Cd2+, MoO42', SeO42', and H2PO4'. Many
ions remain in the cation form regardless of soil redox conditions: Li+, K+, Na+, Rb+, Cs+ (alkali
cations), Be2+, Mg2+, Ca2+, Sr2^ Ba2+ (alkaline earth cations), and select trace elements, including
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-4
-------
Chapter 6.0 Risk Characterization
Zn2+, Cu2+, Ni2+, Pb2+, Cd2+, and MoO42". Similarly, many anions occur as halides (F~, Cr, Br, T)
in terrestrial soils (Bohn et al., 2001).
Flooding a soil (e.g., rice paddies) causes the redox potential to decrease as the soil
becomes reducing, as little oxygen dissolves in water and soil organisms consume the oxygen.
The soil pores become filled with water or gases are formed in the soil under anaerobic
conditions. With the reducing environment, some metalloids are reduced to chemical forms
different than those found in normal aerobic soils. In particular, As(V) as arsenate (AsO43~) is
reduced to the more mobile As(III) as arsenite (AsOs3"), which increases the arsenic in the soil
solution. This is important in the case for phytotoxicity of arsenic; flooded rice is the crop plant
found to be most sensitive to excessive soil arsenic. The higher concentration of AsOs3" in
flooded soils compared to AsO43" in aerobic soil allows much easier plant uptake and injury from
the soil arsenic. Uptake of some other ions may be increased in reducing soils, but the potential
for toxicity of other metals is not increased by reducing conditions as found with arsenic.
Sorption is a chemical process that buffers the partitioning of trace metals between solid
and liquid phases in soils and byproducts. Iron, aluminum, and manganese oxide soil minerals
are important sinks for trace metals in soil and byproduct-amended soils (Essington and
Mattigod, 1991; Lombi et al., 2002; Hettiarachchi et al., 2003). Trace metal sorption by the oxide
surface is a pH-dependent process; protons compete with cations for sorption. The adsorption of
metal cations by the oxide surfaces increases to almost 100% with increasing pH (McKenzie,
1980). In contrast, oxyanion adsorption generally decreases with increasing pH.
Trace metal cations can also sorb to soil organic matter (SOM) and other forms of
humified natural organic matter (NOM). Strong adsorption by NOM in byproducts (through the
formation of metal chelates) reduces solubility of several trace metals in soil (Adriano, 2001).
Sorption of trace metals by SOM or NOM increases with pH because protons compete less well
with increasing pH. Trace metal sorption by NOM is reduced less at lower pH than is trace metal
ion sorption on iron and manganese oxides.
Trace metal cations also form sparingly soluble precipitates with phosphate, sulfides, and
other anions (Lindsay, 2001; Langmuir et al., 2005). Trace metal precipitation is highly pH
dependent and increases with pH for many trace metal cations. Arsenate and other trace metal
oxyanions can form insoluble precipitates with multivalent cations, including aluminum,
calcium, and iron. Trace metal precipitation affects the amount of trace metal in solution (i.e.,
availability and mobility).
6.3.3 Soil-Plant Barrier
The potential risk that diverse trace metals in soils pose to the feed- and food-chain has
been thoroughly examined over the last several decades. One purpose of that investigation has
been to understand the risk from application of biosolids, livestock manure, and other trace metal
contamination sources to soil. During this period, the soil-plant barrier concept was introduced to
communicate how metal addition rate and chemistry, soil chemistry, and plant chemistry affected
the risk to plants and animals from metals in soil amendments (Chaney, 1980; 1983; Langmuir et
al., 2005). The soil barrier protects by way of soil chemical processes that limit the availability of
metals for uptake, while senescence due to phytotoxicity further reduces the chances that
excessively contaminated plants will be consumed (i.e., plant barrier). This concept is based on
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-5
-------
Chapter 6.0 Risk Characterization
much experience in veterinary toxicology and agronomy. Reactions and processes that influence
the soil-plant barrier include the following:
• Solid adsorbent sources (e.g., iron, aluminum, and manganese oxyhydroxides and organic
matter) in soil amendments have adsorptive surfaces that influence soil chemistry
• Adsorption or precipitation of metals in soils or in roots limit uptake-translocation of
most metals to shoots
• The phytotoxicity of some elements (e.g., aluminum, arsenic, boron, chromium III,
copper, fluorine, manganese, nickel, zinc) limits the concentrations of these metals in
plant shoots to levels chronically tolerated by livestock and humans
• The food-chain transfer of an element may not constitute a risk, but the direct ingestion of
the contaminated soil may cause risk from arsenic, fluorine, lead, and some other
elements under poor management conditions if the soils are highly contaminated
• The soil-plant barrier does not restrict the transfer of soil selenium, molybdenum, and
cobalt well enough to protect all animals (selenium, molybdenum) or ruminant livestock
(cobalt), or cadmium to subsistence rice consumers or cadmium in the absence of the
usual 100-fold greater concentrations of zinc than the concentrations of cadmium.
A summary of the trace metal tolerances by plants and livestock is presented in
Appendix A, Table A-l. It should be noted that the National Research Council (NRC, 1980)
committee, which identified the maximum levels of trace metals in feeds tolerated by domestic
livestock, based its conclusions on data from toxicological-type feeding studies in which soluble
trace metal salts had been mixed with practical or purified diets to examine the animals' response
to the dietary metals. If soil or some soil amendment is incorporated into the diet, metal
solubility and bioavailability are much smaller than in the tests relied on by the NRC (1980). For
example, it has been noted that until soil exceeds about 300 mg Pb kg"1, animals show no
increased body lead burden from ingesting the soil (Chaney and Ryan, 1993). Other metals in
equilibrium with poorly soluble minerals or strongly adsorbed in ingested soils are often much
less bioavailable than they would be if they were added to the diet as soluble salts.
6.3.4 Interactions Among Constituents
The toxicity to animals of biosolids or manure-applied metals is an example of how the
interaction between metals affects their toxicity. Specifically, copper deficiency-stressed animals
are more sensitive to dietary zinc than animals fed with copper-adequate diets. Biosolids-
fertilized crops are not low in copper, reducing animal sensitivity to zinc levels (Chaney,
1983).47 Similarly, copper toxicity to sensitive ruminant animals is substantially reduced by
increased dietary levels of cadmium, iron, molybdenum, zinc, and SO42", or sorbents such as
SOM. In contrast with the predicted toxicity from copper in ingested swine manure or biosolids,
reduced liver copper concentrations have been found in cattle or sheep that ingested biosolids,
unless the ingested biosolids exceeded about 1,000 mg Cu kg"1 (Chaney and Ryan, 1993).
Similarly, zinc in plants inhibits the absorption of cadmium by animals, as plant sulfate inhibits
47 Chaney (1983) also found that zinc phytotoxicity further protects livestock (including the most sensitive
ruminants) against excessive zinc in forages: Plant senescence from phytotoxicity reduces the chances that
excessively contaminated plants will be consumed by animals.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-6
-------
Chapter 6.0 Risk Characterization
absorption of plant selenium. Interactions that reduce risk are evident in many trace element
issues.
Interactions can also limit toxicity and risk. For example, cadmium bioavailability is
strongly affected by the presence of normal background levels of zinc in soils (100- to 200-fold
cadmium level); zinc inhibits the binding of cadmium by soil, but also inhibits cadmium uptake
by roots, cadmium transport to shoots, and cadmium transport to storage tissues. Furthermore,
zinc in foods significantly reduces cadmium absorption by animals (Chaney et al., 2004).
Increased zinc levels in spinach and lettuce reduced the absorption of cadmium in these leafy
vegetables by Japanese quail (McKenna et al., 1992a). Also, increased zinc in forage diets
strongly inhibited cadmium absorption and reduced liver and kidney cadmium concentrations in
cattle (Stuczynski et al., 2007).
6.3.5 Highly Exposed Populations
Risk assessment for wildlife is similar to that of livestock; because of their limited range,
the diets of some species (e.g., plants, soil invertebrates, small mammals) can originate entirely
from the soil or plants grown on a site. Because these species have higher exposures than most
wildlife, they are used as the highly exposed populations. In cases involving wildlife in
unmanaged ecosystems, maximal plant residues may exceed those allowed on managed
farmland—wildlife may eat sick plants that would not be harvested by a commercial grower.
Evaluation of the literature on wildlife exposure to trace metal-contaminated soils indicates that
animals that consume earthworms are the highly exposed populations (Brown et al., 2002).
Cadmium has received much study because of extensive human cadmium disease in
nations where subsistence rice farmers consume locally grown rice for their lifetime (Chaney et
al., 2004). The disease results from chronic exposure to food-borne cadmium. Basic studies on
the bioavailability of food cadmium have indicated that rice promotes cadmium absorption by
inducing iron and zinc deficiency in the subsistence rice farm families because of the very low
levels and low bioavailability of iron and zinc in polished rice (Reeves and Chaney, 2002). A
diet deficient in iron and zinc causes much more of the cadmium to be absorbed than in other
diets tested (Reeves and Chaney, 2004). Several epidemiological studies have found no evidence
of human cadmium disease from garden foods grown on Zn+Cd rich smelter or mine waste
contaminated garden soils (Chaney et al., 2004).
Cobalt is another unusual case in that ruminant livestock are at risk from dietary cobalt at
much lower crop cobalt levels. Cobalt is essential for vitamin B12 synthesis by rumen bacteria.
Crops can accumulate at least 25 mg Co kg"1 dry weight before even sensitive crops are injured
by the absorbed cobalt, but ruminants can tolerate no more than about 10 mg Co kg"1 dry weight
(DW) diets (Keener et al., 1949; Becker and Smith, 1951; Corrier et al., 1986; NRC, 1980). In
practice, no case of cobalt toxicity has been reported, apparently because excessive levels of
cobalt in soil are rare. It remains theoretically possible for cobalt in soil to poison ruminants. In
the case of serpentine soils geochemically enriched with both nickel and cobalt, the nickel
inhibits the uptake of cobalt and the soil properties limit the uptake of both nickel and cobalt, and
the potential adverse effects of cobalt to plants or animals have never been observed.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-7
-------
Chapter 6.0 Risk Characterization
6.4 Spent Foundry Sand Product Risks
Spent foundry sand has been found to be useful in making fertile soil mixtures for many
agricultural and horticultural uses. The present evaluation considered a high-end use: a 20-cm
layer of manufactured soil containing 50% SFS by dry weight in the blend. Such blends often
contain soil, composts, manure, and other ingredients that provide a rooting mixture for diverse
plants. These soils are used for yards, gardens, institutional lawns, and other instances where
existing soils have been disturbed or have very low fertility and fail to support plant growth.
Uses of SFS in manufactured soils are mostly at lower rates than the rates assumed in the
present risk assessment. Evaluation of SFS alone (i.e., not blended with organic additives) as a
replacement soil was considered, but research has shown that for many SFSs, this is not feasible.
SFS without treatment tends to form a cemented solid material (De Koff et al., 2008). Often this
is due to the presence of sodium bentonite in the SFS, which causes the cementation reaction and
"sealing" of the soil (Dungan et al., 2007). This can be corrected through the addition of soluble
calcium salts. The usefulness of SFS alone is also restricted by its limited particle size. Soil-
related beneficial uses of SFS generally use SFS as a small fraction of a mixed soil. Under the
expected conditions (i.e., SFS as a component of manufactured soil), no risks were identified in
the literature.
Under aerobic conditions, long-term exposures to metals in SFS-manufactured soil will
continue to be low as it weathers. Over time, the sand and clays present in SFS are reduced in
size by physical processes and/or dissolution, while organic byproducts will be broken down to
elemental forms, mainly through biological processes. The trace metals in a SFS-manufactured
soil are not normally bioavailable, as they are bound within the matrix of minerals or sorbed to
organic matter or metal oxides. Even exposing pure iron, steel, and aluminum SFSs to acid
conditions (e.g., TCLP, SPLP) did not cause significant quantities of trace metals to be released
into leachates. Given the pH range of SFS (neutral to slightly alkaline), the presence of
aluminum, iron, and manganese will decrease the availability of trace metal cations due to the
adsorption on oxide surfaces. Metal oxides, such as iron and manganese, are important in
regulating the partitioning of trace metals between solution and solid phases in soils (Basta et al.,
2005). Trace metal cations and oxyanions, which are generally more mobile and bioavailable
than elemental forms, can also be expected to sorb to organic matter and form insoluble
precipitates. Because an SFS-manufactured soil will become more "soil-like" with time,
elements released due to weathering and mineralization are likely to behave like those in native
soils.
6.5 PAHs, Dioxins, Furans, and Dioxin-Like PCBs in SFS
6.5.1 PAHs
Chapter 2 points out that the majority of the PAHs that were found at concentrations
above detection limits were the 2- and 3-ring PAHs (i.e., acenaphthene, acenaphthylene,
anthracene, fluorene, naphthalene, and phenanthrene). Anthracene, fluorene, naphthalene, and
phenanthrene were the most prevalent PAHs, detected in >79% of the SFSs (Dungan, 2006).
Also detected above the MDLs, though in only a few sands, were benz[a]anthracene, chrysene,
fluoranthene, and pyrene.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-8
-------
Chapter 6.0 Risk Characterization
The 95th percentile concentrations for 11 PAHs in SFS were compared to (1) the
Residential SSLs adjusted to also address home gardener produce ingestion pathways (Adjusted
SSL), (2) the inhalation screening level concentrations for benz[a]anthracene, chrysene,
dibenz[a,h]anthracene, and naphthalene, the only PAHs for which inhalation health benchmarks
were available, and (3) Eco-SSLs for total Low Molecular Weight PAHs and total High
Molecular Weight PAHs (see Section 4.4.3 for a discussion of these categories). As seen in
Table 6-1, in all cases, the 95th percentile constituent concentrations in SFS were below the
corresponding Adjusted SSL, with most cases, the 95th percentile constituent concentrations
being orders of magnitude below the corresponding Adjusted SSL. When aggregated by
molecular weight category, the 95th percentile constituent concentrations of Low and High
molecular weight PAHs were similarly below their respective Eco-SSLs.
Based on this comparison, the presence of these PAH compounds in SFS are unlikely to
cause adverse human or ecological health impacts at levels of concern when SFS is used in SFS-
manufactured soils, soil-less potting media, or road base..
Table 6-1. Comparison of PAH Concentrations in SFS to Screening Criteria (mg kg'1)
Constituent
Low Molecular Weight PAHs a - Total
Acenaphthene
Acenaphthylene
Anthracene
Fluorene
Naphthalene
Phenanthrene
High Molecular Weight PAHs a - Total
B enz [a] anthracene
Chrysene
Dibenz[a,h]anthracene
Fluoranthene
Pyrene
SFS
95%-ile
7.59
0.34
0.20
0.88
0.73
3.89
1.56
0.95
0.14
0.04
0.08
0.21
0.48
ConcMs
3.79
0.17
0.10
0.44
0.37
1.94
0.78
0.48
0.07
0.02
0.04
0.10
0.24
Adjusted
SSL
N/A
350
N/A
1,700
230
3.8
N/A
N/A
0.15
1.5
0.15
230
170
Inhalation
Screening
Level
N/A
N/A
N/A
N/A
N/A
60,300
N/A
N/A
4,020
221
402
N/A
N/A
Eco-SSL
29
N/A
N/A
N/A
N/A
N/A
N/A
1.1
N/A
N/A
N/A
N/A
N/A
N/A = no benchmark available.
a Low Molecular Weight PAHs are composed of fewer than four condensed aromatic ring structures, and High
Molecular Weight PAHs are composed of four or more condensed aromatic ring structures (EPA, 2007e).
6.5.2 PCDDs, PCDFs, and Dioxin-like PCBs
As described in Chapter 2, except for 1,2,3,7,8,9-HxCDF, most PCDD and PCDF
congeners were detected, but not in all SFSs. Concentrations of the PCDD congeners ranged
from <0.01-44.8 ng kg'1, with 1,2,3,4,6,7,8,9-OCDD being found at the highest concentration in
all of the SFSs. Expressed in terms of TEQs, the total dioxin concentrations ranged from 0.01-
3.13 ng TEQ kg"1, with an average concentration of 0.58 ng TEQ kg"1. However, because PCB-
81 and mono-ort/zo-substituted PCBs were not measured, the PCB contribution to the total TEQ
concentration is not known. Nevertheless, the highest total dioxin concentration (expressed as a
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-9
-------
Chapter 6.0 Risk Characterization
toxic equivalency value) of 3.13 ng TEQ kg"1 is about 100 times lower than the 300 ng TEQ kg"1
limit developed by EPA for biosolids (U.S. EPA, 2002e). The biosolids matrix has a significantly
higher organic carbon content relative to the SFSs; however, SFS-manufactured soils will
presumably also contain organic amendments and nutrients at levels that are beneficial to the
soil.
The maximum concentration for total TEQs48 was compared to (1) the Residential SSL
adjusted to also address home gardener produce ingestion pathways (Adjusted SSL), and (2) the
inhalation screening level concentration derived for the manufacturing scenario. No ecological
health benchmarks were available for PCDDs, PCDFs and dioxin-like PCBs; therefore the
potential for adverse ecological impacts from exposure to these SFS constituents was not
evaluated. As seen in Table 6-2, the maximum total TEQ was at least an order of magnitude
below the soil and inhalation screening levels. Also, the concentrations of TCDD-TEQ in SFS
were below background levels in U.S. agricultural soils, and well below levels in urban soils
(Rogowski and Yake, 2005; Andersson and Ottesen, 2008). Furthermore, the highest total dioxin
concentration was about 100 times lower than the biosolids limit. Based on the above
information, exposure to levels of PCDDs, PCDFs, and dioxin-like PCBs found in SFS is
unlikely to cause adverse human health impacts when SFS is used in SFS-manufactured soils,
soil-less potting media, or road base.
Table 6-2. Comparison of Total Dioxin TEQ Concentrations in SFS to Screening Criteria
(mg TEQ kg"1)
PCDDs, PCDFs, and
Co-planar PCBs
Total dioxin TEQ
Maximum SFS
Concentration
3.1E-06
ConcMs
1.6E-06
Adjusted SSL
4.9E-06
Inhalation
Screening Level
2.01E-02
6.6 Phenolics in SFS
As discussed in Chapter 2, the phenolics that were detected in the majority of the SFSs
included phenol, 2-methylphenol, 3- and 4-methylphenol, and 2,4-dimethylphenol. In general,
phenol was found at the highest concentration, followed by 2-methylphenol and then 3- and 4-
methylphenol and 2,4-dimethylphenol. Phenol was present in 35 of the 39 silica-based samples
from iron, steel, and aluminum foundries at concentrations ranging from 0.11-46.1 mg kg"1.
The 95th percentile concentrations for these five phenolics in SFS were compared to (1)
the human health SSLs for soil ingestion, and (2) the inhalation screening level concentrations
for the three compounds for which inhalation health benchmarks were available. No ecological
health benchmarks were available for the phenolic compounds found in SFS; therefore the
potential for adverse ecological impacts from exposure to phenolics in SFS was not evaluated.
As shown in Table 6-3, high-end phenolic concentrations in SFS are multiple orders of
magnitude below ingestion SSLs. Concentrations of phenolics in SFS were also orders of
magnitude below inhalation screening levels for those constituents with available inhalation
health benchmarks.
48
Due to a small data set (10 data points), it was decided to use the maximum value rather than the 95th percentile.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-10
-------
Chapter 6.0 Risk Characterization
Table 6-3. Comparison of Phenolic Concentrations in SFS to Screening Criteria
(mgkg-1)
SFS Constituent
4-Chloro-3 -methylphenol
2,4-Dimethylphenol
2-Methylphenol
3- and 4-Methylphenol
Phenol
95%-ile SFS
Concentration
0.09
5.60
8.76
3.59
22.1
Concurs
0.05
2.80
4.38
1.79
11.1
Adjusted SSL
620
120
310
310
1,800
Inhalation
Screening Level
N/A
N/A
Capped
Capped
Capped
N/A = no benchmark available.
Capped = Screening modeling estimates indicated risks below levels of concern at concentrations above
1,000,000 mg kg"1 (i.e., SFS could be comprised entirely of this constituent and still not cause risk).
Based on the above information, concentrations of these phenolic compounds in SFSs are
unlikely to cause adverse impacts to human health when SFS is used in SFS-manufactured soils,
soil-less potting media, or road base.
6.7 Metals and Metalloids in SFS
This section brings together previously presented information related to metals in SFS,
their behavior in soil, and results of screening and unitized risk-related modeling. Subsections for
the eight metals that were considered in the home gardener scenario screening (antimony,
arsenic, chromium (III), cobalt, copper, iron, manganese, and nickel) summarize information
comparing metal concentrations in SFS to screening criteria and modeling results to evaluate the
potential for adverse human health and ecological effects. Constituent-specific total
concentrations data for each sample can be found in Appendix B, Table B-19. Specific
subsections for each metal compare background concentrations in native soils to concentrations
in SFS to illustrate the similarity to native soils, as appropriate. Each subsection then describes
other factors that will affect the metal's mobility in soil, bioavailability to plants, and toxicity to
plants. These factors include processes that affect the dynamics of metal behavior associated
with SFS soil applications (e.g., sorption mechanisms), as well as metal-specific characteristics
that will limit or prevent certain exposure pathways from being completed (e.g., the soil-plant
barrier). Lastly, a lines-of-evidence section integrates this information and presents conclusions
regarding the potential risk associated with each of the eight metals evaluated in Phase II.
In addition to these detailed sections, information on other metal and metalloid
constituents found in SFS are summarized, essentially distilling all of the information presented
earlier in the report into a concise discussion of risk conclusions.
6.7.1 Antimony
The total antimony concentrations (see Table 2-4) in silica-based iron, steel, and
aluminum SFSs collected in June 2005 ranged from a minimum of <0.04 mg kg"1 to a maximum
of 1.71 mg kg"1 (using EPA method 3051 A), with a 95th percentile value of 1.23 mg Sb kg"1.
Using the SPLP and water extraction, the antimony results were all below the detection limit of
0.04 mg L"1 (Dungan and Dees, 2009) (Table 2-12, Table 2-13). Sample-specific SPLP and water
extract leachate data can be found in Appendix B, Tables B-13 through B-18.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-11
-------
Chapter 6.0 Risk Characterization
6.7.1.1 Comparison to Screening Levels
The relevant screening levels include Eco-SSLs49, the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL), the tapwater screening level (Tapwater SL), and the MCL
for drinking water. These screening levels typically reflect study data on highly bioavailable
forms of antimony:
• Eco-SSL (soil invertebrates): 78 mg kg"1 soil
• Eco-SSL (mammals): 0.27 mg kg"1 soil
• Adjusted SSL (noncancer): 3.1 mg kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
• Tapwater SL (noncancer): 0.0078 mg L"1
• MCL: 0.006 mg L"1
Comparing the 95th percentile total concentration of antimony in SFS to the SSLs shows
that, in a 1:1 manufactured soil blend (i.e., 50% SFS and 50% organic components, by weight),
the concentration of antimony in manufactured soil is below the Eco-SSL for soil invertebrates,
but exceeds the Eco-SSL for small insectivorous mammals. The 95th percentile antimony
concentration is well below the corresponding Adjusted SSL; at a 50% blend, even the maximum
concentration of antimony in SFS-manufactured soil would be below the Adjusted SSL. There
were no samples above the detection limit for the SPLP and water extraction tests. Although the
lack of detections suggests that antimony is unlikely to leach from SFS-manufactured soils at
levels of concern, the detection limits are above the Tapwater SL and MCL for antimony.
6.7.1.2 Modeling Results
Based on the comparison with screening levels, the groundwater ingestion pathway and
ecological exposure were further evaluated. The groundwater ingestion pathway evaluation used
one half the analytical method detection limit (0.02 mg L"1). The 90th percentile risk screening
results for dry climate were virtually zero (see Chapter 5, Section 5.2.2). The peak 90th
percentile risk screening results for central tendency and wet climates were 1.8E-3 and 5.9E-3
mg L"1, respectively, both below the Tapwater SL and MCL (7.8E-3 mg L"1 and 6.0E-3 mg L"1,
respectively).
The 95th percentile antimony concentration in SFS-manufactured soil (0.62 mg kg"1 DW)
was above the Eco-SSL for small mammals (0.27 mg kg"1 DW). Therefore, there was an
evaluation of the critical assumptions associated with the ecological hazard screen. One such
assumption was that 100% of the small mammal diet originated from the raised home garden
(e.g., for antimony, the shrew was the target species). As discussed in Section 5.3.8.2, the
percentage of the diet attributable to the home garden was adjusted to better reflect the behavior
of the shrew and provide a more realistic scenario for the usage of the home garden as part of the
shrew habitat. This refined ecological modeling estimated that up to a concentration of 4.1 mg
antimony kg"1 SFS (i.e., three times the 95th percentile antimony concentration in SFS), the
49 The Eco-SSL development process includes a number of very conservative modeling assumptions (e.g., metal
exists in most toxic form or highly bioavailable form, high food ingestion rate, high soil ingestion rate). Soil
concentrations above Eco-SSLs are not necessarily of concern, but need further study; constituents with soil
concentrations below Eco-SSLs need no further study.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-12
-------
Chapter 6.0 Risk Characterization
potential for adverse ecological effects would be below levels of concern. This suggests that
adverse ecological effects from antimony in SFS are unlikely for the home gardener scenario.
6.7.1.3 Soil Background Concentrations
Background concentrations of antimony in U.S. and Canadian soils range from 0.14-
2.3 mg kg"1, with a median value of 0.6 mg kg"1 (Smith et al., 2005). As illustrated in Figure 6-1,
the distribution of antimony in U.S. soils is shifted to the right of the distribution of antimony in
SFS. With a maximum SFS value of 1.71 mg kg"1, a 95th percentile value of 1.23 mg kg"1, and a
median SFS value of 0.17 mg kg'^Dayton et al., 2010), the majority of SFS-manufactured soils
would fall below median soil background concentrations. Therefore, the addition of SFS-
manufactured soil is likely to have little effect on the background soil concentrations of antimony
and, in many cases, the concentration of antimony in soil may decrease due to dilution.
14-
12-
I1"
,5 8
O
i- H
0)
a
£ 4-
2-
SFS Dataset, 2009
N=39
nnn n n
D
0.0 0.5 1.0 1.5 2.0 2.5
SFS Sb Concentration, mg kg'1
Ali-
aS-
CD
I25-
re
(0 20-
0
5-
n.
-
|
USGS Dataset, 2005
N=254
H n^n ^n inn.
0.0 0.5 1.0 1.5 2.0 2.5
Soil Sb Concentration, mg kg
Figure 6-1. Concentration distributions of antimony in SFS (top)
and U.S. and Canadian soils (bottom).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-13
-------
Chapter 6.0 Risk Characterization
6.7.IA Additional Factors
Although antimony is not an essential nutrient for plants (e.g. Kabata-Pendias, 2001), it is
generally considered to be readily taken up by plants. The few studies that have been published
on the phytotoxicity of antimony indicate that antimony is moderately phytotoxic (Pais and
Benton Jones, 1997). The lack of reference materials are likely responsible for a lack of
sufficient data for EPA to establish an Eco-SSL for terrestrial plants.
6.7.1.5 Lines of Evidence
The 95th percentile concentration of antimony in SFS (1.23 mg kg"1) falls well within the
range of typical background concentrations of antimony in U.S. and Canadian soils (Smith et al.,
2005). Therefore, the addition of SFS-manufactured soils to native soils (home gardens) would
not be expected to result in significant changes with regard to antimony concentrations.
The 90th percentile screening probabilistic modeling results for the groundwater ingestion
pathway were virtually zero for the dry climate, and were below the Tapwater SL and MCL
(0.0078 mg L"1 and 0.006 mg L"1, respectively) for central tendency and wet climates.
The risk screening results for ecological receptors showed that the 95th percentile
concentration of antimony in SFS was below the Eco-SSL for soil invertebrate receptors, but
exceeds the Eco-SSL for the most sensitive mammalian receptor group, the shrew. Even though
the Eco-SSL for mammals (0.27 mg kg"1 DW) was below the median background concentration
for antimony in the US and Canada (0.6 mg kg"1 DW), refined probabilistic modeling was
conducted to determine if quantitative estimates of ecological hazard would be above levels of
concern. The approach described in Section 5.3.8 resulted in an SFS-specific ecological
screening level for antimony of 4.1 mg kg"1 SFS (dry weight), three times the 95th percentile
antimony concentration in SFS.
Based on the results of the risk screening and probabilistic screening modeling, and
similarity with background concentrations, the levels of antimony in SFS are unlikely to cause
adverse effects to human health and ecological receptors when SFS is used in SFS-manufactured
soils, soil-less potting media, or road base.
6.7.2 Arsenic
The total arsenic concentrations (see Table 2-4) in silica-based SFSs from iron, steel, and
aluminum foundries collected in June 2005 (39 detects in 39 samples) ranged from a minimum
of 0.13 mg kg"1 to a maximum of 7.8 mg kg"1 (using EPA method 3051 A), with a 95th percentile
value of 6.44 mg kg"1 (Dayton et al., 2010). The SPLP leach test data for these same samples (22
of 39 detects) ranged from below the detection limit of 0.001 mg L"1 to a maximum of 0.098 mg
L"1, with a mean value of 0.007 mg L"1. The concentrations in water extracts from the same
samples (23 detects in 39 samples), ranged from below the detection limit of 0.001 mg L"1 to a
maximum of 0.018 mg L"1, with a mean value of 0.005 mg L'^Dungan and Dees, 2009).
Sample-specific SPLP and water extract leachate data can be found in Appendix B, Tables B-13
through B-18
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-14
-------
Chapter 6.0 Risk Characterization
6.7.2.1 Comparison to Screening Levels
The relevant screening levels include Eco-SSLs, the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL), the tapwater screening level (Tapwater SL), and the MCL
for drinking water. These screening levels typically reflect study data on highly bioavailable
forms of arsenic:
• Eco-SSL (plants): 18 mg kg'1 soil
• Eco-SSL (mammals): 45 mg kg"1 soil
• Adjusted SSL (cancer): .43 mg kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways, as well as a target risk level of 1E-5)
• Tapwater SL (cancer): 4.5E-4 mg L"1
• MCL: 0.01 mgL"1.
Comparing the 95th percentile total concentration of arsenic in SFS to the SSLs suggests
that, in a 1:1 manufactured soil blend (i.e., 50% SFS and 50% organic components, by weight),
the concentration of arsenic in soil would be well below any of the identified ecological
screening criteria. The 95th percentile arsenic concentration is also below (though not an order of
magnitude below) the Adjusted SSL for the soil pathways; in a 50% blend, even the maximum
concentration of arsenic from an SFS-manufactured soil would be below the Adjusted SSL.
However, the comparison of the SPLP data from the 23 SFS samples that exceeded the detection
limit of 0.001 mg L"1, along with the water extract samples, indicates that the 95th percentile
arsenic concentrations associated with these tests would exceed both the Tapwater SL and the
MCL.
6.7.2.2 Modeling Results
The soil manufacturing scenario (inhalation of fugitive dust emissions by nearby
residents) and the home gardener scenario (the groundwater ingestion pathway, and ingestion of
soil and home-grown produce) were evaluated. For the inhalation exposure pathway, the
screening results indicated that, up to a concentration of 40.2 mg kg"1 SFS, the potential for
adverse human health impacts from arsenic in SFS-manufactured soil would be below levels of
concern.
For the groundwater ingestion pathway, the 90th percentile probabilistic risk screening
results were above the lowest screening level (i.e. the Tapwater SL) in the Wet and Central
Tendency climates. However, more refined probabilistic modeling of the groundwater pathway
found that the risk due to the ingestion of drinking water would be below the levels of concern
up to a concentration of 59 mg kg"1 SFS.
The soil/produce pathway refined probabilistic results indicated that, up to a
concentration of 8.0 mg kg"1 SFS, the risk due to the consumption of home-grown fruits and
vegetables along with incidental soil ingestion would be below levels of concern.
For the home gardener scenario, separate target SFS screening concentrations were
developed for the soil/produce and the groundwater pathways based on analyses that showed that
these exposures are not likely to occur within the same timeframe.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-15
-------
Chapter 6.0 Risk Characterization
6.7.2.3 Soil Background Concentrations
The range of background concentrations of arsenic in U.S. soils is broad, ranging from
<0.1-93 mg kg"1 (Kabata-Pendias, 2001). The geometric mean of arsenic in surficial soils has
been estimated at 5.8 mg kg"1 (Shacklette and Boerngen, 1984) and more recent studies on
Canadian and U.S. surficial soils estimate that the median concentration of arsenic is 5.0 mg kg"1
(Smith et al., 2005). With a maximum SFS value of 7.79 mg kg"1, a 95th percentile value of 6.44
mg kg"1, and a median value of 1.05 mg kg"1, almost all arsenic concentrations in SFS fall below
the median soil background concentrations (Dayton et al., 2010). Given the importance of site-
specific soil properties—particularly the iron and aluminum content in soil—the comparison
between arsenic concentrations in SFS and arsenic background concentrations in soil suggests
that arsenic concentrations in SFS overlap with the low end of the background concentration
range, with the 95th percentile value in SFS slightly higher than the average soil background
level. It is expected that nearly 95% of the SFS samples would have arsenic concentrations that
were below the median national background soil arsenic level. Figure 6-2 demonstrates these
points graphically.
14-
12-
SFS Dataset, 2009
N=39
o.
E
n
(0
o
si
E
012345678 91011121314151617181920
SFS As Concentration, mg kg
-1
45
USGS Dataset, 2005
N=254
012345678 91011121314151617181920
Soil As Concentration, mg kg'1
Figure 6-2. Concentration distributions of arsenic in SFS (top)
and U.S. and Canadian soils (bottom).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-1
-------
Chapter 6.0 Risk Characterization
6.7.2.4 Additional Factors
Arsenic is a constituent of most plants, although little is known about its biochemical role
(Kabata-Pendias, 2001). The arsenic concentration in plants grown on uncontaminated soils
varies from 0.009-1.5 mg kg"1 DW, with leafy vegetables falling into the upper end of the range
and fruits falling into the lower end of the range. Some authors have shown that the uptake of
arsenic depends upon the form of arsenic in the soil; for the radish, the order of uptake is As0rg
» As(V) > As(III) (Tlustos et al., 1998). There are some reports that indicate the linear uptake
of arsenic in soil considers both the soluble and total arsenic forms; however, more recent
research has shown that AsCV" enters plant roots on the phosphate transporter (Zhao et al.,
2009). Although some plant species have been shown to tolerate high levels of arsenic in the
tissues, the residue tolerance has generally been established around 2 mg kg"1 DW for plant
species that are neither highly sensitive nor highly tolerant (Kabata-Pendias, 2001). Phytotoxicity
appears to vary with the soil type; "heavy" soils with high organic matter content and
vermiculitic clay as the predominant clay tend to significantly reduce the toxicity of arsenic to
plants (Woolson et al., 1973).
The chemical reactions of arsenic in soils are thought to be controlled largely by the
oxidation state, with the As(V) and As(III) forms dominant at the typical oxidation potential (Eh)
and pH ranges of soil. The bioavailability of arsenic in soil is significantly reduced in the
presence of hydrated iron and aluminum oxides.50 A change in the redox potential of the soil to
flooded anaerobic conditions results in the greater desorption of As(III), the more highly
bioavailable form; flooded arsenic contaminated soils are known to cause arsenic phytotoxicity
to rice, but not to other crops. In aerobic soils, As(V) predominates, and solubility can be
increased by high additions of phosphate. In short, the chemistry and behavior of arsenic in soil
is a highly complex, multivariate phenomenon that depends greatly on soil characteristics,
especially soil pH and the redox potential, and the presence of other metals that form arsenical
complexes that are generally not available to plants.
Given the complexities of arsenic behavior in soil, an additional analysis was performed
that examined the impact of soil water partitioning coefficient (Kd) distributions on SFS
screening levels as discussed in Appendix G, Attachment E. As described in Section 5.3, the
home gardener scenario assumed that the properties and characteristics of the SFS-manufactured
soil mimicked those of natural soil in the area. Accordingly, the SFS-specific screening levels
were developed based on soil Kd values from U.S. EPA 2005. The resulting screening levels for
the soil/produce and groundwater pathways were 8.0 mg As kg"1 SFS and 59 mg As kg"1 SFS,
respectively. Under the Kd analysis, source modeling was also performed with an SFS waste-
specific Kd distribution developed using the full set of whole waste/1 eachate pairs presented in
Appendix B (i.e., the SFS total waste concentration was divided by the corresponding leachate
concentration). Release estimates developed using the waste-specific Kds represent releases from
SFS and so are not likely to accurately reflect releases from SFS-manufactured soil. While not
used to generate recommended SFS-specific screening levels, these estimates represent a
bounding study. The goal of this effort was to better characterize the uncertainty associated with
the SFS arsenic screening levels. Table 6-4 compares the soil-Kd based SFS Screening Levels
and the bounding material-specific Kd screening levels. As seen from this table, the lowest soil-
50 To reflect this reduction, the exposure estimates developed for incidental ingestion of soil were adjusted using the
EPA's default relative bioavailability (RBA) value of 60% (U.S. EPA, 2012b).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-1
-------
Chapter 6.0 Risk Characterization
Kd screening level (i.e., 8.0 mg kg"1 SFS, for the soil/produce pathway) is nearly identical to the
lowest screening level generated using the material-specific Kd distribution (i.e. 7.7 mg kg"1
SFS, for the groundwater pathway). The similarity between the recommended screening level
and the bounding material-specific estimate fosters a high level of confidence that an SFS-
specific screening level generated using the soil Kd distribution will be protective of human
health under a range of pathways and environmental conditions.
Table 6-4. Home Gardening 90th Percentile Modeled SFS-specific Screening Levels
for Arsenic
Pathway
Soil/Produce
Groundwater
Arsenic SFS Screening Levels (mg kg -1)
Based on Soil Kd
Distribution
8.0
59
Bounding Estimate:
Material-Specific Kd
Distribution
9.5
7.7
6.7.2.5 Lines of Evidence
Based on the results of the comparison of total arsenic concentrations from SFS to Eco-
SSLs, arsenic concentrations in SFS are unlikely to cause adverse health effects to ecological
receptors.
For the home gardener scenario, the results of the probabilistic groundwater screening
modeling showed that the 90th percentile exposure concentration in water could be above the
lowest screening value in the Wet and Central Tendency climates. More refined, yet still
conservative groundwater modeling found that the risk due to the ingestion of drinking water
would be below the levels of concern up to an SFS arsenic concentration of 59 mg kg"1 SFS,
which is well above the 95th percentile SFS concentration of 6.44 mg kg"1 SFS.
For the ingestion of home-grown produce and the incidental ingestion of soil, the most
conservative modeled SFS-specific screening concentration of 8.0 mg kg"1 SFS is even above the
maximum arsenic concentration in SFS, suggesting that human exposure to arsenic via the
ingestion of vegetables and fruit grown in SFS-manufactured soil will be below levels of
concern. The conservative nature of the refined modeling (e.g., allowing simultaneous, high
consumption rates for multiple produce types) is such that arsenic concentrations in SFS are
unlikely to cause adverse health impacts even at produce consumption rates.
The screening modeling analyses also evaluated inhalation risks to receptors living
adjacent to a soil manufacturing facility (the most conservative of the inhalation exposure
scenarios). This modeling generated allowable arsenic concentrations more than an order of
magnitude above the 95th percentile and maximum arsenic concentrations found in SFS samples.
Therefore, because (1) the arsenic concentration in SFS is below all Eco-SSLs; (2)
probabilistic modeling found that the potential for adverse health impacts from use of SFS-
manufactured soil are below levels of concern in all evaluated exposure pathways; and (3)
arsenic concentrations in SFSs are typically below average background soil concentrations,
arsenic in silica-based SFS from iron, steel, and aluminum foundries is unlikely to cause adverse
effects to human health or ecological receptors when SFS is used in manufactured soil, soil-less
potting media, and road base.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-2
-------
Chapter 6.0 Risk Characterization
6.7.3 Chromium
The total chromium concentrations (see Table 2-4) in silica-based SFSs from iron, steel,
and aluminum foundries collected in June 2005 (38 detects in 39 samples) ranged from a
minimum of <0.5 mg kg"1 to a maximum of 115 mg kg"1 (using EPA method 3051 A), with a 95th
percentile value of 109 mg kg"1 (Dayton et al., 2010). The SPLP and water extract leach test data
for these same samples were below the quantitative detection limits of 0.01 mg L"1 and 0.02 mg
L"1, respectively, for all samples (Dungan and Dees, 2009). Sample-specific SPLP and water
extract leachate data can be found in Appendix B, Tables B-13 through B-18.
6.7.3.1 Comparison to Screening Levels
The relevant screening levels include Eco-SSLs, the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL), the tapwater screening level (Tapwater SL), and the MCL
for drinking water. These screening levels typically reflect study data on highly bioavailable
forms of chromium (III):
• Eco-SSL (mammals): 34 mg kg"1 soil
• Adjusted SSL (noncancer): 12,000 mg kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
• Tapwater SL (noncancer): 16 mg L"1
• MCL: 0.1 mg L"1 (based on total Cr)
Comparing the 95th percentile total concentration of chromium in SFS to the SSLs
suggests that, in a 1:1 manufactured soil blend (i.e., 50% SFS and 50% organic components, by
weight), the concentration of chromium in SFS-manufactured soil would be above the Eco-SSL
for small insectivorous mammals. However, this same concentration is below the Adjusted SSL
for soil pathways; in a 50% blend, even the maximum concentration of chromium in SFS-
manufactured soil would be below the Adjusted SSL. The SPLP and water extract leach data
were all well below the Tapwater SL and MCL screening levels.
6.7.3.2 Modeling Results
The 95th percentile chromium III concentration in SFS-manufactured soil (109 mg kg"1
DW) was above the Eco-SSL for small mammals (34 mg kg"1 DW). This prompted a refinement
of the assumptions associated with the ecological hazard screen. For chromium this involved
refining the assumption that 100% of the small mammal diet originated from the home garden
(for chromium, the shrew was the target species). As discussed in Section 5.3.8.2, the percentage
of the diet attributable to the home garden was adjusted to better reflect the behavior of the shrew
and provide a more realistic scenario for the usage of the home garden as part of the shrew
habitat. Refined ecological modeling estimated that up to a trivalent chromium concentration of
510 mg kg"1 SFS (i.e., almost five times the 95th percentile trivalent chromium concentration in
SFS), the potential for adverse effects to even the most sensitive ecological receptors would fall
below levels of concern. Therefore, adverse ecological effects from chromium in SFS are
unlikely for the home gardener scenario.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-3
-------
Chapter 6.0 Risk Characterization
6.7.3.3 Soil Background Concentrations
The range of background concentrations of chromium in U.S. soils is broad, ranging from
3-5,320 mg kg"1, with a median value of 27 mg kg'^Smith et al., 2005). As illustrated in Figure
6-3, the distribution of chromium concentrations in SFS is similar to that of background soils;
however, the median concentrations for SFS is roughly 5 times lower than the median
concentration in background soils. Given this comparison, the addition of SFS to soil is not
expected to result in a significant change with regards to chromium concentrations.
20-
8 15-
SL
ro
OT
•5 10'
E
3 5-I
SFS Dataset, 2009
N=39
n nn
n n nn
20 40 60 80
SFS Cr Concentration, mg kg
100 120
-1
40-
(A
"a
30-
.? 2°H
| '
z 10-
-T~
20
USGS Dataset, 2004
N=254
j^
nrir
40
60 80
Soil Cr Concentration, mg kg
100 120
-1
Figure 6-3. Concentration distributions of chromium in SFS (top)
and U.S. and Canadian soils (bottom).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-4
-------
Chapter 6.0 Risk Characterization
6.7.3.4 Additional Factors
Chromium III is not believed to be an essential nutrient for plants, although some studies
have reported a stimulatory effect. Chromium is not readily taken up by plants, as there is a
relatively low rate of absorption, largely attributed to the mechanism of uptake in plant roots. As
with many metals, the content of chromium in plants is dependent on the concentration of
soluble chromium in soils, the soil type, and the plant species (Kabata-Pendias, 2001). Pais and
Benton Jones (1997) estimated average concentrations of chromium in plants to be 0.02 to 0.2
mg kg"1, with phytotoxic concentrations averaging 10 to 15 mg kg"1, and upper phytotoxic
concentrations at > 150 mg kg"1 in soil. In terms of edible plants and crop species, average
concentrations of total chromium in foods range from 0.05 mg kg"1 (apple) to 0.2 mg kg"1
(wheat) (Pais and Benton Jones, 1997). As evident from these data, chromium has been reported
in varying ranges. However, some studies have documented that concentrations in plants may
actually be an artifact of soil contamination issues related to sampling techniques rather than
uptake by plants (e.g., Gary and Kubota, 1990; Grubinger et al., 1994; and Gary et al., 1994).
6.7.3.5 Lines of Evidence
The 95th percentile chromium concentration in SFS (109 mg kg"1) falls well within the
range of typical background concentrations of chromium for U.S. and Canadian soils (Smith et
al., 2005). Therefore, the addition of SFS-manufactured soils to native soils (home gardens)
would not be expected to result in significant changes to chromium concentrations.
The evaluation found the 95th percentile concentration of chromium in SFS to be below
the health-based benchmarks for human receptors, but exceeded the Eco-SSL for small
mammals. However, refined ecological modeling demonstrated, with a high degree of
confidence that the risk to the target ecological receptor (shrew) would be below levels of
concern. The approach described in Section 5.3.8 resulted in an SFS-specific ecological
screening level for chromium III of 510 mg kg"1 DW, more than 100 times higher than the 95th
percentile chromium concentration in SFS.
Based on the results of the screening comparison, the refined ecological modeling, and
the similarity with background concentrations, chromium levels in in SFS are unlikely to cause
adverse effects to human health and ecological receptors when SFS is used in SFS-manufactured
soils, soil-less potting media, or road base.
6.7.4 Cobalt
The total cobalt concentrations (see Table 2-4) in silica-based SFSs from iron, steel, and
aluminum foundries collected in June 2005 (28 detects in 39 samples) ranged from a minimum
of <0.5 mg kg"1 to a maximum of 6.62 mg kg"1 (using EPA method 3051 A), with a 95th percentile
value of 5.99 mg kg"1 (Dayton et al., 2010). No leach test data were available for cobalt.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-5
-------
Chapter 6.0 Risk Characterization
6.7.4.1 Comparison to Screening Levels
The relevant screening levels include Eco-SSLs and the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL). These screening levels typically reflect study data on highly
bioavailable forms of cobalt:
• Eco-SSL (terrestrial plants): 13 mg kg"1 soil
• Eco-SSL (mammals): 230 mg kg"1 soil
• Adjusted SSL (noncancer): 2.3 g kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
Comparing the 95th percentile total concentration of cobalt in SFS (5.99 mg kg"1 DW)to
the lowest Eco-SSL (13 mg kg"1 DW) indicates that the concentration of cobalt in SFS-amended
soil would be below the Eco-SSL for terrestrial plants (and substantially below that for
mammals). This cobalt concentration in SFS-manufactured soil exceeded the Adjusted SSL for
the soil ingestion pathways. No leachate data were available for cobalt in SFS and, therefore,
cobalt was not evaluated via the groundwater pathway.
6.7.4.2 Modeling Results
The soil manufacturing scenario (inhalation of fugitive dust emissions by nearby
residents), and the home gardener scenario (ingestion of home-grown produce, and incidental
ingestion of garden soil) were evaluated. For the inhalation exposure pathway, the screening
modeling results indicate that up to a cobalt concentration of 2,010 mg kg"1 SFS (i.e., more than
100 times higher than the 95th percentile concentration of cobalt in SFS), the potential for
adverse human health impacts would be below levels of concern.
With respect to the home garden scenario, the results of the refined modeling indicate
that up to a cobalt concentration of 21 mg kg"1 SFS (i.e., over three times the 95th percentile
concentration if cobalt in SFS), the use of SFS in manufactured soil is unlikely to cause adverse
human health impacts.
6.7.4.3 Soil Background Concentrations
The range of background concentrations of cobalt in U.S. and Canadian soils is broad,
ranging from 0.5-143.4 mg kg"1, with a median value of 7.1 mg kg'^Smith et al., 2005). As
illustrated in Figure 6-4, the composition of SFS with respect to cobalt appears to be
substantially below U.S. soils, suggesting that the addition of SFS to soil would nearly always
dilute cobalt levels in native soils.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-6
-------
Chapter 6.0 Risk Characterization
35
30
I 25
Q.
$ 20
& 15
E
Z 10
SFS Dataset, 2009
N=39
10 20 30 40 50 60
SFS Co concentration, mg kg'1
100-
M
1
E
60-|
o
1 40-
Z 20-
USGS Dataset, 2005
N=254
0
-------
Chapter 6.0 Risk Characterization
Plant uptake of cobalt is highly dependent on environmental conditions, such as soil
factors, and varies widely across plant species. For instance, legumes have been shown to
accumulate more cobalt than grasses or grain crops. Moreover, soil texture has been cited as one
of the most significant parameters controlling cobalt concentrations in plants. Some plants,
termed hyperaccumulators, have developed a tolerance mechanism and are able to accumulate
high concentrations of cobalt. In terms of edible plants, cobalt content has been shown to vary
from 8 ppm (e.g. apples) to 100 ppm (e.g. cabbage) (DW). Studies from different countries
report average cobalt concentrations in clover range from 0.10 to 0.57 ppm (DW), while grass
concentrations range from 0.03 to 0.27 ppm (Kabata-Pendias, 2001 and references within).
6.7.4.5 Lines of Evidence
The distribution of cobalt concentrations in SFS is below the distribution in native soils;
the 95th percentile SFS concentration (5.99 mg kg"1) is below the background concentration
median of 7.1 mg kg"1, suggesting that the addition of SFS will tend to dilute rather than increase
the level of cobalt in soils.
For the ingestion of home-grown produce and the incidental ingestion of SFS-
manufactured soil, the most conservative SFS-specific screening concentration for cobalt (i.e., 21
mg kg"1 SFS) is well above the 95th percentile concentration of cobalt in SFS. The conservative
nature of the refined screening modeling for these exposure pathways fosters a high level of
confidence that an SFS-specific concentration of 21mg kg"1 is protective of human health.
Based on the results of the comparison of total cobalt concentrations in SFS with
screening criteria, and probabilistic modeling, cobalt concentrations in SFS are unlikely to cause
adverse effects to human health and ecological receptors when SFS is used in SFS-manufactured
soils, soil-less potting media, or road base.
6.7.5 Copper
The total copper concentrations (see Table 2-4) in silica-based iron, steel, and aluminum
SFSs collected in June 2005 (39 of 39 detects) ranged from a minimum of <0.5 mg kg"1 to a
maximum of 137 mg kg"1 (using EPA method 3051 A), with a 95th percentile value of 107 mg Cu
kg"1 (Dayton et al., 2010). The SPLP leach test data for these same SFSs, from all three sampling
events (June 2005, September 2005, July 2006) were below the quantitative detection limit of
0.07 mg L"1 for all samples. The concentrations in water extracts from the same samples (June
2005 with 2 detects, September 2005 with 0 detects, July 2006 with 1 detect), ranged from <0.07
mg L"1 to a maximum of 1.06 mg L"1, with mean values of 0.070, 0.035, and 0.041 mg L"1 across
the sampling schemes, respectively (Dungan and Dees, 2009). Sample-specific SPLP and water
extract leachate data can be found in Appendix B, Tables B-13 through B-18.
6.7.5.1 Comparison to Screening Levels
The relevant screening levels include Eco-SSLs, the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL), the tapwater screening level (Tapwater SL), and the MCL
for drinking water. These screening levels typically reflect study data on highly bioavailable
forms of copper (Table 4-12, Table 7-1, and Table 4-2, respectively):
• Eco-SSL (terrestrial plants): 70 mg kg"1 soil
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-8
-------
Chapter 6.0 Risk Characterization
• Eco-SSL (soil invertebrates): 80 mg kg"1 soil
• Eco-SSL (mammals): 49 mg kg"1 soil
• Adjusted SSL (noncancer): 310 mg kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
• Tapwater SL (noncancer): 0.62 mg L"1
• MCLil.Smgl/1
Comparing the 95th percentile total concentration of copper in SFS to the SSLs indicates
that, in a 1:1 manufactured soil blend (i.e., 50% SFS and 50% organic components, by weight),
the concentration of copper in SFS-manufactured soil would fall below the Eco-SSLs for
terrestrial plants and soil invertebrates, but exceed the Eco-SSL for mammals. The copper
concentration in SFS-manufactured soil is well below the corresponding Adjusted SSL for soil
pathways; at a 50% blend, even the maximum concentration of copper from an SFS-
manufactured soil would be below the Adjusted SSL.
Comparing the 95th percentile leachate concentration of copper in SFS to the Tapwater
SL and MCL, the concentration of copper in SFS-manufactured soil is well below both relevant
human health water screening levels.
6.7.5.2 Modeling Results
Given the results of the screening comparison for ecological receptors, probabilistic
screening modeling was performed and predicted copper exposure concentrations were
compared to the Eco-SSLs. As discussed in Section 5.3.8.2, the percentage of the diet
attributable to the home garden was adjusted to better reflect the behavior of the shrew and
provide a more realistic scenario for the usage of the home garden as part of the shrew habitat.
The refined ecological modeling results indicate that up to a copper concentration of 160 mg kg"1
SFS, the risk posed to ecological receptors would be below levels of concern (see Table 5-14).
As this is higher than the 95th percentile copper concentration in SFS (i.e., 107 mg kg"1 SFS), this
indicates that copper found in SFS is below levels of concern for ecological receptors.
6.7.5.3 Soil Background Concentrations
Background concentrations of copper in U.S. and Canadian soils range from 1.7-
81.9 mg kg"1, with a median value of 12.7 mg kg"1 (Smith et al., 2005). As illustrated in Figure
6-5, the distribution of Cu concentrations in background soils is similar to the distribution of
concentrations in SFS (e.g., the respective medians are within a factor of 2). However, the tail of
the SFS distribution is characterized by higher concentrations than the tail of the distribution for
background soils (see Table 7-1). Nevertheless, the addition of SFS-manufactured soil would not
be expected to result in significant changes in the Cu concentration in native soils.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-9
-------
Chapter 6.0 Risk Characterization
18-
16-
g 14-
-12-
(0
w 10-
1
I
z 4-
2-
0-
SFS Dataset
N=39
1
mlnm...nfl. ....... i
0 20 40 60 80 100 120 140 160 180 200
SFS Cu Concentration, mg kg'1
.92
g"
re
CD
-Q
E
60-
50-
40-
30-
20-
10-
USGS Dataset, 2005
N=254
0 20 40 60 80 100 120 140 160 180 200
Soil Cu Concentration, mg kg '
Figure 6-5. Concentration distributions of copper in SFS (top)
and U.S. agricultural soils (bottom).
6.7.5.4 Additional Factors
Copper exists normally in soil, primarily as complexed forms of low molecular weight
organic compounds, such as humic and fulvic acids (Pais and Benton Jones, 1997). Copper is an
essential micronutrient for plants and, under normal conditions, its sufficiency range is 5-30 mg
kg"1 (DW) (Pais and Benton Jones, 1997). Copper is important for photosynthesis, respiration,
carbohydrate distribution, and protein metabolism, as well as nitrogen fixation processes
(Kabata-Pendias, 2001). Similar to other metals, there is a variation in tolerance to copper among
different plant species. Copper uptake depends mainly on the type of copper species (i.e. the
oxide form of copper, largely coming from anthropogenic sources, is more bioavailable than
copper coming from pedogenic sources). However, once copper has been absorbed by plant
roots, relatively little is expected to be transported to plant tops (Pais and Benton Jones, 1997). In
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-10
-------
Chapter 6.0 Risk Characterization
fact, copper has a relatively low mobility inside plant bodies compared to other elements; most
of it will remain in the root and leaf tissues until they senesce, and only small amounts may
move to young organs.
The distribution of copper inside plants varies, but the general trend is that translocation
to leaves and other organs occurs predominantly when there is an abundance of copper available,
and the plant is undergoing intensive growth (Kabata-Pendias, 2001). Average concentration
ranges of copper in various foods include: vegetables: 0.1 (for celery root) - 3.2 (for garlic
cloves) ppm FW; fruits: 0.3 (for grapes) - 4 (avocadoes) ppm FW; cereals: 0.3 (oats, whole
grain) - 13 (rye, whole grain) ppm FW; and nuts: 0.2 (fresh coconut meat) - 23.8 (shelled Brazil
nuts) (Kabata-Pendias, 2001).
6.7.5.5 Lines of Evidence
The 95th percentile copper concentration in SFS (107 mg kg"1) falls well within the range
of typical background concentrations of copper in U.S. and Canadian soils (Smith et al., 2005).
Therefore, the addition of SFS-manufactured soils to native soils (home gardens) would not be
expected to result in significant changes to copper concentrations.
The screening comparison indicated that copper in SFS-manufactured soil is below levels
of concern for human exposures, but exceeded the Eco-SSL for small mammals. Refined
ecological modeling demonstrated, with a high degree of confidence that the risk to the target
ecological receptor (shrew) would be below levels of concern. The approach described in Section
5.3.8 resulted in an SFS-specific ecological screening level for copper of 159 mg kg"1 SFS,
which is above the 95th percentile copper concentration in SFS.
Based on the results of the screening comparison for human health, the refined ecological
modeling, and the similarity with background concentrations, copper levels in SFS-manufactured
soil are unlikely to cause adverse effects to human health or ecological receptors when SFS is
used in SFS-manufactured soils, soil-less potting media, or road subbase.
6.7.6 Iron
The total iron concentrations (see Table 2-4) in silica-based SFSs from iron, steel, and
aluminum foundries collected in June 2005 (39 detects) ranged from a minimum of 1.28 g kg"1 to
a maximum of 64.4 g kg"1 (using EPA method 3051 A), with a 95th percentile value of 57.1 g kg"1
(Dayton et al., 2010). No leach test data were available for iron.
6.7.6.1 Comparison with Screening Levels
The relevant screening levels include the default Residential soil screening level for the
protection of human health, adjusted to also address home gardener produce ingestion pathways
(Adjusted SSL). Screening levels typically reflect study data on highly bioavailable forms of
iron:
• Adjusted SSL (noncancer): 5.5 g kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
Comparing the 95th percentile total iron in SFS to the Adjusted SSL indicates that, in a
1:1 manufactured soil blend (i.e., 50% SFS and 50% organic components, by weight), the iron
concentration in SFS-manufactured soil would exceed the Adjusted SSL. Iron was therefore
evaluated under the Phase II probabilistic risk modeling.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-11
-------
Chapter 6.0 Risk Characterization
6.7.6.2 Modeling Results
The refined modeling results demonstrate that up to an iron concentration of 150 g kg"1
SFS (i.e., almost three times the 95th percentile iron concentration in SFS), adverse human health
effects are unlikely to occur.
6.7.6.3 Soil Background Concentrations
The range of iron background concentrations in U.S. and Canadian soils is broad, ranging
from 3.8-87.7 mg kg"1, with a median value of 19.2 mg kg'^Smith et al., 2005). As illustrated in
Figure 6-6, the iron concentration in SFS would generally be lower than the iron concentration
in native soils. The 95th percentile and maximum iron concentrations in SFS are, respectively,
both below the corresponding background concentrations, and the median value for SFS is
roughly 5 times lower than the median in native soils. This strongly suggests that the addition of
SFS-manufactured soils would generally have a diluting effect on the iron concentrations in soil.
SFS Dataset, 2009
N=39
10 20 30 40 50 60 70 80
SFS Fe Concentration, g kg'
60-
USGS Dataset, 2004
N=254
10 20 30 40 50 60 70 80
Soil Fe Distribution, g kg'1
Figure 6-6. Concentration distributions of iron in SFS (top)
and U.S. and Canadian soils (bottom).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-12
-------
Chapter 6.0 Risk Characterization
6.7.6.4 Additional Factors
Iron is an essential micronutrient for all life. The behavior of iron and iron oxides in
terrestrial systems is quite complex and specific to the characteristics (e.g., carbon content of
soil) of the environment. Iron deficiency in many crops worldwide has led to numerous
investigations over the past several decades in order to better understand and mitigate iron
deficiencies in important crop plants. Iron deficiency has been associated mostly with alkaline
soils, the presence of organic matter, soils with high Zn concentrations, the presence of
bicarbonate anion (HCO's), and have been noted especially in arid or semi-arid regions (Pais and
Benton Jones, 1997; Kabata-Pendias, 2001). As an essential nutrient, iron is readily taken up by
plants, usually in the form of the Fe2+ cation. When bound to a bio-chelating agent, Fe3+ uptake
can also take place. The ability for plant roots to reduce Fe3+ to Fe2+ is one of the most
fundamental processes in the absorption of iron in most plants. At excessive soluble iron
concentrations, it can be phytotoxic to plants. Phytotoxicity is most likely to occur on strongly
acidic soils, on acid sulfate soils, or flooded soils (Kabata-Pendias, 2001).
The normal iron content in plants ranges from 20 to 100 mg kg"1, with a sufficiency range
of 5-500 mg kg"1 (DW) (Pais and Benton Jones, 1997). Iron content in common foods ranges
from approx. 8 to 40 mg kg"1 (Pais and Benton Jones, 1997), although higher concentrations in
food plants have also been documented (e.g. some grasses and clover with concentrations up to
1000 ppm DW) (Kabata-Pendias, 2001). Kabata-Pendias (2001) summarize concentrations of
iron in common food crops, with all values in ppm (FW): vegetables, 3 (celery root) - 31
(spinach); fruits, 1 (apples, honey melon) - 11 (black currant); cereals, 3 (barley pearls) - 37 (rye,
whole grain); nuts, 11 (hazelnuts) - 47 (almonds).
6.7.6.5 Lines of Evidence
Iron is well documented as an essential micronutrient for all life, hence the general lack
of health and environmental benchmarks for use in the screening comparison. The concentration
distribution for iron in SFS indicates that, relative to native soils, SFS would not contribute iron
content at a level that would approach phytotoxicity, even for acidic soils. The refined modeling
generated SFS-specific screening levels orders of magnitude above concentrations found in SFS.
Based on these results, iron levels in SFS soil are unlikely to cause adverse effects to human
health or ecological receptors when SFS is used in SFS-manufactured soils, soil-less potting
media, or road subbase.
6.7.7 Manganese
The total manganese concentrations in silica-based iron, steel, and aluminum SFSs
collected in June 2005 (39 of 39 detects) ranged from a minimum of 5.6 mg kg"1 to a maximum
of 707 mg kg"1 (using EPA method 3051 A), with a 95th percentile value of 670 mg kg"1 (Dayton
et al., 2010). No leach test data were available for manganese.
6.7.7.1 Comparison with Screening Levels
The relevant screening levels include Eco-SSLs, and the default Residential soil
screening level for the protection of human health adjusted to also address home gardener
produce ingestion pathways (Adjusted SSL). These screening levels typically reflect study data
on highly bioavailable forms of manganese:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-13
-------
Chapter 6.0 Risk Characterization
• Eco-SSL (plants): 220 mg kg'1 soil
• Eco-SSL (soil invertebrates): 450 mg kg"1 soil
• Eco-SSL (mammals): 4,000 mg kg"1 soil
• Adjusted SSL (noncancer): 1,800 mg kg"1 soil (Residential SSL, adjusted to also address
produce ingestion pathways)
Comparing the 95th percentile total concentration of manganese in SFS to the SSLs
suggests that in a 1:1 manufactured soil blend, concentrations of manganese in SFS-
manufactured soil would be below the Eco-SSLs for soil invertebrates and mammals, but exceed
the Eco-SSL for plants. The 95th percentile manganese concentration in SFS-manufactured soil is
well below the corresponding Adjusted SSL for the soil pathways; at a 50% blend, even the
maximum manganese concentration in SFS-manufactured soil would be below the Adjusted
SSL.
6.7.7.2 Modeling Results
Based on the results of the comparison screening levels, the soil manufacturing scenario
(inhalation of fugitive dust emissions by nearby residents) and ecological receptors in the home
gardener scenario were evaluated. For the inhalation exposure pathway, modeling results
indicated that, up to a manganese concentration of 1,005 mg kg"1, the potential for adverse
human health effects would be below levels of concern. For the home gardener scenario, the
refined ecological modeling results indicated that, up to a manganese concentration of 1,000 mg
kg"1 SFS, ecological exposures would be below levels of concern.
The 95th percentile manganese concentration in SFS-manufactured soil (335 mg kg"1
DW) was above the Eco-SSL for terrestrial plants (220 mg kg"1 DW). This prompted an
evaluation of the critical assumptions associated with the ecological hazard screen. One such
assumption was that 100% of the manganese in SFS-manufactured soil would be available for
plant uptake. To better represent the bioavailable fraction of manganese, the total manganese
concentration in soil was adjusted by the pore water/total ratio as described in Section 5.3.8.2,
creating a reasonably conservative estimate for the soil concentration that would be comparable
with soil concentrations used in deriving the Eco-SSL for terrestrial plants. The refined
ecological modeling results indicate that up to a manganese concentration of 1,000 mg kg"1 SFS,
the potential for adverse effects to even the most sensitive ecological receptors would be below
levels of concern. Therefore, adverse ecological effects from manganese in SFS are unlikely to
occur for the home gardener scenario.
6.7.7.3 Soil Background Concentrations
Manganese is one of the most abundant trace elements in the lithosphere; its common
range in U.S. soils is 20-3,000 mg kg"1 DW, with a mean value of 490 mg kg"1 DW (Kabata-
Pendias, 2001). Studies on U.S. and Canadian surficial soils estimate that the median
concentration of manganese is 490 mg kg"1 DW, with a range of 56-3,120 mg kg"1 DW (Smith
et al., 2005). As illustrated in Figure 6-7, the composition of SFS with respect to manganese
appears to be very similar to U.S. soils, suggesting that the addition of SFS to soil will not, in
general, result in a significant change in soil manganese concentrations. In fact, the beneficial
use of SFS would nearly always dilute manganese levels in the amended soils.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-14
-------
Chapter 6.0 Risk Characterization
20
18
16--,
W 1A
-------
Chapter 6.0 Risk Characterization
affinity of these metals to manganese oxides significantly reduces the bioavailability of other
trace element metals (i.e., copper, lead, zinc) to plants.
Manganese is readily taken up from the soil and translocated within plants, and there is
ample evidence that manganese uptake is metabolically controlled in a way that is similar to
other divalent cation species, such as Mg2+ and Ca2+ (Kabata-Pendias, 2001). Because
manganese is easily taken up by plants in its soluble form, the manganese concentrations in
plants show a negative relationship with increasing soil pH and a positive relationship with soil
organic matter (Kukurenda and Lipski, 1982). An excess of phytoavailable manganese is
associated with strongly acid soils (pH < 5.5) and anaerobic conditions. Plant nutrient sufficient
manganese ranges from 30-300 mg kg"1 (Kabata-Pendias, 1992). However, even though the
manganese deficiency level for most plants ranges from 15-25 mg kg"1, toxicity from manganese
is highly variable due to great differences in species sensitivity, as well as the differences in soil
characteristics, especially soil pH management (Andersson, 1987). Natural manganese
phytotoxicity is one of the reasons that farmers must apply limestone periodically to correct and
maintain pH near 6.5. Because the pH of SFS ranges from neutral to slightly alkaline, exceeding
the highly conservative Eco-SSL for plants (95th percentile SFS concentration) is not necessarily
a valid indicator for adverse effects in plants. In reality, at the typical application rates and pH
that would be expected for SFS-manufactured soils used in home gardens, only a fraction of the
manganese in SFS would be readily available to plants. Also, as discussed in Chapter 2, plant
growth studies have found no negative impacts to plants grown in SFS or manufactured soils that
include SFS (Dungan and Dees, 2007; Hindman et al., 2008; Dayton et al., 2010).
6.7.7.5 Lines of Evidence
For the home gardener scenario, the 95th percentile and maximum manganese
concentrations in SFS-manufactured soil are below the Adjusted SSL for soil pathways. This
indicates that manganese concentrations in SFS-manufactured soil are unlikely to cause adverse
human health effects.
The results of the refined ecological modeling resulted in SFS-specific ecological
screening levels for manganese ranging from 1,000 mg kg"1 SFS (90th percentile, soil
invertebrates) to 9,500 mg kg"1 SFS (50th percentile, mammals). These SFS-specific ecological
screening levels are well above even the maximum manganese concentration found in SFS.
Given the similarity between the concentration distribution of manganese in SFS and soil
background levels, and no evidence of manganese toxicity in SFS plant growth studies, adding
SFS to soil would not increase the likelihood of developing manganese-toxic conditions.
Based on the similarity in concentration distributions for manganese in SFS and
background soils, as well as the results of the screening and risk modeling, manganese
concentrations in SFS are unlikely to cause adverse effects to human health and ecological
receptors when SFS is used in SFS-manufactured soils, soil-less potting media, or road subbase.
6.7.8 Nickel
The total nickel concentrations in silica-based iron, steel, and aluminum SFSs collected in
June 2005 ranged from a minimum of 1.1 mg kg"1 to a maximum of 117 mg kg"1 (using EPA
method 3051 A), with a 95th percentile value of 102 mg Ni kg"1 (Dayton et al., 2010). Using the
SPLP leaching test, only one sample was above the detection limit of 0.05 mg L "*, with a value
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-16
-------
Chapter 6.0 Risk Characterization
of 0.238 mg L"1. The concentrations in water extracts from the same samples (1 detect in 39
samples) were almost all below the detection limit of 0.05 mg L"1; nickel was detected in one
sample at the detection limit of 0.05 mg L"1 (Dungan and Dees, 2009). Sample-specific SPLP
and water extract leachate data can be found in Appendix B, Tables B-13 through B-18.
6.7.8.1 Comparison with Screening Levels
The relevant screening levels include Eco-SSLs, the default Residential soil screening
level for the protection of human health adjusted to also address home gardener produce
ingestion pathways (Adjusted SSL), and the Tapwater SL. The screening levels typically reflect
studies based on nickel soluble salts:
• Eco-SSL (plants): 38 mg kg'1 soil
• Eco-SSL (soil invertebrates): 280 mg kg"1 soil
• Eco-SSL (mammals): 130 mg kg"1 soil
• Adjusted SSL (noncancer): 150 mg kg"1 soil (soil ingestion SSL, adjusted to also address
produce ingestion pathways)
• Tapwater SL (noncancer): 0.3 mg L"1
Comparing the 95th percentile total concentration of nickel in SFS to the SSLs suggests
that, in a 1:1 manufactured soil blend the concentration of nickel would fall below the Eco-SSLs
for soil invertebrates and mammals, but exceed the Eco-SSL for plants. This same nickel
concentration in SFS-manufactured soil would be below the Adjusted SSL. Comparison of the
SPLP and water extract data indicates that nickel concentrations associated with these tests
would fall below the Tapwater SL.
6.7.8.2 Modeling Results
Based on the results of the comparison with screening levels, the soil manufacturing
scenario (inhalation of fugitive dust emissions by nearby residents) and ecological exposure in
the home gardener scenario were further evaluated. For the inhalation exposure pathway, the
screening results indicate that, up to a nickel concentration of 1,005 mg kg"1, adverse human
health effects are unlikely.
As discussed in Section 5.3.8, the phytotoxicity of metals depends on the soluble soil
fraction and, therefore, the actual hazard posed to terrestrial plants depends on the amount of
metal that can desorb from SFS particles and become available in the soluble fraction. To better
represent the bioavailable fraction of nickel, the total nickel concentration in soil was adjusted by
the pore water/total ratio as described in Section 5.3.8.2, creating a reasonably conservative
estimate for the soil concentration that would be comparable with soil concentrations used in
deriving the Eco-SSL for terrestrial plants. The refined ecological modeling results indicate that
up to a nickel concentration of 290 mg kg"1 SFS (i.e., almost twice the 95th percentile nickel
concentration in SFS), adverse impacts to ecological receptors would be unlikely.
6.7.8.3 Soil Background Concentrations
The background concentrations of nickel in soil range from <5-150 mg kg"1 soil, with
mean values on the order of 15-35 mg kg"1 soil across a wide range of U.S. and Canadian soils
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-17
-------
Chapter 6.0 Risk Characterization
(Smith et al., 2005). As illustrated in Figure 6-8, average nickel concentrations in SFS are well
within this range. The 95th percentile nickel concentration in SFS of 102 mg kg"1 falls within this
normal background range. Given the importance of site-specific soil properties such as pH level,
the comparison between nickel concentrations in SFS and soil background suggests that average
concentrations overlap significantly, and that the median concentrations of nickel in SFS are very
similar to median concentrations of nickel in native soils.
30
25-
V>
2 20
Q.
E
re
W 15
E
10
SFS Dataset, 2009
N=39
L0_
0 10 20 30 40 50 60 70 80 90 100110120130140150
SFS Ni Concentration, mg kg'1
80
tn
0)
70-
60-
50-
E
re
w 40 J
fc 30
ja
USGS Dataset, 2005
N=254
0 10 20 30 40 50 60 70 80 90 100110120130140150
Soil Ni Concentration, mg kg'1
Figure 6-8. Concentration distributions of nickel in SFS (top)
and U.S. and Canadian soils (bottom).
6.7.8.4 Additional Factors
Recent research on nickel shows that this metal is an essential nutrient for plants (e.g.,
Wood et al., 2004). Nickel is readily and rapidly taken up by plants, and up to phytotoxic levels
in plant tissue, there is a positive correlation between soluble soil nickel concentrations and plant
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-18
-------
Chapter 6.0 Risk Characterization
concentrations of nickel. The soil pH appears to be the controlling factor with regard to nickel
mobility, bioavailability, and toxicity; increasing soil pH can significantly reduce the nickel
content and reduce the potential for plant toxicity. In soils that are near neutral pH, nickel can
undergo rapid reaction to form less soluble and less bioavailable forms. When soluble nickel
compounds are mixed with soils, the nickel hydrated cations rapidly enter into the soil chemistry,
forming adsorbed forms on iron and manganese oxides and chelated forms with soil organic
matter (e.g., Singh and Jeng, 1993). Then other soil minerals dissolve and nickel reacts to form
new soil minerals, such as nickel-silicates and nickel-aluminum layered double hydroxides
(LDHs - see Appendix A for a more detailed discussion of LDHs). These prevent leaching and
strongly limit potential uptake or phytotoxicity of nickel in contaminated soils with 1,000 mg
kg"1 nickel or higher (Kukier and Chaney, 2004; Siebielec et al., 2007). Therefore, because SFS
and manufactured soils are near neutral pH, the bioavailability of nickel is likely to be very low.
Although the transport and storage of nickel seem to be metabolically controlled, nickel
is mobile in plants and is likely to be accumulated in both the leaves and seeds (Kabata-Pendias,
2001). The mechanism of nickel toxicity in plants is poorly understood, although restricted
growth and injury (e.g., chlorosis) have been observed for decades. In general, concentrations in
plants of 10-100 mg kg"1 (DW) have been shown to be phytotoxic. Sensitive species are affected
at lower foliar concentrations (e.g., 10-30 mg kg"1), while rare nickel hyperaccumulators can
contain nickel concentrations well into the thousands of mg kg"1. Typical nickel concentrations in
produce (fruits and vegetables) are found in the range of 0.6-3.7 mg kg"1 (DW), although plants
grown at nickel-contaminated sites may accumulate significantly higher levels of nickel
depending on the adaptation of plants, the form of the nickel in the contaminated soils, and other
site-specific soil characteristics (especially the pH).
6.7.8.5 Lines of Evidence
The results of the screening comparisons for human health indicate that nickel levels in
SFS were below levels of concern for the groundwater pathway and soil/produce pathways.
Therefore, nickel concentrations in SFS are unlikely to cause adverse human health effects
through dermal contact with or ingestion of groundwater, soil, and home-grown produce.
The inhalation hazard to nearby residents was shown to be well below a level of concern,
with modeled inhalation screening concentrations close to 100 times above the 95th percentile
nickel concentration in SFS. Therefore, nickel concentrations in SFS are unlikely to cause
adverse human health effects through inhalation.
Refined ecological modeling results in SFS-specific ecological screening levels ranging
from 290 mg Ni kg"1 (90th percentile, mammals) to 5,100 mg Ni kg"1 (50th percentile, terrestrial
plants). These SFS-specific ecological screening levels are above even the maximum
concentration of Ni found in SFS.
Given the similarity between the concentration distribution of nickel in SFS and soil
background levels, adding SFS to soil would not significantly alter the nickel content in native
soils.
Based on the similarity in concentration distributions for nickel in SFS and background
soils, as well as the results of screening comparisons and screening modeling, nickel
concentrations in SFS are unlikely to cause adverse effects to human health and ecological
receptors when SFS is used in SFS-manufactured soils, soil-less potting media, or road subbase.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-19
-------
Chapter 6.0 Risk Characterization
6.7.9 Other Metals
6.7.9.1 Lines of Evidence
Appendix A describes a substantial body of research on the behavior of metals in soils
with respect to mobility (e.g., sorption and desorption), bioavailability (e.g., metal species,
oxides), phytotoxicity (e.g., soil levels that damage plants), and toxicity to animals and soil
invertebrates (e.g., nature and severity of potential effects). This information is critical in
determining whether or not these other metal constituents in SFS pose a potential risk to human
health and the environment when beneficially used in soil-related activities.
To complement the information provided in Appendix A, Table 6-5 presents a summary
of the available data on various metals with respect to their potential for release to the
environment at levels of concern. The table compares a 1:1 manufactured soil blend using the
95th percentile concentration in SFS with the Residential SSL adjusted to also address home
gardener produce ingestion pathways (Adjusted SSL), the inhalation screening level, and the 50th
percentile background concentration in soil. This constitutes a conservative comparison because
(1) actual soil blends are likely to include less than 10% SFS (Personal communication,
USDA/ARS51), so the 1:1 blend is highly unlikely, and (2) the SSLs make very conservative
assumptions with respect to exposure (e.g., 100% of incidentally ingested soil comes from the
SFS-manufactured soil). The concentrations in the 1:1 SFS-soil blend do not exceed the
ingestion or inhalation SSLs for any constituent; therefore, it appears highly unlikely that either
of these pathways will pose a significant risk to human health. The limited leach test data suggest
that the metals that were tested (barium, beryllium, cadmium, lead and zinc) do not pose
significant risks via the groundwater pathway and, in fact, only one of the metals (barium) was
present above the detection limit in the SPLP leach test.
Finally, comparing the soil blend to the 50th percentile background concentrations
suggests that molybdenum is present at levels in SFS that might result in an increase in the soil
concentration. However, these concentrations are still well within the range of background
levels, and moreover, the research discussed in Appendix A strongly suggests that, at the
concentrations shown in Table 6-5, the availability and toxicity of molybdenum would be very
low under a wide range of soil conditions.
51 Personal communication, April 2009, Timothy Taylor, U.S. EPA, with Rufus Chaney, USDA-ARS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-20
-------
Chapter 6.0 Risk Characterization
Table 6-5. Summary of Other SFS Metal Concentrations and Relevant Screening Criteria
(mg kg"1 unless otherwise noted)
SFS
Constituent
AlCgkg-1)
B
Ba
Be
Cd
Mo
Pb
Se
Tl
V
Zn
95%-ile of
1:1
Soil: SFS
blend
5.6
10.1
6.9
0.19
0.1
10.9
7.65
0.10
0.05
4.95
36.1
95%-ile
SPLP/
ASTM
—
—
0.37
0.02
0.01
—
0.11
—
—
—
0.22
Above
Adjusted
SSL?
No
No
No
No
No
No
No
No
No
No
No
Above
Inhalation
SSL?
No
—
—
No
No
—
—
—
—
—
—
Above
Eco-SSL?
—
—
No
No
No
—
No
No
No
No
No
Above
Ground
water
Screen?"
—
—
No
BDL
BDL
—
BDL
—
—
—
BDL
Above
50%-ile
Background?
No
NA
No
No
No
Yes
No
No
No
No
No
Above
95%-ile
Background?
No
NA
No
No
No
Yes
No
No
No
No
No
BDL = below detection limit.
NA = not available.
a All groundwater screening levels used in this assessment are listed in Table 4-2.
6.8 Uncertainty Characterization
The goal of this report was to bring together risk screening modeling and the best
available science to provide industry, consumers, and regulatory agencies with the scientific
basis to determine whether certain soil-related beneficial use applications of SFS are appropriate
and protective of human health and the environment. This lines of evidence approach, therefore,
includes two basic components that will be discussed in this uncertainty characterization: (1)
uncertainties associated with the conduct and interpretation of the risk screening modeling, and
(2) uncertainties associated with the state-of-the-science research on the behavior of metals and
other SFS contaminants in soils.
6.8.1 Risk Screening Modeling
In the Guidance for Risk Characterization developed by EPA's Science Policy Council
(U.S. EPA, 1995c), EPA defined the high end of the risk distribution as being at or above the
90th percentile risk or hazard estimate generated during Monte Carlo simulation. The high end of
the risk distribution for risk screening modeling refers only to hypothetical individuals living
within the areas of "economic feasibility" for SFS use that may
• Live near roadways that were constructed with SFS as a component of the subbase
• Live near facilities that manufacture soils and soil-less media by blending SFS with other
ingredients
• Incorporate SFS-manufactured soils into their home garden and consume a large fraction
of fruits and vegetables from the home garden.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
6-21
-------
Chapter 6.0 Risk Characterization
The conceptual model for each of these three scenarios was described in Chapter 3,
Section 3.1.4. At a minimum, the risk screening modeling was designed to ensure that 90% of
the individuals associated with these high-end exposure scenarios would not be exposed to
constituents in SFS above the screening levels or benchmarks. In addition, the risk screening
modeling also used conservative ecological screening criteria, the Eco-SSLs, to ensure that
ecological receptors (e.g., plants, soil invertebrates, and mammals) are not exposed to constituent
levels above the criteria levels. However, the receptors considered in this assessment are
hypothetical, and the modeling reflects exposures that are almost certain to be well above the
90th percentile of the distribution. In fact, given the conservative nature of the modeling, the
modeling results provide bounding estimates of risk that fall at the extreme tail of the
distribution. Therefore, this discussion is focused on better understanding the key sources of
conservatism in the input data and scenario assumptions that EPA developed to ensure that the
modeling results would not underestimate the potential risks associated with SFS. There are
considerable uncertainties in the modeling risk estimates. However, these estimates are
conservative by design, and the uncertainties in the assumptions and selection of input data bias
the risk predictions heavily toward the overestimation of risk.
Roadway Subbase. The use of SFS as a component in roadway subbase was addressed
through the evaluation of subbase-relevant exposure pathways (i.e., groundwater ingestion and
inhalation of fugitive dust) in a use scenario likely to cause greater exposure - SFS-manufactured
soil use in a home garden.52 Once in place as subbase, the only exposure pathway of potential
concern would be leaching of constituents into the subsurface following fracturing of the road
surface (allowing rainwater infiltration through the underlying materials) or mounding of a high
water table. For almost all constituents, the leach test data (except perhaps that from the ASTM
shake method) provide extreme conditions that will not occur under the roadway. Even under
these conditions, very few constituents had leach test results above detection limits. For those
constituents that demonstrated an ability to leach from SFS, the groundwater screening showed
that the potential for these constituents to reach receptors at levels of concern is extremely low.
Thus, the demonstration of low teachability even under extreme conditions, along with the
conservative groundwater modeling provides a high level of confidence that this pathway will
not be of concern.
Similarly, the inhalation screening modeling used a series of conservative assumptions
ranging from the emission factors to placing the receptors in the downwind plume of the
maximum air concentration. These bounding results demonstrated that the protective
concentrations of chemical constituents found in SFS were higher—in many cases orders of
magnitude higher—than the actual constituent concentrations found in SFS. Due to the transitory
nature of storage piles of SFS during roadway construction, the pathways associated with
delivery to nearby streams (after windblown emissions and runoff) were considered to be
essentially incomplete. That is, as with other typical roadway construction components, the
storage piles are not retained for sufficient periods to result in a significant mass transport to
local waterbodies. These materials are valuable, and it was assumed that storage piles would
exist for a few days (at most) before being incorporated into the subbase. The relatively large
SFS grain size and very low leaching potential of constituents in SFS further supports the
52 Though the groundwater modeling was performed for the manufactured soil use scenario rather that road subbase,
modeled inputs (e.g., distance to drinking water well) were more conservative than road subbase inputs. The
findings from the manufactured soil scenario are therefore also protective of the road subbase use scenario.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-22
-------
Chapter 6.0 Risk Characterization
contention that (1) very little mass of material could be transported from a storage pile, and (2)
the constituents found in the SFS are tightly bound in the sand matrix and not very available at
environmental pHs in the aquatic environment.
Manufacture of Blended Soils and Soil-Less Potting Media. Soil blending operations
that use SFS pose a potential inhalation risk due to the large volumes of SFS piles that would
likely be required to support such operations. As suggested by the conceptual model, leaching of
constituents, and inhalation following volatile and particulate emissions, are potential concerns.
Thus, the combination of leach test data, inhalation screening, and probabilistic groundwater
modeling was used to screen for potential risks. The manufacture of blended soils presents low
risks to human health at the 95th percentile constituent concentrations found in SFSs. For this
scenario, it was assumed that runoff would not be a pathway of concern because manufacturing
facilities would impose basic controls (e.g., berms) to avoid losing valuable ingredients to the
soil blending process, and generally states require facilities to institute stormwater controls to
prevent significant levels of chemical constituents from being directly discharged into nearby
surface waters. Furthermore, it was assumed that deposition from soil-blending emissions would
not contribute significantly to the surface soil layer and ecological exposures when compared to
SFS use in home gardens. Therefore, given the highly conservative assessment of risks
associated with soil manufacturing, the potential for adverse health effects is considered unlikely.
Use of SFS in Home Gardens. The use of SFS-manufactured soils by home gardeners
could pose potential risks through inhalation, incidental ingestion of the soil, the consumption of
home-grown fruits and vegetables grown in soil containing SFS, or groundwater impacted by
garden leachate.
As shown by the comparison of the 95th percentile constituent concentrations in SFS to
inhalation screening concentrations for the SFS storage pile (see Table 4-4), the inhalation
pathway was screened out by the deterministic modeling of air releases from SFS storage piles.
These results also screened out the inhalation pathway for the home garden scenario because
they represent a scenario in which SFS-manufactured soil was used in a home garden as top
dressing with no mixing or dilution. This is a highly conservative assumption because, in
practice, SFS-manufactured soils will be mixed with native soils, thereby diluting the constituent
concentrations in the SFS. Thus, comparing the 95th percentile constituent concentrations in SFS
with the inhalation pathway screening concentrations demonstrates that the inhalation exposures
for the home garden scenario also will be below levels of concern.
Therefore, a screening modeling scenario was developed for the use of SFS-
manufactured soils in the home garden that addressed both the incidental ingestion of
constituents in SFS, as well as the consumption of contaminated groundwater and produce from
the garden. A Monte Carlo simulation was implemented to assess human and ecological
exposures under the home gardener scenario. As discussed in Section 5.1, the implementation
does not distinguish between uncertainty and variability. In essence, input parameters were
selected to represent variability (e.g., exposure factors), and in some cases, to also represent the
uncertainty in the true parameter value (e.g., soil-specific parameters). Previous chapters of this
document describe how input distributions and input values were developed and used to estimate
risk. Use of these inputs in a national level assessment may result in an underestimation or
overestimation of risk. To ensure that the Monte Carlo simulation was highly conservative and
produced a bounding estimate of risk, several assumptions were built into the modeling scenario.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-23
-------
Chapter 6.0 Risk Characterization
First, the assumption of a 1:1 SFS-soil blend in a single-application "soil replacement"
scenario is conservative. It is possible that this blend could be used to replace the topsoil in small
home gardens, or that this blend could be used multiple times in smaller volumes to amend
existing local soils. However, the amount of SFS required to replace the top soil layer with this
blend in a home garden capable of producing a significant proportion of the home-grown diet of
fruits and vegetables is quite large. The costs of SFS-manufactured soil for a 0.1-acre garden
would be on the order of $2,300 (assuming approximately $21.50 yd"3 delivered). In all
likelihood, these costs would be prohibitive, and the home gardener would use smaller SFS-
manufactured soil application rates or seek alternative methods to improve the physical and
chemical properties of the soil for large gardens.
Second, the consumption rates of fruits and vegetables sampled during the probabilistic
modeling were based on EPA's Exposure Factors Handbook (U.S. EPA, 2011). The distribution
for each category of produce (e.g., exposed vegetables) was based on actual survey data;
however, these distributions are sampled independently, even though there is very likely a
correlation among the consumption of different types of produce. It would be unlikely that a
person would consume a high-end amount of root vegetables and leafy greens and apples that
were all grown from the same garden because (1) all types of produce cannot be grown in the
same season, (2) there are regional characteristics (e.g., soil type, precipitation) that strongly
influence what types of crops can be grown, and (3) there are agronomic limits as to how much
produce can be grown, harvested, and consumed that are not reflected in the exposure factor
data. Thus, the total ingestion risks tend to overestimate the likely consumption of home-grown
produce. For example, in EPA's deterministic risk assessment of chemical pollutants in biosolids
conducted in 1993 (U.S. EPA, 1993), the estimated consumption rate of home-grown fruits and
vegetables was 105 g (WW) d"1 for an average adult (not including tree fruits). In the
probabilistic modeling conducted for this assessment, the total consumption rate of home-grown
fruits and vegetables for the adult at the 90th percentile risk level was approximately 500 g (WW)
d"1 for an average adult. Also, it is not possible to harvest most garden crops for more than a
short period when the crop is ripe, which considerably limits potential exposure to garden foods.
Given the size of the garden required to support such a diet, the costs of delivering SFS-
manufactured would likely reduce the actual exposure by several orders of magnitude due to the
limited garden area. Thus, the results of the home gardener refined modeling should be
considered as an overestimate of the actual risks.
In addition, evaluation of the home gardener groundwater pathway with IWEM and
EPACMTP incorporated several conservative assumptions, including the placement of the
drinking water receptor well adjacent to the edge of the garden. Considering that the U.S. EPA
estimates that only 15% of the U.S. population have their own drinking water sources (U.S. EPA,
2002f) and the fact that modeling identified the 90th percentile groundwater well concentration,
the applied approach ensures that the results of this analysis can be used to confidently determine
if the applications of SFS will be protective of human health and the environment in the United
States.
In summary, the uncertainties associated with the screening and refined risk modeling
bias the results to produce overestimates of the potential risks associated with the three exposure
scenarios of interest. Although the accuracy of the screening modeling could be increased by
making less conservative assumptions and developing additional data inputs for the models, the
modeling results are appropriate for their intended purpose—to ensure with a high level of
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-24
-------
Chapter 6.0 Risk Characterization
confidence that the risk estimates are health protective. Given the level of conservatism in the
modeling assumptions and inputs, the use of SFS in the soil-related applications discussed in this
report would not be expected to pose significant risks to human health or the environment.
Human Health Benchmarks. There is uncertainty inherent in the development of the
human health benchmarks used in this risk assessment. Uncertainty that is typically associated
with human health benchmarks is discussed in detail in EPA's Guidelines for Carcinogen Risk
Assessment (U.S. EPA, 2005c), Methods for Derivation of Inhalation Reference Concentrations
and Application of Inhalation Dosimetry (U.S. EPA, 1994a), and IRIS (U.S. EPA, 2012a). With
regard to the application of human health benchmarks developed by EPA for risk assessment
purposes, U.S. EPA (2005c) states that".. .the primary goal of EPA actions is protection of
human health; accordingly, as an Agency policy, risk assessment procedures, including default
options that are used in the absence of scientific data to the contrary, should be health
protective." Thus, EPA acknowledges the uncertainty associated with the use of point estimates
for human health benchmarks, but also recognizes the Agency's responsibility with regard to the
protection of human health in addressing this uncertainty.
Ecological Benchmarks. There is uncertainty inherent in the development of the
ecological screening level benchmarks used in this risk assessment (i.e., Eco-SSLs). Like their
human toxicity counterparts, Eco-SSLs are conservative screening values. For example, use of
conservative modeling assumptions (e.g., metal exists in most toxic form or highly bioavailable
form, high food ingestion rate, high soil ingestion rate) in the Eco-SSL derivation process leads
to some Eco-SSLs that are below the average background soil concentration. As screening
values, users can be confident that if soil concentrations fall below Eco-SSLs, then no further
evaluation is necessary.
Eco-SSLs for terrestrial plants, soil invertebrates, and small insectivorous mammals were
applied to evaluate exposures to ecological receptors under the home garden scenario. Avian
Eco-SSLs were deemed not applicable to the home garden scenario for several reasons. First, it
is highly likely that the home gardener will adopt measures (e.g., fencing, netting) that would
limit potential exposure for birds. Second, the home ranges for most birds that are either included
or represented by the Eco-SSLs are significantly larger than a 0.1 acre (405 m2) garden. The
woodcock, for example is reported in U.S. EPA 1999c as having a mid-point home range of
857,500 m2. Therefore, the impact attributable to home gardens on reproductive fitness of avian
populations is likely to be negligible.
6.8.2 State-of-the-Science on SFS
This report presents a tremendous amount of information on SFS characteristics, uses,
and the behavior of SFS constituents in the environment, particularly the metals and metalloids.
Where the soil uses are being considered, this information speaks to one important question—
namely, is SFS significantly different than native soils.53 Clearly, the demonstration that SFS is
similar in its composition and properties to that of background soil may question the need for
risk screening modeling. However, there is variability in the properties of SFS and there is
variability in the properties of background soils, and as a result, the use of this information in
answering this core question is associated with some level of uncertainty. There are aspects of
53 A comparison of other materials used in manufactured soils or road base (including native sand) is also relevant,
but beyond the scope of this evaluation.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-25
-------
Chapter 6.0 Risk Characterization
uncertainty that are not addressed in the screening modeling that can only be considered as they
relate to the research and scientific findings presented in this report and appendices. This section
addresses several broader aspects of the uncertainty, given the current state-of-the-science, that
are highly relevant to the interpretation of potential risk associated with the beneficial use of
SFSs in certain soil-related applications. To provide the context for this discussion, four
questions are posed that directly relate to the core question in this assessment.
1. Are the analytical data on SFS representative of SFSs that will be beneficially used in
the soil-related applications addressed in this report?
The analytical data on total constituent concentrations and leach test data were developed
to represent the specific types of sands that have been identified for soil-related beneficial uses.
These sands include SFS from iron, steel, and aluminum foundries that were repeatedly used in
the molding process; though the initial survey included sampling SFS from nonleaded brass
foundries, the risk evaluation did not include SFS from brass or bronze foundries. The data
include SFS samples from 39 foundries in 12 states that were specifically selected to ensure that
the full range of constituents and concentrations for these types of sands were adequately
represented. Given the similarity in molding processes for these types of foundries, both in terms
of the input materials used and the reclamation/reuse practices, the analytical data are believed to
represent the range of constituent concentrations and the distribution of those concentrations in
foundry sand. Nevertheless, it is unknown if the SFS samples from these 39 foundries are
statistically representative of SFS from all iron, steel, and aluminum foundries. The related data
may, therefore, overestimate or underestimate the range and distribution of SFS constituent
concentrations.
2. Are the data presented by Smith et al. (2005) representative of background soil
concentrations of metals in the areas of economic feasibility?
The data presented by Smith et al. (2005) represent the USGS's attempt to systematically
characterize the background concentrations of metals in the U.S. and Canadian soils. The authors
noted that
"The transects were located to cross multiple climatic, topographic,
physiographic, land use, geologic, pedologic, and ecological boundaries. This
imposes rigorous field testing of sampling protocols across a wide range of
conditions. The generated data will allow estimation of geochemical and
microbiological variation at a continental scale." (Smith et al., 2005)
The Smith et al. (2005) data on background concentrations of metals in soil were
compared to a variety of other sources of background data for the United States summarized in
Trace Elements in Soils and Plants-Third Edition (Kabata-Pendias, 2001), as well as in EPA's
Attachment 1-4: Guidance for Developing Ecological Soil Screening Levels (Eco-SSLs)-Review
of Background Concentrations for Metals (U.S. EPA, 2003e). Based on a visual inspection of the
data in these respective sources, the data presented by Smith et al. track well with work
performed by a number of different sources (e.g., U.S. EPA, 2003e included information
developed by states, as well as under the Comprehensive Environmental Response,
Compensation, and Liability Information System [CERCLIS]), particularly with respect to the
minimum, 95th percentile, and mean values for metal concentrations in soil. The overlap in data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-26
-------
Chapter 6.0 Risk Characterization
on metal concentrations suggests a high level of confidence in the use of the Smith et al. (2005)
data to represent background concentrations at a continental scale.
In considering the similarity between SFS and native soils, it is important to recognize
that the use of continental or east-west regional data on background soil concentrations
represents a source of uncertainty. Because the soil concentrations are variable, it is uncertain
whether a specific application of SFS will introduce metals above background levels at a specific
location. From a risk assessment standpoint, it was determined that, independent of background
levels, the introduction of metals associated with soil-related applications of SFS is unlikely to
cause adverse effects to human health and ecological receptors. For the purposes of interpreting
the general impacts of soil-related applications with respect to background, the most appropriate
comparisons are to (1) consider the entire empirical distributions of metals in SFSs and in native
soils, and (2) compare the metal concentrations in SFS products at the high end of the
distribution (i.e., 95th percentile) to robust measures of background soil concentrations (i.e., the
50th percentile). The former provides important insights regarding the nature of the respective
materials with respect to metals, and the latter provides a statistical indication of the probability
of an SFS application exceeding typical background concentrations. Given these comparisons, as
well as the results of the conservative risk screening modeling, the uncertainty inherent in using
background concentrations at scales above what is expected at local levels is not considered to be
significant.
3. How will the soil characteristics affect the bioavailability, mobility, and toxicity of
metals in the soil-related applications of SFS addressed in this report?
As discussed throughout this report, the bioavailability of most metals tends to increase
with decreasing pH, particularly for acidic soils in the range of pH 4. Given the variability in soil
pH, with decreasing pH associated with the use of SFSs in areas that are closer to the East Coast,
evaluating the potential impacts of adding SFS to soils at the low end of the pH range is
associated with some level of uncertainty. With regard to the leaching potential of metals, the
SPLP leach test reflects acid rain conditions, and considering the low levels found, these
conditions are not anticipated to significantly alter the leaching potential of metals in SFS. The
groundwater pathway screening is sufficiently conservative to state with a high degree of
confidence that pH variability will not drive risks due to groundwater ingestion above the levels
of concern.
With regard to the home gardener scenario, if SFS-manufactured soils were applied in
regions with lower pH and assuming that the home gardener did not lime the soil, the uptake and
translocation of metals into plants could be increased. Depending on the form of the metal, this
could result in higher phytotoxicity or accumulation of metal at higher rates for more tolerant
plant species. In addition, the more mobile and toxic metal species may cause adverse effects to
invertebrates in the garden soil. Although these effects could occur, the variability in soil pH is
not regarded as a significant source of uncertainty for the home gardener scenario for three
reasons. First, the size of the garden would have to be relatively large to support the consumption
rates used in the evaluation, and as previously discussed, the economics and physical attributes
of such large SFS applications would prohibit the blend from reaching 50%. Second, it is
reasonable to assume that home gardeners have sufficient experience in cultivating produce to
routinely monitor and improve the quality of their soil; this would almost certainly include
liming in many of the low pH regions in the east, thereby minimizing the impact of potentially
low soil pH on plant health and productivity. Third, in accordance with the soil-plant barrier, soil
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-27
-------
Chapter 6.0 Risk Characterization
acidity at pH <5.2 would result in natural soil aluminum phytotoxicity, thereby preventing plant
growth and protecting the food chain.
4. Are the chemical and physical properties of metals in SFS similar to the chemical and
physical properties in native soils?
With respect to the distribution of metals concentrations in SFS, the data presented in this
report indicate that metal concentrations in SFS are generally relatively low compared to the soil
background levels at a national scale. However, the concentrations of metals do not, by
themselves, indicate whether SFS is similar to soil with respect to how those metals behave.
Specifically, the concentration data do not indicate whether the forms of metals in SFS are more
mobile, bioavailable, or toxic than those same metals in native soils. Although this is a potential
source of uncertainty, three pieces of information strongly suggest that metals in SFS will behave
in a very similar manner as metals in native soils.
First, the leach test data on SFS indicate that even under very acidic conditions
(representative of a landfill), the metals in SFS demonstrate a very low potential to leach out of
this material. Of the very few metals that either demonstrated some leaching potential (arsenic)
or had detection limits above the screening criteria (antimony, beryllium, cadmium), the
conservative risk screening (e.g., using the 95th percentile leach test concentration) demonstrated
that these metals would not pose a significant risk via the groundwater ingestion pathway. Given
the similarity between the background concentrations of these metals and the concentrations in
SFS, this result indicates that the risks to background concentrations should also be very low.
Second, the most commonly used sand is silica sand (silicon dioxide, SiCh) because of its
wide availability and relatively low cost; this material is a component of native soils.54
Section 2.5 describes the "soil-like qualities" of SFS that make this material a valuable soil
amendment; these properties include, for example, desirable chemical (e.g., pH, salinity) and
physical (e.g., texture, water holding capacity) characteristics that are typical of high-quality
soils.
Third, because soil-related applications of SFS are likely to be used in aerobic soils that
are typical of home gardens, it is reasonable to assume that the cationic form of many of the trace
elements in SFS will be the predominant form. As discussed in Appendix A, Section A. 1.1.2,
complexation of trace metals with amorphous iron and manganese hydrous oxides (both of
which are available in SFS) is common in aerobic soils; in addition, the cationic forms of a
number of metals in SFS can be expected to sorb to soil organic matter and other forms of
humified natural organic matter, reducing the solubility of the metals in the soil. The behavior of
metals in SFS added to aerobic soils would, therefore, be expected to be similar to the behavior
of metals already present in the soil. Further, the increased availability of iron and manganese in
SFS may actually decrease the solubility and availability of trace metals originating from both
native soils and SFS due to adsorption on oxides. In consideration of the information on leaching
potential, the soil-like qualities of SFS, and the chemical behavior of metals in SFS once added
to aerobic soils, it appears very likely that the behavior of metals in SFS would be similar, if not
indistinguishable, from the behavior of metals in the native soils to which the SFS is added.
54 Sands, including silica sand, are also frequently used in manufactured soil and road subbase.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 6-28
-------
Chapter 7.0 Findings and Conclusions
7. Findings and Conclusions
This report presents an extensive review of information on SFSs, including analytical
results for metals and metalloids (including both totals and leach test results), PAHs, phenolics,
dibenzodioxins and furans, and dioxin-like PCBs. It also includes deterministic risk screening
model results for the inhalation exposure scenario and probabilistic screening and refined model
results for the home gardener exposure scenario. Taken together, this information provides the
scientific basis for decision makers to determine the appropriate soil-related applications for
certain unencapsulated beneficial uses of SFS. The major findings and conclusions from this
report as they pertain to silica-based SFSs produced by iron, steel, and aluminum foundries, and
their use in manufactured soil, soil-less potting media, and road subbase, are summarized below.
7.1 Beneficial Use of SFS (Chapter 1)
• SFS is a valuable industrial byproduct, and therefore, there are economic and possibly
environmental advantages to identifying which soil-related applications are appropriate
SFS beneficial uses.
• State regulators need access to sound scientific data and analyses to support the decision-
making process regarding the beneficial use of SFS.
7.2 Characterization of SFS (Chapter 2)
• SFS has a number of soil-like qualities that make it an attractive material for use in
roadway subbase, soil-less media, and manufactured soils.
• The concentrations of organic constituents and trace elements (including metals and
metalloids) are, on average, very low in silica-based SFS produced by iron, steel, or
aluminum foundries.
• Published background concentrations of metals in soils provides additional information in
evaluating the scientific basis for considering the implications of adding SFS as soil
amendments.
• The current data on SFS show that the distributions of metal constituents in silica-based
SFS from iron, steel, and aluminum foundries are very similar to the background
distributions of metals in native soils.
• The presence of manganese and iron and the neutral pH of SFS strongly suggest that soil-
related applications will likely reduce the mobility, bioavailability, and toxicity of metal
constituents in SFS and, possibly, metal constituents already in the soil.
• Although applications of SFS in strongly acidic soils (pH <5) could increase the mobility
of metals, this increase would mirror the same increase in natural soil. The common
agricultural practices of testing pH and liming to ensure good crop growth conditions are
expected to preclude highly acidic conditions from occurring.
• Based strictly on a comparison between the SFS and background concentrations of
metals, it is unlikely that the addition of silica-based SFS from iron, steel, and aluminum
foundries would significantly alter the composition of soil.
7.3 Exposure Scenarios Examined (Chapter 3)
• Three exposure scenarios that reflect the unencapsulated beneficial use considerations for
SFS, as well as the potential for complete exposure pathways, included (1) use as subbase
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 7-1
-------
Chapter 7.0 Findings and Conclusions
in roadway construction, (2) use in soil-less potting media, and (3) blending in
manufactured soils.
7.4 Screening of Exposure Pathways (Chapter 4)
• The inhalation pathway screening indicates that even high-end concentrations of the
constituents in SFS were well below screening values for all constituents for which
inhalation benchmarks were available.
• The groundwater ingestion pathway screening indicates that even high-end
concentrations of metal constituents in SFS were below water quality screening criteria
for all constituents for which such criteria were available, except antimony, arsenic,
beryllium, cadmium, and lead.
• The soil ingestion pathway screening indicates that even high-end concentrations of
metal constituents in SFS were below soil screening criteria for all constituents for which
such criteria were available, except antimony, arsenic, chromium III, cobalt, copper, iron,
manganese, and nickel.
7.5 Modeling of Exposures from Home Gardening (Chapter 5)
• Eight metals (antimony, arsenic, chromium III, cobalt, copper, iron, manganese, and
nickel) were evaluated with probabilistic screening modeling and refined modeling.
Arsenic, cobalt, and iron were evaluated for human exposures through the soil/produce
ingestion pathway but, only arsenic was evaluated under the groundwater pathway.
Although concentrations of manganese and nickel in SFS were below their respective
human health screening criteria (described in Chapter 4), they were modeled in the home
gardening scenario because of their high potential for phytotoxicity. Similarly,
concentrations of antimony, trivalent chromium, and copper were below their human
health screening levels, but they were retained for further study due to the potential to
impact small insectivorous mammals.
• One of the more conservative assumptions for the home gardener soil/produce pathway
screening modeling was the addition of exposures across all five produce categories (e.g.,
exposed vegetables), which results in consumption rates for the home gardener that are
well above expected values.
• Investigation of the influence of produce consumption rates suggests that adding across
produce categories is likely more appropriate for the median consumption rates for the
home gardener, and that the use of values at the tail of the exposure factor distributions is
associated with higher levels of uncertainty.
• The refined groundwater modeling used the distribution of the home garden source model
outputs (i.e., leachate fluxes and annual average leachate infiltration rates) as input to the
groundwater model. Coupling the home garden source and groundwater modeling
captured variability in conditions within the SFS economic feasibility areas when
predicting SFS constituent fate and transport in the environment. The conservative nature
of the assessment was maintained through the placement of the drinking water receptor
well 1 m from the edge of the garden in the centerline of the plume.
• Because arsenic has the potential to exhibit nonlinear behavior during transport through
the unsaturated zone as simulated by EPACMTP, it was necessary to ensure the
appropriateness of applying the unitized approach to the groundwater pathway. As a
result, an analysis was performed which demonstrated that arsenic would behave linearly
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 7-2
-------
Chapter 7.0 Findings and Conclusions
in the subsurface under anticipated home garden environmental conditions and at
concentrations found in SFS samples. (Appendix J and Chapter 5).
• An analysis was performed to evaluate anticipated arrival times of peak contaminant
concentrations in the receptor drinking water well. Based on the analysis, it is unlikely
that peak surface and peak groundwater exposures will occur within the same timeframe.
For example, the earliest estimated timeframe for arrival of arsenic in the well spanned
from 29 to 200 years following the application of the SFS-manufactured soil. Given this
timeframe, it is likely that the peak well concentrations will not occur until well past the
receptor's timeframe of residency (i.e., exposure duration). Therefore, surface and
subsurface ingestion exposures would not occur together during the same exposure
period. (Appendix J and Chapter 5).
• The probabilistic modeling for the home gardener scenario demonstrated that, even using
consumption rates at the upper end of the distribution, the estimated exposures were
below health benchmarks.
7.6 Characterization of Risks Associated With SFS Beneficial Use (Chapter
6)
• The assumption of a 1:1 mix for manufactured soil in the home gardener scenario was a
conservative assumption, because this would be cost prohibitive for even small home
gardens. A more likely scenario would be a manufactured soil consisting of 5-10% SFS,
rather than the 50% SFS modeled here. Therefore, this assumption likely overestimates
soil concentrations.
• Evaluating the national-scale beneficial use of SFS in road subbase, soil-less potting
media, and manufactured soil includes numerous sources of variability. However, the
findings from the available multiple lines of inquiry—such as newly available analytical
results for SFS, research on metals behavior in soil (including SFS-specific studies), and
risk screening methods (including modeling), all within the context of well-established
soil science—when used collectively provide a sound scientific basis for determining
appropriate soil-related uses of SFS.
• Given the assumption of high-end concentrations of the metals and other constituents in
SFS, and the application of highly conservative screening techniques, risk screening
models and refined models, the preponderance of the evidence demonstrates that the
evaluated uses of silica-based SFS produced by iron, steel, and aluminum foundries are
unlikely to cause adverse effects to human health and ecological receptors.
Table 7-1 provides a useful data summary for regulatory decision makers and other
stakeholders; the table presents the analytical and background information on metal constituents
in SFS, as well as the HH-SSLs and Eco-SSLs. In addition, the table provides the SFS-specific
modeled screening values for the specific home gardener scenario evaluated in this report, as
well as modeled screening values based on median and high-end consumption by the general
public.55 As shown in this table, the concentrations of metal constituents found in SFS are below
the health-based and ecological screening levels for soil and are present at levels that are similar
to those found in native soils.
55 Chapter 5 discusses the rationale for deriving screening levels based on three different consumption rates.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 7-3
-------
Chapter 7.0 Findings and Conclusions
Table 7-1. Comparing SFS Concentrations to Various Screening Values (mg kg'1 unless otherwise noted)
Elements
Alfekg-1)
As
B
Ba
Be
Cd
Co
Cr(III)
Cu
Fe (g kg'1)
Mn
Mo
Ni
Pb
Sb
Se
Tl
V
Zn
Silica-based Iron, Steel, and
Aluminum Sands"
Max
11.7
7.79
59.4
141
0.60
0.36
6.62
115
137
64.4
707
22.9
117
22.9
1.71
0.44
0.10
11.3
245
95%-ile
11.2
6.44
20.2
17.7
0.38
0.20
5.99
109
107
57.1
670
21.8
102
15.3
1.23
0.20
0.09
9.90
72.1
Median
5.56
1.05
10.0
5.00
0.15
0.05
0.88
4.93
6.22
4.26
54.5
0.50
3.46
3.74
0.17
0.20
0.04
2.88
5.00
Manuf.
Soil
5.6
3.22
10.1
8.85
0.19
0.10
3.00
54.5
53.5
28.9
335
10.9
51.0
7.65
0.62
0.10
0.05
4.95
36.1
Human Screening Values
SSLd
77
6.7«
16,000
15,000
160
70
23
120,000
3,100
55
1,800
390
1,500
400
31
390
0.78
390
23,000
Modeled Consumption Rates0
Home
Gardener
—
8.0
-
—
—
-
22
—
—
160
-
-
—
—
-
-
—
—
--
Gen. Pop.
Median
—
30
-
—
—
-
58
—
—
230
-
-
—
—
-
-
—
—
--
Gen. Pop.
High
—
9.1
-
—
—
-
21
—
—
150
-
-
—
—
-
-
—
—
--
Eco Screening Values
Eco-
SSLse
ND
18
ND
330
21
0.36
13
34
49
ND
220
ND
38
56
0.27
0.52
ND
280
79
Modeled
(SFS-
Specific)
—
40
-
—
—
-
-
510
159
-
1000
-
290
—
4.1
-
—
—
--
USDAf
—
-
-
—
—
-
-
—
200
-
-
-
200
—
-
-
—
—
300
U.S. and Canadian
Surface Soilsb
Max
87.3
18.0
ND
1800
4.0
5.2
143.4
5320
81.9
87.7
3,120
21.0
2,314
244.6
2.3
2.3
1.8
380
377
95%-ile
74.6
12.0
ND
840
2.3
0.6
17.6
70.0
30.1
42.6
1,630
2.16
37.5
38.0
1.39
1.0
0.7
119
103
Median
47.4
5.0
ND
526
1.3
0.2
7.1
27.0
12.7
19.2
490
0.82
13.8
19.2
0.60
0.3
0.5
55
56
— = No modeled value was generated because constituent was screened out of further study in an earlier stage of the evaluation. If a constituent screened out based on human
health SSL and had no Eco-SSL, the constituent was considered to have screened out for both human and eco.
ND = No Data.
a Source: Dayton et al. (2010).
b Source: Smith et al. (2005).
c See Chapter 5 for a detailed discussion of how the modeled values were generated.
d Concentrations of SFS constituents in manufactured soil (a 1:1 blend) were compared to an order-of-magnitude below the SSLs listed here, as discussed in Chapter 4,
Section 4.4.3. Values are from EPA Regional Screening Tables (http://www.epa.gov/reg3hwmd/risk/human/rb-concentration_table/index.htm'l. Unless otherwise noted, all
values are based on noncarcinogenic impacts.
e Concentrations of SFS constituents in manufactured soil (a 1:1 blend) were compared to theEco-SSLs, as discussed in Chapter 4, Section 4.4.3.
f See Appendix C for an explanation of USDA Phytotoxicity Screening Values for copper, nickel, and zinc.
g Based on carcinogenic risk, set at the standard EPA Office of Resource Conservation and Recovery risk target level of IE-OS.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
7-4
-------
Chapter 8.0 Agency Policy on the Beneficial Use of Silica-Based SFS
8. Agency Policy on the Beneficial Use of Silica-Based Spent
Foundry Sands from Iron, Steel and Aluminum Foundries
The beneficial use of SFS, when conducted in an environmentally sound manner, can
contribute positive environmental and economic benefits. Environmental benefits can include
energy savings, reduced greenhouse gas emissions, and water savings. Economic benefits can
include job creation in the beneficial use industry, reduced costs associated with SFS disposal,
increased revenue from the sale of SFS, and savings from using SFS in place of more costly
materials.
Spent foundry sand has been used as a substitute for virgin sand in certain markets. In
this risk assessment, the EPA and USDA have focused on a number of these markets.
Approximately 10 million tons of SFS are produced annually, with only 26% of these SFSs
being beneficially used beyond the foundry. Table 8-1 shows the beneficial uses (EPA, 2008c)
of SFS that were evaluated in this risk assessment. When beneficially using SFS it is particularly
important to check with your State Agency, which may have specific requirements pertaining to
such activities.
Table 8-1. Quantity SFS Beneficially used, by Market (tons)
Beneficial Use Market
Road construction (excluding asphalt)
Top soil mix/horticulture
SOURCE: U.S. EPA (2008c), Table ES-1
Quantity Beneficially Used
144,288
220,949
An EPA analysis (EPA, 2008c) provides estimates of the environmental benefits that can
be achieved with the beneficial applications that were studied in this risk assessment. The
analysis calculated environmental benefits per 1,000 cubic yards of SFSs and then extrapolated
these benefits to the total amount of SFSs used in a specific application.
Table 8-2. Primary Environmental Benefits of Beneficial use of SFS, by Market
Avoided Impacts
Energy Consumption
(megajoules)
Water consumption
(1,000 gallons)
CO2 Emissions
(metric tons)
Road Base Use
Extrapolated to 144,288
tons of SFS
17,800,000
3,000
1,500
Manufactured Soil Use
Extrapolated to 220,949
tons of SFS
27,900,000
4,800
2,500
SOURCE: U.S. EPA (2008c), Table ES-3
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
8-1
-------
Chapter 8.0 Agency Policy on the Beneficial Use of Silica-Based SFS
This risk assessment concluded that the beneficial uses of silica-based SFSs from iron,
steel and aluminum foundry operations when used in manufactured soils, soil-less potting media
and roadway subbase, are protective of human health and the environment. Based on this
conclusion, and the available environmental and economic benefits, the EPA and USDA support
the beneficial use of silica-based SFSs specifically from iron, steel and aluminum foundry
operations when used in manufactured soils, soil-less potting media and roadway subbase. The
EPA and USDA believe that these beneficial uses provide significant opportunities to advance
Sustainable Materials Management (SMM) (http://www.epa.gov/smm).
Any conclusions drawn by this risk assessment should be understood within the
limitations and scope of the evaluation, including the following:
• Only silica-based SFS from iron, steel and aluminum foundries are evaluated. In contrast,
SFS from leaded brass and bronze foundries often qualify as RCRA hazardous waste.
Also, there weren't sufficient data to characterize SFS from non-leaded brass foundries
and SFS containing olivine sand, and therefore these SFSs are not evaluated in this risk
assessment.
• In addition to SFS, foundries can generate numerous other wastes (e.g., unused and
broken cores, core room sweepings, cupola slag, scrubber sludge, baghouse dust,
shotblast fines). This assessment, however, applies only to SFS as defined in the
assessment: molding and core sands that have been subjected to the metalcasting process
to such an extent that they can no longer be used to manufacture molds and cores. To the
extent that other foundry wastes are mixed with SFS, the conclusions drawn by this
assessment may not be applicable.
• Samples from 39 foundries (totals and pore water data from 39 samples, and leachate
data from 108 samples) were used to represent silica-based SFS from all iron, steel, and
aluminum foundries in the U.S. Because the foundries were not chosen randomly, there
is uncertainty regarding whether the data are statistically representative of SFS from all
iron, steel, and aluminum foundries. However, these foundries were specifically selected
to ensure that the full range of constituents and their concentrations were adequately
represented, and the analytical data from these samples are the best available for
characterizing SFS constituents.
• Analytical data were available for 25 metals, 16 PAHs, 17 phenolics, and 20 dioxins and
dioxin-like compounds. USDA analyzed for organic compounds that are major binder
components (i.e., phenolics) or might be generated during thermal degradation of
chemical binders and other organic additives (i.e., PAHs, dioxins, furans), because these
constituents present the greatest hazard if at elevated levels in the environment. Review
of the scientific literature for evidence of additional organic compounds present in SFS
indicated that they were well below levels of concern.
• Screening and modeling evaluated those constituents for which toxicity benchmarks
exist.
• Evaluated beneficial uses include manufactured soil, soil-less growth media and road
subbase. The home garden using SFS-manufactured soil was modeled because it
demonstrated the greatest potential for exposure.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 8-2
-------
Chapter 8.0 Agency Policy on the Beneficial Use of Silica-Based SFS
The home garden scenario evaluated a single eight-inch deep application of SFS-
manufactured soil (comprised of 50% SFS) to a 0.1-acre garden.
Additional information can be found at the following web-sites:
EPA's (http://www.epa.gov/solidwaste/conserve/imr/foundry/index.htm),
American Foundry Society (AFS) (http://www.afsine.org),
Industry Recycling Starts Today (FIRST)
(http://www.afsinc.org/government/AFSFirst.cfm?ItemNumber=7887&&navItemNumbe
r=528)
Federal Highway Administration (FHWA)
(https://www.fhwa.dot.gov/publications/research/infrastructure/structures/97148/fsl.cfm)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 8-3
-------
Chapter 8.0 Agency Policy on the Beneficial Use of Silica-Based SFS
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 8-4
-------
Chapter 9.0 References
9. References
40 CFR 261.24. Code of Federal Regulations. Title 40, Section 261.24. Toxicity Characteristic.
Available at http://www.gpo.gov/fdsys/ (accessed 27 June 2014).
Abichou, T., T.B. Edil, C.H. Benson, and H. Bahia. 2004. Beneficial use of foundry by-products
in highway construction. Proceedings from the Geo-Institute of the American Society of
Civil Engineers. July 27-31, Los Angeles, CA.
Aceves, M.B., C. Grace, J. Ansorena, L. Dendooven, and P.C. Brookes. 1999. Soil microbial
biomass and organic C in a gradient of zinc concentrations in soils around a mine spoil
tip. Soil Biology and Biochemistry 31:867-876.
Adriano, D.C. 2001. Trace Elements in the Terrestrial Environments: Biogeochemistry,
Bioavailability, and Risks of Heavy Metals. 2nd ed. New York: Springer-Verlag.
AFS (American Foundry Society). 2007. Foundry Industry Benchmarking Survey: Industry
Practices Regarding the Disposal and Beneficial Reuse of Foundry Sand- Results and
Analysis. American Foundry Society, August.
Aitken, M.N., B. Evans, and J.G. Lewis. 1998. Effect of applying paper mill sludge to arable
land on soil fertility and crop yields. Soil Use Management 14:215-222.
Alberg, A.J., R.C. Yung, P. Strickland, and J. Nelson. 2002. Respiratory cancer and exposure to
arsenic, chromium, nickel, and polycyclic aromatic hydrocarbons. Clinics in
Occupational and Environmental Medicine 2:779-801.
Albiach, R., R. Canet, F. Pomares, F., and F. Ingelmo. 2000. Microbial biomass content and
enzymatic activities after the application of organic amendments to a horticultural soil.
Bioresource Technology 75:43-48.
Alexander, M. 1995. How toxic are toxic chemicals in soil. Environmental Science and
Technology 29:2713-2717.
Anderson, C. 1979. Cadmium, lead and calcium content, number and biomass, of earthworms
(Lumbricidae) from sewage sludge treated soils. Pedobiologia 79:309-319.
Andersson, M. 1987. Toxicity and Tolerance of Manganese andiron in Vascular Plants. A
Literature Review. Lund, Sweden: University of Lund.
Andersson, M. and R.T. Ottesen. 2008. Levels of dioxins and furans in urban surface soil in
Trondheim, Norway. Environmental Pollution 752:553-558.
Angle, J.S., and R.L. Chaney. 1991. Heavy metal effects on soil populations and heavy metal
tolerance ofRhizobium meliloti, nodulation, and growth of alfalfa. Water Air and Soil
Pollution 57(58):597-604.
Angle, J.S., M.A. Spiro, A.M. Heggo, M. El-Kherbawy, and R.L. Chaney. 1988. Soil microbial-
legume interactions in heavy metal contaminated soils of Palmerton, PA. Trace
Substances and Environmental Health 22:3 21 -3 3 6.
ASTM International (American Society for Testing and Materials). 2004. Standard Test Method
for Shake Extraction of Solid Waste with Water. Designation: D 3987-85. West
Conshohocken, PA.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-1
-------
Chapter 9.0 References
ATSDR (Agency for Toxic Substances and Disease Registry). 2007. Minimal Risk Levels for
Hazardous Substances. U.S. Department of Health and Human Services, Washington,
DC. Available at http://www.atsdr.cdc.gov/mrls/ (accessed 27 June 2014).
Baath, E. 1989. Effects of heavy metals in soil on microbial processes and populations (A
review). Water Air Soil Pollution ¥7:335-379.
Bailey, B. 2007. Personal communication. Cathy Davis, U.S. EPA, and Bruce Bailey, Vice
President of Technical Affairs, Schmack BioEnergy, Cleveland, OH. June.
Baird, S.J., E.A. Bailey, and D.J. Vorhees. 2007. Evaluating human risk from exposure to
alkylated PAHs in an aquatic system. Human and Ecological Risk Assessment 73:322-
338.
Baker, S.W. 1983. Sands for soil amelioration: Analysis of the effects of particle size, sorting,
and shape. Journal of the Sports Turf Research Institute 59:133-145.
Bartlett, R. J. 1986. Soil redox behavior. In Soil Physical Chemistry. Edited by D.J. Sparks. Boca
Raton, FL: CRC Press.
Basta, N.T., J.A. Ryan, and R.L. Chaney. 2005. Trace element chemistry in residual-treated soil:
Key concepts and metal bioavailability. Journal of Environmental Quality 34:49-63.
Becker, D.E. and S.E. Smith. 1951. The level of cobalt tolerance in yearling sheep. Journal of
Animal Science 10:266-271.
Beyer, L., and K. Mueller. 1995. Combined application of secondary paper mill sludges and
cattle slurry: soil risk or soil improvement. Toxicological and Environmental Chemistry
¥7:243-249.
Beyer, W.N., and C. Stafford. 1993. Survey and evaluation of contaminants in earthworms and
in soils derived from dredged material at confined disposal facilities in the Great Lakes
region. Environmental Monitoring and Assessment 2¥:151-165.
Beyer, W.N., G. Hensler, and J. Moore. 1987. Relation of pH and other soil variables to
concentrations of Pb, Cu, Zn, Cd, and Se in earthworms. Pedobiologia 30:167-172.
Beyer, W.N., G. Miller, and J.W. Simmers. 1990. Trace elements in soil and biota in confined
disposal facilities for dredged material. Environmental Pollution A65:l9-32.
Beyer, W.N., R.L. Chaney, and B.M. Mulhern. 1982. Heavy metal concentrations in earthworms
from soil amended with sewage sludge. Journal of Environmental Quality 77:381-385.
Bohn, H.L., B.L. McNeal, and G.A. O'Connor. 2001. Soil Chemistry. 3rd ed. New York: John
Wiley & Sons.
Bostrom, C.-E., P. Gerde, A. Hanberg, B. Jernstrom, C. Johansson, T. Kyrklund, A. Rannug, M.
Tornqvist, K. Victorin, and R. Westerholm. 2002. Cancer risk assessment, indicators, and
guidelines for polycyclic aromatic hydrocarbons in the ambient air. Environmental
Health Perspectives 770:451-488.
Brady, N.C., and R.R. Weil. 2007. The Nature and Properties of Soils. 14th ed. New York:
Prentiss Hall.
Broos, K., H. Beyens, and E. Smolders. 2005. Survival of rhizobia in soil is sensitive to elevated
zinc in the absence of host plant. Soil Biology and Biochemistry 37:573-579.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-2
-------
Chapter 9.0 References
Broos, K., M. Uyttebroek, J. Mertens, and E. Smolders. 2004. A survey of symbiotic nitrogen
fixation by white clover grown on metal contaminated soils. Soil Biology and
Biochemistry 3(5:633-640.
Brown, K.W., and R.L. Duble. 1975. Physical characteristics of soil mixtures used for golf green
construction. Agronomy Journal 67:647-652.
Brown, S.L., R.L. Chaney, M. Sprenger, and H. Compton. 2002. Assessing impact to wildlife at
biosolids remediated sites: Soil-Animal pathway. BioCycle 43(8):5Q-58.
Brown, S., R.L. Chaney, J.G. Hallfrisch, and Q. Xue. 2003. Effect of biosolids processing on
lead bioavail ability in an urban soil. Journal of Environmental Quality 32:100-108.
Budavari. 1989. The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals. llth
ed. Rahway, NJ: Merck and Company.
CalEPA (California Environmental Protection Agency). 2005. Chronic Reference Exposure
Levels. Adopted by OEHHA as of February 2005. California Office of Environmental
Health Hazard Assessment. Available at http://www.oehha.ca.gov/air/allrels.html
(accessed 27 June 2014).
CalEPA (California Environmental Protection Agency). 2009. Cancer Potency Values as of
July21, 2009. California Office of Environmental Health Hazard Assessment. Available
at http://www.oehha.ca.gov/risk/pdf/tcdb072109alpha.pdf (accessed 27 June 2014).
Chan, K.Y. 2001. An overview of some tillage impacts on earthworm population abundance and
diversity - implications for functioning in soils. Soil and Tillage Research. 179-191.
Chaney, R.L. 1980. Health risks associated with toxic metals in municipal sludge, pp. 59-83 in
Sludge—Health Risks of Land Application. Edited by G. Bitton, B.L. Damron, G.T. Edds,
and J.M. Davidson. Ann Arbor, MI: Ann Arbor Science Publishers Inc.
Chaney, R.L. 1983. Potential effects of waste constituents on the food chain, pp. 152-240 in
Land Treatment of Hazardous Wastes. Edited by J.F. Parr, P.B. Marsh, and J.M. Kla.
Park Ridge, NJ: Noyes Data Corp.
Chaney, R.L., and J.A. Ryan. 1993. Heavy metals and toxic organic pollutants in MSW-
composts: Research results on phytoavailability, bioavailability, fate, etc. pp. 451-505 in
Science and Engineering of Composting: Design, Environmental, Microbiological and
Utilization Aspects. Edited by H.A.J. Hoitink and H.M. Keener. Columbus, OH: The
Ohio State University.
Chaney, R.L., P.G. Reeves, J.A. Ryan, R.W. Simmons, R.M. Welch, and J.S. Angle. 2004. An
improved understanding of soil Cd risk to humans and low cost methods to remediate soil
Cd risks. BioMetals 77:549-553.
Chang, A.C., A.L. Page, L.J. Lund, J.E. Warneke, and C.O. Nelson. 1989. Municipal sludges and
utility ashes in California and their effects on soils, pp. 125-139 in Inorganic
Contaminants in the Vadose Zone. Edited by O. Bar-Yosef, NJ. Barrow, and J.
Goldshmid. Ecological Studies, Vol. 74. Berlin: Springer-Verlag.
Chapman, P.M. 1999. The role of soil microbial tests in ecological risk assessment. Human and
Ecological Risk Assessment 5:65 7-660.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-3
-------
Chapter 9.0 References
Gary, E.E., D.L. Grimes, S.L. Dallyn, G.A. Pearson, N.H. Peck, and R.S. Hulme. 1994 Plant Fe,
Al, and Cr Concentrations in Vegetables as Influenced by Soil Inclusion. Journal of Food
Quality 77:467-476.
Gary, E.E. and J. Kubota. 1990. Chromium Concentration in Plants: Effects of Soil Chromium
Concentration and Tissue Contamination by Soil. Journal of Agricultural Food
Chemistry 35:108-114.
Clark, R.B., K.D. Ritchey, and V.C. Baligar. 1999. Benefits and constraints for use of FGD
products on agricultural land. International Ash Utilization Symposium. Paper #22.
Center for Applied Energy Research, University of Kentucky, Lexington, KY.
Conder, J.M., and R.P. Lanno. 2000. Evaluation of surrogate measures of cadmium, lead, and
zinc bioavailability to Eiseniafetida. Chemosphere 41:1659-1668.
Corrier, D.E., L.D. Rowe, D.E. Clark and M.F. Hare. 1986. Tolerance and effect of chronic
dietary cobalt on sheep. Veterinary and Human Toxicology 25:216-219.
Crittenden, S.J., T. Eswaramurthy, R.G.M. de Goede, L. Brussaard, and M.M. Pulleman. Effect
of tillage on earthworms over short- and medium-term in conventional and organic
farming. Applied Soil Ecology'. In Press.
Crowley, D., and R.S. Dungan. 2002. Metals: Microbial processes affecting metals. Vol. 4, pp.
1878-1893 in Encyclopedia of Environmental Microbiology. Edited by G. Bitton. New
York: Wiley and Sons, Inc.
Curry, J.P., D. Byrne, and O. Schmidt. 2002. Intensive cultivation can drastically reduce
earthworm populations in arable land. Euro. J. of Soil Biology. 127-130.
Dai, J., T. Becquer, J.H. Rouiller, G. Reversal, F. Bernhard-Reversat, J. Nahmani, and P.
Lavelle. 2004. Heavy metal accumulation by two earthworm species and its relationship
to total and DTPA-extractable metals in soils. Soil Biology and Biochemistry 36:91-98.
Davis, W.B. 1978. Pros and cons of frequent sand topdressing. California Turfgrass Culture
25(4, Fall):25-29.
Dayton, E.A., S.D. Whitacre, R.S. Dungan, andN.T. Basta. 2010. Characterization of physical
and chemical properties of spent foundry sands pertinent to beneficial use in
manufactured soils. Plant and Soil. 329:27-33
de Koff, J., B.D. Lee, and R.S. Dungan. 2008. Amelioration of Physical Strength in Waste
Foundry Green Sands for Reuse as a Soil Amendment. Journal of Environmental Quality
37:2332-2338.
Deng, A., and PJ. Tikalsky. 2008. Geotechnical and leaching properties of flowable fill
incorporating waste foundry sand. Waste Management 25:2161-2170.
Dick, W.A., Y. Hao, R.C. Stehouwer, J.M. Bigham, W.E. Wolfe, D. Adriano, J.H. Beeghly, and
FJ. Haefner. 2000. Beneficial uses of flue gas desulfurization by-products: Examples and
case studies of land application, pp. 505-536 in Land Application of Agricultural,
Industrial, andMunicipal By-Products. Edited by J.F. Power and W.A. Dick. SSSA
Book Series No. 6, Madison, WI.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-4
-------
Chapter 9.0 References
Doran , J.W. and T.B. Parkin. 1996. Quantitative indicators of soils quality: A minimum data set.
pp. 25-37. in Methods for Assessing Soil Quality. J.W. Doran and AJ. Jones (Eds). Soil
Science Society of America Special Publication No. 49. Madison, WI.
Dungan, R.S. 2006. Polycyclic aromatic hydrocarbons and phenolics in ferrous and non-ferrous
waste foundry sands. Journal of Residuals Science & Technology 3:203-209.
Dungan, R.S. 2008. The characterization of trace metals and organics in spent foundry sands
over a one-year period. Journal of Residuals Science & Technology. 5:111-125/
Dungan, R.S., and N.H. Dees. 2006. Metals in waste foundry sands: Assessment with
earthworms. Journal of Residuals Science & Technology 3:177-184.
Dungan, R.S., andN. Dees. 2007. Use of spinach, radish, and perennial ryegrass to assess the
availability of metals in waste foundry sands. Water Air Soil Pollution 753:213-223.
Dungan, R.S., and N. Dees. 2009. The characterization of total and teachable metals in foundry
molding sands. Journal ofEnvironmental Management 90:539-548.
Dungan, R.S., and W.T. Frankenberger, Jr. 1999. Microbial transformations of selenium and the
bioremediation of seleniferous environments. Bioremediation Journal 3:171-188.
Dungan, R.S., and J.B. Reeves, III. 2005. Pyrolysis of foundry sand resins: A determination of
organic products by mass spectrometry. Journal of Environmental Science and Health
^¥0:1557-1567.
Dungan, R.S., and J.B. Reeves, III. 2007. Pyrolysis of carbonaceous foundry sand additives:
Seacoal and gilsonite. Thermochimica Acta 460:60-66.
Dungan, R.S., U. Kukier, and B.D. Lee. 2006. Blending foundry sands with soil: Effect on
dehydrogenase activity. Science of the Total Environment 357:221-230.
Dungan, R.S., B.D. Lee, P. Shouse, and J.P. de Koff. 2007. Saturated hydraulic conductivity of
soils blended with waste foundry sands. Soil Science 70:751-758.
Dungan, R.S., J. Huwe, and R.L. Chaney. 2009. Concentrations of PCDD/PCDFs and PCBs in
spent foundry sands. Chemosphere 75:1232-1235.
Edwards, C.A., and P.J. Bohlen. 1996. Biology and Ecology of Earthworms. 3rd ed. London:
Chapman and Hall.
El-Aziz, R., J.S. Angle, and R.L. Chaney. 1991. Metal tolerance ofRhizobium meliloti isolated
from heavy metal contaminated soils. Soil Biology and Biochemistry 23:795-798.
Essington, M.E., and S.V. Mattigod. 1991. Trace element solid-phase associations in sewage
sludge and sludge-amended soil. Soil Science Society of America Journal 55:350-356.
Fahnline, D.E., and Regan, R.W. 1995. Leaching of metals from beneficially used foundry
residuals into soils, pp. 339-347 in 50th Purdue Industrial Waste Conference
Proceedings. Chelsea, MI: Ann Arbor Press Inc.
FIRST (Foundry Industry Recycling Starts Today). 2004. Foundry Sand Facts for Civil
Engineers. FHWA-IF-04-004. U.S. Department of Transportation, Federal Highway
Administration. Washington, DC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-5
-------
Chapter 9.0 References
Fischer, C., C. Thies, and T. Tsharntke. 2011. Small mammals in agricultural landscapes:
Opposing responses to farming practices and landscape complexity. Biological
Conservation. (144): 113 0-113 6.
Ford, R.G., P.M. Bertsch, and KJ. Farley. 1997. Changes in transition and heavy metal
partitioning during hydrous iron oxide aging. Environmental Science and Technology
37:2028-2033.
Frink, C.R., and GJ. Bugbee. 1989. Ethylene dibromide: Persistence in soil and uptake by
plants. Soil Science 148:303-307.
Gandini, A., and M.N. Belgacem. 1997. Furans in polymer chemistry. Progress in Polymer
Science 22:1203-1379.
Garcia-Gil, J.C., C. Plaza, P. Soler-Rovira, and A. Polo. 2000. Long-term effects of municipal
solid waste compost application on soil enzyme activities and microbial biomass. Soil
Biology and Biochemistry 32:1907-1913.
Gardziella, A., L.A. Pilato, and A. Knop. 2000. Phenolic Resins: Chemistry, Applications,
Standardization, Safety and Ecology. 2nd ed. New York: Springer.
Gee, G.W., and J.W. Bauder. 1986. Particle-size analysis, pp. 383-411 mMethods of Soil
Analysis Physical andMineralogicalMethods. 2nd ed. Edited by A. Klute. Madison, WI:
Soil Science Society of America.
Getz, L.L., and B. McGuire. 2008. Factors Influencing Movement Distances and Home Ranges
of the Short-tailed Shrew (Blarina brevicauda). Northeastern Naturalist. 15(2): 293-302.
Ghosh A, Mukiibi M, Ela W. 2004. TCLP underestimates leaching of arsenic from solid
residuals under landfill conditions. Environ Sci Technol. 38:4677-4682.
Giller, K.E., S.P. Wani, and J.M. Day. 1986. Use of isotope dilution to measure nitrogen fixation
associated with the roots of sorghum and millet genotypes. Plant Soil 90:255-263.
Giller, K.E., E. Witter, and S.P. McGrath. 1998. Toxicity of heavy metals to microorganisms and
microbial processes in agricultural soils: A review. Soil Biology and Biochemistry
30:1389-1414.
Giller, K.E., E. Witter, and S.P. McGrath. 1999. Assessing risks of heavy metal toxicity in
agricultural soils: Do microbes matter? Human and Ecological Risk Assessment 5:683-
689.
Guney, Y., A.H. Aydilek, and M.M. Demirkan. 2006. Geoenvironmental behavior of foundry
sand amended mixtures for highway subbases. Waste Management 26:932-945.
Guo, M., S.K. Papiernik, W. Zheng, and S.R. Yates. 2003. Formation and extraction of persistent
fumigant residues in soils. Environmental Science and Technology 37:1844-1849.
Grubinger, V.P., W.H. Gutenmann, GJ. Doss, M. Rutzke, and D. Lisk. 1994. Chromium in
Swiss Chard Grown on Soil Amended with Tannery Meal Fertilizer. Chemosphere 28,4:
717-720.
Gwin, C.H., W.D. Scott, and R.H. James. 1976. A preliminary investigation of the organic
emissions from green sand pyrolysis. American Industrial Hygiene Association Journal
37:685-689.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-6
-------
Chapter 9.0 References
Ham, R.K., W.C. Boyle, and T.P. Kunes. 1981. Leachability of foundry process solid wastes.
Journal of Environmental Engineering 707:155-170.
Ham, R.K., W.C. Boyle, E.G. Engroff, and R.L. Fero. 1993. Organic compounds in ferrous
foundry process waste leachates. Journal of Environmental Engineering 779:34-55.
Ham, R.K., W.C. Boyle, FJ. Blaha, D. Oman, D. Trainer, T.J. Kunes, D.G. Nichols, and R.R.
Stanforth. 1986. Leachate and groundwater quality in and around ferrous foundry landfill
and comparison to leach test results. American Foundry Society Transactions 94:935-
942.
Hartenstein, E., E.F. Neuhauser, and J. Collier. 1980. Accumulation of heavy metals in the
earthworm Eiseniafoetida. Journal of Environmental Quality 9:23-26.
Hatzinger, P.B., and M. Alexander. 1995. Effect of ageing chemicals in soil upon their
biodegradability and extractability. Environmental Science and Technology 29:537-545.
He, Z.L., X.E. Yang, and PJ. Stoffella. 2005. Trace elements in agroecosystems and impacts on
the environment. Journal of Trace Elements in Medicine and Biology 79:125-140.
Heckman, J.R., J.S. Angle, and R.L. Chaney. 1986. Soybean nodulation and nitrogen fixation on
soil previously amended with sewage sludge. Biology and Fertility of Soils 2:181-185.
Heinz, G.H., DJ. Hoffman, and DJ. Audet. 2004. Phosphorus amendment reduces
bioavailability of lead to mallards ingesting contaminated sediments. Archives of
Environmental Contamination and Toxicology 46:534-541.
Helmke, P. A., W.P. Robarge, R.L. Korotev, and PJ. Schomberg. 1979. Effects of soil-applied
sewage sludge on concentrations of elements in earthworms. Journal of Environmental
Quality 5:322-327.
Helper, J., and M. Sobera. 1999. Thermal degradation of novolac resins by pyrolysis-gas
chromatography with a movable reaction zone. Journal ofChromatographyA833:211-
281.
Hettiarachchi, G.M., J.A. Ryan, R.L. Chaney, and C.M. LaFleur. 2003. Sorption and desorption
of cadmium by different fractions of biosolids-am ended soils. Journal of Environmental
Quality 32:1684-1693
Hindman, J., R. Stehouwer, and K. MacNeal. 2008. Spent foundry sand and compost in blended
topsoil: Availability of nutrients and trace elements. Journal of Residuals Science and
Technology 5:77-86.
Hoffman, F.O., and J.S. Hammonds. 1994. Propagation of uncertainty in risk assessments: The
need to distinguish between uncertainty due to lack of knowledge and uncertainty due to
variability. Risk Analysis 7¥(5):707712.
Holmgren, G.G.S., M.W. Meyer, R.L. Chaney, and R.B. Daniels. 1993. Cadmium, lead, zinc,
copper, and nickel in agricultural soils of the United States of America. Journal
Environmental Quality 22:335-348.
Hoppe, R.A., J. Johnson, J.E. Perry, P. Korb, J.E. Sommer, J.T. Ryan, R.C. Green, R. Durst, and
J. Monke. 2001. Structural and Financial Characteristics of U.S. Farms: 2001 Family
Farm Report. Agriculture Information Bulletin No. AIB768. U.S. Department of
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-7
-------
Chapter 9.0 References
Agriculture, Economic Research Service, Resource Economics Division, Washington,
DC. May.
IARC (International Agency for Research on Cancer). 2003. Handbook of Cancer Prevention,
Volume 8: Fruits and Vegetables. Lyon, France: IARC Press.
Ibekwe, A.M., J.S. Angel, R.L. Chaney, and P. van Berkum. 1995. Sewage sludge and heavy
metal effects on nodulation and nitrogen fixation of legumes. Journal of Environmental
Quality 2¥: 1199-1204.
Ibekwe, A.M., J.S. Angle, R.L. Chaney, and P. van Berkum. 1997. Enumeration and N2 fixation
potential ofRhizobium leguminosarum biovar trifolii grown in soil with varying pH
values and heavy metal concentrations. Agriculture Ecosystems and Environment
67:103-111.
Jager, T., R. Baerselman, E. Dijkman, A.C. De Groot, E.A. Hogendoorn, A. De Jong, J.A.W.
Kruitbosch, W.J.G.M. Peijnenburg. 2003. Availability of polycyclic aromatic
hydrocarbons to earthworms (Eisenia andrei, Oligochaeta) in field-polluted soils and
soil-sediment mixtures. Environmental Toxicology and Chemistry 22:767-775.
Ji, S., L. Wan, and Z. Fan. 2001. The toxic compounds and leaching characteristics of spent
foundry sands. Water Air Soil Pollution 732:347-364.
Jing, J., and S. Barnes. 1993. Agricultural use of industrial by-products. Biocycle 3¥(ll):63-64.
Johnson, D.L., K.C. Jones, C.J. Langdon, T.G. Piearce, and K.T. Semple. 2002. Temporal
changes in earthworm availability and extractability of polycyclic aromatic hydrocarbons
in soil. Soil Biology and Biochemistry 34:1363-1370.
Kabata-Pendias, A. 1992. Trace Elements in Soils and Plants. 1st ed. Boca Raton, FL: CRC
Press.
Kabata-Pendias, A. 2001. Trace Elements in Soils and Plants. 3rd ed. Boca Raton, FL: CRC
Press.
Kauffmann, P., R.W. Regan, and R.C. Voight. 1996. Survey of state environmental regulations
impacting beneficial reuse of foundry residuals. American Foundry Society Transactions
70¥:527-538.
Keener, H.A., G.P. Percival, K.S. Morrow, and G.H. Hill. 1949. Cobalt tolerance in young dairy
cattle. Journal of Dairy Science 30:527-533.
Kelsey, J.W., and M. Alexander. 1997. Declining bioavailability and inappropriate estimation of
risk of persistent compounds. Environmental Toxicology and Chemistry 7(5:582-585.
Kendall, D.S. 2003. Toxicity characteristic leaching procedure and iron treatment of brass
foundry waste. Environmental Science and Technology 37:367-371.
Kester, G.B., R.B. Brobst, A. Carpenter, R.L. Chaney, A.B. Rubin, R.A. Schoof, and D.S.
Taylor. 2005. Risk characterization, assessment, and management of organic pollutants in
beneficially used residual products. Journal of Environmental Quality 34:80-90.
Khalil, M.A., H.M. Abdel-Lateif, B.M. Bayoumi, andN.M. van Straalen. 1996. Analysis of
separate and combined effects of heavy metals on the growth of Aporrectodea caliginosa
(Oligochaeta: Annelida), using the toxic unit approach. Applied Soil Ecology 4:213-219.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-8
-------
Chapter 9.0 References
Kinraide, T.B. 1991. Identity of the rhizotoxic aluminum species. Plant Soil 134:167-178.
Kinkle, B.K., J.S. Angle, and H.H. Keyser. 1987. Long-term effects of metal-rich sewage sludge
application on soil populations of Bradyrhizobium japonicum. Applied Environmental
Microbiology 53:315-319.
Korcak, R.F. 1995. Utilization of coal combustion by-products in agriculture and horticulture.
pp. 107-130 in Agricultural Utilization of Urban and Industrial By-Products. Edited by
D.L. Karlen, RJ. Wright, and W.D. Kemper. ASA Special Publication No. 58. American
Society of Agronomy, Madison, WI.
Krauss, M., W. Wilcke, and W. Zech. 2000. Availability of polycyclic aromatic hydrocarbons
(PAHs) and polychlorinated biphenyls (PCBs) to earthworms in urban soils.
Environmental Science and Technology 34:4335-4340.
Krueger, R.C., R.K. Ham, and W.C. Boyle. 1989. The variability of ferrous foundry waste
leaching characteristics and comparison to landfill unsaturated zone leachate quality.
43rdPurdue Industrial Waste Conference Proceedings. Chelsea, MI: Lewis Publishers,
Inc.
Kukier, U., and R.L. Chaney. 2004. In situ remediation of nickel phytotoxicity for different plant
species. Journal of Plant Nutrition 27(3):465-495.
Kukier, U., C.A. Peters, R.L. Chaney, J.S. Angle and RJ. Roseberg. 2004. The effect of pH on
metal accumulation in two Alyssum species. Journal of Environmental Quality 33:2090-
2102.
Kukurenda, H., and R. Lipski. 1982. Solubility of manganese in different soils and its availability
to plants, Pamiet. Pulawski 76:172 (in Polish).
Kumpiene, J., A. Lagerkvist, and C. Maurice. 2008. Stabilization of As, Cr, Cu, Pb and Zn in soil
using amendments - A review. Waste Management 28:215-225.
Kurtz Bros., Inc. 2007. Personal communication. Rufus Chaney, U.S. Department of Agriculture,
with John Kurtz, Kurtz Bros., Inc. April.
Lakzian, A., P. Murphy, A. Turner, J.L. Beynon, and K.E. Giller. 2002. Rhizobium
leguminosarum bv. viciae populations in soils with increasing heavy metal
contamination: Abundance, plasmid profiles, diversity and metal tolerance. Soil Biology
and Biochemistry 34:519-529.
Langmuir, D.L., P. Chrostrowski, R.L. Chaney, and B. Vigneault. 2005. Issue Paper on
Environmental Chemistry of Metals. U.S. EPA Risk Assessment Forum: Papers
Addressing Scientific Issues in the Risk Assessment of Metals. Available at
http://www.epa. gov/raf/publications/pdfs/ENVCHEMFINAL81904CORRO1 -25-05 .PDF
(accessed 27April 2014).
Lee, I.-S., O.-K. Kim, Y.-Y. Chang, B.-H. Bae, H.-H. Kim, and K.-H. Baek. 2002. Heavy metal
concentrations and enzyme activities in soil from a contaminated Korean shooting range.
Journal ofBioscience andBioengineering 94:406-411.
Lee, K.E. 1985. Earthworms: Their Ecology and Relationships with Soils and Land Use. New
York: Academic Press.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-9
-------
Chapter 9.0 References
Lee, T., and C.H. Benson. 2004. Sorption and degradation of alachlor and metolachlor in ground
water using green sands. Journal of Environmental Quality 33:1682-1693.
Lee, T., and C.H. Benson. 2006. Leaching behavior of green sands from gray-iron foundries used
for reactive barrier applications. Environmental Engineering Science 23:156-170.
Lee, T., C.H. Benson, and G.R. Eykholt. 2004a. Waste green sands as reactive media for ground
water contaminated with trichloroethylene (TCE). Journal of Hazardous Materials
109:25-36.
Lee, T., J.-W. Park, and J.-H. Lee. 2004b. Waste green sands as reactive media for the removal
of zinc from water. Chemosphere 5(5:571-581.
Leidel, D.S., M. Novakowski, D. Pohlman, Z.D. MacRunnels, and M.H. MacKay. 1994.
External beneficial reuse of spent foundry sand. American Foundry Society Transactions
102:235-243.
Lexmond, T.M. 1980. The effect of soil pH on copper toxicity to hydroponically grown maize.
Netherlands Journal of Agricultural Science. 29: 209-230.
Lindsay, B.J., and TJ. Logan. 2005. Agricultural reuse of foundry sand. Journal of Residuals
Science and Technology 2:3-12.
Lindsay, W.L. 2001. Chemical Equilibria in Soils. Caldwell, NJ: The Blackburn Press.
Lindsay, W.L., and W.A. Norvell. 1978. Development of a DTP A soil test for zinc, iron,
manganese, and copper. Soil Science Society of America Journal ¥2:421-428.
Lock, K., and C.R. Janssen. 2001. Ecotoxicity of zinc in spiked artificial soil versus
contaminated field soils. Environmental Science and Technology 35:4295-4300.
Lock, K., and C.R. Janssen. 2003. Influence of ageing on zinc bioavailability in soils.
Environmental Pollution 126:371-374.
Lombi, E., F. Zhao, G. Zhang, B. Sun, W. Fitz, H. Zhang, and S.P. McGrath. 2002. In situ
fixation of metals in soils using bauxite residue: Chemical assessment. Environmental
Pollution 775:435-443.
Lytle, C.A., W. Bertsch, and M. McKinley. 1998a. Determination of novolac resin thermal
decomposition products by pyrolysis-gas chromatography-mass spectrometry. Journal of
Analytical and Applied Pyrolysis 45:121-131.
Lytle, C.A., W. Bertsch, andM.D. McKinley. 1998b. Determination of thermal decomposition
products from a phenolic urethane resin by pyrolysis-gas chromatography-mass
spectrometry. Journal of High Resolution Chromatography 27:128-132.
Ma, W.-C. 1982. The influence of soil properties and worm-related factors on the concentration
of heavy metals in earthworms. Pedobiologia 24:109-119.
Ma, W.-C. 1984. Sublethal toxic effects of copper on growth, reproduction, and litter breakdown
activity in the earthworm Lumbricus rubellus, with observations on the influence of
temperature and soil pH. Environmental Pollution 33:207-219.
Ma, W.-C. 1987. Heavy metal accumulation in the mole, Talpa europea, and earthworms as an
indicator of metal bioavailability in terrestrial environments. Bulletin of Environmental
Contaminant Toxicology 39:933-939.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-10
-------
Chapter 9.0 References
Mastral, A.M., Call en, M.S., and Garcia, T. 2000. Toxic organic emissions from coal
combustion. Fuel Processing Technology (57:1-10.
McBride, M.B. 1994. Environmental Chemistry of Soils. New York: Oxford University Press.
McBride, M., Sauve, S. and Hendershot, W. 1997. Solubility control of Cu, Zn, Cd and Pb in
contaminated soils. European Journal of Soil Science 48: 337-346.
McCoy, E.L. 1998. Sand and organic amendment influences on soil physical properties related to
turf establishment. Agronomy Journal 90:411-419.
McGrath, S.P., P.C. Brookes, and K.E. Giller. 1988. Effects of potentially toxic metals in soils
derived from past applications past applications of sewage sludge on nitrogen fixation by
Trifolium repens L. Soil Biology and Biochemistry 20:415-424.
McKeague, J.A., and J.H. Day. 1966. Dithionite- and oxalate-extractable Fe and Al as aids in
differentiating various classes of soils. Canadian Journal of Soil Science 46:13-22.
McKenna, I.M., R.L. Chaney, S.H. Tao, R.M. Leach Jr., and P.M. Williams. 1992. Interactions
of plant zinc and plant species on the bioavailability of plant cadmium to Japanese quail
fed lettuce and spinach. Environmental Research 57:73-87.
McKenzie, R.M. 1980. The adsorption of lead and other heavy metals on oxides of manganese
and iron. Australian Journal of Soil Research 75:61-73.
Miller, D.H. and L.L. Getz. 1977. Factors influencing local distribution and species diversity of
forest small mammals in New England. Can. J. Zool. 55: 806-814.
Morgan, J.E., and A.J. Morgan. 1988. Earthworms as biological monitors of cadmium, copper,
lead and zinc in metalliferous soils. Environmental Pollution 54:123-138.
Morgan, J.E., and A.J. Morgan. 1999. The accumulation of metals (Cd, Cu, Pb, Zn and Ca) by
two ecologically contrasting earthworm species (Lumbricus rubellus and Aporrectodea
caliginosd): Implications for ecotoxicological testing. Applied Soil Ecology 73:9-20.
Naik, R.T., V.M. Patel, D.M. Parikh, and M.P. Tharaniyil. 1994. Utilization of used foundry
sand in concrete. Journal of Materials in Civil Engineering 6:254-263.
Naystrom, P., J. Lemkow, and J. Orkas. 2004. Waste foundry sand - a resource in composting
and soil production. Foundry Trade Journal 775(3615): 188-189
NLM (National Library of Medicine). 2006. Hazardous Substance Data Bank. National
Institutes of Health, Bethesda, MD. Available at http://toxnet.nlm.nih.gov/cgi-
bin/sis/htmlgen?HSDB (accessed 27 June 2014).
NRC (National Research Council). 1980. Mineral Tolerance of Domestic Animals. Washington,
DC: National Academy of Sciences.
Obbard, J.P., D. Sauerbeck, and K.C. Jones. 1994. Dehydrogenase activity of the microbial
biomass in soils from a field experiment amended with heavy metal contaminated sewage
sludge. Science of the Total Environment 142:157-162.
Oliveira, T.D., M. Bertrand, and J. Roger-Estrade. 2012. Short-term effects of ploughing on the
abundance and dynamics of two endogeic earthworm species in organic cropping systems
in northern France. Soil and Tillage Research. 76-84.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-11
-------
Chapter 9.0 References
Overcash, M., R.C. Sims, J.L. Sims, and J.K.C. Nieman. 2005. Beneficial reuse and
sustainability: The fate of organic compounds in land-applied waste. Journal of
Environmental Quality 34:29-41.
Owusu, Y.A. 1982. Physical-chemical study of sodium silicate as a foundry sand binder.
Advances in Colloid and Interface Science 75:57-91.
Pais, I, and J. Benton Jones. 1997. The Handbook of Trace Elements. Boca Raton, FL, St. Lucie
Press.
Palmer, W.G., R.H. James, and W.J. Moorman. 1985. Analysis of emissions collected from four
types of iron casting molds. American Industrial Hygiene Association Journal 46:724-
730.
Peijnenburg, W., Baerselman, R. de Groot, A., Jager, T., Leenders, D., Posthuma, L. and Van
Veen, R. 2000. Quantification of metal bioavailability for lettuce (Lactuca sativaL.) in
field soils. Archives of Environmental Contamination and Toxicology 39: 420-430.
Phillips, V.R., N. Kirkpatrick, I.M. Scotford, R.P. White, and R.G.O. Burton. 1997. The use of
paper-mill sludges on agricultural land. Bioresource Technology (50:73-80.
Pietz, R.I., J.R. Peterson, J.E. Prater, and D.R. Zenz. 1984. Metal concentrations in earthworms
from sewage sludge-amended soils at a strip mine reclamation site. Journal of
Environmental Quality 13:651-654.
Posthuma, L., R. Baerselman, R.P.M. Van Veen, and E.M. Dirven-Van Breemen. 1997. Single
and joint toxic effects of copper and zinc on reproduction of Enchytraeus crypticus in
relation to sorption of metals in soils. Ecotoxicology and Environment Safety 38:108-
121.
Puglisi, E., F. Cappa, G. Fragoulis, M. Trevisan, and A.M. Del Re. 2007. Bioavailability and
degradation of phenanthrene in compost amended soils. Chemosphere (57:548-556.
Rana, S.V.S., and Y. Verma. 2005. Biochemical toxicity of benzene. Journal of Environmental
Biology 2(5:157-168.
Reeves, P.G., and R.L. Chaney. 2002. Nutritional status affects the absorption and whole-body
and organ retention of cadmium in rats fed rice-based diets. Environmental Science and
Technology 36:2684-2692.
Reeves, P.G., and R.L. Chaney. 2004. Marginal nutritional status of zinc, iron, and calcium
increases cadmium retention in the duodenum and other organs of rats fed a rice-based
diet. Environmental Research 96:311-322.
Regan, R.W., Sr., R.C. Voigt, W.T. Paletski, and R.P. Massell. 1994. Chemical characterizations
of spent molding sands: Environmental issues. American Foundry Society Transactions
103:749-756.
Reichman, S.M. 2002. The Response of Plants to Metal Toxicity: A Review Focusing on Copper,
Manganese and Zinc. Melbourne, Australia: Australian Minerals & Energy Environment
Foundation, Occasional Paper No. 14.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-12
-------
Chapter 9.0 References
Renella, G., A.M. Chaudri, and P.C. Brooks. 2002. Fresh additions of heavy metals do not model
long-term effects on microbial biomass and activity. Soil Biology and Biochemistry
34:121-124.
Riediker, S., S. Ruckstuhl, MJ.-F. Suter, A.M. Cook, and W. Giber. 2000. p-Toluenesulfonate in
landfill leachates: Leachability from foundry sands and aerobic biodegradation.
Environmental Science and Technology 34:2156-2161.
Roberts, T.M., W. Gizyn, and T.C. Hutchinson. 1974. Lead contamination of air, soil, vegetation
and people in the vicinity of secondary lead smelters. In Trace Substances in
Environmental Health, Vol. 8. Edited by D.D. Hemphill. Columbia, MO: University of
Missouri.
Rogowski, D.L., and W. Yake. 2005. Typical dioxin concentrations in agricultural soils of
Washington state and potential sources. Environmental Science and Technology
39:5170-5176.
Rother, J.A., J.W. Millbank, and I. Thornton. 1983. Nitrogen fixation by white clover (Trifolium
repens) in grassland soils contaminated with cadmium, lead, and zinc. Journal of Soil
Science 34:127-136.
Sauve, S., Cook, N., Hendershot, W.H. and McBride, M.B. 1996. Linking plant tissue
concentrations and soil copper pools in urban contaminated soils. Environmental
Pollution. 44: 193-210.
Sauve, S., McBride, M.B., Norvel, W.A., and Hendershot, W.H. 1997. Copper solubility and
speciation of in situ contaminated soils: effects of copper level, pH and organic matter.
Water, Air and Soil Pollutions: 100: 133-149.
Saxton, K.E., WJ. Rawls, J.S. Romberger, and R.I. Papendick. 1986. Estimating generalized
soil-water characteristics from texture. Soil Science Society of America Journal 50:1031-
1036.
Scott, W.D., C.E. Bates, and R.H. James. 1977. Chemical emissions from foundry molds.
American Foundry Society Transactions 95:203-208.
Scott, W.D., R.H. James, and C.E. Bates. 1976. Foundry air contaminants from green sand
molds. American Industrial Hygiene Association Journal 37:335-344.
Semple, K.T., A.W.J. Morriss, and G.I. Paton. 2003. Bioavailability of hydrophobic organic
contaminants is soils: fundamental concepts and techniques for analysis. European
Journal of Soil Science 54:809-819.
Shacklette, H.T., and J.G. Boerngen. 1984. Element Concentrations in Soils and Other Surficial
Materials of the Conterminous United States. U.S. Geological Survey, Professional Paper
1270. U.S. Gov. Printing Office, Washington, DC.
Siebielec, G., R.L. Chaney, and U. Kukier. 2007. Liming to remediate Ni contaminated soils
with diverse properties and a wide range of Ni concentration. Plant Soil 299:117-130.
Simard, R.R., R. Baziramakenga, S. Yelle, and J. Coulombe. 1998. Effects of de-inking paper
sludges on soil properties and crop yields. Canadian Journal of Soil Science 75:689-697.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-13
-------
Chapter 9.0 References
Singh, B.R., and A.S. Jeng. 1993. Uptake of zinc, cadmium, mercury, lead, chromium, and
nickel by ryegrass grown in a sandy soil. Norwegian Journal of Agricultural Science
7:147-157.
Smith, D.B., W.F. Cannon, L.G. Woodruff, R.B. Garrett, R. Klassen, I.E. Kilburn, J.D. Horton,
H.D. King, M.B. Goldhaber, and J.M. Morrison. 2005. Major- and Trace-Element
Concentrations in Soils from Two Continental-Scale Transects of the United States and
Canada. Open-File Report 2005-1253. U.S. Department of the Interior, U.S. Geological
Survey, Reston, VA. Available at http://pubs.usgs.gov/of/2005/1253/pdf/OFR1253.pdf
(accessed 27 June 2014).
Smith, S.R. 1997. Rhizobium in soils contaminated with copper and zinc following the long-term
application of sewage sludge and other organic wastes. Soil Biology and Biochemistry
29:1475-1489.
Sobera, M., and J. Helper. 2003. Pyrolysis-gas chromatography-mass spectrometry of cured
phenolic resins. Journal of Chromatography A993:131-135.
Sparks, D.L. 2003. Environmental Soil Chemistry. 2nd ed. San Diego, CA: Academic Press.
Spurgeon, D.J., and S.P. Hopkin. 1996. Effects of metal-contaminated soils on the growth,
sexual development, and early cocoon production of the earthworm Eisenia fetida, with
particular reference to zinc. Ecotoxicology and Environment Safety 35:86-95.
Stanforth, R., D. Nichols, T. Kunes, D. Oman, D. Trainor, R. Ham, W. Boyle, and F. Blaha.
1988. Foundry waste landfills and ground water quality. American Foundry Society
Transactions 9(5:873-880.
Sterrett, S.B., R.L. Chaney, C.E. Hirsch, and H.W. Mielke. 1996. Influence of amendments on
yield and heavy metal accumulation of lettuce grown in urban garden soils.
Environmental Geochemistry and Health 18:135-142.
Stevenson, FJ. 1982. Nitrogen organic forms, pp. 625-642 in Methods of Soil Analysis. Part 2.
Chemical and Microbiological Properties. Edited by A.L. Page. Agronomy Monograph
No. 9, 2nd Edition. Soil Science Society of America (SSSA), Madison, WI.
Stevenson, M. 1996. The taxing problem of waste sand disposal. Foundry Trade Journal
770(3525):580-583.
Stout, W.L., J.L. Hern, R.F. Korcak, and C.W. Carlson. 1988. Manual for Applying Fluidized
Bed Combustion Residue to Agricultural Lands. USDA-ARS No. 74. National Technical
Information Service, Springfield, VA.
Stuczynski, T.I., G. Siebielec, W.L. Daniels, G.C. McCarty, and R.L. Chaney. 2007. Biological
aspects of metal waste reclamation with sewage sludge. Journal of Environmental
Quality 36:1154-1162.
Sullivan, T.P., D.S. Sullivan, andH.M.A. Thistlewood. 2012. Abundance and diversity of small
mammals in response to various linear habitats in semi-arid agricultural landscapes. J. of
Arid Environments. (83): 54-61.
Swartz, W.E., and L.T. Kardos. 1963. Effects of compaction on physical properties of sand-soil-
peat mixtures at various moisture contents. Agronomy Journal 55:1'-10.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-14
-------
Chapter 9.0 References
Talmage, S.S., and B.T. Walton. 1993. Food chain transfer and potential renal toxicity of
mercury to small mammals at a contaminated terrestrial field site. Ecotoxicology 2:243-
256.
Tang, J., H.-H. Liste, and M. Alexander. 2002. Chemical assays of availability to earthworms of
polycyclic aromatic hydrocarbons in soil. Chemosphere 48:35-42.
Taylor, D.H., and G.R. Blake. 1979. Sand content of sand-soil-peat mixtures for turf grass. Soil
Science Society of America Journal ¥3:394-398.
Tikalsky, P.J., H.U. Bahia, A. Deng, and T. Snyder. 2004. Excess Foundry Sand
Characterization and Experimental Investigation in Controlled Low-strength Materials
and Hot-Mixing Asphalt, Final Report. Prepared for U.S. Department of Energy, Contract
No. DE-FC36-01ID13974.
Tlustos, P., J. Balik, J. Szakova, and D. Pavlikova. 1998. The accumulation of arsenic in radish
biomass when different forms of As were applied in the soil. Rostlinnd Vyroba 44(1)
(Cz).
Trivedi, P., and L. Axe. 2000. Modeling Cd and Zn sorption to hydrous metal oxides.
Environmental Science and Technology 34:2215-2223.
U.S. ACE (Army Corps of Engineers). 1984. Engineering and Design—Pavement Criteria for
Seasonal Frost Conditions. EM 1110-3-138. Department of the Army, Corps of
Engineers, Office of the chief of Engineers, Washington, DC. Available at
http://www.publications.usace.army.mil/Portal s/76/Publications/EngineerManuals/EM_l
110-3-138.pdf (accessed 27 June 2014).
USDA (U.S. Department of Agriculture). 1993. Soils: Soil Texture Calculator. U.S. Department
of Agriculture, Natural Resources Conservation Service, Soil Survey Division,
Washington, DC. Available at
http://www.nrcs. usda.gov/wps/portal/nrcs/detail/national/nedc/training/soil/? cid=nrcsl42
p2_054167 (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1991a. Risk Assessment Guidance for Superfund
Volume I: Human Health Evaluation Manual Supplemental Guidance "Standard Default
Exposure Factors" Interim Final. Office of Emergency and Remedial Response. March.
OSWER Directive 9285.6-03.
U.S. EPA (Environmental Protection Agency). 1991b. EPA Region 3 Guidance on Handling
Chemical Concentration Data Near the Detection Limit in Risk Assessments. Region 3,
Philadelphia, PA. November.
U.S. EPA (Environmental Protection Agency). 1992a. Workbook of Screening Techniques for
Assessing Impacts of Toxic Air Pollutants. EPA-454/R-92-024. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Research Triangle
Park, NC.
U.S. EPA (Environmental Protection Agency). 1992b. Screening Procedures for Estimating the
Air Quality Impact of Stationary Sources, Revised. EPA-454/R-92-019. U.S.
Environmental Protection Agency, Research Triangle Park, NC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-15
-------
Chapter 9.0 References
U.S. EPA (Environmental Protection Agency). 1993. Standards for the use or disposal of sewage
sludge. 40 CFR Parts 257, 403, and 503. Federal Register 55:9248-9415.
U.S. EPA (Environmental Protection Agency). 1994a. Methods for Derivation of Inhalation
Reference Concentrations and Application of Inhalation Dosimetry. EPA/600/8-90/066F.
U.S. Environmental Protection Agency, Office of Research and Development, Office of
Health and Environmental Assessment, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1994b. Revised Interim Soil Lead Guidance for
CERCLA Sites andRCRA Corrective Action Facilities. OSWER Directive #9355.4-12.
EPA/540/F-98/030. U.S. Environmental Protection Agency, Office of Solid Waste and
Emergency Response, Washington, DC. August.
U.S. EPA (Environmental Protection Agency). 1995a. Compilation of Air Pollutant Emission
Factors: Volume I: Stationary Point and Area Sources. U.S. Environmental Protection
Agency, Office of Air Quality Planning & Standards, Technology Transfer Network,
Clearinghouse for Inventories & Emissions Factors, Research Triangle Park, NC.
Available at http://www.epa.gov/ttn/chief/ap42/index.html (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1995b. SCREENS Model User's Guide. EPA-
454/B-95-004. U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards, Emissions, Monitoring, and Analysis Division, Research Triangle Park,
North Carolina Available at http://www.epa.gov/scramOO 1 /userg/screen/screen3d.pdf
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1995c. Guidance for Risk Characterization.
Science Policy Council. U.S. Environmental Protection Agency, Office of the Science
Advisor, Science Policy Council, Washington, DC. February. Available at
http://www.epa.gov/osa/spc/pdfs/rcguide.pdf (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1995d. User's Guide for the Industrial Source
Complex (ISC3) Dispersion Models. Volume II: Description of Model Algorithms. EPA-
454/B-95-003b. Emissions, Monitoring, and Analysis Division, Office of Air Quality
Planning and Standards, Research Triangle Park, NC. September.
U.S. EPA (Environmental Protection Agency). 1996. Soil Screening Guidance: Fact Sheet. EPA-
540/F-95/041. U.S. Environmental Protection Agency, Office of Solid Waste and
Emergency Response, Washington, DC. July. Available at
http://www.epa.gov/superfund/health/conmedia/soil/pdfs/fact_sht.pdf (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 1997a. EPA 's Composite Model for Leachate
Migration with Transformation Products. EPACMTP: User's Guide. Office of Solid
Waste, Washington, DC. Available online at:
http://www.epa.gov/osw/nonhaz/industrial/tools/cmtp/index.htm (accessed 27 June 2014)
U.S. EPA (Environmental Protection Agency). 1997b. "Health Effects Assessment Summary
Tables (HEAST). Fiscal Year 1997 Update". Office of Solid Waste and Emergency
Response. EPA-540-R-97-036. PB97-921199. July.
U.S. EPA (Environmental Protection Agency). 1997c. Exposure Factors Handbook, Volume II
FoodIngestion Factors. EPA/600/P-95/002Fa. U.S. Environmental Protection Agency,
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-16
-------
Chapter 9.0 References
Office of Research and Development, Washington, DC. August. Available at
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid= 12464 (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1998. Clarification to the 1994 Revised Interim
Soil Lead (Pb) Guidance for CERCLA Sites andRCRA Corrective Action Facilities.
OSWER Directive #9200.4-27P. EPA/540/F-98/030. U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, DC. August.
U.S. EPA (Environmental Protection Agency). 1999a. Contract Laboratory Program Statement
of Work for Inorganic Analysis, Multi-Media, Multi-Concentration. Document number
ILM04.0b. U.S. Environmental Protection Agency, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1999b. Air Module Pre- and Postprocessor
Background and Implementation for the Multimedia, Multipathway, andMultireceptor
Risk Assessment (3MRA) for HWIR99. U.S. Environmental Protection Agency, Office of
Solid Waste and Emergency Response, Office of Solid Waste, Washington, DC. October.
Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/pdf/risk/modules/s0048.pdf
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 1999c. Data Collection for the Hazardous Waste
Identification Rule, Section 13.0 Ecological Receptors and Habitats. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Office of Solid
Waste, Washington, DC. October. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/pdf/risk/data/s0043.pdf (accessed
27 June 2014).
U.S. EPA (Environmental Protection Agency). 2000. Risk Characterization Handbook. U.S.
Environmental Protection Agency, Office of Research and Development, Office of
Science Policy, Washington, DC. Available at http://www.epa.gov/spc/pdfs/rchandbk.pdf
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2002a. Industrial Waste Management Evaluation
Model (IWEM) User's Guide. EPA530-R-02-013. U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, DC. August.
Available at http://www.epa.gov/epawaste/nonhaz/industrial/tools/iwem/index.htm
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2002b. Industrial Waste Management Evaluation
Model (IWEM) Technical Background Document. EPA530-R-02-012. U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response,
Washington, DC. August. Available at
http://www.epa.gov/epawaste/nonhaz/industrial/tools/iwem/index.htm (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 2002c. Supplemental Guidance for Developing
Soil Screening Levels for Superfund Sites. OSWER 9355.4-24. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC.
December. Available at
http://www.epa.gov/superfund/health/conmedia/soil/pdfs/ssg_main.pdf (accessed 27 June
2014).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-17
-------
Chapter 9.0 References
U.S. EPA (Environmental Protection Agency). 2002d. Beneficial Reuse of Foundry Sand: A
Review of State Practices and Regulations. U.S. Environmental Protection Agency,
Sector Strategies Division, Office of Policy, Economics, and Innovation. Washington
DC. Available at
http://www.epa.gov/sectors/sectorinfo/sectorprofiles/metalcasting/reuse.pdf (accessed 27
June 2014).
U.S. EPA (Environmental Protection Agency). 2002e. 2002 Notice of data availability, 40 CFR
Part 503. Standards for the use or disposal of sewage sludge. Federal Register 67:40554-
40576. U.S. Environmental Protection Agency, Washington, DC.
U.S. EPA (Environmental Protection Agency). 2002f Drinking Water From Household Wells.
U.S. Environmental Protection Agency, Office of Water, Washington, DC. Available at
http://www.epa.gov/privatewells/pdfs/household_wells.pdf (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2003a. Human Health Toxicity Values in
Superfund Risk Assessments. OSWER Directive 9285.7-53. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC.
December 5. Available at http://www.epa.gov/oswer/riskassessment/pdf/hhmemo.pdf
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2003b. Technical Background Document for the
Sewage Sludge Exposure and Hazard Screening Assessment. U.S. Environmental
Protection Agency, Office of Water, Washington, DC. 822-B-03-001. December.
Available at
http://water.epa.gov/scitech/wastetech/biosolids/upload/sewagesludge_background.pdf
(accessed 27 June 2014)
U.S. EPA (Environmental Protection Agency). 2003c. Multimedia, Multipathway, and
Multireceptor Risk Assessment (3MRA) Modeling System Volume I: Modeling System and
Science. U.S. Environmental Protection Agency, Office of Solid Waste, Washington DC.
530-D-03-001a. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/risk03.htm (accessed 31
December 2013).
U.S. EPA (Environmental Protection Agency). 2003d. Multimedia, Multipathway, and
Multireceptor Risk Assessment (3MRA) Modeling System Volume II: Site-Based,
Regional, and National Data. U.S. Environmental Protection Agency, Office of Solid
Waste, Washington DC. 530-D-03-001a. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/risk03.htm (accessed 31
December 2013).
U.S. EPA (Environmental Protection Agency). 2003e. Attachment 1-4 to Guidance for
Developing Ecological Soil Screening Levels (Eco-SSLs): Review of Background
Concentrations for Metals. OSWER Directive 92857-55. U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, DC. November.
Available at http://www.epa.gov/ecotox/ecossl/SOPs.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2003f. EPACMTP Technical Background
Document. Office of Solid Waste, Washington, DC. Available online at:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-18
-------
Chapter 9.0 References
http://www.epa.gov/osw/nonhaz/industrial/tools/ctntp/index.httn (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 2003g. EPACMTP Parameters/Data Document.
Office of Solid Waste, Washington, DC. Available online at:
http://www.epa.gov/osw/nonhaz/industrial/tools/cmtp/index.htm (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 2003h. Addendum to the EPACMTP Technical
Background Document. Office of Solid Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). 2004. Risk Assessment Guidance for Superfund
Volume I: Human Health Evaluation Manual (Part E, Supplemental Guidance for
Dermal Risk Assessment). Final. U.S. EPA, Office of Superfund Remediation and
Technology Innovation, Washington, DC. Available at
http://www.epa.gov/oswer/riskassessment/ragse/index.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2005a. Partition Coefficients for Metals in
Surface Water, Soil, and Waste. EPA/600R-05/074. U.S. Environmental Protection
Agency, Office of Research and Development. July. Available at
http://www.epa.gov/athens/publications/reports/Ambrose600R05074PartitionCoefficients
^df (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2005b. Human Health Risk Assessment Protocol
for Hazardous Waste Combustion Facilities. EPA 530-R-05-006. U.S. Environmental
Protection Agency, Office of Solid Waste, Washington DC. Available (with supporting
documentation in a self-extracting file) at
http://www.epa.gov/osw/hazard/tsd/td/combust/risk.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2005c. Guidelines for Carcinogen Risk
Assessment. EPA/630/P-03/001B. U.S. EPA, Risk Assessment Forum. March.
Washington, DC. Available at http://www.epa.gov/ttn/atw/cancer_guidelines_final_3-25-
05.pdf (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2006. Provisional Peer Reviewed Toxicity Values
for Aluminum (CASRN 7429-90-5). U.S. Environmental Protection Agency, National
Center for Environmental Assessment, Superfund Technical Support Center, Cincinnati,
OH. October 23. Available at http://hhpprtv.ornl.gov/quickview/pprtv.php (accessed 27
June 2014).
U.S. EPA (Environmental Protection Agency). 2007a. Test Methods for Evaluating Solid Wastes,
Physical/Chemical Methods, SW-846. U.S. Environmental Protection Agency, Office of
Solid Waste and Emergency Response, Washington, DC. Available at
http://www.epa.gov/epaoswer/hazwaste/test/main.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2007b. Metals Framework for Risk Assessment.
EPA 120/R-07/001. U.S. Environmental Protection Agency, Washington, DC. March.
Available at http://www.epa.gov/raf/metalsframework/pdfs/metals-risk-assessment-
final.pdf (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2007d. Method 3051 A: Microwave Assisted Acid
Digestion of Sediments, Sludges, Soils, and Oils, Test Methods for Evaluating Solid
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-19
-------
Chapter 9.0 References
Wastes, Physical/Chemical Methods, SW-846. U.S. Environmental Protection Agency,
Office of Hazardous Waste, Washington, DC. Available at
http://www.epa.gov/osw/hazard/testmethods/sw846/pdfs/3051a.pdf (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 2007e. Ecological Soil Screening Levels for
Polycyclic Aromatic Hydrocarbons (PAHs): Interim Final OSWER 9285.7-78. U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response,
Washington, DC. June. Available at http://www.epa.gov/ecotox/ecossl/ (accessed 27 June
2014).
U.S. EPA (Environmental Protection Agency). 2008a. Child-Specific Exposure Factors
Handbook (Final Report). U.S. Environmental Protection Agency, Office of Research
and Development, National Center for Environmental Assessment, Washington, DC.
October. Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid= 199243
(accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2008b. SuperfundChemical Data Matrix
(SCDM). U.S. Environmental Protection Agency, Office of Emergency Response and
Remediation, Washington, DC. Available at
http://www.epa.gov/superfund/sites/npl/hrsres/tools/scdm.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2008c. Waste and Materials—Flow Benchmark
Sector Report: Beneficial Use of Secondary Materials—Foundry Sand, Final Report.
U.S. Environmental Protection Agency, Washington, DC. February 12.
U.S. EPA (Environmental Protection Agency). 2009. Mid Atlantic Risk Assessment Screening
Table: Generic Tables. U.S. Environmental Protection Agency, Region III. Available at
http://www.epa.gov/reg3hwmd/risk/human/rb-
concentration table/Generic Tables/index.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2011. Exposure Factors Handbook: 2011
Edition. EPA/600/R-090/052F. U.S. Environmental Protection Agency, Office of
Research and Development, Washington, DC. September. Available at
http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=236252 (accessed 27 June 2014)
U.S. EPA (Environmental Protection Agency). 2012a. Integrated Risk Information System
(IRIS). Online database. U.S. Environmental Protection Agency, Office of Research and
Development, National Center for Environmental Assessment, Washington, DC. March.
Available at http://www.epa.gov/iris/index.html (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2012b. Recommendations for Default Value for
Relative Bioavailability of Arsenic in Soil. OSWER 9200.1-113. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC.
Available online at
http://www.epa.gov/superfund/bioavailability/pdfs/Arsenic%20Bioavailability%20POLI
CY%20Memorandum%2012-20-12.pdf. (Accessed 27 June 2014)
U.S. EPA (Environmental Protection Agency). 2013a. Conditional Exclusions from Solid Waste
and Hazardous Waste for Solvent-Contaminated Wipes. Final Rule. 78 FR 46448-46485.
Available online at http://www.gpo.gov/fdsvs/pkg/FR-2013-07-31/pdf/2013-18285.pdf
(Accessed 27 June 2014)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-20
-------
Chapter 9.0 References
U.S. EPA (Environmental Protection Agency). 2013b. Technical Review Workgroup
Recommendations Regarding Gardening and Reducing Exposure to Lead-Contaminated
Soils. OSWER 9200.2-142. U.S. Environmental Protection Agency, Office of Solid
Waste and Emergency Response, Washington, DC. Available online at
http://www.epa.gov/superfund/lead/products/FINAL%20TRW%20Lead%20Committee
%20Gardening%20Recommendations.pdf (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2014. Provisional Peer Reviewed Toxicity
Values for Superfund (PPRTV). Environmental Protection Agency, Office of Superfund
Remediation and Technology Innovation, Washington, DC. Available online at:
http://hhpprtv.ornl.gov/quickview/pprtv_papers.php (accessed 27 June 2014)
Van den Berg, M., L.S. Birnbaum, A. Bosveld, B. Brunstrom, P.M. Cook, M. Feeley, J. Giesy,
A. Hanberg, R. Hasegawa, S.W. Kennedy, T. Kubiak, J.C. Larsen, F.X. Rolaf van
Leeuwen, and A. Lien. 2006. Toxic Equivalency Factors (TEFs) for PCBs, PCDDs,
PCDFs for Humans and Wildlife. Environmental Health Perspectives 106:775-792.
van Zyll de Jong, C.G. 1983. Handbook of Canadian Mammals. National Museums of Canada.
pp.210.
Vasquez-Murrieta, M.S., I. Migueles-Gardufio, O. Franco-Hernandez, B. Govaerts, and L.
Dendooven. 2006. C and N mineralization and microbial biomass in heavy-metal
contaminated soil. European Journal of Soil Biology ¥2:89-98.
von Mersi, W., and F. Schinner. 1991. An improved and accurate method for determining the
dehydrogenase activity of soils with iodonitrotetrazolium chloride. Biology and Fertility
ofSoils 11:2\6-220.
Wang, M.-J., S.P. McGrath, and K.C. Jones. 1995. Chlorobenzenes in field soil with a history of
multiple sewage sludge applications. Environmental Science and Technology 29:356-
362.
Wang, Y., H. Huang, F.S. Cannon, R.C. Voight, S. Komarneni, and J.C. Furness. 2007.
Evaluation of volatile hydrocarbon emission characteristics of carbonaceous green sand
foundries. Environmental Science and Technology 41:2957-2963.
Wang, G., and L.L. Getz. 2007. State-space models for stochastic and seasonal fluctuations of
vole and shrew populations in east-central Illinois. Ecological Modelling. 207: pp 189-
196.
Wang, M., and V. Grimm. 2007. Home range dynamics and population regulation: An
individual-based model of the common shrew Sorex araneus. Ecological Modelling.
(205): 397-409.
Winkler, E.S., and A. A. Bol'shakov. 2000. Characterization of Foundry Sand Waste. Technical
Report No. 31. Chelsea Center for Recycling and Economic Development, University of
Massachusetts, Lowell, MA.
Wolt, J.D. 1994. Soil Solution Chemistry: Applications to Environmental Science. John Wiley
and Sons, New York.
Wood, B.W., C.C. Reilly, and A.P. Nyscepir. 2004. Mouse-ear of Pecan: A Nickel Deficiency.
Horticultural Science 39(6): 123 8-1242.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-21
-------
Chapter 9.0 References
Woolson, E.A., J.H. Axley, and P.C. Kearney. 1973. The chemistry and phytotoxicity of arsenic
in soils. II. Effects of time and phosphorous. Soil Science Society of America Journal
37:254.
Wright, R.J., W.D. Kemper, P.O. Millner, J.F. Power, and R.F. Korcak. 1998. Agricultural Uses
of Municipal, Animal, and Industrial Byproducts. Conservation Research Report No. 44.
U.S. Department of Agriculture, Agricultural Research Service, Washington, DC.
Available at http://www.ars.usda.gov/is/np/agbyproducts/agbyintro.htm (accessed 19
March 2012).
Yang, G., S. Wang, R. Zhou and S. Sun. 1983. Endemic selenium intoxication of humans in
China. American Journal of Clinical Nutrition 3 7:872-881.
Yang, Y.-J., R.S. Dungan, A.M. Ibekwe, C. Valenzuela-Solano, D.M. Crohn, andD.E. Crowley.
2003. Effect of organic mulches on soil microbial communities one year after application.
Biology and Fertility of Soils 35:273-281.
Zanetti, M.C., and S. Fiore. 2002. Foundry processes: The recovery of green moulding sands for
core operations. Resources, Conservation and Recycling 35:243-254.
Zhang, C., L. Huang, T. Luan, J. Jin, and C. Lan. 2006. Structure and function of microbial
communities during the early stages of revegetation of barren soils in the vicinity of a
Pb/Zn smelter. Geoderma 136:555-565.
Zhao, F.J., J.F. Ma, A.A. Meharg, and S.P. McGrath. 2009. Arsenic uptake and metabolism in
plants. New Phytologist 757:777-794.
Zibilske, L.M., W.M. Clapham, and R.V. Rourke. 2000. Multiple applications of paper mill
sludge in an agricultural system: Soil effects. Journal of Environmental Quality 29:1975-
1981.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 9-22
-------
Appendix A: Fundamental Concepts
Appendix A
Fundamental Concepts
Trace Elements in Byproduct-Treated Soils
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix A: Fundamental Concepts
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix A: Fundamental Concepts
Appendix A
Fundamental Concepts Regarding Trace Elements
in Byproduct-Treated Soils
This section of the report was included to help readers better understand the potential for
trace elements in SFS to react in soils and interact in the environment. For decades, researchers
have been working to better understand the potential for soil elements to support growth of
plants and livestock or to become deficient for or phytotoxic for plants or soil organisms. Much
of soil toxicology is based on studies that add soluble metal salts to soils that are cropped
immediately to examine adverse effects. But the added metals quickly react with the adsorbent
surfaces or precipitate in the soil, greatly reducing phytoavailability. Thus realistic assessment of
risk from chronic exposure to trace elements in soils benefits from a deeper understanding of the
metal species found in soils and their longer-term behavior.
A.1 Chemical Reactions in Soils
Soils contain all elements at concentrations dependent on the parent rocks from which the
soil is derived. Elements may also reach soils as components of fertilizers, manures, byproducts,
aerosols, etc., and hence may exist in varied chemical forms. If elements reach soils in elemental
forms, they corrode/oxidize depending on the redox characteristics of the element and the soil.
For example, Ag and Cu are found in metallic form in some reducing soils, but usually oxidize in
aerobic soils over time. Some elements (e.g., metallic Pb, Zn, and Ni) oxidize slowly, while
others oxidize more rapidly. A few persist for long periods depending on the particle size of the
element that reached the soil (smaller particles have higher surface area and react more rapidly),
or redox status of the soil. Flooded peat soils may provide a reducing soil environment that will
allow metallic or metal sulfide particles to persist for long periods.
Another aspect of reactions of trace elements in residuals with soils is the unusually low
reactivity of some metal oxides such as NiO. This compound was emitted by some Ni refineries
and found to persist for decades in aerobic soils (McNear et al., 2007). Studies showed that the
dissolution of NiO is inherently slow, with a half-life of 6.5 years at a pH of 6 (Ludwig and
Casey, 1996).
For a material such as SFS, the trace elements are present as (1) oxidized equilibrium
forms in the input sand and (2) some metallic particles and oxidized forms of the elements used
in producing castings at a foundry. Iron and steel may remain partly in the metallic forms for
some time, but will eventually oxidize and enter soil equilibria.
For the remaining discussion, we will assume that elements in a residual are the ionized
forms in equilibrium with aerobic soils rather than the elemental state which could enter soils
from some sources. The ionized forms are more mobile, and thus potentially more toxic than the
elemental forms, so risk assessment for the ionized forms is appropriate. In this case, the element
will have reacted with redox buffering parts of the soil and with adsorptive or chelation surfaces
of the soil. In a normal aerobic soil, most elements are present as hydrated or complexed cations
or anions in equilibrium, either bound to the soil surfaces or precipitated as minerals (Langmuir
et al., 2004) (e.g., Zn2+, Cu2+, Ni2+, Pb2+, Cd2+, MoO42-, SeO42-, NO3-, SO42; H2PO4-). Many ions
are so easily oxidized that they remain the cation regardless of soil redox conditions: Li+, K+,
Na+, Rb+, Cs+ (alkali cations), Be2+, Mg2+, Ca2+, Sr2+, Ba2+ (alkaline earth cations). Similarly,
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-l
-------
Appendix A: Fundamental Concepts
reducing soils are not reducing enough to alter the form of halogen ions in soils: F", Cl~, Br", I",
although it is possible to reduce iodide to iodine in strongly reducing soils. Most elements in
soils are not transformed to organic compounds with covalent linkage, but those that are
transformed have very changed properties.
Mercury is transformed by soil microbes and some plants into Hg°, HgS, organic matter
bound Hg, CFb-Hg, or even (CFb^Hg. The Hg° is volatile and can be emitted from the soil;
most Hg° emission from soils is induced by light striking the surface soil. The methyl-Hg forms
are volatile and lipophilic, and can bioaccumulate in organisms. But the fraction of soil Hg in the
methylated forms is quite small.
Flooding a soil causes the redox potential to rise as the soil becomes reducing because
little O2 dissolves in water and soil organisms consume the O2. The soil pores become filled with
water or gases formed in the soil under anaerobic conditions. With the reducing environment,
some elements are reduced to chemical forms different than those found in normal aerobic soils.
In particular, arsenate (AsO43~) is reduced to the more soluble and more phytotoxic arsenite
(AsOs3"). This is important because flooded rice is the crop plant found to be most sensitive to
excessive soil As; the higher concentration of AsCb3" in flooded soils compared to AsCV" in
aerobic soil allows much easier plant uptake and injury from soil As. Uptake of some other
elements may also be increased in reducing soils, but without an increased phytotoxicity as
demonstrated for As.
Soil Mn is the cation most altered by soil reduction. Mn is usually present in aerobic soils
as MnO2 and not available to plant roots except where roots reduce the MnO2 to Mn2+. In
flooded soils, Mn2+ can be greatly increased; Mn2+ is not strongly adsorbed by soils and can
accumulate to high levels and become phytotoxic to sensitive plant species. Draining the soil
allows rapid oxidation of the Mn2+ to MnCh if the soil pH is higher than 5.5 (the oxidation is
catalyzed by soil microbes).
A. 1.1 Reactions Over Time
Time is an important variable when assessing soil chemistry and risk from trace element
exposure. Most microelements react more strongly with soil over time (Logan and Chaney,
1983; Basta et al., 2005). This is shown by how the plant availability or extractability of an
element changes with time after a soluble salt of the element is added to soils. There are several
kinds of reactions: hydrolysis (or precipitation), chelation by organic matter, chemisorption on
Fe and Mn oxide surfaces, and formation of new solid phases. These reactions are nicely
illustrated by the reactions of Ni with mineral and organic soils. When soluble Ni compounds are
mixed with soils, the Ni-hydrated cations rapidly form adsorbed forms on Fe and Mn oxides and
chelated forms with soil organic matter (SOM). Other soil minerals then dissolve, and Ni reacts
to form new soil minerals such as Ni silicates and Ni-Al layered double hydroxides (LDHs).1
The overall process is illustrated by Singh and Jeng (1993), who tested Ni reactions with soil
over time when they grew ryegrass in a greenhouse annually for 3 years in large pots using Ni-
salt applications to an acidic sandy soil. Phytotoxicity was not observed at the highest soil nickel
application (50 mg Ni kg"1), even though shoot nickel reached nearly 50 mg kg"1 dry shoots in
1 Ni-Al LDHs were discovered only recently when extended X-ray absorption fine structure spectroscopy (EXAFS)
was applied to the reactions of elements with soils (Sparks, 2003).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-2
-------
Appendix A: Fundamental Concepts
the first year. In subsequent years, oat shoot nickel declined from 48 mg kg"1 in Year 1 to 18 mg
kg"1 in Year 2 to 8 mg kg"1 in Year 3. Such a decline is expected because the freshly added Ni
requires time to equilibrate with soil adsorption sites and form more stable Ni silicate and LDH
compounds in soil.
Using physical-chemical methods not available earlier (e.g., EXAFS), research conducted
during the past decade has provided additional information on how water-soluble Ni2+ reacts
with soils and with specific minerals. During such reaction/speciation tests, the ionic strength of
the soil solution influences the strength of adsorption; high ionic strength inhibits adsorption and
could thus promote the formation of inorganic compounds in soil. In one set of experiments,
Scheidegger and colleagues (1996a, 1996b, 1997, and 1998), Scheidegger and Sparks (1996),
Ford et al. (1999), and Sheinost et al. (1999) added Ni2+ to clays and minerals and used
Synchrotron radiation after varied amounts of time to examine the formation of LDHs (e.g.,
nickel aluminum hydroxide) and Ni silicates. The higher ionic strength of these tests (0.1 M
KNCb) led to the formation of LDHs if the clays and minerals released Al, and to the formation
of Ni-silicate crystalline materials if the clays and minerals released silicate ions. However, when
Elzinga and Sparks (2001) used a lower ionic strength, the relative proportion of adsorption (or
chemisorption; specific adsorption) increased, and the formation of surface-induced precipitates
decreased.
This work demonstrated important aspects of the reactions of Ni with soils in that slow
reactions over time converted added Ni2+ to forms of Ni that were much less soluble or
phytoavailable. This is further illustrated by Scheckel and Sparks (2001), who examined mineral
samples that had been reacting with Ni2+ for 1 hour to 2 years. The longer the reaction period, the
lower the water solubility or acid extractability of the adsorbed or precipitated insoluble Ni
species. For example, after Ni2+ equilibrated with several minerals, extractability was as high as
98% for the 1-hour equilibrated materials and as low as 0 for the 2-year equilibrated materials.
The increase in stability of the Ni surface precipitates with increasing residence time in their
studies was attributed to three aging mechanisms: (1) Al-for-Ni substitution in the octahedral
sheets of the brucite-like hydroxide layers, (2) Si-for-nitrate exchange in the inter-layers of the
precipitates, and (3) Ostwald ripening of the precipitate phases. We believe these findings are
complementary with the work of Bruemmer and colleagues (1988), who found adsorption to
strengthen with time of reaction, following a diffusion-type process. The comparatively insoluble
chemical forms of Ni formed during the prolonged reactions of Ni2+ with soil were simply more
ordered Ni silicates and LDHs, not Ni2+ adsorbed within nanopores in the surfaces of Fe and Mn
oxides.
Ni, Co, and Zn have also been found to form LDH compounds over time after addition to
soils or contamination in the field (Ford and Sparks, 2000; Voegelin et al., 2002; Voegelin and
Kretzschmar, 2005). At low soil pH, Zn is much less likely to generate LDH forms than Ni, but
at neutral pH, the Zn-LDH formed and must contribute to the ability of limestone to remediate
Zn-contaminated soils. Voegelin and Kretzschmar (2005) tested formation of mixed LDH with
both Zn and Ni and found that the mixed LDH were not as stable (to pH 3 extraction) as Ni-LDH
without the presence of high levels of Zn. In any regard, the formation of LDH metal compounds
in soils helps explain the very strong difference in response of plants to added soluble metal salts
compared to pre-equilibrated metals from different sources. This "metal reacts more strongly
with time" response was evident in a study even 30 years ago on the availability of fertilizer Zn
added to soils. Based on this study, added Zn becomes less plant available over time and re-
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-3
-------
Appendix A: Fundamental Concepts
fertilization may be required for soils with the highest ability to inactivate added soluble Zn
fertilizers such as ZnSO4 (Boawn, 1974 and 1976).
A.1.2 Sorption in Aerobic Soils
Sorption is a chemical process that buffers the partitioning of trace elements between
solid and liquid phases in soils and byproducts. Fe, Al, and Mn oxide soil minerals are important
sinks for trace elements in soil and byproduct-amended soil (Essington and Mattigod, 1991;
Lombi et al., 2002; Hettiarachchi et al., 2003). Trace element sorption by the oxide surface is a
pH-dependent process; protons compete with cations for sorption. The adsorption of metal
cations by the oxide surfaces increases to nearly 100% with increasing pH (McKenzie, 1980). In
contrast, oxyanion adsorption generally decreases with increasing pH. Differences between
adsorption and desorption isotherms typically reveal significant hysteresis (Hettiarachchi et al.,
2003), providing evidence that this process is not simply a competitive ion-exchange reaction
between metal ions and protons or hydroxyls. Some adsorbed metals are strongly bonded and not
readily desorbed from these oxide surfaces. Some research suggests that the increasingly strong
sorption and lower phytoavailability results from the trace elements moving to nano-sized pores
in Fe and Mn oxides (Bruemmer et al., 1988).
Trace element sorption by oxides shows Fe and Mn oxides have a much greater
adsorption capacity compared to Al oxides and clay minerals (Brown and Parks, 2001).
Molecular-scale X-ray spectroscopic studies show that the strong bonding of Cu, Co, Cr, Mn, Ni,
Cd, Pb, and Zn to these oxide surfaces is due to formation of inner-sphere surface complexes and
formation of metal hydroxide precipitate phases (Brown and Parks, 2001; Sparks, 2003). New
solids found after trace element ion reactions with soil materials, including metal silicates and
mixed double hydroxides with Al, can substantially reduce element solubility and availability
(Scheckel and Sparks, 2001). Sorption by Fe and Mn metal oxides is a major mechanism for
removal of trace element cations (i.e., Cd, Cr, Cu, Pb, Hg, Ni, Zn) and trace element oxyanions
(i.e., AsO43; AsO33', SeO42; SeO32; MoO42', WO42', VO42', CrO42') from aqueous solution (e.g.,
soil solution) (Stumm, 1992; Sparks, 2003).
Trace element cations also sorb to SOM and other forms of humified natural organic
matter (NOM). Strong adsorption to NOM in byproducts by formation of metal chelates reduces
the solubility of several trace elements in soil (Adriano, 2001). Sorption of trace elements to
SOM or NOM increases with pH because protons compete less well at increasing pH. At lower
pH, trace element sorption by NOM is reduced less than is trace element sorption to Fe and Mn
oxides.
Trace element cations form sparingly soluble precipitates with phosphate, sulfides, and
other anions (Lindsay, 2001; Langmuir et al., 2004). Trace element precipitation is highly pH
dependent and increases with pH for many trace element cations. AsO43" and other trace element
oxyanions can form insoluble precipitates with multivalent cations, including Fe, Al, and Ca.
The resulting trace element minerals (i.e., precipitates) may control the amount of trace element
in solution (i.e., availability and mobility).
Byproducts typically contain components (NOM; Fe, Mn, and Al oxides; and anions such
as phosphate and silicate) that can adsorb or precipitate trace elements. Many types of
byproducts (e.g., biosolids, manure, municipal solid waste compost, coal combustion residuals)
with a wide range of properties have been applied to agricultural land and have modified the
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-4
-------
Appendix A: Fundamental Concepts
adsorption properties of soils (Power and Dick, 2000; Basta et al., 2005). Trace element
phytoavailability is affected by the sorption capacity and properties (e.g., pH, salinity) of both
the soil and the byproduct. Sandy soils with low Fe and Mn oxide content and low SOM have
low sorption capacities and will have greater trace element phyto- and bioavailabilities than
loamy or clayey soils with greater amounts of sorbents (i.e., reactive oxides, SOM), provided the
soils have similar pH values. Similarly, byproducts with low Fe and Mn oxide content and low
NOM have low trace element sorption capacities and higher potential element availabilities as
compared to byproducts with high Fe and Mn oxide and NOM.
Byproduct-soil mixtures would have intermediate sorption properties between that of the
soil and byproduct and, perhaps, intermediate phytoavailabilities if other properties (e.g., pH)
were similar. As the loading rate of the byproduct increases, the byproduct-soil mixture will be
increasingly affected by the sorption properties of the byproduct. Some byproducts have greater
amounts of these sorbents than soil and can increase the sorption capacity of soils for trace
elements. Added to soil in sufficient amounts, a high-sorbent byproduct can dominate the trace
element binding chemistry of the soil-byproduct mixture (Basta et al., 2005; Kukier et al., 2010).
This phenomenon is illustrated in Figure A-l, which shows the results of a Cd
phytoavailability bioassay using Romaine lettuce grown on a Christiana fine sandy loam soil
with no amendment (control), with 224 t
ha"1 of a digested biosolid, or with 672 t
ha"1 of a biosolids compost applied over 25
years before the test was conducted.
During this test, all soils were adjusted to a
pH of 6.5, and five rates of soluble Cd
were applied. Lettuce uptake of Cd was
linear, with increasing added soluble Cd,
but the slope of this uptake was reduced
up to 90% by the historic amendment with
high-Fe biosolids or biosolids compost.
These two amendments were rich in Fe
and phosphate, and it is believed that an
Fe-P-NOM complex provides the
persistent high Cd binding. It seems likely
that inclusion of Fe oxide in organic P-rich
byproducts can readily reduce trace
element cation phytoavailability
(Hettiarchchi et al., 2003; Basta et al.,
2005; Kukier etal., 2010).
80
E 50-
O 40-
0)
o
3 30-
20-
10-
Hayden Farm Plots
Biosolids
Experiment
Beltsville, MD
Control
224 t/ha Heat-Treated
13.4 ppm Cd
10
12
14
Soil Total Cd, mg/kg DW
Figure A-l. The effect of historic biosolids
applications on the phytoavailability of applied
Cd salt to Romaine lettuce.
A.2 Soil-Plant Barrier Limits Risks from Trace Elements in Soils or Soil
Amendments
The potential risk that diverse trace elements in soils pose to the feed- and food-chain has
been intensively examined during the past 35 years. One purpose of the investigation has been to
understand the risk from application of biosolids, livestock manure, and other trace element
contamination sources to soil.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
A-5
-------
Appendix A: Fundamental Concepts
During this period, the "Soil-Plant Barrier" concept was introduced to communicate how
element addition rate and chemistry, soil chemistry, and plant chemistry affected the risk to
plants and animals from elements in soil amendments (Chaney, 1980 and 1983). This concept is
based on long experience in veterinary toxicology and agronomy. Reactions and processes
related to the Soil-Plant Barrier include the following:
1. Solid adsorbent sources (e.g., Fe, Al, and Mn oxyhydroxides and organic matter) in soil
amendments may have adsorptive surfaces that influence soil chemistry.
2. Adsorption or precipitation of elements in soils or in roots limits uptake-translocation of
most elements to shoots.
3. The phytotoxicity of Zn, Cu, Ni, Mn, As, B, Al, F, and other elements limits
concentrations of these elements in plant shoots to levels chronically tolerated by
livestock and humans.
4. Food-chain transfer of an element may not constitute a risk, but the direct ingestion of
highly contaminated soil may cause risk from Pb, As, F, and some other elements if the
soil is poorly managed.
5. The Soil-Plant Barrier does not restrict transfers of soil Se, Mo, and Co well enough to
protect all animals from elements (e.g., Se, Mo) or ruminant livestock (e.g., Co). In
addition, the soil-plant barrier does not restrict transfer of Cd in rice and, as a result,
subsistence rice consumers may be at risk in situations of moderate Cd contamination
because of the physiology of paddy rice and for garden crops where Cd contamination
occurs without the usual 100-fold greater Zn contamination.
A summary of trace element tolerances by plants and livestock is presented in Table A-l.
Please note that the National Research Council (NRC; 1980) committee that identified the
maximum levels of trace elements in feeds tolerated by domestic livestock based its conclusions
on data from toxicological-type feeding studies in which soluble trace element salts had been
mixed with practical or purified diets to examine animal response to the dietary elements. If soil
or some soil amendment is incorporated into diets, element solubility and bioavailability very
likely are much lower than in the tests relied on by NRC (1980). For example, Chaney and Ryan
(1993) noted that animal body Pb burden from ingesting the soil does not increase until the soil
Pb concentration exceeds approximately 300 mg Pb kg"1. Other elements, in equilibrium with
poorly soluble minerals or strongly adsorbed in ingested soils, are often much less bioavailable
than they would be if they were added to the diet as soluble salts.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-6
-------
Appendix A: Fundamental Concepts
Table A-l. Maximum Tolerable Levels of Dietary Minerals for Domestic Livestock
in Comparison With Levels in Forages
Element
As
(inorganic)
B
Cdc
Cr3*
Co
Cu
F
Fe
Mn
Mo
Ni
Pbc
Se
V
Zn
"Soil-Plant
Barrier"
Yes
Yes
Fails
Yes
Fails
Likely
Yes
Yes
Likely
Fails
Likely
Yes
Fails
Yes
Likely
Level in Plant Foliage"
(mg kg -1 Dry Foliage)
Normal
0.01-1.0
7-7.5
0.1-1
0.1-1
0.01-0.3
3-20
1-5
30-300
15-150
0.1-3
0.1-5
2-5
0.1-2
0.1-1
15-150
Phytotoxic
3-10
75
5-700
20
25 - 100
25-40
-
-
400 - 2,000
100
50 - 100
-
100
10
500 - 1,500
Maximum Levels Chronically Tolerated1"
(mg kg -1 Dry Diet)
Cattle
50
150
0.5
(3,000)
10
100
40
1,000
1,000
10
50
30
(2)
50
500
Sheep
50
(150)
0.5
(3,000)
10
25
60
500
1,000
10
(50)
30
(2)
50
300
Swine
50
(150)
0.5
(3,000)
10
250
150
3,000
400
20
(100)
30
2
(10)
1,000
Chicken
50
(150)
0.5
3,000
10
300
200
1,000
2,000
100
(300)
30
2
10
1,000
a Based on literature summarized in Chaney (1983).
b Based on NRC, 1980. Continuous long-term feeding of minerals above the maximum tolerable levels may cause
adverse effects. NRC estimated the levels in parentheses by extrapolating between animal species when data were
not available for an animal.
c NRC based the maximum levels chronically tolerated of Cd or Pb in liver, kidney, and bone in foods for humans
rather than simple tolerance by the animals. Because of the simultaneous presence of Zn, Cd in animal tissues is
less bioavailable than Cd salts added to diets and the maximum levels chronically tolerated should have been
higher than listed.
The chemistry of elements in soils is affected by the presence of ions, which can cause
precipitation of the element, organic matter, and sesquioxides, which, in turn, can adsorb
elements; redox changes, which affect the chemical species of the elements present; and similar
factors. Soils are usually in a relatively restricted pH range of 5.5 to 8 for high-producing soils
and as wide ranging as 5 to 9 in nearly all soils in the general environment. Some soil
amendments have a pH greater than 8, but soils thus amended absorb atmospheric CO2, which
returns the soil pH to no higher than calcareous soil levels.
Many elements (e.g., Ti, Fe3+, Pb, Hg, Al, Cr3+, Ag, Au, Sn, Zr, and rare earth elements
[e.g., Ce] that serve as a label for soil contamination of plants and diets) are so insoluble in
aerobic soils between a pH of 5.5 and 8 that they do not cause risk to animals even when soils
with relatively high concentrations are ingested by livestock. This is especially well illustrated by
Cr uptake by plants growing on high Cr-mineralized serpentine soils (Gary and Kubota, 1990);
soil contained more than 10,000 mg Cr kg"1, but all Cr measured in plant samples could be
explained by soil particle contamination of the plant sample (based on Ti and other element
concentrations). Cr was actually added to diets as a non-absorbed index cation to follow
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
A-7
-------
Appendix A: Fundamental Concepts
absorption of other nutrients along the gastrointestinal tract or the timing of movement (Irwin
and Crampton, 1951; Raleigh et al., 1980). Direct soil ingestion could provide exposure and must
be considered separately from plant uptake.
When present at very high concentrations in soil, some elements that are not accumulated
by plants to levels of concern for livestock or wildlife (e.g., F, Pb, As, and Zn) may still pose a
risk because of absorption from ingested soil. These same elements may comprise risk to
earthworm-consuming wildlife (e.g., shrews, moles, badgers, woodcocks) from soil ingestion,
but not plant uptake. The earthworm is consumed without depuration of internal soil, exposing
the predator to high levels of dietary soil—perhaps 35% of dry weight (Beyer and Stafford,
1993). This high soil ingestion rate makes earthworm-consuming wildlife among the most highly
soil-exposed animals, and is an important consideration in risk assessment of soil contaminants
such as Cd, which can also bioaccumulate in earthworm tissues.
Freshly applied trace element salts are not in equilibrium with soil and have a greater
bioavailability than they would exhibit upon equilibrating with the soil over time. The
phytoavailability and bioavailability of trace elements may also be reduced if the metals are
adsorbed, chelated, or precipitated before ingestion by children or grazing livestock.
A.3 Toxicity or Prevention of Toxicity by Interaction Among Trace
Elements
The toxicity to animals of biosolids- or manure-applied Cu or Zn is an example of the
interaction between elements impacting element toxicity. Cu deficiency-stressed animals are
more sensitive to dietary Zn than animals fed Cu-adequate diets, but biosolids-fertilized crops are
not low in Cu, so ordinarily Zn phytotoxicity protects all livestock against excessive Zn in
forages, including the most sensitive ruminants. Similarly, Cu toxicity to sensitive ruminant
animals is substantially reduced by increased dietary levels of Zn, Cd, Fe, Mo, and SO42" or
sorbents such as SOM. In contrast with the predicted increase in liver Cu concentrations and
toxicity from Cu in ingested swine manure or biosolids, reduced liver Cu concentrations have
been found in cattle or sheep unless the ingested biosolids exceeded approximately 1,000 mg Cu
kg1.
Interactions can also limit toxicity and risk. For example, Cd bioavailability is strongly
affected by the presence of the normal geogenic levels of Zn (100- to 200-fold Cd level); Zn
inhibits binding of Cd by soil, but also inhibits Cd uptake by roots, Cd transport to shoots and Cd
transport to storage tissues. Furthermore, Zn in foods significantly reduces Cd absorption by
animals (Chaney et al., 2004). Increased Zn in spinach and lettuce reduced absorption of Cd in
these leafy vegetables by Japanese quail (McKenna et al., 1992), and increased Zn in forage diets
strongly inhibited Cd absorption and reduced liver and kidney Cd concentration in cattle
(Stuczynski et al., 2007).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-8
-------
Appendix A: Fundamental Concepts
A. 4 References
Adriano, D.C. 2001. Trace Elements in the Terrestrial Environments: Biogeochemistry,
Bioavailability, and Risks of Heavy Metals. 2nd ed. New York: Springer-Verlag.
Basta, N.T., J.A. Ryan, and R.L. Chaney. 2005. Trace element chemistry in residual-treated soil:
Key concepts and metal bioavailability. Journal of Environmental Quality 34:49-63.
Beyer, W.N., and C. Stafford. 1993. Survey and evaluation of contaminants in earthworms and
in soils derived from dredged material at confined disposal facilities in the Great Lakes
region. Environmental Monitoring and Assessment 24:151-165.
Boawn, L.C. 1974. Residual availability of fertilizer zinc. Soil Science Society of America
Journal 38:800-803.
Boawn, L.C. 1976. Sequel to "residual availability of fertilizer zinc." Soil Science Society of
America Journal 40:467-468.
Brown, G.E., Jr., and G.A. Parks. 2001. Sorption of trace elements on mineral surfaces: Modern
perspectives from spectroscopic studies, and comments on sorption in the marine
environment. International Geology Review 43:963-1073.
Bruemmer, G.W., J. Germ, and K.G. Tiller. 1988. Reaction kinetics of the adsorption and
desorption of nickel, zinc, and cadmium by goethite. I. Adsorption and diffusion of
metals. European Journal of Soil Science 39:37-52.
Gary, E.E., and J. Kubota. 1990. Chromium concentration in plants: Effects of soil chromium
concentration and tissue contamination by soil. Journal of Agricultural and Food
Chemistry 38:108-114.
Chaney, R.L. 1980. Health risks associated with toxic metals in municipal sludge. Pp. 59-83 in
Sludge Health Risks of Land Application. Edited by G. Bitton, B.L. Damron, G.T.
Edds, and J.M. Davidson. Ann Arbor, MI: Ann Arbor Science Publishers.
Chaney, R.L. 1983. Potential effects of waste constituents on the food chain. Pp 152-240 in
Land Treatment of Hazardous Wastes. Edited by J.F. Parr, P.B. Marsh, and J.M. Kla.
Park Ridge, NJ: Noyes Data Corp.
Chaney, R.L., and J.A. Ryan. 1993. Heavy metals and toxic organic pollutants in MSW
composts: research results on phytoavailability, bioavailability, fate, etc. Pp. 451-505 in
Science and Engineering of Composting: Design, Environmental, Microbiological and
Utilization Aspects. Edited by H.A.J. Hoitink and H.M. Keener. Ohio State Univ.,
Columbus, OH.
Chaney, R.L., P.G. Reeves, J.A. Ryan, R.W. Simmons, R.M. Welch, and J.S. Angle. 2004. An
improved understanding of soil Cd risk to humans and low cost methods to remediate soil
Cd risks. BioMetals 17:549-553.
Elzinga, E.J., and D.L. Sparks. 2001. Reaction condition effects on nickel sorption mechanisms
in illite-water suspensions. Soil Science Society of America Journal 65:94-101.
Essington, M.E., and S.V. Mattigod. 1991. Trace element solid-phase associations in sewage
sludge and sludge-amended soil. Soil Science Society of America Journal 55:350-356.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-9
-------
Appendix A: Fundamental Concepts
Ford, R.G., and D.L. Sparks. 2000. The nature of Zn precipitates formed in the presence of
pyrophyllite. Environmental Science and Technology 34:2479-2483.
Ford, R.G., A.C. Scheinost, K.G. Scheckel, and D.L. Sparks. 1999. The link between clay
mineral weathering and the stabilization of Ni surface precipitates. Environmental
Science and Technology 33:3140-3144.
Hettiarachchi, G.M., J.A. Ryan, R.L. Chaney, and C.M. LaFleur. 2003. Sorption and desorption
of cadmium by different fractions of biosolids-am ended soils. Journal of Environmental
Quality 32:1684-1693
Irwin, M.I., and E.W. Crampton. 1951. The use of chromic oxide as an index material in
digestion trials with human subjects. Journal of Nutrition 43:77-85.
Kukier, U., R.L. Chaney, J.A. Ryan, W.L. Daniels, R.H. Dowdy and T.C. Granato. 2010.
Phytoavailability of cadmium in long-term biosolids amended soils. Journal of
Environmental Quality 39:519-530
Langmuir, D.L., P. Chrostrowski, R.L. Chaney and B. Vigneault. 2004. Issue Paper on
Environmental Chemistry of Metals. US-EPARisk Assessment Forum: Papers
Addressing Scientific Issues in the Risk Assessment of Metals. U.S. Environmental
Protection Agency, National Center for Environmental Assessment. Available at
http://www.epa.gov/raf/publications/pdfs/ENVCHEMFINAL81904CORRO1 -25-05 .PDF
(accessed 19 March 2012).
Lindsay, W.L. 2001. Chemical Equilibria in Soils. Caldwell, NJ: The Blackburn Press.
Logan, T.J., and R.L. Chaney. 1983. Utilization of municipal wastewater and sludge on land-
metals. Pp. 235-295 in Utilization of Municipal Wastewater and Sludge on Land. Edited
by T.L. Gleason, III, I.E. Smith, Jr. I.K. Iskandar, and L.E. Page. Riverside, CA:
University of California.
Lombi, E., F. Zhao, G. Zhang, B. Sun, W. Fitz, H. Zhang, and S.P. McGrath. 2002. In situ
fixation of metals in soils using bauxite residue: Chemical assessment. Environmental
Pollution 118:435-443.
Ludwig, C., and W.H. Casey. 1996. On the mechanisms of dissolution of bunsenite [NiO(s)] and
other simple oxide minerals. Journal of Colloid and Interface Science 178:176-185.
McKenna, I.M., R.L. Chaney, S.H. Tao, R.M. Leach, Jr., andF.M. Williams. 1992. Interactions
of plant zinc and plant species on the bioavailability of plant cadmium to Japanese quail
fed lettuce and spinach. Environmental Research 57:73-87.
McKenzie, R.M. 1980. The adsorption of lead and other heavy metals on oxides of manganese
and iron. Australian Journal of Soil Research 18:61-73.
McNear, D.H., Jr., R.L. Chaney, and D.L. Sparks. 2007. The effects of soil type and chemical
treatment on nickel speciation in refinery enriched soils: A multi-technique investigation.
Geochimica et Cosmochimica Acta 71:2190-2208.
National Research Council. 1980. Mineral Tolerance of Domestic Animals. Washington, DC:
National Academy of Sciences.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-10
-------
Appendix A: Fundamental Concepts
Power, J.F., and W.A. Dick. 2000. Land Application of Agricultural, Industrial, and Municipal
By-Products. SSSA Book Series No. 6. Soil Science Society of America, Madison, WI.
Raleigh, R.J., RJ. Kartchner, and L.R. Rittenhouse. 1980. Chromic oxide in range nutrition
studies. Oregon State University's Agricultural Experiment Station Bulletin 641:1-41.
Scheckel, K.G., and D.L. Sparks. 2001. Dissolution kinetics of nickel surface precipitates on clay
mineral and oxide surfaces. Soil Science Society of America Journal 65:685-694.
Scheidegger, A.M., and D.L. Sparks. 1996. Kinetics of the formation and the dissolution of
nickel surface precipitates on pyrophyllite. Chemical Geology 132:157-164.
Scheidegger, A.M., D.G. Strawn, G.M. Lamble, and D.L. Sparks. 1998. The kinetics of mixed
Ni-Al hydroxide formation on clay and aluminum oxide minerals: A time-resolved XAFS
study. Geochimica et Cosmochimica Acta 62:2233-2245.
Scheidegger, A.M., G.M. Lamble, and D.L. Sparks. 1997. Spectroscopic evidence for the
formation of mixed-cation hydroxide phases upon metal sorption on clays and aluminum
oxides. Journal of Colloid and Interface Science 186:118-128.
Scheidegger, A.M., M. Fendorf, and D.L. Sparks. 1996a. Mechanisms of nickel sorption on
pyrophyllite: Macroscopic and microscopic approaches. Soil Science Society of America
Journal 60:1763-1772.
Scheidegger, A.M., G.M. Lamble, and D.L. Sparks. 1996b. Investigation ofNi sorption on
pyrophyllite: An XAFS study. Environmental Science and Technology 30:548-554.
Scheinost, A.C., R.G. Ford, and D.L. Sparks. 1999. The role of Al in the formation of secondary
Ni precipitates on pyrophyllite, gibbsite, talc, and amorphous silica: A DRS study.
Geochimica et Cosmochimica Acta 63:3193-3203.
Singh, B.R., and A.S. Jeng. 1993. Uptake of zinc, cadmium, mercury, lead, chromium, and
nickel by ryegrass grown in a sandy soil. Norwegian Journal of Agricultural Science
7:147-157.
Sparks, D.L. 2003. Environmental Soil Chemistry. 2nd ed. San Diego, CA: Academic Press.
Stuczynski, T.I., G. Siebielec, W.L. Daniels, G.C. McCarty, and R.L. Chaney. 2007. Biological
aspects of metal waste reclamation with sewage sludge. Journal of Environmental
Quality 36:1154-1162.
Stumm, W. 1992. Chemistry of the Solid-Water Interface: Processes at the Mineral-Water and
Particle-Water Interface in Natural Systems. New York: John Wiley & Sons.
Voegelin, A., and R. Kretzschmar. 2005. Formation and dissolution of single and mixed Zn and
Ni precipitates in soil: Evidence from column experiments and extended x-ray absorption
fine structure spectroscopy. Environmental Science and Technology 39:5311-5318.
Voegelin, A., A.C. Scheinost, K. Biihlmann, K. Barmettler, and R. Kretzschmar. 2002. Slow
formation and dissolution of Zn precipitates in soil: A combined column-transport and
XAFS study. Environmental Science and Technology 36:3749-3754.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-l 1
-------
Appendix A: Fundamental Concepts
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications A-12
-------
Appendix B: Spent Foundry Sand Characterization Data
Appendix B
Spent Foundry Sand Characterization Data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix B: Spent Foundry Sand Characterization Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-l. Element-Specific Concentrations in Spent Foundry Molding Sands Collected June 2005 (Concentrations in mg kg"1)
FIN"
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Ag
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
Al
4,379
2,137
1,732
1,983
1,896
996
<311
805
522
532
3,543
2,153
1,961
2,184
2,304
1,013
3,042
1,100
1,998
10,048
<311
3,944
1,980
1,247
2,164
1,906
4,799
1,849
<311
1,788
321
1,148
<311
1,325
1,852
2,406
1,681
<311
1,595
1,630
<311
954
1,813
As
1.4
0.11
0.46
0.23
0.64
0.16
0.04
0.77
0.38
0.83
2.4
0.36
1.1
1.2
2.0
0.85
2.0
1.3
1.5
0.84
0.57
4.8
1.2
1.2
0.72
1.5
3.0
0.95
0.38
0.79
0.41
0.82
0.13
0.74
1.2
1.9
1.3
0.13
0.52
0.87
0.12
0.97
0.58
B
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
Ba
95.0
<8.7
28.1
<8.7
12.2
126.2
<8.7
<8.7
<8.7
<8.7
21.6
<8.7
16.4
11.2
15.5
10.0
19.8
12.3
19.2
14.7
<8.7
24.5
19.7
<8.7
18.8
19.6
25.4
22.9
75.7
27.8
<8.7
13.6
<8.7
12.1
20.8
17.2
14.4
27.4
151
14.9
<8.7
<8.7
37.2
Be
<1.2
3.1
<1.2
<1.2
<1.2
1.4
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
1.6
<1.2
<1.2
2.2
<1.2
<1.2
<1.2
<1.2
1.92
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
Cd
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
Co
<0.84
95.3
<0.84
<0.84
<0.84
41.4
<0.84
<0.84
<0.84
<0.84
0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
1.20
<0.84
<0.84
1.7
<0.84
<0.84
<0.84
<0.84
1.8
<0.84
1.8
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
Cr
9.4
57.3
2.1
4.3
5.7
149
4.3
7.1
1.4
20.2
3.3
2.3
1.5
2.3
2.8
2.8
3.1
2.4
38.0
4.2
5.9
40.4
2.0
6.1
2.5
2.3
50.0
4.2
22.6
1.7
8.7
2.0
<1.0
2.1
<1.0
2.1
2.4
1.6
8.3
1.5
<1.0
1.6
6.9
Cu
97.6
27.0
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
34.24
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
37.7
<23.1
<23.1
75.6
<23.1
<23.1
<23.1
<23.1
61.8
<23.1
<23.1
<23.1
<23.1
<23.1
70.0
3318
<23.1
<23.1
<23.1
<23.1
60.3
<23.1
<23.1
<23.1
<23.1
Fe
8,914
44,320
1,535
2,020
6,354
20,410
549
5,709
1,206
7,630
3,292
606
1,980
2,750
2,643
2,206
3,237
2,029
20,210
2,575
6,364
29,680
1,987
4,678
2,260
2,276
29,550
2,613
3,394
2,420
2,593
2,744
<352
1,556
3,333
3,625
3,021
740
4,004
1,781
1,969
2,169
3,877
Mg
1,535
51,574
1892
<720
<720
46,366
<720
<720
<720
<720
<720
<720
<720
1,389
810
<720
915
<720
<720
1054
<720
1,080
<720
<720
<720
<720
1,656
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
797
<720
<720
<720
<720
Mn
145
671
65.0
<45.0
137
509
<45.0
73.6
<45.0
60.6
69.7
<45.0
<45.0
<45.0
56.7
<45.0
98.3
<45.0
206
76.6
<45.0
595
<45.0
45.3
<45.0
<45.0
499
<45.0
<45.0
<45.0
<45.0
<45.0
<45.0
<45.0
<45.0
<45.0
63.5
<45.0
117
<45.0
<45.0
<45.0
74.9
Mo
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
9.6
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
4.7
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
Ni
6.5
2,328
2.4
3.4
5.3
1,022
2.2
3.5
<1.2
6.1
3.6
<1.2
2.8
2.4
2.5
1.7
3.2
1.8
13.2
5.1
2.9
20.6
1.9
3.1
2.0
2.2
29.8
2.4
36.0
1.9
12.3
1.6
<1.2
8.9
1.7
2.3
2.5
14.7
107
1.4
2.6
1.5
8.6
Pb
12.7
<7 7
<7 7
<7.7
<7.7
<7.7
<7.7
<7.7
<7.7
<7.7
<7 7
<7 7
<7.7
<7.7
<7.7
<7.7
<7.7
<7.7
<7 7
<7 7
<7.7
25.7
<7.7
<7.7
<7.7
<7. 7
8.5
<7 7
<7 7
<7.7
<7.7
<7.7
<7.7
19.0
<7. 7
<7. 7
<7 7
<7 7
<7.7
<7.7
<7.7
<7.7
<7. 7
Sb
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
V
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
9.1
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
Zn
54.8
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
33.7
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
179
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
44.2
1640
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
" FIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-l
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-2. Element-Specific Concentrations in Spent Foundry Molding Sands Collected September 2005 (Concentrations in mg kg"1)
FIN3
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
Ag
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
Al
3,496
1,740
1,304
1,482
353
<311
655
512
520
2,114
2,264
1,865
2,206
1,143
3,173
1,044
3,574
6,940
<311
3,267
2,705
2,150
2,103
1,865
2,431
2,500
Oil
2,120
1,213
<311
513
2,072
2,529
1,811
1,823
<311
1,162
1,856
As
2.4
2.0
0.42
1.8
0.18
1.4
0.83
2.0
2.1
0.64
1.9
2.1
1.5
1.5
2.2
3.4
1.9
0.66
5.1
2.7
2.4
1.2
2.0
2.2
2.1
0.39
1.7
1.5
0.82
0.78
2.6
3.0
1.0
0.67
0.13
1.1
1.4
B
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
Ba
38.2
23.7
<8.7
9.6
39.7
<8.7
<8.7
<8.7
<8.7
13.1
<8.7
15.1
<8.7
28.8
19.2
10.1
27.1
19.0
<8.7
20.4
29.3
12.0
18.7
19.6
15.1
31.2
68.4
30.0
14.5
<8.7
<8.7
25.9
20.9
13.1
15.8
24.4
10.0
72.5
Be
1.3
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
3.5
<1.2
<1.2
1.6
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
Cd
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
Co
1.1
<0.84
<0.84
<0.84
9.10
<0.84
<0.84
<0.84
1.07
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
2.9
<0.84
<0.84
1.26
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
Cr
21.4
5.1
4.0
5.6
25.2
7.9
9.0
1.3
51.6
2.0
2.5
2.1
2.5
3.7
2.9
2.7
196
7.5
3.5
32.5
2.6
4.0
2.6
2.1
13.0
5.5
7.0
2.0
11.0
2.5
3.5
1.5
2.4
2.3
1.7
<1.0
1.7
5.7
Cu
115
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
85.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
114
<23.1
<23.1
53.5
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
14,360
14,220
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
Fe
22,989
5,635
2,516
5,754
4,558
727
4,938
1,071
21,860
2,073
904
4,170
3,357
2,520
3,942
2,391
60,020
3,132
5,386
20,340
3,343
4031
2,265
2,739
10,358
3,810
1,087
2,704
3,564
789
1,704
4,196
3,962
3,048
1,760
4,926
2,743
2,647
Mg
1,295
1,946
<720
<720
26,994
<720
<720
<720
<720
<720
<720
<720
1,678
<720
971
<720
1,310
1,267
<720
954
1,031
<720
<720
<720
979
813
<720
859
<720
<720
<720
807
<720
<720
<720
<720
<720
<720
Mn
199
139
<45.0
121
184
<45.0
81.6
<45.0
149
<45.0
<45.0
<45.0
<45.0
45.1
94.6
<45.0
920
135
<45.0
458
<45.0
62.2
<45.0
<45.0
89.9
57.0
<45.0
<45.0
<45.0
<45.0
<45.0
<45.0
50.6
67.1
<45.0
<45.0
<45
<45
Mo
<4.4
<4.4
<4.4
<4.4
<4.4
9.2
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
19.8
<4.4
<4.4
6.1
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
Ni
17.0
3.7
3.5
6.4
139
3.1
4.5
<1.2
18.0
2.0
<1.2
3.2
2.5
2.2
3.3
2.0
36.7
6.9
2.0
15.4
2.8
3.7
1.9
2.4
20.6
3.0
15.7
1.9
14.0
34.5
21.1
2.1
2.6
2.1
<1.2
5.9
<1.2
7.4
Pb
18.4
<7.7
<7.7
<7.7
<7.7
<7.7
<7.7
<7 7
<7.7
<7.7
<7 7
<7.7
<7 7
<7.7
<7.7
<7 7
11.0
<7.7
<7.7
14.0
<7.7
<7.7
<7.7
<7 7
<7.7
<7.7
<7 7
<7.7
<7 7
20.6
28.9
<7 7
<7.7
<7.7
<7 7
<7.7
<7.7
<7 7
Sb
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
<4.5
V
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
19.3
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
<7.4
Zn
88.2
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
87.8
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
986
1732
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
<33.4
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2, 15, 32, 38, and 39.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-2
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-3. Element-Specific Concentrations in Spent Foundry Molding Sands Collected July 2006 (Concentrations in mg kg'1)
FIN"
1
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
Ag
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
<17.6
Al
3,431
1,780
1,072
<311
<311
816
370
650
2,197
1,416
1,870
2,086
981
3,068
1,044
2,092
4,680
<311
6,189
2,777
1,936
2,075
1,592
2,741
1,795
387
1,792
410
<311
1,681
2,739
1,269
911
1,298
1,612
799
1,092
As
2.0
1.0
0.47
0.24
0.07
0.68
0.31
1.2
1.1
0.34
1.0
1.3
0.68
1.4
1.8
0.72
0.85
0.47
4.9
1.4
1.1
0.47
2.0
1.6
0.67
0.17
0.66
0.27
0.08
1.2
1.5
1.0
3.0
0.34
0.70
0.77
0.59
B
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
<19.2
Ba
45.5
24.1
<8.7
<8.7
<8.7
<8.7
<8.7
<8.7
28.8
<8.7
23.0
9.5
19.4
53.1
12.3
25.5
12.9
<8.7
49.1
27.1
12 2
20.6
18.4
26.1
21.1
110
27.0
<8.7
<8.7
14.9
22.3
12.1
60.3
149
15.1
13.8
39.3
Be
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
<1.2
2.5
<1.2
<1.2
<1.2
<1.2
Cd
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
<5.9
Co
<0.84
<0.84
<0.84
6.1
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
2.2
<0.84
<0.84
<0.84
<0.84
1.2
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
<0.84
9.1
<0.84
<0.84
<0.84
<0.84
Cr
5.0
<1.0
2.2
2.9
5.3
6.8
<1.0
30.5
2.1
1.6
<1.0
2.2
4.0
4.0
2.2
12.7
7.6
2.9
32.1
2.7
3.0
2.4
1.9
8.5
4.5
8.0
1.9
2.5
<1.0
5.7
2.4
1.7
132
5.6
1.6
1.3
14.3
Cu
31.3
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
78.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
47.0
<23.1
63.2
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
<23.1
38.5
4668
<23.1
<23.1
111
25.7
<23.1
<23.1
<23.1
Fe
5,265
1,575
1,542
4,287
566
4,549
931
15,596
2,530
710
1,841
2,970
2,770
3,751
2,263
7,727
2,473
3,760
25,310
2,999
3,115
2,209
2,543
4,748
4,703
1,047
2,578
1,696
<352
4,339
2,862
2,281
45,120
3,162
1,628
2,787
1,682
Mg
<720
2,218
<720
15,990
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
906
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
<720
16,566
<720
<720
<720
<720
Mn
80.7
46.5
<45.0
59.7
<45.0
62.6
<45.0
128
<45.0
<45.0
<45.0
<45.0
<45.0
81.4
<45.0
99.0
83.8
<45.0
483
<45.0
60.8
<45.0
<45.0
66.7
80.8
<45.0
<45.0
<45.0
<45.0
<45.0
50.6
<45.0
845
85.2
<45.0
<45.0
<45.0
Mo
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
<4.4
54.6
<4.4
<4.4
<4.4
<4.4
Ni
3.2
<1.2
1.9
110.8
2.0
4.6
<1.2
10.5
3.4
<1.2
2.9
2.5
2.5
4.7
2.1
4.8
7.1
<1.2
17.7
3.3
3.6
2.6
2.6
7.0
3.1
8.6
2.1
3.0
<1.2
16.3
3.4
2.5
189
15.1
2.0
<1.2
3.9
Pb
10.6
<7 7
<7 7
^77
<7.7
9.6
<7 7
19.6
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-4. Polycyclic Aromatic Hydrocarbon Concentrations in Spent Foundry Molding Sands Collected June 2005 (Concentration in mg kg"1)
FINa
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Acenaph-
thene
<0.04
<0.04
<0.04
0.04
<0.04
0.07
<0.04
0.40
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.04
0.09
<0.04
<0.04
<0.04
11.7
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
2.9
0.26
0.19
0.04
<0.04
0.06
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.42
<0.04
<0.04
<0.04
|g
2 %
< •£
0.29
<0.03
0.13
<0.03
<0.03
0.09
0.09
0.09
0.05
<0.03
0.15
<0.03
<0.03
<0.03
0.04
<0.03
0.09
<0.03
0.10
<0.03
<0.03
<0.03
<0.03
0.08
0.07
<0.03
0.10
0.03
<0.03
<0.03
<0.03
<0.03
0.09
0.17
0.06
<0.03
0.19
<0.03
<0.03
0.26
0.03
<0.03
<0.03
Anthracene
0.13
<0.03
0.11
0.16
<0.03
0.37
0.11
0.31
0.38
<0.03
0.87
<0.03
0.81
0.53
0.57
0.52
0.25
0.08
0.36
0.89
<0.03
0.95
0.37
0.15
0.27
<0.03
0.67
0.70
0.11
0.14
0.06
0.61
0.04
0.91
0.09
<0.03
0.55
0.31
<0.03
0.90
0.11
0.18
<0.03
Benz[a]-
anthracene
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.17
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.30
<0.10
<0.10
<0.10
<0.10
<0.10
0.19
<0.10
<0.10
<0.10
Benzo[b]-
fluoranthene
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
Benzo[k]-
fluoranthene
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
Benzo[ghi]-
perylene
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Benzo[a]-
pyrene
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
Chrysene
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
0.18
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
0.30
<0.08
<0.08
<0.08
<0.08
<0.08
0.22
<0.08
<0.08
<0.08
Dibenz[a,h]-
anthracene
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
Fluoran-
thene
< 0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.50
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.43
< 0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
Fluorene
0.15
0.22
0.18
0.07
0.27
0.15
0.16
0.33
0.49
<0.04
0.83
<0.04
0.71
0.41
0.29
0.36
0.35
0.07
0.28
0.69
0.18
0.42
0.10
0.31
0.21
0.21
0.49
0.12
0.05
0.08
0.06
0.38
0.05
2.58
0.37
<0.04
0.67
0.04
0.09
0.75
0.12
0.15
<0.04
oi „
iai
!a £
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Naphthalene
4.2
0.39
0.75
0.09
0.41
48.1
0.59
0.67
0.53
0.14
2.1
<0.03
3.3
0.45
2.1
0.86
0.94
0.66
0.66
1.1
0.31
0.89
0.54
1.1
0.74
0.73
3.0
1.1
0.22
0.26
<0.03
6.8
27.6
5.3
0.25
3.5
2.4
<0.03
0.16
0.65
32.9
0.95
0.27
Phenan-
threne
0.46
0.49
0.70
0.29
0.52
0.45
0.23
0.72
0.76
0.08
1.6
<0.03
1.5
0.42
0.62
0.71
0.73
0.30
0.62
1.2
0.45
1.2
0.17
0.70
0.32
0.57
1.1
0.69
0.10
0.15
0.06
0.60
0.06
2.2
0.62
0.99
1.8
<0.03
0.13
1.2
0.45
0.36
0.09
Pyrene
0.31
0.07
0.10
0.08
0.31
0.06
0.06
0.53
0.19
<0.03
<0.03
<0.03
0.35
<0.03
<0.03
0.29
0.26
<0.03
0.21
0.46
0.19
0.23
<0.03
0.05
<0.03
<0.03
0.35
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
0.43
0.27
0.27
0.41
<0.03
<0.03
0.30
<0.03
<0.03
<0.03
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-4
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-5. Polycyclic Aromatic Hydrocarbon Concentrations in Spent Foundry Molding Sands Collected September 2005 (Cone, in mg kg"1)
FIN3
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
Acenaph-
thene
<0.04
<0.04
<0.04
0.06
<0.04
0.09
0.05
<0.04
<0.04
0.11
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.09
<0.04
<0.04
0.17
<0.04
<0.04
<0.04
<0.04
<0.04
0.06
<0.04
<0.04
<0.04
<0.04
<0.04
0.12
<0.04
0.18
<0.04
0.04
<0.04
Acenaph-
thylene
0.17
<0.03
<0.03
0.06
0.28
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
0.03
<0.03
<0.03
<0.03
0.08
<0.03
<0.03
<0.03
0.04
0.10
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
0.32
0.25
0.11
<0.03
0.12
0.09
<0.03
<0.03
0.20
Anthracene
0.24
0.18
0.33
0.62
0.38
0.27
0.52
0.69
<0.03
0.99
<0.03
0.43
0.84
0.49
0.44
<0.03
0.36
0.83
0.23
0.62
0.54
0.97
0.33
0.50
0.28
0.14
0.35
<0.03
0.06
0.46
0.45
0.36
0.35
0.66
0.48
0.53
0.53
0.05
Benz[a]-
anthracene
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.20
0.13
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
Benzo[b]-
fluoranthene
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
Benzo[k]-
fluoranthene
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
Benzo[ghi]-
perylene
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Benzo[a]-
pyrene
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
Chrysene
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
0.11
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
Dibenz[a,h]-
anthracene
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
0.17
<0.16
<0.16
Fluoranthene
<0.06
<0.06
<0.06
0.10
0.06
<0.06
1.0
0.11
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.23
<0.06
<0.06
<0.06
<0.06
Fluorene
0.25
<0.04
<0.04
0.50
0.38
0.25
0.11
0.47
<0.04
0.55
<0.04
0.44
0.53
0.36
0.11
0.36
0.25
0.56
<0.04
0.45
0.41
1.2
0.19
0.49
0.10
0.08
0.16
0.71
0.11
0.54
0.84
0.26
<0.04
0.88
0.41
0.38
0.24
0.16
Indeno[l,2,3-
cdjpyrene
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Naphthalene
1.6
0.26
0.34
0.84
8.3
0.43
0.16
0.38
0.39
1.5
<0.03
1.9
0.59
0.55
0.34
0.27
0.51
0.66
<0.03
0.81
0.09
1.4
0.43
0.89
0.19
0.95
<0.03
0.74
0.36
9.8
0.74
0.29
1.1
2.2
1.1
14.6
0.68
0.10
Phenan-
threne
1.4
0.18
0.37
0.62
0.43
0.29
1.29
0.94
0.94
1.4
<0.03
0.91
1.0
0.77
0.39
1.1
0.40
0.97
0.20
0.64
0.57
1.9
0.45
1.3
0.54
0.14
0.29
1.1
0.26
0.66
0.90
0.57
0.88
1.7
0.77
0.65
0.55
0.23
=
2
£
0.06
<0.03
<0.03
0.24
<0.03
0.06
0.86
0.20
<0.03
0.18
<0.03
0.12
0.47
<0.03
<0.03
0.16
0.07
0.47
<0.03
0.23
0.22
0.80
0.09
0.33
<0.03
0.07
<0.03
0.49
<0.03
0.23
0.27
<0.03
0.10
0.16
0.04
0.13
<0.03
<0.03
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2, 15, 32, 38, and 39.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-5
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-6. Polycyclic Aromatic Hydrocarbon Concentrations in Spent Foundry Molding Sands Collected July 2006 (Concentrations in mg kg"1)
FINa
1
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
Acenaph-
thene
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.11
<0.04
0.10
0.11
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.40
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
0.06
<0.04
0.05
<0.04
<0.04
0.25
0.04
<0.04
*a
1 £
3*
0.03
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
<0.03
0.14
<0.03
0.07
<0.03
<0.03
<0.03
0.25
<0.03
<0.03
<0.03
<0.03
0.07
0.16
<0.03
0.05
0.04
<0.03
<0.03
<0.03
0.05
0.08
0.10
<0.03
0.07
<0.03
<0.03
0.33
<0.03
<0.03
Anthracene
0.09
0.10
0.10
0.11
<0.03
0.25
0.13
0.07
0.69
<0.03
0.25
0.18
0.16
0.10
0.60
0.13
0.37
<0.03
0.17
0.19
0.56
0.04
0.23
0.15
<0.03
0.17
<0.03
0.05
0.20
0.35
<0.03
0.27
0.10
0.07
0.60
0.46
0.10
Benz[a]-
anthracene
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.15
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.15
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
Benzo[b]-
fluoranthene
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
<0.12
Benzo[k]-
fluoranthene
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
Benzo[ghi]-
perylene
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Benzo[a]-
pyrene
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
<0.20
1
X
S
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
Dibenz[a,h]-
anthracene
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
<0.16
Fluoranthene
<0.06
<0.06
<0.06
<0.06
<0.06
0.18
<0.06
<0.06
0.33
<0.06
0.06
<0.06
0.10
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
0.09
<0.06
<0.06
Fluorene
0.14
0.14
<0.04
0.39
0.06
0.28
<0.04
0.08
0.64
<0.04
0.32
0.24
0.14
0.07
0.56
0.16
0.30
<0.04
0.17
0.12
1.0
0.06
0.42
0.14
0.05
0.07
0.07
0.04
0.46
0.83
0.22
0.41
<0.04
<0.04
0.41
0.18
<0.04
Indeno[l,2,3-
cd]pyrene
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Naphthalene
0.65
0.67
0.07
42.2
0.16
0.35
<0.03
0.03
2.8
<0.03
0.63
0.14
0.50
0.17
0.21
0.43
0.64
<0.03
0.64
0.30
1.9
0.08
0.89
1.0
0.53
0.08
0.12
0.17
9.6
2.0
5.8
0.67
0.03
0.05
0.54
0.41
0.03
Phenanthrene
0.30
0.39
0.18
0.18
0.09
0.43
<0.03
0.19
1.9
<0.03
0.80
0.64
0.43
0.22
0.80
0.43
1.4
0.09
0.61
0.31
1.1
0.37
0.44
0.44
0.74
0.18
0.25
0.10
0.46
1.6
0.48
0.81
0.11
0.21
0.73
0.43
0.11
v
£
0.07
0.09
0.06
0.06
<0.03
0.19
<0.03
0.06
0.29
<0.03
0.11
0.11
0.10
0.06
0.09
0.10
0.27
0.04
0.12
0.08
0.73
0.09
0.13
0.13
0.04
0.06
0.08
0.03
0.05
0.24
0.09
0.12
0.05
<0.03
0.06
0.05
0.05
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2,5,15, 32, 35, and 41.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-6
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-7. Phenolics Concentrations in the Spent Foundry Molding Sands Collected June 2005 (Concentrations in mg kg"1)
FIN"
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
2-Sec-Butyl-
4,6-Dinitro-
phenol
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
4-Chloro-3-
Methylphenol
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
0.31
<0.18
<0.18
0.81
<0.18
<0.18
2-Chloro-
phenol
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
2,4-Dichloro-
phenol
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
<0.13
2,6-Dichloro-
phenol
< 0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
2,4-Dimethyl-
phenol
0.70
<0.08
<0.08
0.51
<0.08
4.1
<0.08
<0.08
0.36
<0.08
3.85
<0.08
7.5
1.4
0.68
1.1
0.79
<0.08
0.64
12.30
<0.08
<0.08
0.09
0.46
<0.08
<0.08
0.53
0.16
0.65
<0.08
<0.08
1.2
0.25
4.57
0.16
0.23
2.3
0.48
<0.08
2.1
0.30
0.50
<0.08
2,4-Dinitro-
phenol
<0.24
<0.24
<0.24
<0.24
<0.24
0.86
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
1-
^J 5
^1
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-8. Phenolics Concentrations in the Spent Foundry Molding Sands Collected September 2005 (concentrations in mg kg"1)
FIN3
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
N
Hi
ap vi £
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-9. Phenolic Concentrations in the S]
FINa
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
i'l-
£ o 1
3 * £
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-10. Element Concentrations in the Toxicity Characteristic Leaching Procedure (TCLP) Extracts
from the Spent Foundry Molding Sands Collected June 2005 (Concentrations in mg L"1)
FIN3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Ag
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
As
< 0.001
< 0.001
0.002
0.002
0.011
< 0.007
< 0.007
< 0.007
< 0.007
0.001
0.004
2.4
0.003
0.001
0.005
< 0.007
0.006
< 0.007
< 0.007
< 0.007
< 0.007
< 0.007
0.003
0.001
0.002
0.004
< 0.007
< 0.007
0.003
0.002
< 0.007
0.003
< 0.007
0.001
0.001
0.008
< 0.007
0.002
0.001
0.002
< 0.007
< 0.007
0.003
Ba
<0.86
<0.86
<0.86
<0.86
<0.86
1.1
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
Be
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
0.04
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
0.02
<0.07
<0.07
<0.07
<0.07
<0.07
0.03
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
Cd
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
0.02
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
0.02
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
0.06
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
Cr
< 0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
Cu
0.14
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
0.22
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
<0.70
0.10
<0.70
<0.70
<0.70
<0.70
<0.70
0.20
<0.70
0.19
<0.70
<0.70
<0.70
2.1
3.5
<0.70
<0.70
<0.70
<0.70
0.11
<0.70
<0.70
<0.70
<0.70
Ni
<0.14
0.94
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.18
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.19
<0.14
<0.14
0.25
<0.14
<0.14
<0.14
<0.14
0.44
<0.14
0.61
<0.14
0.20
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.33
1.5
<0.14
<0.14
<0.14
<0.14
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.10
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Sb
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Zn
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
2.5
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
1.7
37.6
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-10
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-ll. Element Concentrations in the TCLP Extracts from the Spent Foundry Molding Sands Collected September 2005
(Concentrations in mg L"1)
FIN3
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
Ag
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
As
<0.001
0.001
0.001
0.018
<0.001
<0.001
<0.001
<0.001
<0.001
0.019
0.003
<0.001
0.002
0.001
0.013
0.001
<0.001
0.001
0.001
<0.001
0.007
0.002
0.003
0.008
<0.001
0.003
0.005
0.004
0.001
<0.001
<0.001
0.003
0.013
0.005
0.003
<0.001
0.001
0.006
Ba
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
Be
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
< 0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
Cu
0.14
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.14
<0.10
0.10
0.11
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
43.9
0.65
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
Ni
0.15
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.25
<0.14
<0.14
0.17
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.30
0.14
0.26
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Sb
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Zn
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
0.58
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
1.32
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
6.5
40.3
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Data were not available for FIN 2, 15, 32, 38, and 39.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-ll
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-12. Element Concentrations in the TCLP Extracts from the Spent Foundry Molding Sands Collected July 2006
(Concentrations in mg L"1)
FINa
1
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
Ag
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
As
0.004
0.003
<0.001
<0.001
<0.001
<0.001
0.001
<0.001
0.004
0.001
0.004
0.006
<0.001
0.007
0.001
<0.001
0.001
<0.001
< 0.001
0.005
0.003
0.003
0.012
<0.001
<0.001
0.002
0.001
<0.001
<0.001
0.001
0.017
0.001
0.007
0.004
0.001
<0.001
0.005
Ba
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
<0.86
Be
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.02
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.06
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
<0.46
Cu
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
5.4
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
Ni
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.27
<0.14
<0.14
<0.14
<0.14
<0.14
<0.14
0.20
<0.14
<0.14
<0.14
0.23
<0.14
<0.14
1.71
<0.14
<0.14
<0.14
<0.14
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
1.1
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Sb
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Zn
<0.41
<0.41
<0.41
<0.41
<0.41
0.68
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
4.49
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
<0.41
42.5
<0.41
<0.41
0.71
<0.41
<0.41
<0.41
<0.41
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Data were not available for FIN 2, 5,15, 32, 35, and41.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-12
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-13. Element-Specific Concentrations in the Synthetic Precipitation Leaching Procedure (SPLP) Extracts
from the Spent Foundry Molding Sands Collected June 2005 (Concentrations in mg L"1)
FIN3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Ag
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
As
3.0E-03
<0.001
9.8E-02
<0.001
1.7E-02
<0.001
<0.001
<0.001
<0.001
<0.001
1.1E-02
<0.001
4.0E-03
9.0E-03
l.OE-02
<0.001
7.0E-03
7.0E-03
<0.001
<0.001
<0.001
<0.001
8.0E-03
3.0E-03
4.0E-03
1.1E-02
<0.001
5.0E-03
3.0E-03
5.0E-03
<0.001
4.0E-03
<0.001
3.0E-03
7.0E-03
9.0E-03
9.0E-03
<0.001
4.0E-03
4.0E-03
<0.001
6.0E-03
7.0E-03
Ba
<0.23
<0.23
2.9E-01
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
2.5E-01
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
4.3E-01
<0.23
<0.23
3.0E-01
2.7E-01
2.9E-01
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
2.7E-01
6.1E-01
2.9E-01
<0.23
<0.23
<0.23
Be
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cu
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
5.5E-01
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
2.4E-01
<0.05
<0.05
<0.05
<0.05
Pb
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
Sb
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
Zn
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
3.4E-01
3.1E+00
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications B-13
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-14. Element-Specific Concentrations in the SPLP Extracts from the Spent Foundry Molding Sands Collected September 2005
(Concentrations in mg L"1)
FINa
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
Ag
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
As
< 0.001
1.7E-02
< 0.001
2.4E-02
< 0.001
< 0.001
< 0.001
4.9E-03
< 0.001
1.9E-02
3.5E-03
3.9E-03
1.6E-02
< 0.001
1.5E-02
9.6E-03
1.8E-03
2.7E-03
< 0.001
< 0.001
2.3E-02
9.0E-03
l.OE-02
1.9E-02
< 0.001
1.1E-02
5.2E-03
1.1E-02
< 0.001
< 0.001
< 0.001
1.5E-02
1.4E-02
1.5E-02
6.8E-03
< 0.001
1.2E-02
1.3E-02
Ba
<0.23
2.5E-01
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
2.6E-01
<0.23
<0.23
<0.23
<0.23
<0.23
3.7E-01
Be
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cu
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
7.5E-01
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
8.9E-02
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
6.4E-02
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
5.0E-02
Pb
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
Sb
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
Zn
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
1.6E+00
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2, 15, 32, 38, and 39.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-14
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-15. Element Concentrations in the SPLP Extracts from the Spent Foundry Molding Sands Collected July 2006
(Concentrations in mg L"1)
FIN3
1
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
Ag
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
As
5.3E-03
9.5E-03
1.7E-03
<0.001
<0.001
<0.001
2.2E-03
<0.001
4.0E-03
1.8E-03
4.3E-03
1.1E-02
<0.001
5.6E-03
7.4E-03
3.4E-03
2.0E-03
<0.001
2.5E-03
1.2E-02
6.7E-03
6.2E-03
1.1E-02
2.2E-03
3.8E-03
2.8E-03
4.7E-03
<0.001
<0.001
1.6E-03
1.7E-02
1.1E-02
<0.001
3.4E-03
4.5E-03
4.7E-03
7.5E-03
Ba
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
3.2E-01
<0.23
<0.23
<0.23
<0.23
<0.23
3.9E-01
<0.23
<0.23
3.1E-01
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
<0.23
6.3E-01
<0.23
<0.23
3.9E-01
Be
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cu
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
1.7E+00
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
<0.21
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
7.0E-02
<0.05
<0.05
<0.05
Pb
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
2.8E-01
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
<0.08
Sb
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
Zn
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
4.0E+00
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
<0.18
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2,5, 15, 32, 35, and 41.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-15
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-16. Element Concentrations in Water Extracts from the Spent Foundry Molding Sands Collected June 2005
(Concentrations in mg L"1)
FINa
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Ag
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
As
4.0E-03
<0.003
8.0E-03
<0.003
1.8E-02
<0.003
<0.003
<0.003
<0.003
< 0.003
1.1E-02
< 0.003
5.0E-03
l.OE-02
l.OE-02
<0.003
7.0E-03
5.0E-03
< 0.003
< 0.003
< 0.003
< 0.003
9.0E-03
< 0.003
6.0E-03
1.1E-02
< 0.003
4.0E-03
3.0E-03
5.0E-03
< 0.003
4.0E-03
< 0.003
< 0.003
8.0E-03
1.1E-02
9.0E-03
< 0.003
4.0E-03
4.0E-03
< 0.003
6.0E-03
6.0E-03
Ba
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
Be
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cu
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
1.1E+00
3.0E-01
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.05
<0.05
4.6E-02
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Pb
<0.11
<0.11
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-17. Element Concentrations in the Water Extracts from the Spent Foundry Molding Sands Collected September 2005
(Concentrations in mg L"1)
FINa
1
3
4
5
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
40
41
42
43
Ag
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
As
< 0.003
1.1E-02
1.8E-03
2.3E-02
< 0.003
< 0.003
< 0.003
5.1E-03
< 0.003
1.9E-02
4.1E-03
6.5E-03
1.7E-02
1.5E-03
1.6E-02
7.4E-03
< 0.003
< 0.003
< 0.003
< 0.003
2.4E-02
1.1E-02
1.1E-02
1.9E-02
< 0.003
1.3E-02
4.7E-03
1.2E-02
< 0.003
< 0.003
< 0.003
1.8E-02
1.7E-02
1.8E-02
9.0E-03
< 0.003
1.3E-02
1.3E-02
Ba
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
Be
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cu
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
7.0E-02
2.2E-01
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Pb
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
<0.11
Sb
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
<0.04
Zn
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
<0.22
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
No data were available for FIN 2, 15, 32, 38, and 39.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-17
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-18. Element Concentrations in the Water Extracts from the Spent Foundry Molding Sands Collected July 2006
(Concentrations in mg L"1)
FINa
1
3
4
6
7
8
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
36
37
38
39
40
42
43
Ag
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
As
6.6E-03
7.8E-03
< 0.003
< 0.003
< 0.003
< 0.003
< 0.003
< 0.003
4.7E-03
2.6E-03
4.8E-03
1.3E-02
< 0.003
7.3E-03
8.0E-03
3.1E-03
2.5E-03
< 0.003
< 0.003
1.4E-02
6.7E-03
5.8E-03
1.2E-02
4.0E-03
3.0E-03
2.8E-03
4.6E-03
< 0.003
< 0.003
< 0.003
1.7E-02
1.1E-02
< 0.003
2.9E-03
5.1E-03
5.3E-03
6.7E-03
Ba
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
<0.24
Be
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Cu
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
<0.07
8.0E-02
<0.07
<0.07
<0.07
<0.07
Ni
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Pb
<0.11
<0.11
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-19. Summary of Total Elemental Content of 43 Spent Foundry Molding Sands Using a Microwave-Assisted Aqua Regia Digest
(U.S. EPA, 3051a)
FINa
Units
LOQ
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Al
gkg-1
0.001
11.7
3.44
8.09
4.95
7.57
1.70
0.219
3.00
1.59
1.51
5.78
4.06
6.15
5.11
7.87
4.23
7.33
4.82
6.02
11.2
0.705
6.24
6.66
5.73
3.57
7.19
10.5
6.33
0.572
7.31
1.57
2.44
0.650
5.14
7.89
7.98
6.28
1.11
5.65
4.57
0.193
2.72
5.46
As
mg kg4
0.1
3.72
0.395
1.13
0.633
2.14
0.498
0.126
1.22
0.363
0.972
2.55
0.241
2.11
1.05
3.19
0.921
3.01
1.62
3.85
0.993
1.26
7.79
1.24
2.14
0.585
2.54
6.44
0.899
0.335
0.770
0.767
0.767
0.223
2.09
2.25
3.25
2.56
0.164
0.578
0.256
0.256
0.664
0.771
B
mg kg4
20.0
10
10
10
10
20.2
10
10
10
10
10
10
10
10
59.4
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
Ba
mg kg'1
10.0
5
5
5
5
5
120
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
141
5
5
5
17.7
Be
mg kg'1
0.1
0.383
0.05
0.148
0.05
0.157
0.05
0.05
0.101
0.05
0.05
0.272
0.05
0.242
0.096
0.252
0.122
0.264
0.114
0.207
0.110
0.05
0.227
0.147
0.151
0.369
0.186
0.328
0.171
0.05
0.205
0.05
0.141
0.05
0.127
0.226
0.282
0.599
0.05
0.110
0.177
0.05
0.153
0.171
Ca
gkg-1
0.005
3.05
1.09
1.22
1.20
0.975
1.32
0.285
3.23
1.61
1.17
1.72
0.503
2.09
3.10
2.59
1.57
1.83
1.59
2.78
2.36
0.393
3.10
2.70
1.27
2.65
2.79
2.57
2.51
0.405
2 92
0.993
2.12
0.075
1.69
4.09
2.05
1.88
0.370
1.89
2.07
0.094
0.835
3.13
Cd
mg kg'1
0.04
0.16
0.05
0.05
0.051
0.08
0.103
0.02
0.199
0.02
0.02
0.085
0.02
0.063
0.051
0.05
0.02
0.062
0.02
0.112
0.099
0.02
0.36
0.051
0.043
0.055
0.046
0.19
0.02
0.061
0.067
0.067
0.02
0.103
3.79
0.066
0.119
0.078
0.02
0.074
0.02
0.02
0.02
0.055
Co
mg kg'1
0.5
1.74
92.5
0.25
0.806
1.11
42.1
0.25
1.25
0.25
1.82
1.01
0.25
0.952
0.25
1.29
0.856
1.14
0.25
5.88
0.958
0.25
5.99
0.65
1.15
0.25
0.794
6.62
0.633
2.74
0.880
1.31
0.25
0.25
0.25
1.02
0.981
1.44
0.25
1.07
0.524
0.25
0.781
0.808
Cr
mg kg'1
0.5
17.4
49.3
0.25
5.52
15.7
134
19.1
16.4
1.98
25.1
3.27
4.70
2.68
2.98
4.78
8.64
4.21
5.26
115
4.93
14.8
95.1
3.59
7.85
2.93
3.79
87.1
5.39
109
2.94
16.9
2.96
1.36
4.17
2.60
3.50
3.83
8.87
34.0
2.59
0.97
2.50
18.2
Cu
mg kg'1
0.5
82.6
7.04
0.25
8.42
13.0
23 2
3.26
32.9
2.94
46.0
4.69
0.25
3.57
2.60
3.13
8.22
4.03
5.99
88.2
12.6
15.1
137
2.11
12.5
3.16
4.76
107
2.11
46.3
14.5
23.9
2.92
85.2
3805
4.57
6.46
6.22
21.0
72.4
2.15
3.46
2.84
12.8
Fe
gkg-1
0.005
12.0
54.7
2.90
3.08
12.8
27.0
2.88
9.68
1.58
10.4
2.91
1.28
3.49
3.62
4.37
4.51
4.87
4.26
55.7
3.49
13.4
57.1
3.64
6.70
3.16
3.96
64.4
4.25
5.49
4.22
7.66
3.61
0.536
2.94
5.98
5.47
4.87
1.29
5.73
2.33
3.13
3.00
5.70
K
mg kg'1
50.0
610
321
215
618
248
203
25
268
204
253
544
153
647
376
547
343
374
419
292
370
25
531
403
396
224
416
1300
251
1780
453
114
189
25
445
437
337
242
328
289
266
25
299
316
Mg
gkg4
0.002
3.20
287
3.02
1.03
1.16
124
0.103
0.963
0.713
0.521
1.14
0.192
1.37
1.95
1.79
1.27
1.72
1.16
1.50
1.83
0.129
1.60
1.65
1.06
1.11
1.62
2.29
1.55
0.236
1.91
0.545
1.31
0.060
1.39
1.96
1.51
1.32
0.305
1.47
1.06
0.050
0.629
1.28
Mn
mg kg'1
0.5
237
639
52.4
34.6
288
570
14.1
107
20.5
79.4
65.4
24.1
37.4
38.5
67.1
54.5
114
30.0
482
109
90.6
707
40.2
93.7
33.7
34.6
670
51.1
61.2
40.4
84.2
21.8
14.0
18.7
46.5
58.1
119
28.4
161
20.9
5.56
27.0
110
Mo
mg kg'1
1.0
1.85
0.5
0.5
0.5
1.94
2.49
0.5
1.98
0.5
2.90
0.5
0.5
0.604
0.5
0.5
0.5
1.02
0.5
22.9
0.5
3.33
21.8
0.5
1.41
0.5
2.51
19.7
0.5
10.6
0.5
3.10
0.5
0.5
0.5
0.5
0.5
0.68
0.5
6.64
0.5
0.5
0.5
2.11
Na
gkg4
0.02
1.39
0.305
1.38
0.280
1.82
0.320
0.01
0.225
0.165
0.140
1.46
0.01
0.820
0.560
1.13
0.305
1.37
1.16
0.975
1.22
0.160
0.983
1.39
0.930
1.71
1.16
1.11
1.26
0.408
1.93
0.347
1.46
0.01
0.800
1.27
1.85
1.02
0.630
1.31
0.837
0.01
0.605
1.37
Ni
mg kg'1
0.5
6.60
2560
3.47
3.46
5.64
1160
11.7
6.39
1.11
8.73
2.90
1.32
3.85
2.62
3.30
2.82
4.09
3.43
42.9
5.92
11.5
49.4
2.29
4.24
2.13
2.68
117
2.84
81.5
2.51
23.1
2.15
1.02
15.3
2.53
2.78
3.04
36.0
102
1.61
2.18
2.06
19.6
P
mg kg'1
5.0
78.6
17.6
54.6
81.2
34.3
36.4
11.1
49.4
36.6
44.9
55.9
36.0
56.0
46.0
62.0
57.5
58.5
46.9
71.4
64.9
18.3
82.5
47.4
73.4
49.5
47.1
85.9
50.9
32.7
59.3
17.0
41.0
20.2
176
66.6
58.8
53.8
17.8
96.6
42.5
5.41
44.4
58.9
Pb
mg kg'1
1.0
15.3
3.03
4.46
2.22
4.87
2.55
0.50
5.06
1.41
1.63
4.38
1.28
3.04
3.26
5.02
2.22
5.92
2.35
7.04
2.20
1.90
22.9
3.28
7.07
4.14
3.74
8.63
4.00
1.10
3.91
2.41
2.84
1.79
20.8
4.64
5.83
4.30
1.40
4.88
3.63
0.5
3.13
4.30
S
gkg4
0.05
1.64
0.152
0.588
0.302
0.352
0.248
1.31
0.234
0.139
0.137
0.813
0.025
0.616
0.145
0.826
0.528
0.957
0.674
0.564
0.557
0.190
1.18
0.713
0.430
0.591
0.898
1.13
0.850
0.119
0.802
2.04
0.660
0.025
0.545
0.641
0.940
0.684
0.025
0.327
0.646
0.025
0.320
0.397
Sb
mg kg'1
0.04
0.353
0.094
0.277
0.12
0.08
0.11
0.15
0.21
0.02
0.31
0.06
0.17
0.16
0.17
0.09
0.10
0.14
0.93
1.23
0.02
0.24
1.04
0.150
1.71
0.72
0.14
0.65
0.25
0.11
0.17
0.16
0.17
0.10
0.74
0.29
0.12
0.24
0.07
0.26
0.25
0.08
0.26
0.16
Se
mg kg'1
0.4
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.438
0.2
0.2
0.2
0.2
0.2
1.15
59.5
0.2
0.2
0.2
0.2
0.2
0.06
0.2
0.2
0.2
Tl
mg kg'1
0.04
0.083
0.02
0.045
0.02
0.051
0.02
0.02
0.02
0.02
0.02
0.063
0.02
0.040
0.02
0.082
0.02
0.070
0.061
0.043
0.02
0.02
0.089
0.051
0.02
0.02
0.049
0.090
0.056
0.02
0.041
0.02
0.047
0.02
0.065
0.062
0.096
0.049
0.02
0.02
0.056
0.02
0.02
0.02
V
mg kg'1
1.0
6.60
2.10
1.21
1.20
2.45
3.11
0.5
4.02
2.14
3.13
2.13
6.25
3.09
4.98
4.95
3.24
2.88
3.39
11.3
4.42
2.47
9.03
2.69
3.92
1.84
2.98
9.90
2.48
2.73
3.80
1.32
1.47
1.27
2.88
2.96
2.59
3.91
0.5
2.48
2.44
1.18
2.90
2.62
Zn
mg kg'1
10.0
63.7
46.3
5
22.8
24.5
26.0
5
40.2
12.1
5
5
5
5
5
5
14.4
5
72.1
5
34.6
5
245
5
14.3
5
5
30.2
5
44.5
5
5
5
34.7
2474
5
5
5
5
17.5
11.4
5
5
13.2
aFIN = foundry identification number; see Table 2-2 for details.
Entries in italics were below the limit of quantification (LOQ) and were calculated as 0.5 times the LOQ.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-19
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-20. Total Metal Concentrations in Spinach Leaves Grown on Spent Foundry Sand Blends
(Concentrations in mg kg"1)
Element
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
Control
10.1
79.9
<0.12
<0.01
0.24
0.72
<0.01
4.74
68.1
8,511
25.0
0.97
2.06
<0.28
<0.09
27.0
±
±
±
±
±
±
±
±
±
±
±
2.3*
33.7
0.16
0.10
0.50
8.8
3,242
4.1
0.14
0.13
1.4
IGS-1
50.3
51.7
1.35
<0.01
0.20
0.40
<0.01
3.07
41.9
3,475
28.9
0.65
1.23
3.96
<0.09
29.9
±
±
±
±
±
±
±
±
±
±
±
±
±
12. 3a
6.2
0.90
0.15
0.17
l.OSa
12.7
l,466a
7.2
0.45
0.50
1.22a
2.9
IGS-2
65.4
47.4
1.80
<0.01
0.47
0.72
<0.01
3.62
58.2
5,515
69.6
0.55
1.19
<0.28
<0.09
38.1
±
±
±
±
±
±
±
±
±
±
±
±
28.7a
6.7
1.23
0.27
0.24
0.70
10.0
1,198
16. 5a
0.65
0.36
3.7
AGS-1
34.4
58.1
2.20
<0.01
1.00
<0.01
<0.01
6.15
73.9
3,339
262
0.70
<0.05
<0.28
<0.09
84.0
±
±
±
±
±
±
±
±
±
±
18.7
39.4
1.62a
0.81
0.57
17.0
l,315a
43a
0.53
12.3a
AGS-2
25.5
41.3
3.55
<0.01
0.39
0.67
<0.01
10.9
59.5
6,009
119
0.98
1.59
<0.28
<0.09
50.2
±
±
±
±
±
±
±
±
±
±
±
±
7.6
3.1
0.26a
0.16
0.16
1.3a
4.7
548
12a
0.11
0.25
4.5a
NBS-1
9.94
74.2
0.79
<0.01
0.70
1.38
<0.01
6.87
110
10,182
54.8
0.81
9.91
<0.28
<0.09
76.3
±
±
±
±
±
±
±
±
±
±
±
±
2.57
13.3
0.53
0.26
0.32
1.02
20a
835a
13.0
0.55
6.47
15. 8a
NBS-2
14.9
57.6
1.52
<0.01
0.30
0.96
<0.01
3.85
110
12,758
32.0
0.89
3.78
<0.28
<0.09
22.4
±
±
±
±
±
±
±
±
±
±
±
±
1.2
8.8
0.17
0.22
0.16
0.34a
18a
1,988
3.5
0.07
0.89a
1.8
Sufficiency Range
(Jones et al., 1991)
25-60
5-25
60-200
6,000-10,000
30-250
25-100
* Average value of four replicates ± standard deviation.
"a" values indicate a significant difference than the control (PO.05).
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Note: AGS = aluminum green sand, IGS= iron green sand, NBS = steel no-bake sand
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-20
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-21. Total Element-Specific Concentrations in Radish Globes and Leaves Grown on Spent Foundry Sand Blends
(Concentrations in mg kg"1)
Element
Control
IGS-1
IGS-2
AGS-1
AGS-2
NBS-1
NBS-2
Sufficiency
Range (Jones et
al., 1991)
Radish Globes
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
52.7
24.9
<0.12
<0.01
<0.01
0.58
0.38
5.48
139
2,112
18.4
<0.07
1.46
<0.28
<0.09
17.8
±
±
±
±
±
±
±
±
±
±
12.5*
2.9
0.08
0.12
1.50
23
471
1.8
0.16
3.2
319
27.0
3.78
<0.01
<0.01
0.08
0.31
2.43
256
1,162
11.5
1.27
<0.05
<0.28
<0.09
16.3
±
±
±
±
±
±
±
±
±
±
±
275
6.2
2.84
0.16
0.48
1.75
296
449
4.3
0.52a
3.2
258
26.0
4.19
<0.01
<0.01
0.28
<0.01
2.26
132
1,313
23.0
1.33
0.17
<0.28
<0.09
17.4
±
±
±
±
±
±
±
±
±
±
±
89
1.6
1.46
0.12
0.60
80
192
11.2
0.38a
0.34
6.9
2,393
30.2
5.46
<0.01
<0.01
<0.01
0.70
7.05
652
2,497
104
<0.07
<0.05
<0.28
<0.09
18.7
±
±
±
±
±
±
±
±
±
l,157a
2.5
6.30
0.81
8.88
232a
437
37a
12.6
111
45.3
<0.12
<0.01
<0.01
0.11
0.22
13.49
438
1847
22.7
<0.07
<0.05
<0.28
<0.09
19.9
±
±
±
±
±
±
±
±
±
499
27.6
0.21
0.44
8.28
197
486
5.2
3.4
55.6
32.7
<0.12
<0.01
<0.01
0.78
<0.01
4.08
165
3704
14.0
<0.07
5.13
<0.28
<0.09
15.9
±
±
±
±
±
±
±
±
±
111
9.9
0.38
1.22
93
2,332
4.2
2.37a
2.6
270
34.1
<0.12
<0.01
<0.01
1.13
<0.01
1.39
178
3,614
17.1
<0.07
4.81
<0.28
<0.09
18.8
±
±
±
±
±
±
±
±
±
55
11.3
0.69
1.72
61
1,619
3.9
3.91
2.7
Radish Leaves
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
9.9
98.8
<0.12
<0.01
0.11
1.17
<0.01
7.10
131
8,907
88.2
5.00
3.93
<0.28
<0.09
29.2
±
±
±
±
±
±
±
±
±
±
±
19.8
12.4
0.22
0.42
3.30
58
1,060
15.2
1.05
1.27
4.3
58.7
49.2
8.12
<0.01
0.21
0.03
<0.01
2.75
106
3,315
60.6
6.28
<0.05
<0.28
<0.09
24.1
±
±
±
±
±
±
±
±
±
±
±
44.0
5.5
0.61a
0.42
0.07
0.94a
27
299a
11.3
1.77
2.7
75.9
66.0
14.0
<0.01
<0.01
0.26
<0.01
3.22
109
4,842
143
7.84
0.28
<0.28
<0.09
28.8
±
±
±
±
±
±
±
±
±
±
±
88.1
7.5
1.9a
0.19
0.53a
36
1120a
7a
1.48a
0.55
5.2
288
91.8
<0.12
<0.01
<0.01
<0.01
<0.01
0.75
214
4,353
192
10.7
<0.05
<0.28
<0.09
31.6
±
±
±
±
±
±
±
±
214a
8.5
1.49a
65
407a
33a
2.0a
4.5
118
81.2
6.20
<0.01
<0.01
<0.01
<0.01
4.86
134
3,774
128
6.67
<0.05
<0.28
<0.09
25.3
±
±
±
±
±
±
±
±
±
119
64.8
0.56a
0.82
65
l,254a
18a
0.82
3.1
<2.00
100.8
3.10
<0.01
<0.01
2.29
<0.01
4.75
137
12,357
125
2.87
19.7
<0.28
<0.09
37.5
±
±
±
±
±
±
±
±
±
±
10.5
2.38a
1.59
1.22
41
3,134a
13a
0.12
11. 7a
8.3
47.2
88.5
8.14
<0.01
0.19
2.39
<0.01
3.75
245
10,212
86.8
4.26
7.07
<0.28
<0.09
30.0
±
±
±
±
±
±
±
±
±
±
±
±
94.5
9.2
1.55a
0.38
1.20
1.16a
90
2,341
3.8
0.41
2.98
3.2
25-125
5-25
50-200
5,000^5,000
50-250
25-100
* Average value of four replicates ± standard deviation.
"a" values indicate a significant difference than the control (PO.05).
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Note: AGS = aluminum green sand, IGS= iron green sand, NBS = steel no-bake sand
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-21
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-22. Total Metal Concentrations in Perennial Ryegrass Grown on Spent Foundry Sand Blends
(Concentrations in mg kg"1)
Element
Control
IGS-1
IGS-2
AGS-1
AGS-2
NBS-1
NBS-2
Sufficiency Range
(Jones et al., 1991)
Harvest 1
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
<2.00
113
<0.12
<0.01
0.09
0.92
<0.01
5.61
49.4
3,806
100
1.51
1.66
<0.28
<0.09
34.5
±
±
±
±
±
±
±
±
±
±
34.3*
0.11
0.05
1.29
1.0
155
4
0.24
0.30
1.7
17.8
132
12.7
<0.01
0.19
0.41
<0.01
7.97
64.3
2,576
121
3.54
2.49
<0.28
<0.09
30.8
±
±
±
±
±
±
±
±
±
±
±
±
35.5
136
2.7a
0.30
0.09a
1.55a
10.8
939
12
0.54a
0.31a
3.6
<2.00
57.6
7.85
<0.01
0.14
0.35
<0.01
9.53
63.1
3,231
200
5.01
2.30
<0.28
<0.09
25.8
±
±
±
±
±
±
±
±
±
±
±
7.7
1.81a
0.19
0.07a
2.08a
12.2
884
23a
0.49a
0.58
3.6a
<2.00
296
1.49
<0.01
0.09
<0.01
<0.01
13.8
87.2
1,872
192
9.69
<0.05
<0.28
<0.09
46.1
±
±
±
±
±
±
±
±
±
60.2a
2.97
0.19
l.Oa
11.7
2,167
25a
l.21a.
4.0a
<2.00
42.7
6.53
<0.01
0.25
0.33
<0.01
11.1
61.9
2,736
191
4.27
2.19
<0.28
<0.09
26.9
±
±
±
±
±
±
±
±
±
±
±
6.2
1.33a
0.16
O.OSa
0.9a
10.2
753
20a
0.56a
0.16
2.7a
<2.00
120
1.78
<0.01
0.19
0.58
<0.01
7.96
59.0
4,978
123
1.25
3.35
<0.28
<0.09
45.6
±
±
±
±
±
±
±
±
±
±
±
28
2.07
0.22
0.06a
l.lOa
7.0
1569
11
0.84a
0.77a
3.9a
<2.00
79.8
8.59
<0.01
0.44
0.34
<0.01
5.01
48.3
4,380
135
2.38
2.35
<0.28
<0.09
23.5
±
±
±
±
±
±
±
±
±
±
±
7.1
0.43a
0.28
0.07a
0.48
2.6
95
4a
0.15
0.30
0.5a
52-922 **
5-17
6-38
97-934
1,600-3,200
30-73
0.5-1.0
14-64
Harvest 2
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
<2.00
68.9
<0.12
<0.01
0.11
0.61
<0.01
11.2
71.4
6,755
68.8
1.76
1.60
<0.28
<0.09
23.2
±
±
±
±
±
±
±
±
±
±
11.6
0.04
0.08
1.0
8.9
732
9.2
0.26
0.31
3.4
<2.00
46.3
10.9
<0.01
0.16
0.16
<0.01
7.70
59.7
3,709
119
4.84
1.27
<0.28
<0.09
33.7
±
±
±
±
±
±
±
±
±
±
±
5.8a
2.9a
0.12
O.OSa
1.79
6.8
161a
7
1.25a
0.36
6.2
<2.00
37.2
7.71
<0.01
0.05
0.14
<0.01
9.22
56.5
4,188
271
5.33
1.11
<0.28
<0.09
22.9
±
±
±
±
±
±
±
±
±
±
±
9.5a
2.55a
0.06
O.lla
2.33
12.7
585a
62a
2.04a
0.53
5.3
<2.00
184
4.14
<0.01
0.04
<0.01
0.1
11.4
75.4
4,552
326
6.79
0.48
<0.28
<0.09
48.3
±
±
±
±
±
±
±
±
±
±
±
13.0a
0.58a
0.09
0.1
3.4
18.2
600a
9a
1.70a
0.34a
12.1a
<2.00
28.2
13.2
<0.01
0.06
0.14
<0.01
10.0
56.6
4,321
320
3.20
1.03
<0.28
<0.09
24.0
±
±
±
±
±
±
±
±
±
±
±
12.6s.
3.7a
0.08
O.lOa
4.5
17.0
682a
73a
1.56
0.40
11.1
<2.00
63.3
<0.12
<0.01
0.13
0.45
<0.01
9.06
62.4
5,678
75
1.64
4.82
<0.28
<0.09
42.2
±
±
±
±
±
±
±
±
±
±
5.3
0.10
0.02a
0.31
1.4
287a
5
0.12
0.94a
3. la
<2.00
39.8
2.98
<0.01
0.12
0.07
<0.01
6.75
54.8
5,542
119
1.82
1.73
<0.28
<0.09
22.1
±
±
±
±
±
±
±
±
±
±
±
3.4a
0.24
0.11
O.OSa
0.29
2.6
23 la
9
0.34
0.24
1.9
52-922 **
5-17
6-38
97-934
1.600-3.200
30-73
0.5-1.0
14-64
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-22
-------
Appendix B: Spent Foundry Sand Characterization Data
Element
Control
IGS-1
IGS-2
AGS-1
AGS-2
NBS-1
NBS-2
Sufficiency Range
(Jones et al., 1991)
Harvest 3
Al
B
Ba
Be
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Ni
Pb
V
Zn
<2.00
43.1
<0.12
<0.01
0.06
0.23
<0.01
4.29
57.7
5,092
57.3
1.03
0.75
<0.28
<0.09
13.8
±
±
±
±
±
±
±
±
±
±
1.9
0.07
0.03
0.63
4.4
511
2.6
0.22
0.06
0.6
<2.00
43.3
17.2
<0.01
0.09
0.04
<0.01
6.37
51.8
4,373
99.4
5.50
1.04
<0.28
<0.09
34.5
±
±
±
±
±
±
±
±
±
±
±
6.9
2.8a
0.07
0.07a
0.90a
4.0
387
22.0a
1.66
0.15a
5.5a
<2.00
44.2
16.7
<0.01
0.09
0.11
<0.01
7.75
59.3
4,698
221
6.39
1.14
<0.28
<0.09
29.6
±
±
±
±
±
±
±
±
±
±
±
5.3
4.1a
0.09
0.08a
1.06a
7.8
522
23a
0.94a
0.29a
4.5a
<2.00
96.4
3.34
<0.01
0.25
<0.01
0.1
8.80
61.7
3,719
374
5.97
<0.05
<0.28
<0.09
48.9
±
±
±
±
±
±
±
±
±
±
11. 3a
0.50
0.16
0.1
l.20a.
6.3
359a
17a
l.Oa
4.2a
<2.00
35.3
22.5
<0.01
0.10
0.16
<0.01
9.15
56.5
5,062
280
3.96
1.01
<0.28
<0.09
33.4
±
±
±
±
±
±
±
±
±
±
±
5.4
4.7a
0.09
0.02a
0.58a
5.5
864
14a
0.18a
0.08a
6.4a
<2.00
44.3
<0.12
<0.01
0.04
0.24
<0.01
4.86
45.3
5,313
40.5
0.75
3.71
<0.28
<0.09
31.8
±
±
±
±
±
±
±
±
±
±
2.3
0.05
0.04
0.33
1.6a
236
4.7
0.52
0.23a
2.4a
<2.00
34.5
<0.12
<0.01
0.10
<0.01
<0.01
2.56
38.0
4,405
61.5
0.25
0.94
<0.28
<0.09
15.8
±
±
±
±
±
±
±
±
±
2.5
0.07
0.79a
1.8a
696
6.7
0.51
0.12
1.0
52-922 **
5-17
6-38
97-934
1,600-3,200
30-73
0.5-1.0
14-64
* Average value of four replicates ± standard deviation.
** Non-essential element
"a" values indicate a significant difference than the control (P<0.05).
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
Note: AGS = aluminum green sand, IGS= iron green sand, NBS = steel no-bake sand
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-23
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-23. Lettuce Tissue Elemental Content for 10 Spent Foundry Sands and a Silica Sand Control
Plant Nutrient Tissue Adequacy Levels, Elemental Normal and Toxic Levels
Element
As
B
Ba
Ca
Cd
Co
Cu
Fe
K
Mg
Mn
Mo
N
Na
Ni
P
S
Zn
units
mgkg-1
mg kg'1
mgkg-1
gkg-1
mg kg'1
mgkg-1
mg kg'1
mgkg-1
gkg-1
gkg-1
mg kg'1
mgkg-1
gkg-1
gkg-1
mg kg'1
gkg-1
gkg-1
mg kg'1
Control
1.43
142
0.676
2.09
<0.2
2.82
43.6
201
24.5
4.52
141
<2
49.8
0.661
5.68
9.70
3.05
91.0
Spent Foundry Sands
1
<1
75.9
1.10
4.53
0.72
<2
34.2
148
40.9
2.49
131
<2
31.7
17.4
<2
4.23
3.95
55.1
3
<1
57.1
1.68
2.65
0.413
<2
15.9
89.6
34.2
1.99
65.2
3.94
27.5
28.9
<2
4.83
2.48
94.8
4
<1
101
1.18
4.66
0.387
<2
21.3
140
24.2
2.97
92.3
<2
21.3
6.33
<2
3.84
1.71
28.2
10
<1
57.1
1.16
3.64
1.14
<2
38.9
215
30.6
2.63
93.8
3.04
22.5
7.80
10.3
3.37
1.92
29.1
16
<1
82.9
1.31
4.73
0.29
<2
16.2
176
41.0
2.99
102
<2
31.9
9.85
2.92
5.20
2.75
36.8
20
<1
94.6
1.03
3.69
0.485
<2
10.4
220
38.6
2.53
118
7.18
50.8
29.7
<2
3.84
4.68
64.1
24
<1
78.7
1.68
3.26
1.88
<2
21.7
215
36.1
2.30
72.2
<2
33.0
21.2
<2
5.82
3.18
63.5
25
<1
56.3
1.09
3.21
0.368
<2
19.4
98.1
37.7
1.55
65.6
3.52
26.4
28.0
<2
4.10
2.68
68.2
28
<1
65.4
0.786
3.83
0.393
<2
21.8
271
38.7
2.33
99.2
<2
33.5
20.8
<2
4.05
3.38
63.8
40
<1
75.3
1.13
3.85
0.412
<2
18.3
135
41.9
2.06
45.9
<2
37.3
19.2
<2
4.75
3.78
99.0
Nutrient
Adequate
7-75c
5-30d
5-30a
50-150d
14-35d
3-10d
30-3003
0.25-5.03
17-50d
2.0-5.0d
1.5-5.0d
27-1503
Normal
Toxic
Range
1.0-1.7"
0.1-2.4e
0.02-1.03
0.1-5.03
5-20a
>100b
>500a
5-30a
15-503
20-1003
>800b
10-503
10-1003
>500C
Note: Other elements evaluated were below the detection limit, including Be, Tl, and V (<0.2 mg kg"1) and Cr, Pb, Sb, and Se (<2 0.2 mg kg"1).
a Kabata-Pendias, 2001
b Adriano,2001
0 Pais and Jones, 1997
d Johnson etal., 2000
e Bowen, 1979
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-24
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-24. Total Element-Specific Concentrations in Eiseniafetida After 28 Days in the Spent Foundry Sand Blends
(Average Concentrations of Four Replicates in mg kg * Dry Biomass)
Sample
Ratio
(%)
Al
B
Ba
Be
Ca
Cd
Co
Cr
Cu
Fe
Mg
Mn
Mo
Na
Ni
Pb
V
Zn
Control
100
<256.3
<34.6
<11.0
<0.5
3,310.6
1.3
1.8
<2.1
10.2
<307.2
<51.6
20.9
<1.9
3,181.8
<1.8
<49.7
<4.1
43.2
AGS-1
10
<256.3
<34.6
<11.0
<0.5
2,206.0
0.7
1.2
<2.1
<4.4
<307.2
<51.6
19.8
<1.9
3,083.5
<1.8
<49.7
<4.1
<38.6
30
<256.3
<34.6
<11.0
<0.5
2,938.0
0.7
0.8
<2.1
<4.4
<307.2
<51.6
22.2
<1.9
2,748.0
<1.8
<49.7
<4.1
<38.6
50
<256.3
<34.6
<11.0
<0.5
2,631.1
1.1
1.3
<2.1
7.6
<307.2
<51.6
20.8
<1.9
2,719.3
<1.8
<49.7
<4.1
45.1
AGS-2
10
<256.3
<34.6
<11.0
<0.5
2,839.1
1.6
1.9
<2.1
10.5
<307.2
<51.6
24.3
<1.9
2,844.3
<1.8
<49.7
<4.1
<38.6
30
369.0
<34.6
<11.0
<0.5
2,847.8
1.5
1.7
<2.1
7.2
<307.2
<51.6
18.6
<1.9
2,506.4
<1.8
<49.7
<4.1
41.1
50
662.7
<34.6
<11.0
<0.5
2,621.5
1.3
1.6
<2.1
7.1
<307.2
<51.6
23.8
<1.9
2,787.0
<1.8
<49.7
<4.1
43.8
IGS-1
10
<256.3
<34.6
<11.0
<0.5
2,292.0
1.2
1.6
<2.1
6.9
<307.2
<51.6
15.6
<1.9
2,712.2
<1.8
<49.7
<4.1
<38.6
30
<256.3
<34.6
<11.0
<0.5
2,100.9
1.0
1.3
<2.1
6.1
<307.2
<51.6
13.3
<1.9
2,463.3
<1.8
<49.7
<4.1
38.8
50
<256.3
<34.6
<11.0
<0.5
1,445.6
0.7
0.9
<2.1
<4.4
<307.2
<51.6
17.4
<1.9
2,265.8
<1.8
<49.7
<4.1
<38.6
IGS-2
10
<256.3
<34.6
<11.0
<0.5
2,779.1
0.9
1.1
<2.1
<4.4
<307.2
<51.6
16.3
<1.9
2,635.0
<1.8
<49.7
<4.1
40.5
30
<256.3
<34.6
<11.0
<0.5
2,632.0
1.2
1.5
<2.1
7.4
<307.2
<51.6
15.8
<1.9
2,770.9
<1.8
<49.7
<4.1
<38.6
50
<256.3
<34.6
<11.0
<0.5
2,312.2
0.8
1.2
<2.1
<4.4
<307.2
<51.6
16.3
<1.9
3,024.1
<1.8
<49.7
<4.1
<38.6
NBS
10
<256.3
<34.6
<11.0
<0.5
2,170.6
1.5
1.7
<2.1
8.6
<307.2
<51.6
17.3
<1.9
2,428.7
<1.8
<49.7
<4.1
53.4
30
<256.3
<34.6
<11.0
<0.5
1,585.2
0.8
1.1
<2.1
6.5
<307.2
<51.6
13.8
<1.9
2,677.6
<1.8
<49.7
<4.1
42.0
50
<256.3
<34.6
<11.0
<0.5
2,561.3
1.0
1.8
<2.1
7.0
<307.2
<51.6
17.5
<1.9
2,779.2
<1.8
<49.7
<4.1
45.1
BGS
10
<256.3
<34.6
<11.0
<0.5
2,529.4
1.4
1.3
<2.1
104.1
<307.2
<51.6
6.9
<1.9
2,380.1
<1.8
<49.7
<4.1
75.9
Entries in italics with the less than symbol (<) identify samples that were below the quantitative detection limit shown for that element.
For brass green sand (BGS), the earthworm biomass was insufficient to run metal analyses on 30% and 50% BGS blends.
Note: ASG = aluminum green sand, IGS= iron green sand, NBS = steel no-bake sand
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-25
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-25. Particle-Size Distribution, U.S. Department of Agriculture Textural Class
and Bulk Density for Spent Foundry Sand
FINa
1
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
% Sand (0.05-2 mm)
82.7
79.3
94.6
87.2
98.3
99.9
94.8
98.6
99.8
82.9
98.9
89.1
91.7
85
95.1
84.6
94
90.5
91.6
99
89.9
89.6
92.1
88.1
89.3
76.6
87.7
98.1
89.5
97
90.1
99.1
89
86.7
82.9
86.6
97.7
91
89.6
100
93.3
87.9
% Silt (2-50 fun)
7.7
9.4
2.1
3.6
1.7
0.1
2.3
0.8
0.2
7.0
1.1
2
1.1
5
1
4.3
0.5
4.1
0.9
0
5.7
1.8
3.7
4.6
3.5
16.9
3.9
1.9
3.9
2.1
3.8
0.9
4.5
5.3
7.5
4.9
1.6
2.2
2.8
0
9
4.1
% day (<2 um)
9.6
11.3
3.3
9.2
0
0
2.9
0.6
0
10.1
0
8.9
7.2
10
3.9
11.1
5.5
5.4
7.5
1
4.4
8.6
4.2
7.3
7.2
6.5
8.4
0
6.6
0.9
6.1
0
6.5
8
9.6
8.5
0.7
6.8
7.6
0
4.7
8
Textural Class
Loamy sand
Sandy loam
Sand
Loamy sand
Sand
Sand
Sand
Sand
Sand
Loamy sand
Sand
Loamy sand
Sand
Loamy sand
Sand
Loamy sand
Sand
Sand
Sand
Sand
Sand
Sand
Sand
Loamy sand
Sand
Loamy sand
Loamy sand
Sand
Sand
Sand
Sand
Sand
Sand
Loamy sand
Loamy sand
Loamy sand
Sand
Sand
Sand
Sand
Sand
Loamy sand
Bulk Density
(g cm 3)
1.60
1.57
1.66
1.61
1.66
1.66
1.66
1.66
1.66
1.59
1.66
1.62
1.66
1.6
1.66
1.58
1.66
1.66
1.66
1.66
1.66
1.63
1.66
1.65
1.66
1.64
1.63
1.66
1.66
1.66
1.66
1.68
1.65
1.63
1.6
1.62
1.66
1.66
1.66
1.66
1.66
1.63
a FIN = foundry identification number; see Table 2-2 for details.
No data were available for FIN 3.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-26
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-26. Pore Water Content (mg kg"1) of Spent Foundry Sand Measured in a 1:1 Deionized Water: Spent Foundry Sand Extraction
PQL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Al
0.2
0.1
0.1
39.8
0.1
1,720
0.1
0.1
0.248
0.1
0.225
75.9
0.1
0.782
437
3.89
0.1
1,107
55.5
111
0.982
0.1
0.1
1.79
145
1,847
0.1
0.1
65.1
9.9
74.3
0.1
266
0.322
0.1
0.1
1,116
366
37.6
1,418
841
0.1
193
264
As
0.02
0.01
0.01
0.054
0.01
0.162
0.01
0.01
0.01
0.024
0.01
0.033
0.01
0.022
0.109
0.066
0.01
0.059
0.048
0.01
0.01
0.01
0.01
0.071
0.01
0.116
0.082
0.01
0.019
0.049
0.047
0.01
0.055
0.01
0.01
0.086
0.124
0.09
0.021
0.089
0.074
0.01
0.076
0.059
B
0.1
0.519
0.135
1.1
1.37
4
3.96
0.295
0.172
0.208
0.273
4.42
0.393
0.534
42.2
0.36
0.267
0.449
0.598
0.357
1.67
0.118
0.206
0.428
0.531
1.49
0.56
0.828
0.842
0.208
0.7
3.54
3.31
0.345
0.223
0.888
0.786
0.852
0.134
1.17
0.546
0.138
0.416
1.13
Ba
0.02
0.06
0.082
0.778
0.083
2.85
0.376
0.159
0.01
0.01
0.01
0.049
0.01
0.026
0.486
0.01
0.022
0.119
0.05
0.578
0.01
0.01
0.01
0.01
0.068
4.5
0.01
0.036
0.068
0.336
0.032
0.05
0.16
0.03
0.03
0.205
1.21
0.136
0.104
0.979
0.465
0.01
0.067
0.12
Be
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
Ca
0.02
81.8
25.2
4.37
211
94
60
261
5
14.8
65
27.2
31.1
25.6
44.8
16.4
38.2
107
17.5
12.2
45.7
4.5
49.6
25.3
20.8
124
36.9
47
32.5
8.87
26.3
245
35.6
4.73
41.9
44.6
84.9
35.3
7.8
80.7
75.1
16.8
16
27.4
Cd
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.023
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
Co
0.02
0.01
0.01
0.01
0.01
0.47
0.01
0.029
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.054
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.033
0.01
0.01
0.048
0.026
0.01
0.01
0.01
Cr
0.02
0.01
0.01
0.01
0.01
0.29
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.054
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.083
0.01
0.01
0.01
0.245
0.01
0.01
0.039
0.01
0.01
0.01
0.059
0.01
0.01
0.09
0.056
0.01
0.01
0.01
Cu
0.02
0.027
0.01
0.01
0.01
0.176
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.107
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.083
0.01
0.01
0.01
0.832
0.07
0.01
0.012
0.469
1.14
0.01
0.087
0.01
0.217
1.7
0.081
0.01
0.01
0.09
Fe
0.02
1.14
0.01
7.75
0.01
402
0.01
0.01
0.01
0.01
0.01
15.6
0.01
0.242
141
0.935
0.04
251
11.7
46.5
0.206
0.01
0.01
0.376
45.7
323
0.01
0.01
28.5
8.22
12.7
0.01
51.7
0.57
0.57
0.01
252
98.6
9.02
272
158
0.01
37.5
46.9
K
0.02
22.1
37.7
13.9
54.6
24.8
25.3
27.3
19.6
17.6
16.9
45.9
15.2
50.6
25.3
25.9
29
22
30.8
16.4
32
15
34.1
27.8
23.6
39.3
27.8
53
19.3
854
25.6
19.1
23.1
12.5
38.2
35.3
53.3
19.1
130
90.6
30
10.6
17.2
38.6
Mg
0.02
5.65
29.4
11.8
14.9
235
64.2
25.9
2.69
1.1
0.993
13.2
4.65
2.11
103
2.36
13.5
187
12.3
24.2
4.24
0.464
9.7
1.84
20
313
1.17
18.6
29.6
1.652
14
27.4
54.1
1.62
15.3
74.4
156
63.6
8.42
261
165
2.83
32.3
43.2
Mn
0.02
0.01
0.01
0.01
0.114
1.58
0.01
1.78
0.01
0.01
0.01
0.059
0.047
0.01
1.02
0.022
0.01
0.01
0.034
0.298
0.02
0.022
0.404
0.01
0.01
0.777
0.01
0.01
0.117
0.15
0.054
8.64
0.126
0.154
0.127
0.162
0.817
0.222
0.146
3.02
0.203
0.104
0.093
0.239
Mo
0.02
0.108
0.041
0.252
0.024
0.246
2.85
0.01
0.01
0.047
0.044
0.229
0.01
0.01
0.116
0.156
0.01
0.119
0.079
0.126
0.111
0.048
0.092
0.13
0.105
0.136
1.12
0.067
0.117
0.518
0.104
0.01
0.089
0.054
0.063
0.145
0.183
0.132
0.143
0.55
0.099
0.01
0.116
0.207
Na
0.02
456
165
321
167
577
56
0.635
116
66.4
10.8
480
2.17
242
337
357
120.5
461
279
271
429
68.7
281
498
215
659
407
374
155
123
468
147
345
1.71
180
526
531
237
594
472
287
10.9
204
383
Ni
0.02
0.01
0.01
0.01
0.01
0.219
0.01
0.373
0.01
0.01
0.01
0.01
0.01
0.01
0.059
0.01
0.01
0.01
0.01
0.042
0.01
0.01
0.01
0.01
0.025
0.135
0.01
0.021
0.03
0.402
0.01
0.01
0.035
0.01
0.01
0.032
0.066
0.023
0.202
2.9
0.046
0.01
0.01
0.085
P
0.02
0.053
0.026
0.841
0.02
0.543
0.278
0.017
0.061
0.514
0.114
0.08
0.01
0.01
1.62
0.224
0.391
0.345
4.39
0.169
0.106
0.01
0.037
0.249
0.585
1.02
0.218
0.071
0.46
4.89
1.17
0.01
0.406
0.036
0.661
0.453
3.89
0.712
1.54
8.84
0.598
0.01
0.522
0.782
Pb
0.05
0.025
0.025
0.025
0.025
0.148
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.07
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.205
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.025
0.169
0.087
0.025
0.025
0.025
S
0.02
4.31
84
156
107
198
147
376
43.1
33.4
44.2
221
28.1
167
125
256
115
261
171
117
248
34.1
291
343
97.4
223
300
275
104
14.5
368
518
204
1.2
66.6
264
238
101
27
138
122
12.2
68.7
241
Se
0.02
0.01
0.01
0.01
0.01
0.022
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.039
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.647
0.01
0.01
0.01
0.01
0.024
0.01
0.01
0.01
0.01
Tl
0.02
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
V
0.02
0.07
0.07
0.07
0.07
0.054
0.07
0.07
0.07
0.035
0.07
0.07
0.028
0.07
0.1
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.07
0.099
0.07
0.07
0.016
0.186
0.07
0.07
0.039
0.07
0.07
0.07
0.084
0.053
0.07
0.063
0.087
0.07
0.07
0.03
Zn
0.02
0.07
0.07
0.123
0.07
1.54
0.07
0.045
0.07
0.07
0.07
0.096
0.07
0.07
0.426
0.07
0.07
0.271
0.031
0.3
0.07
0.07
0.07
0.07
0.309
1.88
0.07
0.07
0.176
0.071
0.023
0.065
0.07
0.07
2.25
0.617
1.44
0.216
0.055
0.819
0.273
0.07
0.172
0.22
Entries in italics were below the practical quantitation limit (PQL) and were calculated as 0.5 times the PQL.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-27
-------
Appendix B: Spent Foundry Sand Characterization Data
Table B-27. Tentative Gas Chromatography-Mass Spectrometry Characterization
of Products in the Pyrolyzates
Peak
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Retention Time
6:10
12:47
13:07
15:11
16:22
17:18
17:20
19:53
20:47
20:51
20:55
21:52
23:25
24:25
24:27
24:51
26:42
26:48
27:13
27:32
27:52
28:03
28:45
28:52
29:01
30:28
30:37
31:30
31:33
32:08
32:09
32:50
33:25
33:35
34:10
35:25
36:40
37:05
38:54
Assignment
Ethylbenzene
1 -Ethyl-2-methylbenzene
Phenol
1 -Propynylbenzene
2-Methylphenol
1 -Ethoxy-4-methylbenzene
3- and 4-Methylphenol
2,5-Dimethylphenol
2,3-Dimethylphenol
Naphthalene
2-Ethenyl-l,3,5-trimethylbenzene
2-Methyl-8-propyldodecane
2-Ethyl-4-methylphenol
2-Methylnaphthalene
(E)-5-Tetradecene
1 -Methylnaphthalene
2,6-Dimethylheptadecane
2-Ethenylnaphthalene
2-Ethylnaphthalene
1 ,7-Dimethylnaphthalene
1 ,2-Dimethylnaphthalene
1 ,8-Dimethylnaphthalene
Acenaphthylene
Tridecanol
4-Methyloctadecane
Dibenzofuran
1 ,6,7-Trimethylnaphthalene
3 -Ethyl- 1 -methylnaphthalene
1 ,4,5-Trimethylnaphthalene
Fluorene
1 ,4,6-Trimethylnaphthalene
l,2-Dimethyl-4-(phenylmethyl)-benzene
1 -Methyl-7-( 1 -methylethyl)-naphthalene
2,6,10-Trimethylpentadecane
1 ,6-Dimethyl-4-( 1 -methylethyl)-naphthalene
1-Nonadecane
Anthracene
2,6,10,14-Tetramethylhexadecane
(Z)-9-Octadecenal
Molecular Weight
106
118
94
116
108
136
108
122
122
128
146
226
136
142
196
142
268
154
156
156
156
156
152
200
268
168
170
170
170
166
170
196
184
254
198
266
178
282
266
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-28
-------
Appendix B: Spent Foundry Sand Characterization Data
14
10
22
16
19
27 32
14
10
16
20
J
30
10
14
22
Time (min)
Figure B-l. Gas chromatogram of pyrolysis products from fresh green sand
at a) 500°C, b) 750°C, and c) 1000°C.
The fresh green sand contained 92% silica sand, 4% sodium bentonite, 2% calcium bentonite, and 2% seacoal
(w/w). Assignments of the labeled peaks are shown in Table B-27.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
B-29
-------
Appendix B: Spent Foundry Sand Characterization Data
References
Adriano, D.C. 2001. Trace Elements in the Terrestrial Environments: Biogeochemistry,
Bioavailability, and Risks of Heavy Metals. 2nd ed. New York: Springer-Verlag.
Bowen, H.J.M. 1979. Environmental Chemistry of the Elements. New York: Academic Press.
Johnson, G.V., R.R. Raun, H. Zhang, and F.A. Hattey. 2000. Oklahoma Soil Fertility Handbook.
Division of Agricultural Sciences and Natural Resources, Oklahoma State University,
Stillwater, OK.
Kabata-Pendias, A. 2001. Trace Elements in Soils and Plants. 3rd ed. Boca Raton, FL: CRC
Press.
Jones, Jr., J.B., B. Wolf, and H.A. Mills. 1991. Plant Analysis Handbook: A Practical Sampling,
Preparation, Analysis, and Interpretation Guide. Athens, GA: Micro-Macro Publishing,
Inc.
Pais, I, and J.B. Jones, Jr. 1997. The Handbook of Trace Elements. Boca Raton, FL: St. Lucie
Press.
U.S. EPA (Environmental Protection Agency). 2007. Method 3051 A: Microwave Assisted Acid
Digestion of Sediments, Sludges, Soils, and Oils, Test Methods for Evaluating Solid
Wastes, Physical/Chemical Methods, SW-846. U.S. Environmental Protection Agency,
Office of Hazardous Waste, Washington, DC. Available at
http://www.epa.gov/osw/hazard/testmethods/sw846/pdfs/305 la.pdf (accessed 19 March
2012).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications B-30
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
Appendix C
Explanation of USD A Eco Screening Values
for Cu, Ni, and Zn
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
Appendix C
Explanation of USDA Eco Screening Values for Cu, Ni, and Zn
Inspection of Cu, Ni, and Zn concentrations in silica-based iron, steel and aluminum
SFSs reveals a few samples with concentrations higher than the 95th percentile of U.S. and
Canadian background soils. The 95th percentile was used to represent reasonable maximum
background levels in soils, which have caused no known adverse effects in the environment
(Scheckel et al., 2009; Chaney 2010). This does not mean that the 95th percentile of background
is the beginning of potential toxicity; rather, that without more evaluation, we are not
comfortable suggesting that the higher levels are free from concern about possible adverse
effects.
The issue of some Ecological-Soil Screening Levels (Eco-SSLs) that were lower than
some SFS samples and considerably lower than 95th percentile soil background levels has been
noted. We have discussed the source of the Eco-SSL values, which are based on the worst case
for each element. For Cu, Ni, and Zn, the source is acidic soils, low in clay, Fe, Mn and Al
oxides, and organic matter, as well as the fresh addition of soluble metal salts (which have
immediate, near 100% bioavailability, but react with soils over time to less bioavailable forms)
in toxicological tests. USDA argues that these conditions are more severe than the environment
where SFS and manufactured soils containing SFS would be used. Thus, alternative, more
realistic limits were developed for Cu, Ni, and Zn, and an explanation of the derivation of these
less conservative levels than the 95th percentile and Eco-SSL values was required.
It is important to note that the matrix of manufactured soils containing SFS is near neutral
pH, with organic matter (typically 5-10% or higher) and balanced fertility ready to be sold as a
topsoil. Furthermore, if soil pH is allowed to fall to below 5.5 (for Mn2+) or 5.2 (for A13+) over
time due to acidic rainfall and/or use of ammoniacal fertilizers, the soil will eventually become
Mn, or Al and Mn phytotoxic and prevent growth of garden crops and even lawn grasses. Many
garden crops fail at pH 5.5, which is still well above the worst case of the EcoSSL baseline for
metals (pH 4.0). Most garden crops perform much better at a pH ranging from 6.5 to 7 than at
lower pH; gardeners are advised to maintain soil pH in this range.
Copper: Cu is strongly bound by soil organic matter even at relatively low pH. Copper
phytotoxicity has occurred in locations where mine wastes were dispersed, or where excessive
fungicidal sprays were applied to trees growing in very strongly acidic, sandy, low organic
matter soils. As shown in Table C-l, Cu levels in some agricultural soils have risen above the
geochemical background levels from long term applications of Cu fertilizers and Cu-pesticides.
Some peat soils require the addition of as much as 100 kg Cu ha"1 to achieve adequate Cu
fertility for vegetable crops susceptible to Cu deficiency. Because field phytotoxicity of Cu to
sensitive crops has not been observed until acidic sandy soils approach well over 200 mg Cu
kg"1, we conclude that 200 mg Cu kg"1 in a land-applied byproduct such as SFS is not a source of
concern for ecological receptors.
Nickel: Soil Ni is transformed to insoluble solid phases at soil pH levels appropriate for
crop production. Even added soluble Ni salts rapidly convert to insoluble solids, and those
become decreasingly bioavailable over time as additional reactions occur with silicates (Scheckel
and Sparks, 2001). Nickel has been shown to be phytotoxic in highly acidic soils surrounding Ni
Risk Assessment of Spent Foundry Sands in Soil-Related Applications C-l
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
smelters in Canada, but not in soils with reasonable pH management. Natural serpentine soils
contain 1000-2000 mg Ni kg"1 and are seldom phytotoxic until pH drops to pH 5.5 or below;
phytotoxicity is readily reversed by simple addition of limestone (Kukier and Chaney, 2004;
Siebielec et al., 2007). Because field phytotoxicity from Ni has been observed only when acidic
soils exceed about 1000 mg kg"1, we conclude that 200 mg Ni kg"1 in a land-applied byproduct
such as SFS is not a source of concern for soil fertility or ecological receptors.
Zinc: Zn is a common soil contaminant because of its widespread commercial use in
products and farm and garden implements. Urban emissions have raised levels of soil Zn in city
centers as well. Zinc toxicity is the most common phytotoxic effect observed in the environment
because of these uses (Chaney, 1993). Most cases of Zn phytotoxicity involved mine wastes, Zn
smelter emissions, burned rubber tires, or pesticide sprays where high levels of Zn accumulated
over time, and the soils were strongly acidic or very highly contaminated. Alkaline soils can
contain over 1000 mg Zn kg"1 with no adverse effects, and even as high as 10,000 mg Zn kg"1
without harming plants or wildlife (USEPA, 2007). An example of home garden metals levels
from the general Baltimore area was published by Mielke et al. (1983) (Table C-2). When some
of the highly Pb- and Zn-contaminated soils were used in pot experiments to test uptake of
metals by lettuce, even soil with 3,490 mg Zn, 5,210 mg Pb and 269 mg Cu kg"1 did not cause
any adverse effects on the lettuce (Sterrett et al., 1993). A plant response test with Montreal,
Quebec, Canada soils similarly found no adverse effects of substantial soil Cu and Zn levels on
plant growth (Tambasco et al., 2000; Ge et al., 2002). Comparing SFS to urban soils shows that
use of manufactured soils in urban gardens will usually provide lower soil Zn levels than
background urban soils. The recognized adverse effect of excessive soil Zn is phytotoxicity if
soil pH falls below 5.5 and especially below 5.0; simply incorporating agricultural limestone
corrects and prevents future Zn phytotoxicity. Added soluble Zn fertilizers react over time to
form solids or adsorbed species with lower phytoavailability such that additional Zn fertilizers
may be required after 5-10 years. Higher soil Zn levels provide a reservoir of plant-available Zn
that roots can access to obtain adequate Zn for plant growth and improve plant quality by
increasing plant Zn concentrations. However, plant accumulation of Zn to levels above about
400-500 mg kg"1 dry leaves causes visibly evident phytotoxicity, but ruminant livestock tolerate
diets with at least 500 mg salt Zn kg"1, and monogastric animals tolerate higher dietary soluble
Zn. Plant storage tissues (grain, fruits, edible roots) contain considerably lower Zn levels than do
leaves. Thus the suggested investigatory limit of 300 mg Zn kg"1 in land-applied SFS is
protective of soil fertility and ecological receptors.
Table C-l. Comparison of USD A Recommended maximum concentration of Cu, Ni, and
Zn in SFS before additional investigation is required with Eco-SSL and 95th percentile of
background U.S. and Canadian soils
(mg kg * DW)
Element
Cu
Ni
Zn
SFS 95th
Percentile
107
102
72.1
SFS
Maximum
137
117
245
USDA
Recommendations
200
200
300
EPA Eco-
SSL
(Plants)
70
38
160
95th
Percentile
(Smith)
30.1
37.5
103
95th
Percentile
(Holmgren)
216
154
170
SOURCES: Holmgren et al. (1993); Smith et al. (2005); U.S. EPA (2007a, 2007b, and 2007c)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
C-2
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
Table C-2. Concentrations of Zn, Cu and Ni extractabSe with 1.0 M HNOs
in 422 Baltimore, MD, area gardens
(mg kg-1 DW)
Element
Cu
Ni
Zn
Minimum
0.7
0.5
0.3
Median
17.2
2.8
92
Mean
25
4.9
211
90th Pereentile
64.4
8.4
521
Maximum
96.7
53.4
4,880
SOURCE: Mielke et al. (1983)
References
Chancy, R.L. 1993. Zinc phytotoxicity. pp. 135-150. In A.D. Robson (ed.) Zinc in Soils and
Plants. Kluwer Academic Publ., Dordrecht.
Chancy, R.L. 2010. Cadmium and zinc. Chapter 17. pp. 409-439. In P. Hooda (Ed.) Trace
Elements in Soils. Blackwell Publ., Oxford, UK.
Ge, Y., P. Murray, S. Sauve and W. Hendershot. 2002. Low metal bioavailability in a
contaminated urban site. Environ. Toxicol. Chem. 21:954-961.
Holmgren, G.G.S., M.W. Meyer, R.L. Chancy and R.B. Daniels. 1993. Cadmium, lead, zinc,
copper, and nickel in agricultural soils of the United States of America. J. Environ. Qual.
22:335-348.
Kukier, U. and R.L. Chancy. 2004. In situ remediation of Ni-phytotoxicity for different plant
species. J. PlantNutr. 27:465-495.
Mielke, H.W., J.C. Anderson, K.J. Berry, P.W. Mielke, R.L. Chancy and M.L. Leech. 1983.
Lead concentrations in inner city soils as a factor in the child lead problem. Am. J. Public
Health 73:1366-1369.
Scheckel, K.G. and D.L. Sparks. 2001. Dissolution kinetics of nickel surface precipitates on clay
mineral and oxide surfaces. Soil. Sci. Soc. Am. J. 65:685-694.
Scheckel, K.G., R.L. Chancy, N.T. Basta and J.A. Ryan. 2009. Advances in assessing
bioavailability of metal(loid)s in contaminated Soils. Adv. Agron. 104:1-52.
Siebielec, G., R.L. Chancy and U. Kukier. 2007. Liming to remediate Ni contaminated soils with
diverse properties and a wide range of Ni concentration. Plant Soil 299:117-130.
Smith, D.B., W.F. Cannon, L.G. Woodruff, R.B. Garrett, R. Klassen, I.E. Kilbum, J.D. Horton,
H.D. King, M.B. Goldhaber, and J.M. Morrison. 2005. Major- and Trace-Element
Concentrations in Soils from Two Continental-Scale Transects of the United States and
Canada. Open-File Report 2005-1253. U.S. Department of the Interior, U.S. Geological
Survey, Reston, VA. Available at http://pubs.usgs.gov/of/2005/1253/pdf/OFR1253.pdf
(accessed 19 March 2012).
Tambasco, G., S. Sauve, N. Cook, M. McBride and W. Hendershot. 2000. Phytoavailability of
Cu and Zn to lettuce {Lactuca sativa) in contaminated urban soils. Can. J. Soil Sci.
80:309-317.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
C-3
-------
Appendix C: Explanation of USD A Eco Screening Values for Cu, Ni, and Zn
U.S. EPA (Environmental Protection Agency), 2007a. Ecological Soil Screening Levels for
Copper. Interim Final. OSWER Directive 9285.7-68. US Environmental Protection
Agency, Washington, DC. (February, 2007).
U.S. EPA (Environmental Protection Agency), 2007b. Ecological Soil Screening Levels for
Nickel. Interim Final. OSWER Directive 9285.7-76. US Environmental Protection
Agency, Washington, DC. (March, 2007).
U.S. EPA (Environmental Protection Agency), 2007c. Ecological Soil Screening Levels for Zinc.
Interim Final. OSWER Directive 9285.7-73. US Environmental Protection Agency,
Washington, DC. (June, 2007).
U.S. EPA (Environmental Protection Agency), 2007d. The Use of Soil Amendments for
Remediation, Revitalization and Reuse. EPA 542-R-07-013. US Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Washington, DC.
December.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications C-4
-------
Appendix D: Meteorological Data
Appendix D
Meteorological Data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix D: Meteorological Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix D: Meteorological Data
Appendix D
Meteorological Data
This analysis reflects 5 years of representative meteorological data, including surface data
and upper-air data. These data were obtained from 41 meteorological stations selected to
represent the nine general climate regions of the continental United States. These observational
data were processed and used as input to the home garden model and to the Industrial Source
Complex, Short-Term Model, Version 3 (ISCST3).1 Using the locations of the economic
feasibility areas and their associated meteorological stations, a subset of the national data were
extracted for use in the SFS analysis. This appendix describes the approach that was applied in
selecting the representative meteorological stations and describes how the meteorological data
were processed for use in the modeling.
D.I Meteorological Station Selection
Forty-one meteorological stations were chosen to represent the nine general climate
regions of the continental United States. The approach used the following three main steps:
1. Identify contiguous areas that are sufficiently similar, as defined by Bailey regions.
Bailey's ecoregions and subregions of the United States (Bailey et al., 1994) are used to
associate coverage areas with meteorological stations. This hierarchical classification
scheme is based primarily on rainfall regimes; subregions are delineated by elevation and
other factors affecting ecology.
2. Select one meteorological station to represent each contiguous area. The station selection
step considered the following parameters:
- Major National Weather Service (NWS) station preferred NWS stations are
expected to have high-quality equipment that is kept in good repair and is suitably
sited.
- Number of years of surface-level meteorological data available More years of
data provide a more realistic long-term estimate of air concentration and
deposition.
- Central location within the area. All other factors being equal, central locations
are more likely to be representative of the entire contiguous area because they
have the smallest average distance from all points in the region.
3. Identify the boundaries of the area to be represented by each meteorological station.
Thiessen polygons, which are created by a geographic information systems (GIS)
procedure that assigns every point on a map to the closest station, were used as the first
step in drawing the boundaries. The borders of adjacent areas that were in different
Bailey ecoregions were adjusted along the Bailey boundaries.
Table D-l lists the selected stations for the continental United States and Figure D-l
shows these stations and their boundaries.
1ISCST3 modeling was not performed specifically for this analysis. National ISCST3 modeling was performed to
support EPA's 503 biosolids program. The SFS analysis applies a subset of the national outputs to estimate
deposition impacts in SFS economic feasibility areas.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-l
-------
Appendix D: Meteorological Data
Table D-l. Surface-Level Meteorology Stations Used, by State
Station
Number
13963
23183
93193
23174
23234
94018
14740
12839
12842
13874
24131
94846
12916
13957
14764
14840
14922
13865
24033
03812
13722
24011
14935
23050
23169
24128
14820
13968
24232
14751
13739
14778
13880
13897
12960
24127
13737
14742
24233
03860
24089
Station Name
Little Rock/ Adams Field
Phoenix/Sky Harbor International Airport
Fresno /Air Terminal
Los Angeles /International Airport
San Francisco /International Airport
Boulder Airport
Hartford/Bradley International Airport
Miami/International Airport
Tampa/International Airport
Atlanta/ Atlanta-Hartsfield International
Boise/ Air Terminal
Chicago/O'Hare International Airport
New Orleans/International Airport
Shreveport/Regional Airport
Portland/International Jetport
Muskegon/County Airport
Minneapolis-St Paul/International Airport
Meridian/Key Field
Billings/Logan International Airport
Asheville/Regional Airport
Raleigh/Raleigh-Durham Airport
Bismarck/Municipal Airport
Grand Island/ Airport
Albuquerque/International Airport
Las Vegas/McCarran International Airport
Winnemucca/WSO Airport
Cleveland/Hopkins International Airport
Tulsa/International Airport
Salem/McNary Field
Harrisburg/Capital City Airport
Philadelphia/International Airport
Williamsport-Lycoming/County Airport
Charleston/International Airport
Nashville/Metro Airport
Houston/Intercontinental Airport
Salt Lake City /International Airport
Norfolk/International Airport
Burlington/International Airport
Seattle/Seattle-Tacoma International Airport
Huntington/Tri-State Airport
Casper/Natrona County International Airport
State
AR
AZ
CA
CA
CA
CO
CT
FL
FL
GA
ID
IL
LA
LA
ME
MI
MN
MS
MT
NC
NC
ND
NE
NM
NV
NV
OH
OK
OR
PA
PA
PA
SC
TN
TX
UT
VA
VT
WA
WV
WY
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
D-2
-------
Appendix D: Meteorological Data
Figure D-l. Meteorological stations and regions
For purposes of this discussion, the contiguous United States was divided into the
following sections: West Coast, Western Mountains, Desert Southwest, Gulf Coast, Southeast,
Middle Atlantic, Northeast, Great Lakes, and Central States. The process of selecting stations
and delineating the region assigned to each station is discussed in these sections.
D.I.I West Coast
The California coast is divided just north of Los Angeles. This northern section is
represented by the San Francisco International Airport (23234).
The southern California coast contains the Los Angeles basin south to the
California/Mexico border. This region is represented by the Los Angeles International Airport
(23174).
The California Central Valley Region, which encompasses the Sacramento Valley to the
north and the San Joaquin Valley to the south, is defined by the Coast Range and Diablo Range
to the west and the Sierra Nevada Mountains to the east. The valley extends south to the northern
rim of the Los Angeles basin. The region is represented by Fresno Air Terminal (93193).
The coastal half of Oregon includes the Pacific Coast, the Central Valley Region, and the
Great Sandy Desert, east to the Columbia Plateau. This region is represented by the station at
McNary Field in Salem, OR (24232).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
D-3
-------
Appendix D: Meteorological Data
The coastal half of Washington is bounded by the edge of the Humid Temperate Domain
to the east, the Washington/Canada border to the north, and the Columbia River to the south.
This region is represented by the Seattle-Tacoma International Airport (24233).
D.1.2 Western Mountains
The Boise Air Terminal (24131) in Idaho represents the northern Rocky Mountains.
Almost all of Nevada and southeastern Oregon are represented by the station at
Winnemucca WSO Airport (24128) in Nevada.
The Salt Lake Basin and the Great Divide Desert in Utah and Colorado are represented
by the station at Salt Lake City International Airport (24127) in Utah.
D.1.3 Desert Southwest
The Desert Southwest is defined by various deserts and mountain ranges. One
distinguishing feature is the transition between low desert in southern Arizona and high desert in
northern Arizona. The southern boundary of this section is the U.S./Mexico border.
Southern Arizona, New Mexico, and western Texas comprise a region of low desert that
is represented by the station at Phoenix/Sky Harbor International Airport (23183). The region is
bounded to the north between Phoenix and Prescott, AZ, along the southern edge of the
Columbia Plateau, which represents the transition from low to high desert.
Southeastern California, southern Nevada, and a small portion of northeastern Arizona
are represented by the station at Las Vegas/McCarran International Airport (23169). This region
is characterized by high desert.
The station at Albuquerque International Airport (23050) represents the mountainous
region of northern Arizona, most of New Mexico, and central Texas.
D.1.4 Gulf Coast
The Texas Gulf Coast is represented by the station at Houston Intercontinental Airport
(12960).
The Central Gulf Coast extends from western Louisiana through the Florida panhandle.
This entire region is part of the Outer Coastal Plain Mixed Forest Province. The station at New
Orleans International Airport (12916) in Louisiana was chosen to represent this region.
The West Coast of the Florida Peninsula is heavily influenced by the Gulf of Mexico,
which has warmer water than the Atlantic Ocean off the East Coast of the Florida Peninsula.
This region of the West Coast of Florida extends from the Florida Panhandle to the southern tip
of Florida. The station at Tampa International Airport (12842) was chosen to represent this
region.
D.1.5 Southeast
The Southeast section extends from the Atlantic coastal region of Florida and the Florida
Keys northward through Georgia and North and South Carolinas. This region has an extremely
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-4
-------
Appendix D: Meteorological Data
broad coastal plain, requiring it to be divided between the coastal region and more inland regions
for Georgia and South Carolina. This region also includes the inland areas of Louisiana,
Mississippi, and Alabama.
The southern tip of Florida includes the Everglades, which have been drained along the
Atlantic Coast to provide land for Miami, Ft. Lauderdale, West Palm Beach, and other coastal
cities. This region north to the Georgia border is represented by the station at Miami
International Airport (12839).
A long stretch of the Southeastern Atlantic Coast extends from the Georgia-Florida
border through Georgia, South Carolina, and the southern portion of North Carolina. The
boundary between the more forested coast and more agricultural inland area forms the western
boundary. The station at Charleston International Airport (13880) represents this region.
The Blue Ridge region is further inland in Georgia and South Carolina. The station at
Atlanta Hartsfield International Airport (13874) represents this region.
The inland areas of Alabama and Mississippi are represented by the station at Meridian
Key Field (13865), which is located in Mississippi near the Alabama border. This area extends
from the Central Gulf Coast region northward to southern Tennessee and westward to the
Mississippi River Valley in western Mississippi.
The inland portion of Louisiana and eastern Texas is part of the Coastal Plain. This
region extends northward to the Ouachita Mountains, which are just south of the Ozark Plateau
in Arkansas. The hill country in eastern Texas is included. This region is represented by the
station at Shreveport Regional Airport (13957) in Louisiana.
D.1.6 Middle Atlantic
The Middle Atlantic section includes coastal areas with bays, sounds, inlets, and barrier
islands; a broad coastal plain; and the southern Appalachian Mountains.
The northern portion of the coastal region of North Carolina, coastal Virginia, and the
Delmarva Peninsula is represented by the station at Norfolk International Airport (13737) in
Virginia.
The Piedmont region of North Carolina, South Carolina, and Virginia is just inland from
the coastal region. The station at Raleigh-Durham Airport (13722) in North Carolina represents
this region.
The southern Appalachian Mountains lie to the west of the Piedmont region of North
Carolina and Virginia. This region extends to the southwest to include a portion of western South
Carolina and northeastern Georgia and to the northeast to include the southeastern portion of
West Virginia. The station at Asheville Regional Airport (03812) in North Carolina was chosen
to represent this region.
The Appalachian Mountains of West Virginia and eastern Kentucky are represented by
the station at Huntington Tri-State Airport (03860) in West Virginia.
The inland region encompassing northern Virginia, part of Maryland, and eastern
Pennsylvania is composed of another section of the Appalachian Mountains. Boundaries are
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-5
-------
Appendix D: Meteorological Data
approximated by the Bailey's Central Appalachian Forest province. The station at
Harrisburg/Capital City Airport (14751) in Pennsylvania represents this region.
The area just to the north of the Chesapeake Bay northward through New Jersey, eastern
Pennsylvania, and New York City is characterized by the Eastern Broadleaf Forest (Oceanic)
Province in the Coastal Plain. The station at Philadelphia International Airport (13739) in
Pennsylvania represents this region.
D.1.7 Northeast
The Northeast section includes Maine and New England. This region is characterized by
forests to the north, large urban areas along the southern Coastal Plain, and the mountain ridges
and valleys of the northern Appalachian Mountains. This section is bounded by the Atlantic
Ocean on the east, the U.S.-Canada border on the north, and the coastal plain of the eastern Great
Lakes to the west.
The station at Bradley International Airport (14740) in Hartford, CT, represents the New
England region, which encompasses Connecticut, Massachusetts, Rhode Island, and a small
portion of Vermont, New Hampshire, and eastern New York.
Northern New England and Maine are represented by the station located at the
International Jetport (14764) in Portland, ME. This region includes Maine and most of New
Hampshire and Vermont.
The station at the International Airport (14742) in Burlington, VT, represents
northeastern New York, Vermont, New Hampshire, and western Maine.
The remainder of the northern Appalachian Mountains in New York and Pennsylvania is
represented by the station at Williamsport-Lycoming (14778) in Pennsylvania. This region is
bounded on the west by the Adirondack Mountains, just to the east of the coastal plain of Lake
Ontario.
D.1.8 Great Lakes
The Eastern Great Lakes divide the United States and Canada. On the U.S. side, the
western portion of New York, a small portion of Pennsylvania, and northeastern Ohio border the
eastern shores of Lake Ontario and Lake Erie. Mountains form the eastern boundary. The
western border is just inland from the western shore of Lake Erie. The station at Hopkins
International Airport (14820) in Cleveland, OH, represents this region.
The Lower Peninsula of Michigan is bordered by the Great Lakes on three sides. As
previously noted, the eastern portion along Lake Erie is represented by the station in Cleveland,
OH. The remainder of the Lower Peninsula of Michigan and the eastern portion of the Upper
Peninsula of Michigan are represented by the station at Muskegon County Airport (14840).
The western shore of Lake Michigan, which includes Green Bay, is formed by the
northeastern portion of Illinois, eastern Wisconsin, and part of the Upper Peninsula of Michigan.
Lake Superior forms the northern boundary of this region, and the western boundary is
formed by the hills to the east of the Wisconsin River and the Upper Mississippi River. Most of
Illinois, western Indiana, eastern Iowa, and northeastern Missouri are included in this region,
which is represented by the station at O'Hare International Airport (94846) in Chicago, IL.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-6
-------
Appendix D: Meteorological Data
D.1.9 Central States
This section includes the Central Lowlands (south of the Great Lakes), the Midwest, and
the Great Plains. The elevation for this section is generally lowest in the Mississippi Valley,
which extends through the Midwest and drains a large portion of the center of the continental
United States. This section also includes other major river valleys, including the Ohio,
Tennessee, and Missouri. This section is bordered on the east by the Appalachian Mountains, on
the west by the Rocky Mountains, on the north by the border with Canada, and on the south by
the Southeast section, Texas, and the Desert Southwest section.
One region includes western Kentucky, central and western Tennessee north of Memphis,
and southeastern Missouri east of the Ozark Plateau, southern Illinois, and southern Indiana. This
region is represented by the station at Nashville Metropolitan Airport (13897) in Tennessee.
A large region is assigned to the station at Adams Field (13963) in Little Rock, AR.
The northern portion of the Midwest includes the portion of Wisconsin west of the Lake
Michigan coastal plain, the western portion of the Upper Peninsula of Michigan, Minnesota, and
the eastern portion of North and South Dakota. This region is represented by the station at
Minneapolis-St. Paul International Airport (14922) in Minnesota.
The Great Plains lie between the Central Lowlands to the east and the Rocky Mountains
to the west. Lands at higher elevations are more grassland and shrubland used for cattle ranges,
whereas the lower elevations are more frequently used for crops. The region that includes most
of North and South Dakotas is represented by the station at Bismarck Municipal Airport (24011)
in North Dakota.
The central portion of Montana is more rugged, but still part of the Great Plains. The
Rocky Mountains form the western and southwestern boundaries of this region, which is
represented by the station at Billings Logan International Airport (24033) in Montana.
The station at Casper/Natrona County International Airport (24089) in Wyoming
represents most of Wyoming, southwestern South Dakota, and northwestern Nebraska.
Most of Colorado, southwestern Nebraska, western Kansas, and the panhandle of
Oklahoma are represented by the station at the Boulder Airport (23062) in Colorado.
The north central portion of the Great Plains includes most of Nebraska, northern Kansas,
western Iowa, southeastern South Dakota, and northwestern Missouri. This region is represented
by the station at Grand Island Airport (14935) in Nebraska.
The southern portion of the Great Plains includes most of Kansas, part of Missouri, and
eastern Oklahoma. This region is represented by the station at Tulsa International Airport
(13968).
D.2 Processing Meteorological Data
Surface Data. Hourly surface meteorological data used in air dispersion and deposition
modeling were processed from the Solar and Meteorological Surface Observation Network
(SAMSON) CD-ROM (U.S. DOC and U.S. DOE, 1993). Variables included the following:
• Temperature
• Pressure
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-7
-------
Appendix D: Meteorological Data
• Wind direction
• Wind speed
• Opaque cloud cover
• Ceiling height
• Current weather
• Hourly precipitation.
Upper-Air Data. Twice-daily mixing-height data were calculated from upper-air data
contained in the radiosonde data of the North America CD-ROM set (NCDC, 1997). This set
contains upper-air data from 1946 through 1996 for most upper-air stations in the United States.
The upper-air data were combined with the SAMSON data to create the mixing-height files.
EPA's Support Center for Regulatory Air Models (SCRAM) bulletin board was also used to
obtain mixing-height data (if available) when mixing-height data could not be successfully
calculated from the radiosonde data. This risk assessment used variable mixing heights that were
based on hourly ceiling height observations used in the ISCST3 air model.
Filling in Missing Data. The program SQAQC identified missing surface data by
searching for incidents of missing data on the observation indicator, opaque cloud cover,
temperature, station pressure, wind direction and wind speed, and ceiling height. Years that were
missing 10% or more of the data were discarded (Atkinson and Lee, 1992). Verification (quality
control [QC]) checks were performed on the SQAQC program by applying it to station data
where the missing data were known.
For years missing less than 10% of the data, missing surface data were filled in by a
program called METFIX. This program fills in up to 5 consecutive hours of data for cloud cover,
ceiling height, temperature, pressure, wind direction, and wind speed. For single missing values,
the METFIX program follows the objective procedures developed by Atkinson and Lee (1992).
For two to five consecutive missing values, other rules were developed because the subjective
methods provided by Atkinson and Lee (1992) rely on professional judgment and could not be
programmed. The METFIX program flagged files where missing data exceeded five consecutive
values. In the few cases where this occurred and the missing data did not constitute 10% of the
file, they were filled in manually using procedures from Atkinson and Lee (1992).
All upper-air files were checked for missing data using a program called QAQC. QAQC
produces a log file containing occurrences of missing mixing height. Verification (QC) checks
were performed on the QAQC program by applying it to station data where the missing data
were known.
Missing mixing heights were filled in by interpolating one to five consecutive missing
values. According to Atkinson and Lee (1992), if there are one to five consecutive missing
values, then the values should be filled in subjectively using professional judgment. Again,
programming these subjective procedures was not feasible, and the program used simple linear
interpolation to automatically fill in these values. Information from Atkinson and Lee (1992) was
used to determine which files should be discarded (i.e., files missing more than five consecutive
missing values or missing 10% or more of the data). After the missing mixing heights were filled
in for all upper-air files, they were checked again for missing data using the QAQC program.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-8
-------
Appendix D: Meteorological Data
Other Input Data. Processing of meteorological data also required the following
site/NWS specific parameters:
• Anemometer height (m)
• Bowen ratio
• Minimum Monin-Obukhov length (m)
• Noontime albedo
• Roughness length (m), surface meteorological station
• Fraction net radiation absorbed by the ground.
• Roughness length (m), area around facility
• Anthropogenic heat flux (W m"2)
Anemometer height was collected from local climatic data summaries (NOAA, 1983).
When anemometer height was not available, the station was assigned the most common
anemometer height from the other stations (6.1 m).
Land-use information is required for determining a number of inputs. To obtain this
information, a GIS determined the land uses within a 3-km radius around each meteorological
station using Geographic Retrieval and Analysis System (GIRAS) spatial data with Anderson
land-use codes (Anderson et al., 1976). A weighted average of these land uses was used to
estimate the Bowen ratio, minimum Monin-Obukhov length, the noontime albedo, the roughness
length at the meteorological station, and the fraction of net radiation absorbed by the ground. The
Bowen ratio is a measure of the amount of moisture at the surface around a meteorological
station. The wetness of a location was determined based on the average annual precipitation
amount. For this analysis, the annual average values were applied. The minimum Monin-
Obukhov length, which is a measure of the atmospheric stability at a meteorological station, was
correlated with the land-use classification. Noontime albedo values were also correlated with
land use around a meteorological station. Table D-2 presents the crosswalk between the
Anderson land-use codes from the GIRAS and the PCRAMMET land-use designations used in
air modeling. Other data used in the ISCST3 modeling are presented in Tables D-3 through D-6.
These are the Bowen ratio (Table D-3), the minimum Monin-Obukhov length (Table D-4),
Albedo values (Table D-5), and surface roughness length (Table D-6).
The surface roughness length is a measure of the height of an obstacle to the wind flow. It
is not equal to, but generally proportional to the physical dimensions of the obstacle. The
roughness length was assumed to be the same at the meteorological station and at the garden site.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-9
-------
Appendix D: Meteorological Data
Table D-2. Relation between Anderson Land-Use Codes
and PCRAMMET Land-Use Codes
Anderson Code and Description"
51 Streams and canals
52 Lakes
53 Reservoirs
54 Bays and estuaries
41 Deciduous forest land
61 Forested wetland
42 Evergreen forest land
43 Mixed forest land
62 Nonforested wetland
84 Wet tundra
21 Cropland and pasture
22 Orchards-groves-vineyards-nurseries-ornamental
23 Confined feeding operations
24 Other agricultural land
3 1 Herbaceous rangeland
32 Shrub and brush rangeland
33 Mixed rangeland
1 1 Residential
12 Commercial and services
13 Industrial
14 Transportation-communication-utilities
15 Industrial and commercial complexes
16 Mixed urban or built-up land
17 Other urban or built-up land
71 Dry salt flats
72 Beaches
73 Sandy areas not beaches
74 Bare exposed rock
75 Strip mines-quarries-gravel pits
76 Transitional areas
81 Shrub and brush tundra
82 Herbaceous tundra
83 Bare ground
85 Mixed tundra
91 Perennial snowfields
92 Glaciers
PCRAMMET Type and Description11
1 Water surface
1 Water surface
1 Water surface
1 Water surface
2 Deciduous forest
2 Deciduous forest
3 Coniferous forest
4 Mixed forest
5 Swamp (nonforested)
5 Swamp (nonforested)
6 Agricultural
6 Agricultural
6 Agricultural
6 Agricultural
7 Rangeland (grassland)
7 Rangeland (grassland)
7 Rangeland (grassland)
9 Urban
9 Urban
9 Urban
9 Urban
9 Urban
9 Urban
9 Urban
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
10 Desert shrubland
a Anderson codes from Anderson and colleagues (1976)
bPCRAMMET codes from U.S. EPA (1995)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
D-10
-------
Appendix D: Meteorological Data
Table D-3. Daytime Bowen Ratio by Land Use and Season
Land-Use Type
Water surface
Deciduous forest
Coniferous forest
Swamp
Cultivated land
(agricultural)
Grassland
Urban
Desert shrubland
Spring
Dry
0.1
1.5
1.5
0.2
1.0
1.0
2.0
5.0
Wet
0.1
0.3
0.3
0.1
0.2
0.3
0.5
1.0
Avg.
0.1
0.7
0.7
0.1
0.3
0.4
1.0
3.0
Summer
Dry
0.1
0.6
0.6
0.2
1.5
2.0
4.0
6.0
Wet
0.1
0.2
0.2
0.1
0.3
0.4
1.0
5.0
Avg.
0.1
0.3
0.3
0.1
0.5
0.8
2.0
4.0
Autumn
Dry
0.1
2.0
1.5
0.2
2.0
2.0
4.0
10.0
Wet
0.1
0.4
0.3
0.1
0.4
0.5
1.0
2.0
Avg.
0.1
1.0
0.8
0.1
0.7
1.0
2.0
6.0
Winter
Dry
2.0
2.0
2.0
2.0
2.0
2.0
2.0
10.0
Wet
0.3
0.5
0.3
0.5
0.5
0.5
0.5
2.0
Avg.
1.5
1.5
1.5
1.5
1.5
1.5
1.5
6.0
Annual Average
Dry
0.575
1.53
1.4
0.65
1.63
1.75
3.0
7.75
Wet
0.15
0.35
0.275
0.2
0.35
0.425
0.75
2.5
Avg.
0.45
0.875
0.825
0.45
0.75
0.825
1.6
4.75
Source: U.S. EPA (1995)
Averages were computed for this effort.
Table D-4. Minimum Monin-Obukhov Length
(Stable Conditions)
Urban Land-Use Classification
Agriculture (open)
Residential
Compact residential/industrial
Commercial (19-40 story buildings)
(>40 story buildings)
Length (m)
2
25
50
100
150
Source: U.S. EPA (1995)
Table D-5. Albedo Values of Natural Ground Covers for Land-Use Types and Seasons
Land-Use Type
Water surface
Deciduous forest
Coniferous forest
Swamp
Cultivated land (agricultural)
Grassland
Urban
Desert shrubland
Spring
0.12
0.12
0.12
0.12
0.14
0.18
0.14
0.3
Summer
0.1
0.12
0.12
0.14
0.2
0.18
0.16
0.28
Autumn
0.14
0.12
0.12
0.16
0.18
0.20
0.18
0.28
Winter
0.2
0.5
0.35
0.3
0.6
0.6
0.35
0.45
Annual
Average
0.14
0.22
0.18
0.18
0.28
0.29
0.21
0.33
Source: U.S. EPA (1995) Average values were computed for this analysis.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
D-ll
-------
Appendix D: Meteorological Data
Table D-6. Surface Roughness Length for Land-Use Types and Seasons (m)
Land-Use Type
Water surface
Deciduous forest
Coniferous forest
Swamp
Cultivated land (agricultural)
Grassland
Urban
Desert shrubland
Spring
0.0001
1.0
1.3
0.2
0.03
0.05
1.0
0.3
Summer
0.0001
1.3
1.3
0.2
0.2
0.2
1.0
0.3
Autumn
0.0001
0.8
1.3
0.2
0.05
0.01
1.0
0.3
Winter
0.0001
0.5
1.3
0.05
0.01
0.001
1.0
0.15
Annual
Average
0.0001
0.9
1.3
0.16
0.07
0.04
1.0
0.26
Source: U.S. EPA (1995) Average values were computed for this analysis.
During daytime hours, the heat flux into the ground is parameterized as a fraction of the
net radiation incident on the ground. This fraction varies based on land use. A value of 0.15 was
used for rural locations. Suburban and urban locations were given values of 0.22 and 0.27,
respectively (U.S. EPA, 1995).
Anthropogenic heat flux is negligible for meteorological stations outside of highly
urbanized locations; however, in areas with high population densities or energy use, such as
industrial facilities, this flux may not always be negligible (U.S. EPA, 1995). For this analysis,
anthropogenic heat flux was assumed to be zero for all meteorological stations.
D.3 References
Anderson, J.R., E.E. Hardy, J.T. Roach, and R.E. Witmer. 1976. A Land Use and Land Cover
Classification System for Use with Remote Sensor Data. U.S. Geological Survey
Professional Paper 964. United States Department of the Interior, Geological Survey,
Washington, DC. Available at http://landcover.usgs.gov/pdf/anderson.pdf. (accessed 12
December 2012).
Atkinson, D., and R.F. Lee. 1992. Procedures for Substituting Values for Missing NWS
Meteorological Data for Use in Regulatory Air Quality Models. U.S. Environmental
Protection Agency, Research Triangle Park, NC.
Bailey, R.G., P.E. Avers, T. King, and W.H. McNab. 1994. Ecoregions and Subregions of the
United States (Bailey's Ecoregion Map). U.S. Department of Agriculture, Forest Service,
Washington, DC.
NCDC (National Climatic Data Center). 1997. Radiosonde Data of North America: 1946-1996.
Version 1.0. Asheville, NC. June.
NOAA (National Oceanic and Atmospheric Administration). 1983. Local Climatological Data,
Annual Summaries for 1982: Part I - ALA - MONT and Part II - NEB - WYO. U.S.
Department of Commerce, National Oceanic and Atmospheric Administration, National
Environmental Satellite Data and Information Service, Asheville, NC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
D-12
-------
Appendix D: Meteorological Data
U.S. DOC (Department of Commerce) and U.S. DOE (Department of Energy) National
Renewable Energy Laboratory. 1993. Solar and Meteorological Surface Observation
Network (SAMSON), 1961-1990. Version 1.0. National Climatic Data Center,
Asheville, NC.
U.S. EPA (Environmental Protection Agency). 1995. User's Guide for the Industrial Source
Complex (ISC3) Dispersion Models. Volume II: Description of Model Algorithms. EPA-
454/B-95-003b. U.S. Environmental Protection Agency, Emissions, Monitoring, and
Analysis Division, Office of Air Quality Planning and Standards, Research Triangle Park,
NC. September.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-13
-------
Appendix D: Meteorological Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications D-14
-------
Appendix E: Soil Data
Appendix E
Soil Data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix E: Soil Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix E: Soil Data
Appendix E
Soil Data
A soils dataset was developed to represent the variability in soil conditions in areas that
produce SFS. This dataset was defined from a national dataset currently used by EPA to support
the 503 biosolids program. Using the locations of the SFS economic feasibility areas, a subset of
the national data was extracted for use in the SFS analysis. This appendix describes the approach
that was applied in developing the national dataset.
A representative sample of soils was generated by overlaying 7,000 points on a soils
layer using a geographic information system (GIS). The 7,000 points were distributed
proportionally to the number of farms located in each meteorological region. The points were
located randomly within each meteorological region, and information on the soil map unit
corresponding to each point was extracted. The predominant texture by depth for the top 20 cm
of soil was determined using soil texture data by layer. Additional details on the data sources and
methods used to collect soil textures and relate them to the hydrologic soil properties needed for
modeling are provided below.
E.I Data Sources
The primary source for soil properties data was the State Soil Geographic (STATSGO)
database. STATSGO is a repository of nationwide soil properties primarily compiled by USD A
from county soil survey data (USDA, 1994). STATSGO includes a l:250,000-scale GIS
coverage that delineates soil map units and an associated database containing soil data for each
STATSGO map unit. (Map units are areas used to spatially represent soils in the database.)
In addition, two compilations of STATSGO data, each keyed to the STATSGO map unit
GIS coverage, and land-use data from the Geographic Information Retrieval and Analysis
System (GIRAS) land-use database were used as convenient sources of average soil properties:
• USSOILS. USSOILS (Schwarz and Alexander, 1995) averages STATSGO data over the
entire soil column for each map unit.
• CONUS. CONUS (Miller and White, 1998) provides average STATSGO data by map
unit and a set of 11 standardized soil layers.
• GIRAS. The GIRAS land-use database (U.S. EPA, 1994) provides comprehensive
landuse data in a digital GIS format for the contiguous 48 states.
Soil properties derived directly from STATSGO, CONUS, or USSOILS data include
organic matter content, Universal Soil Loss Equation (USLE) K (erodibility) and S (slope)
factors, and pH. A complete set of hydrologic soil properties was not available from STATSGO.1
To ensure consistent and realistic values, it was necessary to rely on established, nationwide
relationships between hydrologic properties and soil texture or hydrologic soil group, both of
1 Hydrologic soil properties required for modeling include bulk density, saturated water content, residual water
content, field moisture content, wilting point, saturated hydraulic conductivity, soil moisture coefficient b, and soil
moisture retention parameters alpha and beta.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications E-l
-------
Appendix E: Soil Data
which are available from STATSGO. Sources for these relationships include Carsel and Parrish
(1988), Carsel et al. (1988), and Clapp and Hornberger (1978). These peer-reviewed references
provide a consistent set of correlated hydrologic properties for each soil texture or hydrologic
group. Table E-l lists soil properties collected for this analysis and their data sources.
Table E-l. Summary of Soil of Properties Collected for Sewage Sludge Risk Analysis
Soil Variable
Units
Data Source
Properties Derived from Soil Texture
USDA soil texture
Saturated hydraulic conductivity
Saturated water content
Soil moisture coefficient b
Soil bulk density
Root zone depth
Unitless
cmh"1
LL-1
Unitless
mgL-1
cm
CONUS/STATSGO
Relationship from Carsel and Parrish (1988)
Relationship from Carsel and Parrish (1988)
Relationship from Clapp and Hornberger (1978)
Calculated from saturated water content
Relationship (with land use) from Dunne and Leopold (1978)
Properties Derived from Soil Hydrologic Class
Soil Conservation Service (SCS)
hydrologic class
Field capacity
Wilting point
SCS curve number
Unitless
% (vol.)
% (vol.)
Unitless
CONUS/STATSGO
Relationship from Carsel et al. (1988)
Relationship from Carsel et al. (1988)
Relationship (with land use) from USDA (1986)
Properties Obtained Directly from STATSGO
Fraction organic carbon
Silt content
USLE credibility factor (K)
USLE slope (S)
gg"1
% (wt.)
kgnr2
Degrees
STATSGO
STATSGO
STATSGO
STATSGO
Properties Derived from Slope
USLE slope length (L)
USLE length/slope factor (LS)
m
Unitless
Relationship fromLightle and Weesies (1998)
Calculated from L and S per Williams and Berndt (1977)
Finally, two parameters—root zone depth and Soil Conservation Service (SCS) curve
number (used for recharge calculations)—required site-based land-use data, as well as soil
texture or hydrologic soil group. The land-use data were obtained for each of the 41
meteorological regions from the GIRAS land-use database (U.S. EPA, 1994). Land-use and land-
cover information in GIRAS was mapped and coded using the Anderson classification system
(Anderson et al., 1976), which is a hierarchical system of land-use characterizations. This
nationwide coverage is based on late-1970s to early-1980s satellite images and aerial
photography. The relationships used to convert the land-use and soil data were obtained from
Dunne and Leopold (1978) for root-zone depth and USDA (1986) for the SCS curve number.
E.2 Data Collection
Soil data collection began by overlaying the boundaries of the 41 meteorological regions
onto the STATSGO map units to determine the STATSGO map units and their areas within each
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-2
-------
Appendix E: Soil Data
region. These data were then used to derive predominant soil properties within each
meteorological region, either through direct calculations or by applying established relationships
in lookup tables. Soil model inputs were based on the soil properties of the predominant soil type
(texture and hydrologic group) for each STATSGO map unit having agricultural land use within
the meteorological region.
Twelve common soil textures were collected to develop soil and hydrologic properties.
Using CONUS data for the top three surface soil layers (20 cm) in each STATSGO map unit, the
soil texture of the thickest CONUS layer was considered the predominant texture for the map
unit. The textures were ranked according to predominance across all map units and, when there
were two soil textures with equal depths, the texture with the higher ranking was chosen for that
map unit. For the 303 out of 7,000 map units without one of the 12 common soil textures (e.g.,
those with water or organic matter), the predominant soil texture (i.e., loam) was selected. Soil
column texture was obtained in a similar manner, except that all CONUS layers were used.
Attachment A to this appendix presents the percentage of soil textures within each
meteorological region.
To limit data collection to agricultural soils in each meteorological region, GIS programs
overlayed the STATSGO map units with the GIRAS land-use coverage to determine which map
units (and their respective areas) occur in cropland use and pastureland use (i.e., Anderson land-
use code 21). These data were then processed to create a set of the 12 soil textures, ranked by
percentage of land in agricultural use with each texture, for each region. These textures were
used to derive soil properties for this analysis for each region/texture combination as described in
the next section.
Because certain soil properties were derived from SCS hydrologic soil groups, it was
necessary to develop a hydrologic soil group that would be consistent with the soils of each
texture within a region. To do so, a table of hydrologic soil groups by STATSGO map unit was
created using STATSGO data for hydrologic soil groups by the component soils within the map
unit. Based on the predominant texture for each map unit, hydrologic soil groups for the
component soils with the same texture were averaged across each map unit (weighted by
component percent) using the numeric conversion: Group A=l, Group B=2, Group C=3, and
Group D=4. These values were then averaged again (weighted by map unit area) for each soil
texture occurring in a region. These regional average textures were converted back to letters
using the same conversion, resulting in a hydrologic soil group for each texture occurring within
a meteorological region. A hydrologic soil group applies to the entire soil column and is not
layer-specific.
E.3 Development of Soil Properties
After the distribution of soil textures and their related hydrologic class were determined
for each meteorological region, average soil properties were determined for each soil texture
present in a region by relationships with soil texture or hydrologic class or by extracting the data
for soils of each texture directly from STATSGO.
Soil Properties Based on Relationship with Soil Texture—Several soil hydrologic
properties were derived directly from the soil texture using database lookup tables relating mean
properties to texture class (see Table E-2):
Risk Assessment of Spent Foundry Sands in Soil-Related Applications E-3
-------
Appendix E: Soil Data
Saturated hydraulic conductivity (cm tr1) was determined for both surface soil and the
entire soil column using a national relationship from Carsel and Parrish (1988).
Saturated water content (unitSess) was determined for both surface soil and the entire
soil column using a relationship from Carsel and Parrish (1988).
Bulk density (g cm"3) was calculated for surface soil from saturated water content using
the following equation:
pb = 2.65(l -cp)
(E-l)
Where
pb = Bulk density of the soil (U.S. EPA, 1997)
2.65 = Particle density in g cm"3 (assumed to be quartz)
cp = Saturated water content
Soil moisture coefficient (unitSess) was determined for both the surface soil and the
entire soil column using a relationship from Clapp and Hornberger (1978).
Depth to root zone (cm) was determined using a Dunne and Leopold (1978) table of
rooting depth by vegetation type and soil texture. For each soil texture, a minimum and a
maximum root zone depth (for shallow and deep-rooted crops) were used to represent the
range across cropland and pastureland use. Because Dunne and Leopold (1978) included
only five soil textures, these five textures were mapped across the 12 basic textures used
in this analysis (see Table E-3).
Table E-2. Hydrological Soil Parameters Correlated to Soil Texture
Soil Texture
Clay
Sandy clay
Silty clay
Clay loam
Sandy clay loam
Silty clay loam
Sand
Loamy sand
Sandy loam
Loam
Silt
Silt loam
Saturated
Hydraulic
Conductivity"
(cm hr1)
0.20
0.12
0.02
0.26
1.31
0.07
29.70
14.59
4.42
1.04
0.25
0.45
Saturated Water
Content3 (L I/1)
0.38
0.38
0.36
0.41
0.39
0.43
0.43
0.41
0.41
0.43
0.46
0.45
Bulk Density b
(g cm-3)
1.643
1.643
1.696
1.5635
1.6165
1.5105
1.5105
1.5635
1.5635
1.5105
1.431
1.4575
Soil Moisture
Coefficient c
11.4
10.4
10.4
8.52
7.12
7.75
4.05
4.38
4.90
5.39
—
5.30
a Carsel and Parrish (1988)
b Calculated from WCS using equation from U.S. EPA (1997)
c Clapp and Hornberger (1978)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-4
-------
Appendix E: Soil Data
Table E-3. Depth to Root Zone Values
USDA Soil Texture
Clay
Sandy clay
Silty clay
Clay loam
Sandy clay loam
Silty clay loam
Sand
Loamy sand
Sandy loam
Loam
Silt
Silt loam
Dunne and Leopold
Texture
Clay
Clay loam
Fine sand
Fine sandy loam
Silt loam
Shallow-Rooted Crops
(cm)
25
40
50
50
62
Deep-Rooted Crops
(cm)
67
100
100
100
125
Source: Derived from Dunne and Leopold (1978)
Soil Parameters Based on Relationship with Hydrologic Group—The following soil parameters
are all based on the average hydrologic soil group for each texture within a meteorological
region. Mean values by hydrologic group were obtained using the following relationships (see
Tables E-4 and E-5):
• Soil moisture field capacity (volume %). A single field capacity value was obtained for
each soil group by averaging the layered property values from Carsel et al. (1988).
• Soil moisture wilting point (volume %). A single wilting point value was obtained for
each soil group by averaging the layered property values from Carsel et al. (1988).
• SCS curve number (unitless). Minimum and maximum SCS curve number values were
determined for each regional soil texture based on a USDA (1986) table of curve
numbers by cover type and hydrologic soil group, assuming a good condition pasture-
land use for the minimum and poor-condition cropland use for the maximum. A lookup
table with minimum and maximum SCS curve numbers by hydrologic soil group was
used to assign the appropriate value for each regional soil texture according to its
hydrologic soil group.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-5
-------
Appendix E: Soil Data
Table E-4. Field Capacity and Wilting Point Values
Hydrologic
Group
A
B
C
D
Layer
1
2
3
4
Average
1
2
3
4
Average
1
2
3
4
Average
1
2
3
4
Average
Field
Capacity
9.4
8.1
5.9
5.8
7.3
19.1
18.8
18.7
17.5
18.5
22.5
23.2
22.9
21.3
22.5
24.2
26.3
25.6
24.4
25.1
Wilting
Point
3.1
2.3
2.1
1.9
2.4
8.7
9.3
8.9
8.4
8.8
10.4
12.1
11.9
11.5
11.5
13.8
17.0
16.3
15.1
15.6
Source :Carseletal. (1988)
Table E-5. SCS Curve Number Values by SCS Hydrologic
Soil Group
SCS
Hydrologic
Soil Group
A
B
C
D
SCS Curve Number
Minimum
39
61
74
80
Maximum
72
81
88
91
Source: Derived from USD A (1986)
E.4 Parameters Collected Directly from STATSGO-Based Data Sources
Several variables were obtained directly from STATSGO (Schwarz and Alexander, 1995).
Although these variables were not derived from soil texture, they were extracted and averaged
based only on soil map units with the predominant texture to ensure consistent soil properties.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-6
-------
Appendix E: Soil Data
These variables include the following:
• USLE credibility factor—top 20 cm (tons acre"1) An area-weighted average
erodibility factor for the top 20 cm of soil was calculated from STATSGO data by layer
and component. STATSGO layer data were translated into lvalues using standardized
CONUS layers and calculating a depth-weighted average value. Furthermore, a
component percent-weighted average K was calculated for each CONUS layer across all
components contained in each map unit. The resulting table contains K values by map
unit and standardized CONUS layer. To derive one value for K by map unit for the top 20
cm of soil, a depth-weighted average for the top three CONUS layers was calculated. The
final K value by meteorological region and soil texture was obtained by averaging the
map units for each surface soil texture present within the meteorological region.
• Fraction organic carbon (foe)—top 20 cm (mass fraction). An area-weighted average
foe for surface soils was calculated for each region and soil texture using only the map
units with the predominant surface soil texture of interest within the region. Percent
organic matter for the top 20 cm of soil was obtained from STATSGO organic matter
data by layer and component (Schwarz and Alexander, 1995) and converted to foe by
dividing by 174 (100 x 1.74 g organic matter g'1 of organic carbon) (U.S. EPA, 1997).
Percent organic matter values were translated from STATSGO layer and component into
standardized CONUS layers using the same methodology described for the USLE
erodibility factor K. Then, a depth-weighted average percent organic matter was
calculated for the top three CONUS layers (top 20 cm of soil).
• Silt content—top 20 cm (weight percent). An area-weighted average silt content for
surface soils was derived from STATSGO data for each region and soil texture in the
same manner described for USLE erodibility factor.
The USLE's length slope factor (LS) was derived from STATSGO slope data. Percent
slope was obtained by region and soil texture using only the map units with the predominant
texture of interest. An area-weighted average slope was calculated for each texture occurring in a
region. Length (ft) was then obtained from a Lightle and Weesies (1998) lookup table of default
flow lengths by slope, using slope values rounded to the nearest integer (Table E-6). All slopes
less than 0.5 were given the length corresponding to 0.5, and all slopes greater than 24 were
given the length corresponding to 24. The USLE length/slope factor LS (unitless) was then
calculated using the following equation from Williams and Berndt (1977):
LS = (L/72.6)m (0.065 + 0.0454S + 0.0065S2) (E-2)
Where
L = Flow length
S = Slope in percent
and
m = 0.2 for slope <1%
m = 0.3 for slope >1% and <3%
m = 0.4 for slope >3% and <5%
m = 0.5forslope>5%
Risk Assessment of Spent Foundry Sands in Soil-Related Applications E-7
-------
Appendix E: Soil Data
Table E-6. Default Flow Lengths by Slope
Slope
<0.5
1
2
3
4
5
6
7
8
9
10
11
12
Length
(ft)
100
200
300
200
180
160
150
140
130
125
120
110
100
Slope
13
14
15
16
17
18
19
20
21
22
23
>24
Length
(ft)
90
80
70
60
60
50
50
50
50
50
50
50
Source: Lightle and Weesies (1998)
E.5 References
Anderson, J.R., E.E. Hardy, J.T. Roach, and R.E. Witmer. 1976. A Land Use and Land Cover
Classification System for Use with Remote Sensor Data. U.S. Geological Survey
Professional Paper 964. United States Department of the Interior, Geological Survey,
Washington, DC. Available at http://landcover.usgs.gov/pdf/anderson.pdf
Carsel, R.F., and R.S. Parrish. 1988. Developing joint probability distributions of soil water
retention characteristics. Water Resources Research 24(5):755-769.
Carsel, R.F., R.S. Parrish, R.L. Jones, J.L. Hansen, and R.L. Lamb. 1988. Characterizing the
uncertainty of pesticide leaching in agricultural soils. Journal of Contaminant Hydrology
2:111-124.
Clapp, R.B., and G.M. Hornberger. 1978. Empirical equations for some soil hydraulic properties.
Water Resources Research 14:601 -604.
Dunne, T., and L.B. Leopold. 1978. Water in Environmental Planning. New York: W.H.
Freeman and Company.
Lightle, D.T., and G. Weesies. 1998. "Default Slope Parameters." Memorandum submitted to
Scott Guthrie, Research Triangle Institute. West Lafayette, IN: USD A, Natural Resources
Conservation Service. June.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-8
-------
Appendix E: Soil Data
Miller, D.A., and R.A. White. 1998. A Conterminous United States Multilayer Soil
Characteristics Dataset for Regional Climate and Hydrology Modeling. Earth
Interactions 2: 1-26.
Schwarz, G.E., and R.B. Alexander. 1995. State Soil Geographic (STATSGO) DataBase for the
Conterminous United States, Edition 1.1. Reston, VA. September. Web site:
http://water.usgs.gov/lookup/getspatial7ussoils.
U.S. EPA (Environmental Protection Agency). 1994. 1:250,000 Scale Quadrangles of
Landuse/Landcover GIRAS Spatial Data in the Conterminous United States: Metadata.
U.S. Environmental Protection Agency, Office of Information Resources Management,
Washington, DC. Available at:
http://water.epa.gov/scitech/datait/models/basins/metadata giras.cfm. (accessed 12
December 2012).
U.S. EPA (Environmental Protection Agency). 1997. EPA's Composite Model for Leachate
Migration with Transformation Products. EPACMTP: User's Guide. U.S. Environmental
Protection Agency, Office of Solid Waste, Washington, DC.
USDA (U.S. Department of Agriculture). 1986. Urban Hydrology for Small Watersheds. TR 55
(210-VI-TR-55). U.S. Department of Agriculture, Engineering Division, Soil
Conservation Service, Washington, DC. June.
USDA (U.S. Department of Agriculture). 1994. National STATSGO Database: USDA-NRCS
Soil Survey Division Data Access. U.S. Department of Agriculture, Natural Resources
Conservation Service, Fort Worth, Texas. Web site: http://soildatamart.nrcs.usda.gov/.
Williams, J.R., and H.D. Berndt. 1977. Determining the universal soil loss equation's
lengthslope factor for watersheds. In A National Conference on Soil Erosion - Soil
Erosion: Prediction and Control, Perdue University, West Lafayette, IN, May 24-26,
1976. Ankeny, IA: Soil Conservation Society of America.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications E-9
-------
Appendix E: Soil Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications E-10
-------
May 2009 Peer Review Draft Appendix E - Attachment A: Soil Data
Appendix E
Attachment E-A:
Soil Data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
May 2009 Peer Review Draft Appendix E - Attachment A: Soil Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix E Attachment A: Soil Data
Table E-A-1. Soil Textures for Meteorological Regions
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Albuquerque (23050)
Clay
Clay Loam
Loam
Sand
Sandy Clay Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
12.2
29.4
14.1
12.7
0.2
1.2
9.7
0.8
19.6
Ashville (03812)
Clay Loam
Loam
Sandy Clay
Silty Clay Loam
Silt Loam
Sandy Loam
5.0
30.4
2.2
2.6
44.6
15.2
Atlanta (13874)
Clay
Loam
Loamy Sand
Sand
Sandy Clay
Sandy Clay Loam
Silt Loam
Sandy Loam
0.9
3.2
46.8
3
0.3
0.5
8
36.5
Billings (24033)
Clay
Clay Loam
Loam
Sandy Clay Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
6.8
32
34.9
0.4
3.3
9.1
9.6
3.8
Bismarck (240 11)
Clay
Clay Loam
Loam
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
3.5
4
56.1
3.5
5.3
7.8
9.6
10.2
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Boise (24131)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
0.3
1.9
12.5
0.5
3.7
1.9
67.6
11.6
Boulder (94018)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
2.3
6.3
20.8
2.2
5.6
0.3
7.3
37.2
17.5
Burlington (14742)
Clay
Loam
Loamy Sand
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
7.8
11.4
5.3
1.8
2.9
11.3
20.7
38.7
Caspar (24089)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Sandy Clay Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
5.2
10.5
31
3.8
7.4
2.6
1.6
3.1
13
21.8
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-A-1
-------
Appendix E Attachment A: Soil Data
Table E-A-1. Soil Textures for Meteorological Regions (cont'd)
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Charleston (13880)
Clay
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
0.3
2.2
24
45.6
0.4
1
26.5
Chicago (94846)
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
0.5
5.5
0.8
3.6
10.3
75.9
3.3
Cleveland (14820)
Clay
Loam
Loamy Sand
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
10
7.8
1.2
1.1
1.6
3.6
70.8
3.5
Fresno (93193)
Clay
Clay Loam
Loam
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
22.3
14.8
11.7
1.2
8.1
5.5
35.7
Grand Island (14935)
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
6.3
13.4
0.3
4.4
2.9
26.7
43.8
2.1
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Harrisburg (14751)
Clay Loam
Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
3.4
14.1
0.1
4.4
71.5
6.2
Hartford (14740)
Loam
Loamy Sand
Sand
Silt Loam
Sandy Loam
10
3.4
1.6
44.5
40.3
Houston (12960)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Sandy Clay Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
33.8
8.3
3
0.3
10.1
4.6
2.6
1.3
6.8
29.2
Huntington (03860)
Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
1.3
7.5
5.9
84.8
0.3
Las Vegas (23169)
Clay
Loam
Loamy Sand
Sand
Silty Clay
Silt Loam
Sandy Loam
16.4
11.4
40.5
10.1
1
0.8
19.7
Little Rock (13963)
Clay
Loam
Silty Clay
Silty Clay Loam
Silt Loam
13.6
2.7
9.5
7.8
56.5
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-A-2
-------
Appendix E Attachment A: Soil Data
Table E-A-1. Soil Textures for Meteorological Regions (cont'd)
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Los Angeles (23174)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
2.6
2.4
10.7
14
4.5
0.9
2.9
61.5
Meridian (13865)
Clay
Loam
Loamy Sand
Silt
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
8.6
3.9
1.5
0.3
7.4
4.3
40.7
33.3
Miami (12839)
Clay Loam
Loam
Loamy Sand
Sand
Silt Loam
0.1
4.3
1.5
93.7
0.4
Minneapolis (14922)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Sandy Clay Loam
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
0.7
11.2
32.3
2.9
3.2
0.6
1
13.9
20.8
13.5
Muskegon (14840)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
0.3
0.5
34.3
11.7
7.3
2.3
26.1
17.5
Meteorological Station
Soil Texture
Sandy Loam
(Station number)
Percent of
Total Soil
9.9
Nashville (13897)
Clay
Loam
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
0.5
3.7
0.4
2.7
4.2
85.4
o
J
New Orleans (12916)
Clay
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
8.3
1
10.7
2.4
4.7
29.5
43.4
Norfolk (13737)
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
7.6
10.2
4.8
0.6
14.1
62.6
Philadelphia (13739)
Loam
Sand
Silt Loam
Sandy Loam
22.8
o
J
63.5
10.5
Phoenix (23183)
Clay
Clay Loam
Loam
Sand
Sandy Clay Loam
Silty Clay Loam
Silt Loam
Sandy Loam
5.5
10.2
26.2
0.6
1.3
25.6
1.4
29
Portland (14764)
Loam
Loamy Sand
Silt Loam
Sandy Loam
19.2
5.7
44.2
30.9
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-A-3
-------
Appendix E Attachment A: Soil Data
Table E-A-1. Soil Textures for Meteorological Regions (cont'd)
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Raleigh-Durham (13722
Loam
Loamy Sand
Sand
Sandy Clay
Silty Clay Loam
Silt Loam
Sandy Loam
19.4
18.5
11.5
2.2
1.5
13.9
32.7
Salem (24232)
Clay Loam
Loam
Loamy Sand
Sand
Silt
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
3.5
18.3
0.3
3
1
0.6
31.4
34.1
7.8
Salt Lake City (24127)
Clay Loam
Loam
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
2.8
30
0.9
8.2
47.9
9.3
San Francisco (23234)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
20.1
17.1
33.7
1.7
3.3
6
8.6
9.5
Seattle (24233)
Loam
Loamy Sand
Sand
Silty Clay Loam
Silt Loam
Sandy Loam
11.9
1.5
1.1
5.6
52.4
27.4
Meteorological Station
Soil Texture
(Station number)
Percent of
Total Soil
Shreveport (13957)
Clay
Clay Loam
Loam
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
29
5
4.4
6.5
3
3.9
14.2
34
Tampa (12842)
Loamy Sand
Sand
Sandy Loam
25.9
73
1.1
Tulsa (13968)
Clay
Clay Loam
Loam
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
1.9
2.8
10.3
5.7
0.9
8.8
51.5
18.2
Williamsport (14778)
Loam
Silt Loam
Sandy Loam
11.8
86.1
1.4
Winnemucca (24128)
Clay
Clay Loam
Loam
Loamy Sand
Sand
Silty Clay
Silty Clay Loam
Silt Loam
Sandy Loam
4.2
6.5
24.2
1.8
1.7
5.1
8.2
20.6
27.6
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
E-A-4
-------
Appendix F: Chemical Data
Appendix F
Chemical Data
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix F: Chemical Data
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix F: Chemical Data
Table F-l. Chemical Parameters for Antimony (7440-36-0)
Ecological Assessment
Parameter
Description
Value
Reference
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g ml/1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
6.6 E+00
2.66E-5
Lognormal distribution
- Kd values (min 1 .26.,
max 501, mean 200,
stdev 12.6)
O.OOE+00
1.2E+02
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
Set to zero for
metals
U.S. EPA, 2008
Ecological Benchmark
Eco-SSL
EPA Soil Screening Level (mg kg"1 soil)
Soil Biota: 78
Mammals: 0.27
U.S. EPA, 2014a
Table F-2. Chemical Parameters for Arsenic (7440-38-2)
Human Health Soil/Produce Assessment
Parameter
Description
Value
Reference
Biotransfer Factors
BrExfruit
BrExveg
BrProfruit
BrProveg
BrRoot
KpPar
Soil-to-plant bioconcentration factor, exposed fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, exposed
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, root vegetables
(mg kg"1 DW plant) (mg kg"1 soil)"1
Plant surface loss coefficient, paniculate
(1 yr1)
2.00E-03
l.OOE-02
2.00E-03
2.00E-03
4.60E-03
1.81E+01
Calculated based on
U.S. EPA, 1999
Calculated based on
U.S. EPA, 1999
Calculated based on
U.S. EPA, 1999
Calculated based on
U.S. EPA, 1999
Calculated based on
U.S. EPA, 1999
U.S. EPA, 1997
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g mL"1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day1)
Molecular weight (g mol"1)
5.73E+00
3.25E-05
Lognormal distribution
- Kd values (min 2,
max 19953, mean
1585, stdev 5)
O.OOE+00
7.49E+01
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
Set to zero for
metals
U.S. EPA, 2008
Human Health Benchmark
CSF
RfD
Cancer Slope Factor (mg kg^-d"1)"1
Reference Dose (mg kg"1-d"1)
1.50E+00
3.00E-04
U.S. EPA, 2012
U.S. EPA, 2012
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
F-l
-------
Appendix F: Chemical Data
Table F-3. Chemical Parameters for Chromium III (16065-83-1)
Ecological Assessment
Parameter
Description
Value
Reference
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g ml/1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
7.1E+00
4.6E-05
Lognormal distribution
- Kd values (min 10,
max 50,119, mean
6310, stdev2.5)
O.OOE+00
5.1E+01
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
Set to zero for
metals
U.S. EPA, 2008
Ecological Benchmark
Eco-SSL
EPA Soil Screening Level (mg kg"1 soil)
Mammals: 34
U.S. EPA, 2014a
Table F-4. Chemical Parameters for Cobalt (7440-48-4)
Human Health Soil/Produce Assessment
Parameter
Description
Value
Reference
Biotransfer Factors
BrExfruit
BrExveg
BrProfruit
BrProveg
BrRoot
KpPar
Soil-to-plant bioconcentration factor, exposed fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, exposed
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, root vegetables
(mg kg"1 DW plant) (mg kg"1 soil)"1
Plant surface loss coefficient, paniculate
(1 yr1)
7.0E-03
2.0E-02
7.0E-03
7.0E-03
2.0E-02
1.81E+01
Baes et al., 1984
Baes et al., 1984
Baes et al., 1984
Baes et al., 1984
Baesetal., 1984
U.S. EPA, 1997
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g mL"1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
8.8E+00
4.89E-05
Lognormal distribution
- Kd values (min 0.06,
max 12,589, mean 126,
stdev 15.8)
O.OOE+00
5.8E+01
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
Set to zero for
metals
U.S. EPA, 2008
Human Health Benchmark
RfD
Reference Dose (mg kg^-d"1)
3.00E-04
U.S. EPA, 2014b
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
F-2
-------
Appendix F: Chemical Data
Table F-65. Chemical Parameters for Copper (7440-50-8)
Ecological Assessment
Parameter
Description
Value
Reference
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g ml/1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
8.9E+00
4.68E-05
Lognormal distribution
- Kd values (min 1 .26,
max 3981, mean 316,
stdev 4.0)
O.OOE+00
6.3E+01
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
(from literature
data)
Set to zero for
metals
U.S. EPA, 2008
Ecological Benchmark
Eco-SSL
EPA Soil Screening Level (mg kg"1 soil)
Terr. Plants: 70
Soil Biota: 80
Mammals: 49
U.S. EPA, 2014a
Table F-3. Chemical Parameters for Iron (7439-89-6)
Human Health Soil/Produce Assessment
Parameter
Description
Value
Reference
Biotransfer Factors
BrExfruit
BrExveg
BrProfruit
BrProveg
BrRoot
KpPar
Soil-to-plant bioconcentration factor, exposed fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, exposed
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected fruit
(mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, protected
vegetables (mg kg"1 DW plant) (mg kg"1 soil)"1
Soil-to-plant bioconcentration factor, root vegetables
(mg kg"1 DW plant) (mg kg"1 soil)"1
Plant surface loss coefficient, paniculate
(1 yr1)
l.OOE-03
4.00E-03
l.OOE-03
l.OOE-03
4.00E-03
1.81E+01
Baes et al., 1984
Baesetal., 1984
Baes et al., 1984
Baes et al., 1984
Baes et al., 1984
U.S. EPA, 1997
Chemical Properties
Density
Dw
Kd
ksoil
MW
Density of the chemical (g mL"1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
7.8E+00
4.68E-05
25
O.OOE+00
5.5E+01
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
Baes et al., 1984
Set to zero for
metals
U.S. EPA, 2008
Human Health Benchmark
RfD
Reference Dose (mg kg^-d"1)
7.0E-01
U.S. EPA, 2012
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
F-3
-------
Appendix F: Chemical Data
Table F-3. Chemical Parameters for Manganese (7439-96-5)
Ecological Assessment
Parameter
Density
Dw
Kd
ksoil
MW
Description
Density of the chemical (g ml/1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day"1)
Molecular weight (g mol"1)
Value
7.3E+00
4.48E-05
Lognormal distribution
Kd values (min 251,
max 50,1 19, mean 1585,
stdev 5.0)
O.OOE+00
5.4E+01
Reference
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
Allison, 2003
Set to zero for
metals
U.S. EPA, 2008
Ecological Benchmark
Eco-SSL
EPA Soil Screening Level (mg kg"1 soil)
Terr. Plants: 220
Soil Biota: 450
Mammals: 4000
U.S. EPA, 2014a
Table F-4. Chemical Parameters for Nickel (7440-02-0)
Ecological Assessment
Parameter
Density
Dw
Kd
ksoil
MW
Description
Density of the chemical (g mL"1)
Diffusion coefficient in water (cm2 s"1)
Soil-water partition coefficient (L Kg"1)
Degradation rate for soil (1 day1)
Molecular weight (g mol"1)
Value
8.9E+00
4.90E-05
Lognormal distribution
- Kd values (min 10,
max 794, mean 6310,
stdev 3.2)
O.OOE+00
5.8E+01
Reference
U.S. EPA, 2008
Calculated based on
U.S. EPA, 2001
U.S. EPA, 2005
Set to zero for
metals
U.S. EPA, 2008
Ecological Benchmark
Eco-SSL
EPA Soil Screening Level (mg kg"1 soil)
Terr. Plants: 38
Soil Biota: 280
Mammals: 130
U.S. EPA, 2014a
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
F-4
-------
Appendix F: Chemical Data
F.I References
Allison, J. 2003. "3MRA Kds Checked and Revised." E-mail and attachment (spreadsheet of Kd
for manganese) from Jerry Allison, Allison Geoscience Consultants, to Robert S.
Truesdale, RTI International. June 13.
Baes, C.F., III, R.D. Sharp, A.L. Sjoreen, and R.W. Shor. 1984. A Review and Analysis of
Parameters for Assessing Transport of Environmentally Released Radionuclides Through
Agriculture. ORNL-5786. Prepared for U.S. Department of Energy, Oak Ridge National
Laboratory, Oak Ridge, TN. September.
U.S. EPA (Environmental Protection Agency). 1997. The Parameter Guidance Document. A
Companion Document to the Methodology for Assessing Health Risks Associated with
Multiple Pathways Exposure to Combustor Emissions (InternalDraft). NCEA-0238. U.S.
Environmental Protection Agency, National Center for Environmental Assessment,
Cincinnati, OH. March.
U.S. EPA (Environmental Protection Agency). 1999. Estimating Risk from the Use of
Agricultural Fertilizers (Draft Report). U.S. Environmental Protection Agency, Office of
Solid Waste, Washington DC. August. Available at
http://www.epa.gov/wastes/hazard/recycling/fertiliz/risk/report.pdf (accessed 22 March
2012).
U.S. EPA (Environmental Protection Agency). 2001. WATER9, Air Emission Models for Waste
and Wastewater. Technology Transfer Network Clearinghouse for Inventories &
Emission Factors. U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards, Research Triangle Park, NC. Available at
http://www.epa.gov/ttn/chief/software/water (accessed 22 March 2012).
U.S. EPA (Environmental Protection Agency). 2005. Partition Coefficients for Metals in Surface
Water, Soil, and Waste. EPA/600R-05/074. U.S. Environmental Protection Agency,
Office of Research and Development. July. Available at
http://www.epa.gov/athens/publications/reports/Ambrose600R05074PartitionCoefficients
.pdf (accessed 9 December 2013).
U.S. EPA (Environmental Protection Agency). 2008b. Super fund Chemical Data Matrix
(SCDM). U.S. Environmental Protection Agency, Office of Emergency Response and
Remediation, Washington, DC. Available at
http://www.epa.gov/superfund/sites/npl/hrsres/tools/scdm.htm (accessed 27 June 2014).
U.S. EPA (Environmental Protection Agency). 2012. Integrated Risk Information System (IRIS).
U.S. Environmental Protection Agency, Office of Research and Development,
Washington, DC. Available at http://www.epa.gov/iris/ (accessed 22 March 2012).
U.S. EPA (Environmental Protection Agency). 2014a. Provisional Peer Reviewed Toxicity
Values for Superfund (PPRTV). Environmental Protection Agency, Office of Superfund
Remediation and Technology Innovation, Washington, DC. Available online at:
http://hhpprtv.ornl.gov/quickview/pprtv_papers.php (accessed 28 April 2014)
U.S. EPA (Environmental Protection Agency). 2014b. Ecological Soil Screening Levels. (Eco-
SSL). Environmental Protection Agency, Office of Emergency and Remedial Response,
Risk Assessment of Spent Foundry Sands in Soil-Related Applications F-5
-------
Appendix F: Chemical Data
Washington, DC. Available online at: to: http://www.epa.gov/ecotox/ecossl/ (accessed
April 2014)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications F-6
-------
Appendix G: Home Garden Source Model
Appendix G
Home Garden Source Model
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G: Home Garden Source Model
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G: Home Garden Source Model
Appendix G
Home Garden Source Model
G.I Introduction
For the home gardening scenario, the Land Application Unit (LAU) Module from the
Multimedia, Multipathway, Multi-receptor Risk Assessment (3MRA) modeling system was used
to predict the release of metals from SFS-manufactured soil used in a home garden. In support of
EPA's 503 program, the LAU Module has been modified to simulate chemical losses from farm
fields (rather than land application units) that apply biosolids as a soil amendment. This appendix
describes the modified LAU Module1 (henceforth referred to as the home garden source model
[HGSM]) that was used, primarily, to estimate annual average concentrations of metals in garden
soil based on the predicted losses associated with various environmental processes (e.g.,
overland runoff, particulate emissions, leaching).
The HGSM is based on the Generic Soil Column Model (GSCM), a generalized solution
that was developed to simulate the dynamic changes in chemical constituent mass fate and
transport within the field and near-surface soils in watershed subareas. Governing equations for
the GSCM are similar to those used by Jury and colleagues (1983 and 1990) and Shan and
Stephens (1995). However, the analytical solution techniques used by these researchers were not
applicable to the source emission module developed here because they did not consider
constituent mass loss rates in the surface soil from runoff, wind and water erosion, leaching, and
mechanical processes.
Section G.2 describes the assumptions, governing equations, boundary conditions, and
solution technique that were originally developed as the GSCM. Section G.3 describes the
implementation of the HGSM to the garden scenario; specifically, how the GSCM and various
components (e.g., hydrology, soil erosion, and runoff water quality) are integrated with the local
and regional watersheds. Additional details are included in three attachments: Attachment A
lists and defines all symbols used in Sections G.2 and G.3; Attachment B provides
supplementary information on particulate emission equations; and Attachment C presents the
HGSM input parameters used in the SFS analysis. Attachment D describes the modeling that
was performed to estimate the location-specific dispersion and deposition factors originally
generated to support EPA's biosolids evaluation. The subset of these factors relevant to SFS
economic feasibility areas were mapped to and applied in modeling of the SFS gardening
scenario. Attachment E describes the soil Kd evaluation that was performed to examine the
impact of Kd distributions on SFS screening levels.
G.2 Generic Soil Column Model
G.2.1 Assumptions
The GSCM includes the following assumptions:
• The contaminant partitions to three phases: adsorbed (solid), dissolved (liquid), and
1 The information presented in this appendix on the LAU model is based on U.S. EPA, 1999.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-l
-------
Appendix G: Home Garden Source Model
gaseous (as in Jury et al., 1983 and 1990).
CT=PbCs+0wCL+0aCG (G-l)
where
CT = Total contaminant concentration in soil (g m"3 of soil)
pb = Soil dry bulk density (kg m"3)
Cs = Adsorbed-phase contaminant concentration in soil (g kg"1 of dry soil)
9W = Soil volumetric water content (m3 soil water m"3 soil)
CL = Aqueous-phase contaminant concentration in soil (g m"3 of soil water)
0a = Soil volumetric air content (m3 soil air m"3 soil)
Co = Gas-phase contaminant concentration in soil (g m"3 of soil air)
The contaminant undergoes reversible, linear equilibrium partitioning between the
adsorbed and dissolved phases (as in Jury et al., 1983 and 1990),
Cs=KdCL (G-2)
where Kd is the linear equilibrium partitioning coefficient (m3 kg"1). For inorganic
contaminants Kd is a specified input parameter.2 For organic contaminants,
Kd=foc-Koc (G-3)
where foe is the fraction organic carbon in soil and Koc is the equilibrium partition
coefficient (m3 kg"1), normalized to organic carbon.
The contaminant is in equilibrium between the dissolved and gaseous phases, and follows
Henry's law (as in Jury et al., 1983 and 1990),
CG = H'CL (G-4)
where H is the dimensionless Henry's law constant.
The total contaminant concentration in soil can also be expressed in units of ug of
contaminant mass per g of dry soil (ug g"1):
(G-5)
Pb
Using the linear equilibrium approximations in Equations G-2 through G-5, CT can be
expressed in terms of CL, Cs , or Co:
2 Linear equilibrium partitioning assumes that the sorptive capacity of the soil column solids does not become
exhausted.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-2
-------
Appendix G: Home Garden Source Model
CT=KTLCL=CS=CG (G-6)
Kd H
where
KTL=pbKd+ew+eaH' (G-7)
KTL is the dimensionless equilibrium distribution coefficient between the total and
aqueous-phase constituent concentrations in soil.
• The total water flux or infiltration rate (I, m d"1) is constant in space and time (as in Jury
et al., 1983 and 1990) and > 0. It is specified as an annual average.
• The soil column is an unconsolidated, homogeneous, and porous medium whose basic
properties (p\,,foc, 0W, 6a, and r\ — the total soil porosity) are average annual values,
constant in space.
• Contaminant mass may be lost from the soil column as a result of one or more first-order
loss processes.
• The total chemical flux is the sum of the vapor flux and the flux of the dissolved solute
(as in Jury et al., 1983 and 1990).
• The chemical is transported in one dimension through the soil column (as in Jury et al.,
1983 and 1990).
• The vapor-phase and liquid-phase porosity and tortuosity factors obey the model of
Millington and Quirk (1961) (as in Jury et al., 1983 and 1990) (see Equation G-9 below).
• The modeled spatial domain of the soil column remains constant in volume and fixed in
space with respect to a vertical reference (e.g., the water table).
G.2.2 Governing Mass Balance Equation
Under the previously mentioned assumptions, the governing mass fate and transport
equation can be written as follows:
= DE--VE-kCT (G-8)
Where k (1 d"1) is the total first-order loss rate and DE (m2 d"1) is the effective diffusivity
in soil calculated as follows:
Where Da and Dw (cm2 s"1) are air and water diffusivities, respectively, and 8.64 is a
conversion factor ((m2-s) (cm2-d)"1). DE is the sum of the effective gaseous and water
diffusion coefficients in soil, DE,a, and DE,W, respectively, where
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-3
-------
Appendix G: Home Garden Source Model
_
Tl KTL
and
6D 8.64
The effective solute convection velocity (VE, m d"1) is equal to the water flux corrected
for the contaminant partitioning to the water phase as follows:
^TL
G.2.3 Parameter Estimation Methodologies
Water content (0w) is estimated as a function of the annual average infiltration rate (I, m d"1)
using Equation G-13, from Clapp and Hornberger (1978):
( I y(2SM,,+3)
dw~rl\Q.24Ksat} (G"13'
Where Ksat (cm h"1) is saturated hydraulic conductivity, 8Mb is a unitless exponent
specified by soil-type, and 0.24 ((m-h) (cm-d)"1) is a unit conversion factor.
Volumetric air content is estimated using Equation G-14:
H, Da, and Dw are either estimated as a function of temperature in the soil column (Tsc, °C)
or specified directly as input parameters if pre-adjusted values are available.
G.2.4 Solution Technique
G.2.4.1 Background
The governing equation (Equation G-8) was solved to evaluate the following in a soil
column of depth zsc,
• Total contaminant concentration as a function of time, t, and depth below the surface, z;
and
• Contaminant mass fluxes across the upper (z=0) and lower boundaries (z=zsc) of the soil
column.
A numerical solution of Equation G-8, with zero concentration at the soil surface and
zero release at the bottom of the soil column, was first examined using a straightforward explicit
finite difference approach. This approach resulted in such a high numerical diffusion that it was
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-4
-------
Appendix G: Home Garden Source Model
impossible to distinguish diffusion effects. Subdividing each section into relatively thinner
sections reduced the numerical diffusion to more manageable levels, but also required smaller
time steps, and the computation time became quite long. In addition, the numerical solution was
not stable in extreme situations (e.g., high/low VE or DE).
An alternative solution was developed using a quasi-analytical approach. The quasi-
analytical solution allows for relative computational speed and significantly reduces concern
about numerical diffusion and lack of stability. The tradeoff is a loss of ability to evaluate short-
term trends in concentration and diffusive flux profiles. The alternative solution estimates long-
term (i.e., annual average) contaminant concentration profiles and mass fluxes.
The alternative solution consists of a superposition of analytic solutions of the three
components of the governing equation (Equation G-8) on the same grid. The solution for a
simplified case where the soil column consists of one homogeneous zone whose properties are
uniform in space and time is described below. Adaptations of the solution technique to account
for variations from this simplified case are described in the module-specific sections.
G.2.4.2 Description of Quasi-Analytical Approach
The quasi-analytical approach is a step-wise solution of the three components of the
governing equation (Equation G-8) on the same grid. That is, the following equations are solved
individually:
dC d2C
dt
dCT
= -kC
~, -T V^-1"1'
ot
Boundary conditions of Ci=0 at both the upper and lower boundaries of the soil column
are assumed, although some flexibility exists for specifying the lower boundary condition, as
discussed below.
Equations G-15 through G-17 each have an analytical solution that can be combined to
obtain a pure diffusion solution that moves with velocity VE through the porous medium (lost,
1960). The solution of the general differential equation is then the solution of the diffusive
portion with its time dependence, translating in space with velocity VE, and decaying
exponentially with time.
The first two solutions for a point source are graphically illustrated in Figures G-l and
G-2. If it were possible to compute such point source solutions for each position in the soil
column and each time of interest, then the governing differential equations would be linear and
the contributions at each point could be added to obtain a global solution. That is, each point in
the soil column could be treated as if it were the only point for which there is a nonzero
concentration.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-5
-------
Appendix G: Home Garden Source Model
Figure G-l. Diffusive spreading from a point
source with time, at times 0.01, 0.05, and 0.4.
Figure G-2. Diffusive spreading from a point
source with a constant velocity to the right,
at times 0.01, 0.05, and 0.4.
To make the analysis tractable, instead of a point source, the soil column is divided into
layer sources each of depth dz (i.e., a grid). A layer source can be thought of as multiple point
sources packed closely together. In such a case, Equation G-l5 has a solution for one-
dimensional diffusion, with the concentration at any point and any time given by Equation G-l8
for a layer of width dz centered at z'=0 (lost, 1960):
+ erf
dz/2-z'
(G-l 8)
The concentration profile is assumed to be initially uniform from z'=-dz/2 to z'=+dz/2 and zero
everywhere else. With time, the profile spreads outward and the concentration at the origin
decreases, as shown in Figure G-3 for dz=2. The concentration profile also moves down through
the soil column at velocity VE, as illustrated in Figure G-4. Layer solutions assume uniform
average concentrations within each layer. Thus, the thickness of the layers determines the spatial
resolution available.
Figure G-3. Diffusive spreading from a layer
source with time, at times 0.01, 0.05, and 0.4.
Figure G-4. Diffusive spreading from a layer
source with a constant velocity to the right,
at times 0.01, 0.05, and 0.4.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-6
-------
Appendix G: Home Garden Source Model
The total amount of material, m (g mf2), that has passed any ordinate z1 after time, t, is
given by the integral of the concentration from z1 to oo with one-half leaving to the left (negative
z1 values) and one-half to the right (positive z1 values):
m(
(G-19)
Deriving the integral in Equation G-19 results in
erfc(y)dy -
(G-20)
which is evaluated using the relationship from Abramowitz and Stegun (1970):
J erfc(x)dx = x erfc(jc) —•== exp (- x2)+ constant (G-21)
The fraction of the original mass that diffuses past a boundary at z1 in any time period 0
to t, Df(z',t) (g m"2), is one-half m(z',t) divided by the amount of mass initially present in the
source layer (Cixrdz):
Df(z't}-05
LJJ \Z ,1) — U. J
erfc(y}dy-
(z'-dz/2)/,]4DEt
(G-22)
The fraction of mass that remains in the original layer of width dz after diffusion in the
time period 0 to t, Dfo(t), is
(G-23)
By evaluating all the layer boundaries (z'=0.5 dz, 1.5 dz, 2.5 dz, ...), the amount of
contaminant mass transported to any layer via diffusion after time, t, can be calculated as the
difference between the amount outside the upstream boundary and the amount outside the
downstream boundary. For example, the fraction of mass originally present in the source layer
that ends up in the layer adjacent to the source layer in time, t, is Df(z'=0.5 dz, t) -Df(z'=1.5 dz,
t). The integrated amount of material that has crossed into the adjacent layer and the amount that
remains in the source layer after time, t, are given directly by Equations G-22 and G-23,
respectively, and only have to be computed once for fixed time steps and layer thicknesses.
The amount of mass that diffuses from a given layer out the lower boundary of the soil
column in time, t, can be tracked by multiplying Df(z',t)—evaluated where z1 is at the bottom of
the soil column (z=zsc)—by (Cio -dz) for that layer. Diffusive losses across the bottom boundary
from all the soil column layers are summed to calculate the total diffusive (aqueous- and
gaseous-phase) loss across the bottom boundary, Michd(t) (g m"2), in time, t.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-7
-------
Appendix G: Home Garden Source Model
Likewise, the total diffusive loss out the top of the soil column, Mo(t) (g m"2), is the sum
of the total diffusive losses across the upper boundary from each layer. The volatilization loss
from the surface of the soil column, Mvoi(t) (g m"2), is assumed to be from gaseous-phase
diffusion only and is determined by
where (pE,a/DE) is the fraction of the total diffusive loss from any layer that is due to
diffusion in the gaseous phase in the soil.
It is assumed that mass is not lost across the top soil boundary due to diffusion from the
aqueous phase. To maintain mass balance, mass calculated to be lost this way is added back into
the top soil layer, augmenting the total contaminant concentration there by (Mo(t) DE,W/DE). This
method of approximating Mvoi(t) is justified on the basis of computational efficiency. A more
rigorous treatment would include a mathematical transition layer across which diffusion from the
soil to the air occurs. However, use of such a transition layer would require a more
computationally intensive solution technique, as well as specification of the thickness of the
transition layer.
Without this approximation (i.e., if Mvoi(t)=Mo(t)), Mvoi(t) could be >0 for nonvolatile
contaminants (Da=H=0) because of the possible contribution to Mo from the aqueous-phase
diffusive flux. Estimating Mvoi(t) and augmenting the total contaminant concentration in the
surface layer is considered a reasonable approximation of what actually occurs. That is,
contaminant mass diffuses to the surface in both the aqueous and gaseous phases. While the
contaminant mass in the gas phase volatilizes out the surface of the soil column, the contaminant
mass in the aqueous phase is left behind, concentrating the contaminant mass in surface soil
(approximated here as the surface soil column layer).
To account for decay, Equation G-17 is solved using the technique of separation of
variables (lost, 1960). The solution takes the form
CT=CTOexp(-kt) (G-25)
As Equation G-25 is applied to each layer, the amount of mass lost as a result of first-
order decay in time, t, Mioss (g m"2), can be tracked using
^.(0 = (1 - exp(-to))CVo - dz (G-26)
If multiple first-order loss processes occur (i.e., k=£kj), the fraction of initial mass lost as
a result of each process] is determined using the following equation:
A potential difficulty with the layer solution is that the convection of material leads to an
artificial numerical diffusion because the concentration within each layer can only be expressed
as an average value. This component of numerical diffusion can be avoided completely if the
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-8
-------
Appendix G: Home Garden Source Model
contents of each layer are transferred completely to the next layer at the end of each time step by
making the time step equal to the layer thickness divided by the effective velocity, VE:
dt = ^- (G-28)
V E
The contaminant mass in the bottom layer is convected out of the lower boundary. Total
mass lost due to advection in dt, Micha (g m~2), is simply CTO in the lowest soil column layer
multiplied by dz.
To summarize the overall solution technique, the three processes (diffusion, first-order
losses, and advective transport) are considered separately, in series, and then combined (under
the justification of the superposition principle for linear differential equations) to result in the
chemical concentration vertical profile at the end of a computational time step. Specifically, the
chemical concentration profile after diffusion only is simulated first. Next, the chemical mass in
each computational cell (the mass after diffusion) is decreased to account for first-order loss.
Finally, after sufficient time has elapsed (which may take multiple time steps) for the chemical
mass in a cell to advect (at the sorption-corrected velocity) the thickness of the cell, all remaining
chemical mass translates to the next lower cell. This completes the series solution of the overall
fate and transport governing equation.
G.2.4.2.1 Boundary Conditions
Zero concentration is assumed at the upper boundary of the soil column. This is
consistent with the assumption that the air is a sink for volatilized contaminant mass, but requires
the approximate method for estimating Mvoi(t) described above.
At the lower boundary of the soil column, the flexibility exists with this solution
technique to specify a value between zero and 1 for the ratio (bcm) of the total contaminant
concentration in the soil directly below the modeled soil column and in the soil column. A ratio
of one (bcm=l) corresponds to a zero gradient boundary condition (dCi/dz=0). A ratio of zero
(bcm=0) corresponds to a zero concentration boundary condition (Ci=0).
When bcm is equal to zero, diffusive fluxes at the upper and lower boundaries of the soil
column are calculated directly as previously described. When bcm is >0, a reflection of the soil
column is created. The contaminant concentrations in the reflected soil column cells are set equal
to bcm multiplied by the contaminant concentration in the soil column cell being reflected (i.e.,
the concentration in the first cell of the reflected soil column is set to bcm multiplied by the
contaminant concentration in the lowest cell of the actual soil column). The upward diffusive
flux from the reflected soil column cells (1) offsets the diffusive flux out the lower boundary of
the soil column, (2) increments the contaminant concentrations in the soil column, and (3)
augments the diffusive flux out the upper boundary of the soil column. Hence, when bcm is
equal to 1 (the no diffusion boundary condition), the downward diffusive flux out the bottom
boundary of the soil column is completely offset by the upward diffusive flux across the same
boundary from the reflected soil column cells.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-9
-------
Appendix G: Home Garden Source Model
G.2.4.2.2 Algorithm
The general algorithm for applying the individual solutions to Equations G-15 through
G-17 is as follows for a homogeneous soil column and an averaging time period of 1 year:
1. Specify
- Lower boundary condition multiplier (bcm)
- Initial conditions in soil column (Cio)
- Soil column size (zsc) and properties (pt,, foe, r|, Ksat, 8Mb)
- First-order loss rates (kj)
- Chemical properties (Koc, H', Da, Dw)
- Upper and lower averaging depths (zava, zavt>).
2. Calculate/read Kd, which is internally calculated for organics (Kd = Koe x Foe), and read as
a user input for metals.
3. Subdivide the soil column into multiple layers of depth, dz, that are an integral fraction of
zsc. Calculate the total number of layers, Ndz= zsc/dz.
4. Derive an annual average infiltration rate (I) for the year.
5. Calculate 0W, 6a, KTL, DE, and VE.
6. Calculate the time to cross a single layer at velocity VE (Equation G-28). This is the
convection-based computing time step, dt (see also the note below).
7. Evaluate the fraction of mass that remains in a layer (Equation G-25) and that diffuses
across layer boundaries z'=0.5 dz, 1.5 dz, 2.5 dz,... (Equation G-24) at t = dt. (These
fractions are constant for a fixed dt.)
8. Calculate the amount of mass present in the soil column at the beginning of the year
(MCoii, g m'2).
9. Initialize cumulative mass loss variables (Mvoi, Michd, MM™, and MiOSs,j).
10. Diffusion. Adjust the concentration profile to reflect diffusive fluxes for one time step.
This redistributes material throughout the whole soil column. Increment Mvoi and Michd.
11. First-order losses. Allow the concentration profile to decay in each layer (Equation G-27)
for one time step. Increment mass lost due to all applicable first-order loss processes, j,
Miossj (Equation G-25).
12. Convection. Propagate the concentration profile one layer downstream. Increment Micha.
13. Repeat Steps 10 through 12 until it is time to add and/or remove contaminant mass
(proceed to Step 14) or until the end of the year (proceed to Step 15).
14. To account for the addition of contaminant mass, update the contaminant concentrations
in the affected layers. Track total mass added (Madd, g m"2) and/or removed (Mrem, g m"2).
Begin the algorithm again at Step 10.
15. At end of the year, calculate/report
- Total mass in the soil column (Mcoi2, g m"2)
- Mass balance error for the year (Men-, g m"2):
Merr = Mcol2 -Mcoll -Madd +Mrem +Mvol +Mlcha +Mlchd + ^M^ (G-29)
j
- Annual average total concentration in surface layer
- Annual, deptG-weighted average total concentration (zava < z < zavb )
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-10
-------
Appendix G: Home Garden Source Model
- Annual average volatilization flux (Jvoi, g m"2 d"1):
- Annual average leaching flux (Jich, g m"2 d"1):
T MlcM
-
c
JOJ
16. Begin the algorithm again at Step 4 until mass is no longer added to the soil column and
mass has been depleted from the soil (i.e., Mcoi2=0).
Note that the convection time step cannot be any greater than the length of time between
mass additions or removals (e.g., soil applications). For example, if contaminant mass is added
every 30 days, then this is the maximum time step, regardless of how small the velocity is. This
limited dt is used to calculate the number of time steps required before convective transfer takes
place, and the convective transfer step is performed on an as-needed basis. If the calculated
convective time step in the above example is 60 days, then the convective transfer occurs every
other time step. Over several steps this results in a temporal distortion of the concentrations
within the layers, but the effects average out by the end of the year.
To check the performance of the solution algorithm, Equation G-29 tests if the change in
mass in the system over the year is equal to the difference between mass additions and losses. If
the mass balance error (Merr) is >10"8 g m"2, then a message is written to the warning file.
G.3 HGSM Implementation
G.3.1 Introduction
The HGSM provides annual average contaminant mass flux rates from the surface of the
field and contaminant mass emission rates due to particulate emissions. To ensure transparency,
this report documents all of the major theory, algorithms and functionality implemented in the
HGSM, and identifies those used in this evaluation.3
The HGSM assumes that the home garden is one component of a broader watershed, and
so is affected by runoff and erosion from upslope land areas. The watershed, including the home
garden, is referred to as the "local" watershed and is illustrated in Figure G-6. A local watershed
is defined as that drainage area that just contains the home garden or a portion thereof (there can
be multiple local watersheds) in the lateral (perpendicular to runoff flow) direction, and in which
runoff occurs as overland flow (sheet flow) only. This distinguishes it from the "regional"
watershed, which is modeled when estimates are needed for downslope soil concentrations and
waterbody loadings of chemical constituents that are released from the field. Although the local
watershed extends downslope to the point that runoff flows and eroded soil loads would enter a
3 For other analyses, the HGSM is also used to predict the delivery of chemical constituents to downslope land areas
and waterbodies due to runoff and erosion.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-l 1
-------
Appendix G: Home Garden Source Model
well-defined drainage channel (e.g., a ditch, stream, lake, or some other waterbody), the SFS risk
modeling screening did not "track" the constituent loads once released from the field. Rather, the
HGSM was used only to predict the metal concentrations in soil after losses (e.g., runoff,
erosion) have occurred.
Local watershed
Drainage
divide
• Ihmoff flow
direction
Figure G-5. Regional watershed containing the home garden.
Figures G-6 and G-7 illustrate how the local watershed is conceptualized for the
combined Local Watershed/Soil Column Module (i.e., as a two-dimensional, two-medium
system. The dimensions are longitudinal (i.e., downslope or in the direction of runoff flow) and
vertical (i.e., through the soil column). The media are the soil column and, during runoff events,
the overlying runoff water column. In the longitudinal direction, the local watershed is made up
of a number of land subareas that may have differing surface or subsurface characteristics (e.g.,
land uses, soil properties, and chemical concentrations). For example, subarea 2 might be a home
garden, subarea 1 an upslope area, and subareas 3 through N downslope buffer areas extending
to the waterbody.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-12
-------
Appendix G: Home Garden Source Model
Watershed Divide
Figure G-6. Local watershed.
M
1 Column
Figure G-7. Cross-sectional view of a
watershed for the soil column model.
G.3.2 Hydrology
G.3.2.1 Overview
Hydrologic modeling simulates watershed runoff and groundwater recharge (termed here
as "infiltration"). The hydrology module is based on a daily soil moisture water balance
performed for the root zone of the soil column. At the end of a given day, t, the soil moisture in
the root zone of an arbitrary watershed subarea, i, is estimated as
,^ + Pt + R0^t - R0ht - ETt, - INht (G-33)
where
SM;,t = Soil moisture (cm) in root zone at end of day t for subarea i
SMi,t-i = Soil moisture (cm) in root zone at end of previous day for subarea i
Pt = Total precipitation (cm) on day t
ROi-i,t = Storm runoff (cm) on day t coming onto subarea i from i-1
RO;,t = Storm runoff (cm) on day t leaving subarea i
ET;,t = Evapotranspiration (cm) from root zone on day t for subarea i
IN;,t = Infiltration (groundwater recharge) on day t (cm) for subarea i
Frozen precipitation is treated as rainfall. Runoff, evapotranspiration, and infiltration
losses from the root zone are discussed in subsequent sections. The equations presented in these
sections refer to "day t and subarea i" in accordance with the water balance equation (see
Equation G-33).
G.3.2.2 Runoff
G. 3. 2. 2. 1 Governing Equations
Daily runoff is based on the Soil Conservation Service's (SCS's) widely used "curve
number" procedure (USD A, 1986) and is a function of current and antecedent precipitation and
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-13
-------
Appendix G: Home Garden Source Model
land use. Land use is considered empirically by the curve numbers, which are catalogued by land
use or cover type (e.g., woods, meadow, impervious surfaces), treatment or practice (e.g.,
contoured, terraced), hydrologic condition, and hydrologic soil group.
Runoff depth is calculated by the SCS procedure as
for p>ia (G-34)
P-Ia + S
where
RO = Runoff depth (cm)
P = Precipitation depth (cm)
la = Initial abstraction (threshold precipitation depth for runoff to occur) (cm)
S = Watershed storage (cm)
By experimentation with more than 3,000 soil types and cover crops, the SCS developed
the following relationships for watershed storage as a function of curve number (CN) and initial
abstraction as a function of storage:
5A (G-35)
^ }
CN
Ia = 0.2S (G-36)
Combining Equations G-34 and G-35 results in
RO = -. for p > ^
.P + 0.8S V '
RO = OforP^ 0.2S (G-3 8)
where S is given by Equation G-35. For impervious surfaces (CN=100), it can be
observed that RO=P.
Three antecedent moisture classes (AMCs) were used to adjust the SCS curve numbers as
shown in Table G-l. The growing season is assumed to be June through August (Julian Day 152
to 243) throughout the country.
Curve numbers are typically presented in the literature, assuming average antecedent
moisture conditions (AMC II), and can be adjusted for drier (AMC I) or wetter (AMC III)
conditions as (Chow et al., 1988).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-14
-------
Appendix G: Home Garden Source Model
Table G-l. Antecedent Moisture Classes for SCS Curve Number Methodology
Antecedent Moisture
Class
I
II
III
Total 5-Day Antecedent Rainfall (cm)
Dormant Season
<1.3
1.3 to 2.8
>2.8
Growing Season
<3.6
3.6 to 5.3
>5.3
Source: U.S. EPA (1985b)
These adjustments have the effect of increasing runoff under wet antecedent conditions
and decreasing runoff under dry antecedent conditions, relative to average conditions.
G.3.2.2.2
Implementation
In the conceptual model for the local watershed (Figure G-6), the subareas may have
different land uses and different curve numbers for each subarea. Equation G-37 is nonlinear in
the curve number; therefore, the method by which the SCS procedure is applied to multiple
subareas can make a significant difference in the resulting cumulative runoff values for
downslope subareas. There are essentially two options for implementing the procedure. The first
is based on runoff routing from each subarea to the next downslope subarea. That is, the runoff
depth from subarea 1 would first be calculated from Equation G-37. The cumulative runoff depth
from subareas 1 and 2 would then be calculated by applying Equation G-37 to subarea 2 and
adding (routing) the runoff depth from subarea 1. This would be repeated for all subareas. This
method is not appropriate for the sheet flow assumption of the local watershed and can give
much higher cumulative runoff depths (volumes) than would actually occur under the sheet flow
assumption. (The implicit assumption of the routing method is that the subareas are not
hydrologically connected [e.g., runoff from subarea 1 is captured in a drainage system [non-
sheet-flow] and diverted directly to the watershed outlet without passing through/over downslope
subareas.)
A different, nonrouting method is appropriate for implementing the SCS procedure for
the local (sheet flow) watershed. The method is based on determining composite curve numbers
and is analogous to the nonsoil routing implementation of the Universal Soil Loss Equation
(USLE) soil erosion module presented in Section G.3.3.
CN(I) =
4.2CN(ll)
10-0.058CA/"(//)
,
0.13CAf(//)
(G-39)
(G-40)
The methodology used for implementing this method is illustrated by the following pseudo-code:
FOR i=l,...,N (subareas)
CNeff;=Area-weighted composite CN; for all subareas j,j=l,...,i
Calculate Si from equation (3.2.2-2) using CNeff;
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-l 5
-------
Appendix G: Home Garden Source Model
Calculate ROi from equation (3.2.2-1) using Si. (ROi is the average runoff depth
overall upslope subareas j, j=l,...,i).
Calculate Qi=ROi * WSAi where Q; is cumulative runoff volume and WSAi is
cumulative area.
IF i=l THEN
Hlj=ROi where HI; is subarea-specific runoff depth for subarea I (i.e.,
ROi-ROi-i)
ELSE
Hli=(Qi-Qi-i)/Ai where A; is subarea-specific surface area
IFHli
-------
Appendix G: Home Garden Source Model
The functional relationship in Equation G-41 is assumed here to be linear, so that ET
(cm) is calculated as
ET = min
PET, PET
FC-WP)
(G-44)
PET is estimated as described below.
The more theoretically based modules for daily ET (e.g., the Penman-Monteith equation
[Monteith, 1965]) rely on the availability of significant daily meteorological data, including
temperature gradient between surface and air, solar radiation, wind speed, and relative humidity.
All of these variables may not be readily available for all application sites and, therefore, the less
data-demanding Hargreaves equation was used (Shuttleworth, 1993). The Hargreaves method,
which is primarily temperature-based, has been shown to provide reasonable estimates of
evaporation (Jensen et al., 1990)—presumably because it also includes an implicit link to solar
radiation through its latitude parameter (Shuttleworth, 1993).
The Hargreaves equation is
PET = 0.0023S0A°r5(7 + 17.8)*0.1 (G-45)
where
PET = Potential evapotranspiration (cm d"1)
So = water equivalent of extraterrestrial radiation (mm d"1) and is given as
(Duffie and Beckman, 1980)
AT = Difference in mean monthly maximum and mean monthly minimum air
temperature
T = Mean daily air temperature (°C)
S0 = I5.392dr (tnsSin
-------
Appendix G: Home Garden Source Model
( 9 "\
0 = OA093Sin\ —J -1.405 (G-49)
^365 J
G.3.2.4 Infiltration (Recharge)
Any soil moisture in excess of the soil's field capacity (FC) that is not used to satisfy ET,
is available for gravity drainage from the root zone as infiltration to subroot zones (Dunne and
Leopold, 1978). This infiltration rate will, however, be limited by the root zone soil's saturated
hydraulic conductivity. Accordingly, infiltration is calculated as
= mn
Ksat,(SM-FC)
DRZ
(G-50)
100
where
IN = Infiltration rate (cm d"1)
Ksat= Saturated hydraulic conductivity (cm d"1).
If infiltration is limited by Ksat, the hydrology algorithm includes a feedback loop that
increases the previously calculated runoff volume by the amount of excess soil moisture (i.e., the
water above the field capacity that exceeds Ksat). This adjustment preserves water balance and is
based on the assumption that the runoff curve number method, which is only loosely sensitive to
soil moisture (through the antecedent precipitation adjustment), has admitted more water into the
soil column than can be accommodated by ET, infiltration, and/or increased soil moisture. After
the runoff is increased for this excess, the ET, infiltration, and soil moisture are updated to reflect
this modification and preserve the water balance.
G.3.3 Soil Erosion
G.3.3.1 General
The soil erosion module is based on the USLE, which is an empirical methodology (see,
e.g., Wischmeier and Smith, 1978) based on measured soil losses from experimental field-scale
plots in the United States for approximately 40,000 storms. The USLE predicts sheet and rill
erosion from hillsides upslope of defined drainage channels, such as streams; however, it does
not predict streambank erosion.
Let SL (kg m"2 time"1) denote the eroded soil flux (unit load) from a hillside area over
some time period. SL is predicted by the USLE as the product of the following six variables:
SL = RxKxCxPxLSxSd
Where
R = Rainfall factor (time"1). Accounts for the erosive (kinetic) energy of falling
raindrops, which is essentially controlled by rainfall intensity. The kinetic
energy of an individual storm multiplied by its maximum 30-minute
intensity is sometimes called the erosivity index (El) factor. R factors are
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-18
-------
Appendix G: Home Garden Source Model
developed by cumulating these individual storm El factors. R factors have
been compiled throughout the United States on a long-term annual average
basis.
K = Soil erodibility factor (kg m"2). An experimentally determined property and
is a function of soil type, including particle size distribution, organic
content, structure, and profile. K values are available from soil surveys and
databases (e.g., State Soil Geographic [STATSGO]).
C = Dimensionless "cropping management" factor. Varies between 0 and 1. C
accounts for the type of cover (e.g., sod, grass type, fallow) on the soil, and
is used to correct the USLE prediction relative to the cover type for which
the experimentally determined K values were measured (fallow or freshly
plowed fields).
P = Dimensionless practice factor. Accounts for the effect of erosion control
practices such as contouring or terracing. P is never negative, but could be
>1.0 if land practices actually encourage erosion relative to the original
experimental plots on which K was measured.
LS = Length-slope factor, accounts for the effects of the length and angle of the
slope of a field on erosion losses. LS is calculated by the following equation
from U.S. EPA (1985b):
LS, =.045Xi65AlSin20 + 4.56Sin0 + .065 (G-52)
where
X; = Flow length (m) from the point at which sheet flow originates (the
upslope drainage divide) to the point of interest on the hillside.
0 = Slope angle (degrees), where 0 may be calculated from percent slope,
S, as
9 = arctan(S 1 1 00) (G-53)
and b, the exponent, is determined as a function of S as
b = 0.5, ifS>.05
b = 0.4, if.035
-------
Appendix G: Home Garden Source Model
basin (Shen and Julien, 1993). The sediment delivery ratio is used to
account for deposition of eroded soil from the local watershed in ditches,
gullies, or other depressions.
Vanoni (1975) developed the sediment delivery ratio as a function of watershed
drainage area:
(G-54)
where
Sd = Sediment delivery ratio (dimensionless)
a = Normalized to give Sd = 1.0 for an area of 0.001 mi2 as per Vanoni
(1975) (for area in m2, a=2.67).
A = Sub-basin area (m2)
G.3.3.2 Daily USLE Implementation
The HGSM implements the USLE on a storm event basis using a modified USLE
procedure. This implementation requires determining a daily R value (Rt, d"1) that specifies the
erosivity of each daily storm.
For this evaluation, Rt is supplied from published long-term annual total R values. These
long-term annual total R values (published in the form of isopleths across the country) are
disaggregated down to daily values using the following method:
Given: Long-term annual total R for a site, Rann, (obtained from the isopleths)
Given: Number of years in the simulation, NYR
Given: Hourly time series of precipitation amounts for the complete record of NYR years
1. Compute cumulative R over record, Rtotai=Rann x NYR.
2. Compute cumulative precipitation over NYR years, PPTtotai.
3. For each hourly precipitation value in the record, allocate Rtotai to that hour based on the
fraction of PPTtotai represented by the hourly precipitation. Denote an hourly allocation as
Rhour.
4. For each day of the record, cumulate all Rhour values to the daily total. The result is Rt for
each day of the NYR record.
G.3.3.3 Spatial Implementation
For the local watershed application, the daily USLE is applied spatially to a hillside
comprised of N subareas (see Figure G-6). Pseudo-code for this application is
LET CSLi=Cumulative soil load (kg d"1) for subarea i (i.e., eroded load from subarea i)
and all upslope subareas j, j=l,...,i
LET WSAi=Cumulative land area (m2) upslope of and including subarea i
FOR 1=1,...,N
Keff;=Area-weighted K; for all subareas j, j=l,...,i
Ceff;=Area-weighted C; for all subareas j, j=l,...,i
Peff;=Area-weighted P; for all subareas j, j=l,...,i
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-20
-------
Appendix G: Home Garden Source Model
CSL;=R x WSA; x Keffi x Ceffi x Peff; x LSi x Sdi
NEXT!
G.3.4 Chemical Fate and Transport
G.3.4.1 Runoff Compartment
G.3.4.1.1 Introduction
The module used to estimate chemical and suspended solids concentrations in storm
event runoff is based on mass balances of solids and chemical in the runoff and the top soil
column layer of thickness dz. The soil compartment is external to this module (see Section
G.3.4.2), and results from that compartment are called as needed by the software. Solids and
chemical concentrations in the runoff are assumed to be at steady-state during each individual
runoff event, but can vary among runoff events (i.e., a quasi-dynamic approach). The assumption
of steady-state within each storm event is appropriate for the following reasons:
• Run-time considerations (i.e., maximize the numerical time step).
• Data are not available at the temporal scale to accurately track within-storm event
conditions (e.g., rainfall hyetographs).
• Because of the anticipated relatively small surface areas of the watershed subareas and
the associated relatively small runoff volumes, the actual time to steady-state may not
differ significantly from the 1 day or less implicitly assumed here. (A sensitivity analysis
was performed using a dynamic form of the runoff compartment module that suggested
relatively little difference in soil concentrations as a function of the steady-state versus
dynamic assumption.)
• To the extent that the actual time to steady-state would be >1 day, the module is biased
toward overestimating downslope concentrations and waterbody loads (i.e., it is a
protective assumption from the risk standpoint).
Figure G-8 presents the conceptual runoff quality module, showing the two
compartments and the fate and transport processes considered. Development of mass balance
equations for solids and chemical follow.4
4 Hydrolysis, volatilization, and biodegradation processes are not simulated in the runoff compartment. The
percentage of time that runoff is actually occurring will be sufficiently short that any additional losses from these
processes should be minimal. In addition, these processes are continuously simulated in the surface layer of the
soil column: To also include them in the runoff compartment would be double-counting.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-21
-------
Appendix G: Home Garden Source Model
Runon Flow
Partitioning
Dissolved •< >• Paniculate
RUNOFF
Diffusion
Settling
Resuspension
Runoff Flow
SURFACE SOIL
^*T
Burial/erosion
T
Figure G-8. Runoff quality conceptual model.
G. 3.4.1.2 Solids in Runoff Compartment
A steady-state mass balance of solids in the runoff (i.e., suspended solids from erosion),
written for local watershed subarea i is given by the following equation (in the subsequent
module development, units are presented in general dimensional format (i.e., M[ass]-L[ength]-
T[ime], for simplicity of presentation):
0 = Q',-imi,,-i -Q'mi, -vsiAimli+vriAiM7
(G-55)
where
mi,i=
M2 =
Qi =
QM=
A, =
VSi =
vn =
Qi =
CSL;
P =
p
CSL,
P
(G-56)
(G-57)
Solids concentration (M L"3) in the subarea i runoff (suspended solids)
Solids concentration (M L"3) in the top soil column layer of subarea i
Runoff flow (L3 T"1) leaving subarea i
Runon flow (L3 T"1) from subarea i-1
Surface area (L2) of subarea i
Settling velocity (L T1)
Resuspension velocity (L T"1)
Total runoff flow volume (L3 T"1) (water plus solids) leaving subarea i
= Cumulative soil load leaving subarea i (M T"1)
Particle density (M L'3) (i.e., 2.65 g m'3).
Note: Subscript "1" denotes the runoff compartment, whereas "2" denotes the top soil
column layer compartment.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-22
-------
Appendix G: Home Garden Source Model
The first term in Equation G-55 is the flux of soil across the upslope interface of subarea i. The
second term is the flux of soil across the downslope interface, the third term is an internal sink of
soil due to settling, and the fourth term is an internal source due to resuspension.
G. 3. 4. 1. 3 Solids in Soil Compartment
The HGSM does not consider chemical mass transport among watershed subareas due to
soil erosion because it is based on a single subarea only; therefore, that transport is considered
here. The HGSM assumes that solids mass transport to or from the soil compartment of any
given watershed subarea occurs only in a vertical direction (i.e., there is no downgradient
advection of the top soil column layer). (This is analogous to the assumption of a stationary
sediment bed in stream/sediment quality modules.) The downslope mass transport of soil occurs
due to vertical erosion or resuspension of soil followed by advective transport of the soil in the
runoff water as suspended solids. The transport is described in terms of the following three
parameters: settling, resuspension, and burial/erosion velocities. Under the assumption of no
advective transport of the soil column layer, the steady-state mass balance equation for the
surficial soil layer is
0 = vstmvAt - v^m^A, - vbtmvAt (G-5 8)
where
vb; = Burial/erosion velocity (L T"1).
The first term of Equation G-58 is a source of soil mass to the surficial soil column layer
due to settling from the overlying runoff water. The second term is a sink from resuspension. The
third term is either a source or a sink depending on the sign of the burial/erosion velocity as
described below.
Consider the solids balances in the runoff and soil compartments, Equations G-55
through G-58. These equations involve three parameters (i.e., vs, vr, and vb) and two solids
concentrations (i.e., mi and rm). Which of these five variables is known for arbitrary subarea i? It
can be assumed that the solids concentration in the soil (rm) is a known value — it is simply the
bulk soil density. Consider now the suspended solids concentration in subarea i, mi,;. From the
soil erosion module, the total solids mass fluxes moving across both the upslope and downslope
interfaces of subarea i are known, and these two fluxes are, respectively, the first two terms on
the right side of Equation G-55 mi,; and can then be determined as
(G-59)
where
CSL; = the cumulative soil load leaving subarea i, as determined by the soil erosion
module
Q'l = the cumulative runoff flow volume (including solids' volume) leaving
subarea i, as determined by the runoff quantity model.
Therefore, because the soil concentration (im) is assumed to be known, and the soil
erosion and runoff quantity modules can be used to determine the suspended solids
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-23
-------
Appendix G: Home Garden Source Model
concentrations (the mi,;), Equations G-55 through G-58 can now be considered as two equations
in three unknowns (i.e., vs, vr, and vb).
The settling (vs) and resuspension (vr) parameters reflect processes internal to subarea i,
whereas the burial/erosion parameter (vb) reflects net changes across subarea i and is completely
determined by the difference in the soil fluxes entering and leaving subarea i. This can be
observed by adding the right sides of Equations G-55 and G-58 and setting the result to zero. All
terms involving vs and vr cancel, and the burial/erosion velocity is then derived by
CSLH-CSL,
Atm2
where CSL;-i and CSL; denote the soil fluxes into and out of subarea i, respectively, as
previously discussed. From Equation G-60 it can be observed that, if the soil load
entering subarea i (CSLi-i) is greater than the soil load leaving (CSL;), then the
burial/erosion velocity is positive and soil is being deposited (buried). Conversely, as will
typically be the case, if the load leaving is greater than the load entering, then the
burial/erosion velocity will be negative and erosion is occurring.
With the net soil flux across the subarea having been determined, Equations G-55 and G-
58 are in fact the same equation — the burial velocity term is explicitly shown in Equation G-58
and implicitly shown in Equation G-55. Thus, either Equation G-55 or G-58 represents one
equation in two unknowns (i.e., vs and vr). If one of these is known, then the other can be solved.
Of the two, it would be very difficult to obtain estimates for the resuspension velocity, and the
settling velocity could be assumed similar to, for example, hindered or compaction settling in
sludge thickeners. Accordingly, the following equation determines vr as a function of vs (and vb,
which is determined using Equation G-60) for subarea i:
m2
The settling velocity, vs, is assigned values from a uniform random distribution between
the range 0.05 and 1.0 m d"1, based on observed settling velocities for "mineral" sludges in
sludge thickening experiments.
In summary, with ni2 known and mi calculated from results of the soil erosion and runoff
modules, the solids mass balance equations are used to determine the burial/erosion and
resuspension parameters for subsequent use in the chemical (contaminant) model.
G. 3. 4. 1. 4 Contaminant in Runoff Compartment
As shown in Figure G-8, a steady-state mass balance of contaminant in the runoff results
in the following equation:
(Fd7 . Fd,. }
0 = QiiCv-i ~Q^,, -vsAFp^ +vrtAtFpvErtcv +vdtAp2\—±cv —^^ (G-62)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-24
-------
Appendix G: Home Garden Source Model
where
ci,i = Total contaminant concentration (particulate + dissolved) in runoff in
subarea i (M L"3)
C2,i = Total contaminant concentration in soil (M L"3)
Fpi,; = Fraction particulate in runoff
En = Enrichment ratio
vd; = Diffusive exchange velocity (L T"1)
2term, 2 is used to account for the fact that diffusion of dissolved chemical will only occur
across the interstitial area, not the entire interface area.
Equation G-62 can be used to express ci,i as a function of ci,i-i and C2,i as
^K, ((
Q', + vs^Fpv + vd,A,3>
where C2,i is determined by the HGSM as described in Section G.2. Determination of the
individual terms constituting this equation is described below.
Fpi,; is calculated using the following equation from Thomann and Mueller (1987):
,., fc/^uK* (r ^
Fp, . = —^ LJ-TJ— (G-66)
^
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-25
-------
Appendix G: Home Garden Source Model
where
kd = Chemical-specific partition coefficient (L3 M"1) (Note: kd is divided by
porosity to attain the porosity-corrected kd with units of mass per total
[liquid plus solids] volume.)
Fp2,i is similarly calculated as
(yo>
2'! i + /o v ;
where Fp2 (and Fd2) will be constant among all subareas i.
Fdi,; and Fd2,i are then determined as
=l-Fpv (G-68)
Fd2i=l-Fp2i (G-69)
Assuming that resistance to vertical diffusion is much greater in the soil than in the runoff
(Thomann and Mueller, 1987, p. 548), the diffusive exchange velocity, vd;, can be expressed as
vdt = (G-70)
Lc
where
Dw = Water diffusivity (L2 T1).
Lc = Characteristic mixing length (L) over which a concentration gradient exists;
assumed to be the depth of the runoff volume, including the solids (HI1):
I
The enrichment ratio, En, is used to account for preferential erosion of finer soil particles,
with higher specific surface areas and more sorbed chemical per unit area, as rainfall intensity
decreases. That is, large (i.e. highly erosive) runoff events may result in average eroded soil
particle sizes and associated sorbed chemical loads that do not differ much from the average
sizes/loads in the surficial soil column layer. However, less intense runoff events will erode the
finer materials, and resulting runoff chemical loads could be significantly higher than
represented by the average soil concentration. U.S. EPA (1985b) gives the storm event-specific
enrichment ratio as a power function of sediment discharge flux (M L"2). This formulation results
in
(G-72)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-26
-------
Appendix G: Home Garden Source Model
where
a = 7.39 for CSLi/WSA; in kg ha'1 (U.S. EPA, 1985b).
(CSL; is the event soil load leaving subarea I, and WSA; is the local watershed surface
area from the drainage divide down to and including subarea i.) The enrichment ratio is >
1.0. Should specific values of the sediment discharge (the denominator) result in an
enrichment ratio <1.0, it is reset to 1.0 in the code.
G.3.4.2 Soil Compartment
The GSCM (see Section G.2.2) is coupled to the runoff compartment module (see
Section G.3.4. 1) in this section and applied to the several subareas that constitute the sheet flow
for the local watershed of which the home garden is an integral part. Continuing the chemical
concentration indexing scheme (i.e., subscript "1" denoting runoff compartment, and subscript
"2" denoting surficial soil compartment), let the total (dissolved, paniculate, and gaseous phase)
chemical concentration in the surficial soil column layer of any local watershed subarea i be
denoted as C2,i. C2,i is equivalent to CT. From Section G.2.2 (GSCM), the governing differential
equation for the surface soil layer of subarea i is
riC
(G-73)
where kj represents the first-order rate constant due to process], and does not include
runoff/erosion processes (i.e., biological decay and hydrolysis and wind/mechanical action). The
last term, ss;, is a source/sink term representing the net effect of runoff and erosion processes on
C2,i as shown in Figure G-8. This term is derived by the following equation:
Fd Fd
sst = - - - - - - (G-74)
dz
where vs;, vn, vd;, and vb; denote, respectively, the settling, resuspension, diffusive exchange,
and burial/erosion velocities for subarea i as described in the runoff compartment model. Thus,
the terms comprising ss; are, respectively, a source of chemical due to settling from the overlying
runoff water, a sink of chemical due to resuspension, and a source or sink (depending on the
relative values of Ci,; and C2,i) due to chemical diffusion to and from the runoff.
The burial/erosion mechanism introduces a minor mass balance error into the model. The
module for surface soil/runoff water fate and transport (Section G.3.4.1) is based on a conceptual
model originally developed for use in a stream/sediment application (e.g., Thomann and Mueller,
1987) where the sediment compartment location relative to a reference point below the surface
can move vertically ("float") as burial and erosion occur. In that moving frame of reference,
burial/erosion of contaminant does not introduce a mass balance error because, with respect to
the modeled sediment, this sink/source of contaminant is exogenous to the modeled system (i.e.,
it is coming from/going to outside of the modeled system). There is internal (endogenous) mass
balance consistency within the modeled system. However, the frame of reference is not allowed
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-27
-------
Appendix G: Home Garden Source Model
to float, but is fixed by the elevation of the lower boundary (e.g., top of the vadose zone). Thus,
if a sorbed chemical is eroded from the surface cell, then that surface cell, which is vertically
fixed, must have a "source" that is internal to the modeled soil column to compensate for this
sink or its internal mass balance is not maintained. The magnitude of this mass balance error is
equal to the mass of eroded soil from the surface over the duration of the simulation multiplied
by its average sorbed chemical concentration. In most cases, this error as a percentage of the total
chemical mass in the modeled home garden will be quite small, and that has been confirmed in
multiple executions of the module. Conceptually at least, the GSCM could be designed so that,
after each runoff event, the surficial soil compartment could decrease or increase in size to
accommodate the event's erosion/burial magnitude, while maintaining a fixed vertical reference.
Grouping coefficients of Ci,i and €2,1, Equation G-74 can be rewritten as
55,=a,Cv-*A/-**A' (G-75)
where
Fd
,
a, = - - - !i- (G-76)
dz
dz
(G-78)
»,.,
dz
and kbu,i is the first-order rate constant (1 T"1) associated with the burial/erosion process.
Using Equation G-75, Equation G-73 can be rewritten as
rf'
U
2,i Tr ^2,i \^ i ^ r< 1, r< 1 r< //-> -7n\
^ - VE —^ ~l_kCv + atC^ - b,Cv - kbu£v (G-79)
~ , -c ^ z -C ^
c* cfe cfe
Equation G-79 demonstrates that €2,1 is a function of Ci,;. Similarly, Equation G-65 of the
runoff compartment module demonstrates that Ci,i is a function of €2,1 Thus, €2,1 and Ci,i are
jointly determined at any time, t, by simultaneous solution of their two respective equations.
C2,i at time, t, can be determined by substitution for Ci,i. Using Equation G-65, Ci,i can be
expressed as
CM = g''-lC'-''-1 +—C2)! (G-80)
where
?d2,i (G-81)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-28
-------
Appendix G: Home Garden Source Model
Fd.
- (G-82)
Substituting for Ci,i from Equation G-80 into Equation G-79, the differential equation for
C2,i is now expressed implicitly as a function of Ci,i as
After C2,i at time, t, is determined by solution of Equation G-83, the associated value for
Ci,i can be found from Equation G-80, thus completing the simultaneous solution. (The value for
Ci,i-i [i.e., the runoff concentration in the immediately upslope subarea] will have been
determined previously during the simultaneous solution for the i-1 subarea at time, t.)
To implement the simultaneous solution, Equation G-83 can be simplified to
AT P>2r ^r
(G-84)
\ /
r\. c ^.
dt dz
where
(G-85)
(G-86)
-^-10,,-! (G-87)
k'i is the lumped first-order loss rate, which includes the effects of abiotic hydrolysis
(j=hy), aerobic biodegradation (j=ae), and wind/mechanical activity (j=wd), in addition to
runoff and erosion. kev,i is the storm event (or runoff and erosion) first-order loss rate. khy
and kae are inputs to the module. The last term, ldi-i, is the run-on load from upslope
subareas in g m"3 d"1.
Recall that the GSCM, the governing equation is broken up into three component
equations: diffusion, convection, and first-order losses (Equations G-15 through G-17), and each
equation is solved individually on a grid. In the subsurface layers, the solution technique
described in Section G.2 is applied directly. However, for the surface soil column layer, the first
two-component equations remain the same, but the third equation has been revised to
^f = -*'C2^_1 (G-88)
which has the following analytical solution for C2,i=C°2,i at t=0:
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-29
-------
Appendix G: Home Garden Source Model
C2°,exp(-£'0+&/,_,
l-exp (-£'/)
(G-89)
To track mass losses, the total mass added to the soil column in subarea i in any time
period zero to t due to settling from runoff water, Madd,i (M L"2), is evaluated using
A mass balance on the soil column in time, t, gives
(G-90)
--Madd,,-Mlosv (G-91)
where AM; (M L"2^ is the change in mass in the soil column in subarea i as given by
(C2,i~C2,i) x dz, and MiOSs,i (M L"2) is the total mass lost from the subarea i soil column in
any time period zero to t. By substituting Equation G-89 for €2,1 and Equation G-90 for
Madd,i and rearranging, MiOSS!i when k';=0 is found to equal 0, and the following equation
for Mioss,i was derived for k';>0:
dz
The total mass lost in any time period zero to t from subarea i soil column can be
attributed to specific first-order loss processes, j, M;(t) (M L"2) using
(G-92)
(G-93)
where j is hy for hydrolysis, ae for aerobic degradation, wd for losses due to
wind/mechanical activity, ev for runoff/erosion events, and bu for burial/erosion.
Equation G-80 provides the contaminant concentration in the runoff water at time, t. The
average contaminant concentration in the runoff water (Ci,i) over time zero to t is determined
using
- Q i-i
-
d2
lt -
— C2,,-
d 2
(G-94)
where €2,1 is the time-weighted average contaminant concentration in the soil
compartment over the same time period. Given the short time step (i.e., 1 day) used in the
integration of the local watershed/soil column module, €2,1 is approximated using
(G-95)
where the 0 superscript denotes concentration at the beginning of the day.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-30
-------
Appendix G: Home Garden Source Model
G.3.5 Implementation
G.3.5.1 Overview
The HGSM implementation essentially links the regional watershed, GSCM, and local
watershed at scales that are appropriate to the modeling questions. For example, at the regional
level, the infiltration is assumed constant and convection events occur at regular intervals
throughout the simulation. At the local watershed level, the infiltration rate (I) is allowed to vary
from year to year; as a result, convection events are not "required" to occur at regular intervals
(see Figures G-9 and G-10 for the local watershed logic). To determine the appropriate time to
initiate a convection event, a variable (fadv) tracking the fraction of mass in the bottom soil
column layer that would have convected is incremented by (dt-VE/dz) at the end of every time
step. If fadv is sufficiently close to 1, then a convection event is initiated, and fadv is reset to
zero.
is incremented by fadv multiplied by dz multiplied by CT in the lowest layer, and
CT in the lowest layer is adjusted accordingly. Leachate flux for the final year is then calculated
using Equation G-3 1 .
G.3.5.2 Simulation-Stopping Criterion
For a given local watershed, i, the simulation is stopped in each successive subarea when
the amount of contaminant mass in local watershed i and all upslope subareas j (j
-------
Appendix G: Home Garden Source Model
No No
No
Next year, y = y + 1
For all subareas, get daily and annual average I, Q, CSL
Next subarea, i = i + 1
Get time constant subarea soil column parameters
Calculate annually variable subarea soil column parameters
Calculate time step dt (d) and diffusion fractions
Diffusion: Update CT. Increment MVO|, M|Chd.
First order losses, surface: Calculated daily.
First order losses, subsurface: Update CT. Increment MIOSSJ
(j = ae.an, or hy)
:tion: Propagate CT down as needed. Increment M|Cha.
Output annual average fluxes and
surface CT. Initialize M's.
7
Output annual average load to
waterbody
7
Figure G-9. Overview of algorithm for local watershed/GSCM.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-32
-------
Appendix G: Home Garden Source Model
t'
= t-dt
Yes
Calculate kevj, ldM, d^, and d2j
Update CTi (same as C2,i)
Increment MIOSSJ Q = hy, ae, wd, ev)
No Increment load to waterbody
No
\
r
kevj = ldM = 0
\
r
Update CT,i
I
Increment M|OSSjQ = hy, ae, wd)
Yes
Continue
with
i Flowchart 1,
Figure G-10. Detail on calculation of first-order losses in surface layer.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-33
-------
Appendix G: Home Garden Source Model
G.3.5.3 Leachate Flux Processing
Preliminary module runs during model development demonstrated many cases in which
the convective transfer step occurred less than once per year, sometimes even less than once in
the entire simulation period. In these cases, the leachate flux was nonzero in the years when a
convection event occurred, and zero in years when it did not. This is a limitation of the solution
technique. In reality, leaching occurs more or less continuously over the time between
convection events. To mitigate this limitation, a leachate flux post-processing algorithm was
developed. The entire simulation (0
-------
Appendix G: Home Garden Source Model
Table G-2. Variables Summarizing Contaminant Mass Losses
Variable
fMvol_wmu
fMlch wmu
fMwnd_wmu
fMdeg_wmu
fMrmv_wmu
fMvol_sa
fMlch_sa
fMdeg_sa
fMswl
fMbur1
Definition:
Fraction of the Total Mass Added or Lost Due
to
Volatilization from the home garden
Leaching from the home garden
Wind/mechanical action on the home garden surface
Abiotic and biodegradation within the home garden
Removal from the home garden
Volatilization from the non-home garden subarea soil columns
Leaching from the non-home garden subarea soil columns
Abiotic and biodegradation in the non-home garden subarea
columns
soil
Runoff/erosion from the most downslope subarea
Burial/erosion in all subareas (see kbu in Equation G-87)
a fMbur is the only listed variable that can be negative (indicating a mass gain). This results
from including a burial/erosion term when linking the runoff and soil compartments (see
Figure G-8 and the discussion in Section G.3.4.2)
G.3.6 Output Summary
Table G-3 summarizes the HGSM outputs used in the SFS analysis.
• Emissions to Estimate Air Impacts. All annual time series outputs to ISCST3 are reported
up to and including the last year that there are nonzero volatile or particulate emission
rates (VE or CE).
• Soil Concentrations to Estimate Soil and Food Chain Exposures. The annual time series
of depth-weighted average soil concentration (CTda), used in plant root zone
calculations, is reported until soil concentrations reach zero. The same is true for the
surface soil concentration (CTss) used in SFS evaluation to estimate exposures due to
incidental soil ingestion as well as impact due to particulate emissions.
• Leachate to Estimate Groundwater Impacts with EPACMTP. The annual time series of
LeachFlux is reported until LeachFlux is zero. Annlnfil is reported from year 1 to the last
year that meteorological data are available.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-3 5
-------
Appendix G: Home Garden Source Model
Table G-3. Output Summary for the Home garden Source Model
Variable Name"
Documentation
I
Jvol
CE30
E30
pmf
Jlch
LeachFluxNY
CT
CT
Code
Annlnfil
VE
VERY
VENY
CE
CEYR
CENY
PESO
PE30YR
PE30NY
PMF
PMFYR
PMFNY
LeachFlux
LeachFluxYR
LeachFluxNY
SWLoadChem
SWLoadChemYr
SWLoadChemNY
CTss
CTssYR
CTssNY
CTda
CTdaYR
CTdaNY
SrcSoil
SrcOvl
SrcLeachMet
SrcLeachSrc
SrcVE
SrcCE
NyrMet
Definition
Leachate infiltration rate (annual average; home garden
subarea[s] only)
Volatile emission rate
Year associated with output
Number of years in outputs
Constituent mass emission rate-PM3o
Year associated with output
Number of years in outputs
Eroded solids mass emission rate-PM3o
Year associated with output
Number of years in outputs
Paniculate emission particle size distribution
Year associated with output
Number of years in outputs
Leachate contaminant flux
Year associated with output
Number of years in outputs
Chemical load to waterbody
Year associated with output
Number of years in outputs
Soil concentration in surface soil layer
Year associated with output
Number of years in outputs
Depth-weighted average soil concentration (from zava to
zavb)
Year associated with output
Number of years in outputs
Flag for soil presence (true)
Flag for overland flow presence (true)
Flag for leachate presence when leachate is met-driven
(true)
Flag for leachate presence when leachate is not met-
driven (false)
Flag for volatile emissions presence (true)
Flag for chemical sorbed to particulates emissions
presence (true)
Number of years in the available met record
Units
md'1
g m~2 d"1
Year
Unitless
g m~2 d"1
Year
Unitless
g m~2 d"1
Year
Unitless
Mass fraction
Year
Unitless
g m~2 d"1
Year
Unitless
gd-1
Year
Unitless
ugg'1
Year
Unitless
ugg'1
Year
Unitless
Logical
Logical
Logical
Logical
Logical
Logical
Unitless
When the variable name is used in the code but not in the documentation, the first column is left blank.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-3 6
-------
Appendix G: Home Garden Source Model
G.2.7 Limitations Related to the Use of GSCM
The following limitations are noted for the GSCM:
• The GSCM was developed originally for organic contaminants and assumes that the
partition coefficient, Kd, is linear and is estimated as the product of Koc and foe.
Partitioning for metals involves complex chemistry, including the dynamic effects of
aqueous-phase contaminant concentration, precipitation, dissolution,
adsorption/desorption, and media geochemistry (e.g., oxidation-reduction conditions) on
the value of Kd and the fate and transport behavior of metals in general. This complexity
is not modeled by the GSCM for metals partitioning. Rather, Kd is externally provided as
a randomly sampled value by the chemical properties processor (CPP).
• The algorithm estimates annual average source releases. Some of the inputs (e.g.,
infiltration) are long-term annual averages, whereas others are annual averages.
Therefore, the outputs are not strictly applicable to individual years.
• The model in its current form considers only one contaminant at a time and does not
simulate fate and transport of reaction products. With further model development, it
would be possible to track the production of reaction products in each soil column layer
and use basically the same algorithm that is used for the parent compound to model the
fate of reaction products.
• The chosen solution technique (i.e. sequential solutions to the three-component
differential equations of the governing differential equation) allows computational
efficiency. However, the choice of the order in which these solutions are applied could
result in systematic errors. The size of the errors depends on the relative loss rates
associated with the three processes. For example, if the first-order loss rate due to
degradation was high and those losses were calculated first, then less contaminant mass
would be available for diffusive and advective losses. The current algorithm solves for
diffusive losses first. This is followed by first-order losses and advection, respectively.
G.4 References
Abramowitz, M., and LA. Stegun (eds.). 1970. Handbook of Mathematical Functions. New
York: Dover Publications, Inc.
Chow, V.T., D.R. Maidment, and L.W. Mays. 1988. Applied Hydrology. New York: McGraw-
Hill, Inc.
Clapp, R.B., and G.M. Hornberger. 1978. Empirical equations for some soil hydraulic properties.
Water Resources Research 14:601-605.
Cowherd, C., G.E. Muleski, PJ. Englehart, and D.A. Gillette. 1985. Rapid Assessment of
Exposure to Particulate Emissions from Surface Contamination Sites. EPA/600/8-85/002.
U.S. Environmental Protection Agency, Office of Research and Development, Office of
Health and Environmental Assessment, Washington, DC. February. Available at
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=30001EPV.txt (accessed 12 December
2012).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-37
-------
Appendix G: Home Garden Source Model
Duffie, J.A., and W.A. Beckman. 1980. Solar Engineering of Thermal Processes. New York:
John Wiley & Sons, Inc.
Dunne, T., and L.B. Leopold. 1978. Water in Environmental Planning. New York: W.H.
Freeman and Company.
Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Jensen, M.E., R.D. Burman, and R.G. Allen. 1990. Evapotranspiration and irrigation water
requirements. ASCEManual 70:332.
Jost, W. 1960. Diffusion in Solids, Liquids, Gases. Third Printing (with Addendum). New York:
Academic Press, Inc.
Jury, W.A., W.F. Spencer, and W.J. Farmer. 1983. Behavior assessment model for trace organics
in soil: I. Model description. Journal of Environmental Quality 72(4): 558-5 64.
Jury, W.A., D. Russo, G. Streile, and H. El Abd. 1990. Evaluation of volatilization by organic
chemicals residing below the soil surface. Water Resources Research 2(5(1): 13-20.
January.
Lightle, D.T. and G. Weesies. 1998. Default slope parameters. Memorandum to S. Guthrie,
Research Triangle Institute, Research Triangle Park, NC, from D.T. Lightle and G.
Weesies, U.S. Department of Agriculture, Natural Resources Conservation Service, West
Lafayette, IN. June 8.
Millington, R.J., and J.P. Quirk. 1961. Permeability of porous solids. Transactions of the
Faraday Society 57(7): 1200-1207.
Monteith, J.L. 1965. Evaporation and Environment. Pp. 205-234 in Symposia of the Society for
Experimental Biology: Number XIX. New York: Academic Press, Inc.
Richardson, C.W., G.R. Foster, and D.A. Wright. 1983. Estimation of erosion index from daily
rainfall amount. Transactions oftheASABE 26(1): 153-156.
Shan, C., and D.B. Stephens. 1995. An analytical solution for vertical transport of volatile
chemicals in the vadose zone. Journal of Contaminant Hydrology 18:259-277.
Shen, Hsieh Wen, and Pierre Y. Julien. 1993. Chapter 12: Erosion and sediment transport.
Pp. 12-12 in Handbook of Hydrology. Edited by D.R. Maidment. New York: McGraw-
Hill, Inc.
Shuttleworth, W.J. 1993. Chapter 4: Evaporation. Pp. 4-4 in Handbook of Hydrology. Edited by
D.R. Maidment. New York: McGraw-Hill, Inc.
Thomann, R.V., and J.A. Mueller. 1987. Principles of Surface Water Quality Modeling and
Control. New York: Harper & Row.
USDA(U.S. Department of Agriculture). 1986. Urban Hydrology for Small Watersheds. TR-55.
U.S. Department of Agriculture, Engineering Division, Soil Conservation Service,
Washington, DC. June.
U.S. EPA (Environmental Protection Agency). 1999. Source Modules for Nonwastewater Waste
Management Units (Land Application Units, Wastepiles, and Landfills) Background and
Implementation for the Multimedia, Multipathway, and Multireceptor Risk Assessment
(3mra) For Hwir99. U.S. Environmental Protection Agency, Office of Solid Waste and
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-38
-------
Appendix G: Home Garden Source Model
Emergency Response, Office of Solid Waste, Washington, DC. October. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/pdf/risk/modules/s0056.pd
f (accessed 19 February 2014).
U.S. EPA (Environmental Protection Agency). 1985a. Compilation of Air Pollutant Emission
Factors. Volume I: Stationary Point and Area Sources (Fourth Edition). AP-42. EPA
420-R-85-102. Office of Air and Radiation, Office of Air Quality Planning and
Standards, Research Triangle Park, NC. September. Available at
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=2000KKEA.txt (accessed 13 December
2012).
U.S. EPA (Environmental Protection Agency). 1985b. Water Quality Assessment. A Screening
Procedure for Toxic and Conventional Pollutants in Surface and Ground Water-Part I.
Revised. EPA/600/6-85/002a. Office of Research and Development, Environmental
Research Laboratory, Athens, GA. September.
Vanoni, V.A. (ed.). 1975. Sedimentation Engineering. American Society of Civil Engineers,
New York, NY.
Williams, J.R. 1975. Sediment-yield prediction with universal equation using runoff energy
factor. In Present and Prospective Technology for Predicting Sediment Yields and
Sources. ARS-S-40. U.S. Department of Agriculture, Washington, DC.
Wischmeier, W.H., and D.D. Smith. 1978. Predicting rainfall erosion losses. A guide to
conservation planning. In Agricultural Handbook. 537 Edition. U.S. Department of
Agriculture, Washington, DC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-39
-------
Appendix G: Home Garden Source Model
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-40
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
Appendix G
Attachment G-A:
Symbols, Units, and Definitions
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
Appendix G
Attachment A:
Symbols, Units, and Definitions
Table G-A-1. Symbols, Units, and Definitions1
Symbol
1j
ri
Oa
Qaj
n
Of,
n
t>w,j
Pb
Pbj
pb,wWet
A
at
bcm
b,
C'T
C'T,W
C2,,
CG
CL
CLs°l
CN
Cs
CSL,,t
CT
CTO
di,t
d2,,
Da
Units
—
—
—
—
—
—
gem'3
gem-3
gcnr3
m2
Id'1
—
Id-1
mgg-1
mgg"1
gm-3
gm~3
gm-3
gm-3
Unitless
mgg"1
Kg
gm-3
gm-3
rtfd-1
rtf/d'1
cm2 s"1
Definition
Total porosity where j is a subscript indicating waste, w; waste/soil mixture in the till
zone, till; and soil, s
Total porosity
Soil volumetric air content
Soil volumetric air content where j is a subscript indicating waste, w; waste/soil mixture in
the till zone, till; and soil, s
Soil volumetric water content
Soil volumetric water content where j is a subscript indicating waste, w; waste/soil
mixture in the till zone, till; and soil, s
Soil dry bulk density. Same as m2. (Note: g cnr3=mg nr3)
Dry bulk density where j is a subscript indicating waste, w; waste/soil mixture in the till
zone, till; and soil, s
Wet bulk density of home garden soil amendment
Area of home garden
Calculated parameter (Equation G-76) for subarea i
Lower coil column boundary condition multiplier
Calculated parameter (Equation G-77) for subarea i
Total mass-based contaminant concentration in dry soil
Total mass-based contaminant concentration in incoming dry waste
Contaminant concentration in surface soil grid space in subarea i (equivalent to CT)
Contaminant concentration in gaseous phase in soil
Contaminant concentration in aqueous phase in soil
Contaminant aqueous solubility
SCS runoff module curve number parameter
Contaminant concentration in adsorbed phase in soil
Cumulative soil load leaving subarea i, day t
Total volume-based contaminant concentration in soil
Initial total volume-based contaminant concentration in soil
Calculated parameter (Equation G-81) for subarea i
Calculated parameter (Equation G-82) for subarea i
Diffusivity in air
Based on Table A-l, U.S. EPA, 1999.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-A-1
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
Symbol
DE
DE,a
DE.V
Df
Dfo
DRZ
d.
dt
dv
Dv
dz
ERt
ET,,t
FC,
foe
foCj
H
I
IN,,t
Jlch
J vol
k
K'bu, i
Kd
kj
Koc
Ksat
KTL
L
Id,.!
L
mli
m
Mcoll
Mcol2
Units
rtfd'1
rtfd'1
rtfd'1
—
—
cm
m
d
m
cm2 s"1
m
Unitless
cmd"1
cm
—
—
—
md-1
cmd"1
g m"2 d"1
g m"2 d"1
Id'1
md-1
cm3 g-1
Id-1
cm3 g-1
cmhr1
—
mg yr1
gm-M-1
mg yr1
gnr3
gm-2
gm~2
gm-2
Definition
Effective diffusivity in soil
Effective diffusivity in soil air
Effective diffusivity in soil water
Fraction of original mass in soil column grid space that diffuses past a boundary in time, t
Fraction of original mass in soil column grid space that remains after time, t
Depth of the root zone
Thickness of soil in unmixed home garden till zone
Length of time step in GSCM solution algorithm
Thickness of waste in unmixed home garden till zone
Diffusivity in water
Soil column grid size in GSCM solution algorithm
Erosion chemical enrichment ratio for subarea i
Evapotranspiration from root zone on day t for subarea i
Soil moisture field capacity for subarea i
Fraction organic carbon in soil
Fraction organic carbon where j is a subscript indicating waste, w; waste/soil mixture in
the till zone, till; and soil, s
Dimensionless Henry's law constant
Average annual water infiltration rate
Daily infiltration for subarea i, day t
Annual average leachate flux at lower soil column boundary
Annual average volatilization flux at upper soil column boundary
Total first-order loss rate
First-order rate constant due to burial/erosion for subarea i
Soil-water partition coefficient
Annual average first-order loss rate due to process j, where j indicates hydrolysis, h;
aerobic biodegradation, ae; anaerobic biodegradation, an; storm events in subarea i, ev,i;
and wind/mechanical activity, wd
Equilibrium partition coefficient normalized to organic carbon
Saturated hydraulic conductivity
Equilibrium distribution coefficient between the total (g/m3) and aqueous phase (g/m3)
contaminant concentrations in soil
Bulk waste mass loading rate into WMU
Run-on load to subarea i from subarea i-1
Bulk waste loading rate adjusted for mass losses due to unloading
Suspended solids concentration in runoff water, subarea i
Total amount of material from soil column grid space that has passed a boundary at time, t
Total mass in soil column at start of year
Total mass in soil column at end of year
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-A-2
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
Symbol
M,
Madd
Mrem
Nappl
Ndz
PET,
Pt
Q,,t
Q\t
Kappl
Sd
R0,,t
sd
SMb
SM,,t
t
tbet
vb,
vd,
vr,
C-f
VE
W
WP,
y°p
z
•7
£°sc
Ztill
Units
gm'2
gm2
gm-2
lyr1
—
cmd"1
cm
rtfd-1
rtfd-1
mg nr2 yr1
Unitless
cm
w/w, %
—
cm
d
d
md'1
md'1
md'1
gm-3
md-1
mg nr2
cm
yr
m
m
m
Definition
Annual contaminant mass loss due to process i, where i is a subscript indicating:
Total diffusive loss at the surface, 0
Gas phase diffusive losses (volatilization) at the surface, vol
Aqueous phase leaching due to diffusion, Ichd
Aqueous phase leaching due to advection, Icha
First-order loss process j where j is as defined in kj
Annual mass added to soil column
Annual mass removed from soil column
Number of home garden applications per year
Total number of grid spaces of depth dz in soil column
Potential evapotranspiration for day t
Total precipitation on day t
Runoff flow volume (water only) leaving subarea I, day t
Total runoff flow volume (including solids) leaving subarea i, day t
Home garden waste application rate
Sediment delivery ratio for subarea/watershed i
Stormwater runoff depth leaving subarea i, day t
Weight percent of solids in raw waste applied to home garden
Unitless soil-specific exponent in Equation G-13
Soil moisture in root zone at end of day t for subarea i
Time since start of simulation
Time between LAU waste applications
Burial/erosion velocity for subarea i
Diffusive exchange velocity between runoff and surficial soil
Stormwater runoff resuspension velocity for subarea i
DeptG-weighted average CT at time, t
Effective solute velocity in soil
Average mass of waste added per LAU application
Soil moisture wilting point for subarea i
Last year of operation of home garden
Distance down from soil surface
Total depth of soil column
Distance from soil surface to bottom of home garden till (mixing) zone
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-A-3
-------
Appendix G - Attachment A: Symbols, Units, and Definitions
G-A. References
U.S. EPA (Environmental Protection Agency). 1999. Source Modules For Normastewater Waste
Management Units (Land Application Units, Wastepiles, And Landfills) Background And
Implementation For The Multimedia, Multipathway, AndMultireceptor Risk Assessment
(3mra) For Hwir99. U.S. Environmental Protection Agency, Office of Solid Waste and
Emergency Response, Office of Solid Waste, Washington, DC. October. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/pdf/risk/modules/s0056.pd
f (accessed 19 February 2014).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-A-4
-------
Appendix G - Attachment B: Paniculate Emission Equations
Appendix G
Attachment G-B:
Particulate Emission Equations
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment B: Paniculate Emission Equations
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment B: Paniculate Emission Equations
Attachment B:
Particulate Emission Equations
G-B.l Introduction
The HGSM estimates the annual average, area-normalized emission rate of contaminant
mass adsorbed to particulate matter <30 um in diameter, CE30 (g of contaminant m^d"1), as well
as annual average particle size distribution information in the form of the mass fractions of the
total particulate emissions in four aerodynamic particle size categories: 30-15 um, 15-10 um,
10-2.5 um, and<2.5
Table G-B-1 identifies the various release mechanisms and references for the algorithms
implemented within the model. The SFS analysis only considered emissions due to wind erosion
and tilling. This attachment describes the algorithms and assumptions used to estimate annual
releases for each mechanism:
E30i (g of particulates <30 um in diameter nr2 d'1) The annual average
emissions rate due to release mechanism i, where mechanisms of release considered are
summarized in Table G-B-1
Particle size range mass fractions. The mass fractions of E30; in the aerodynamic
particle size categories previously identified.
Table G-B-1. Summary of Mechanisms of Release of Particulate Emissions
Mechanism
Wind erosion from open area
Vehicular activity
Spreading/compacting or tilling
"3
.a
S
!/5
wd
ve
sc
Home garden
Active
X
X
X
Fallow
X
Algorithm
Reference
Cowherd etal. (1985)
U.S. EPA (1995)
U.S. EPA (1985)
G-B.2 Particulate Emission Rate (E30i) Algorithms and Particle Size Range
Mass Fractions
G-B.2.1 Wind Erosion from Open Fields (E30Wd)
The algorithm for the estimation of PMso emissions due to wind erosion from an open
field is based on the procedure developed by Cowherd et al. (1985). This algorithm was adapted
for implementation in a computer code and is presented in detail here. E30wd is estimated in the
source emission module. The user-specified input parameters are summarized in Table G-B-2.
To account for the fact that home gardens can differ in the degree of vegetation (veg1),
surface roughness height (z'o), and frequency of disturbances per month (fd1), different values are
assigned to these parameters in the following equations according to whether the field is active
or inactive (i.e., fallow). The value assignments are summarized in Table G-B-3 where veg, zo,
and fd are user input values.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-B-1
-------
Appendix G - Attachment B: Paniculate Emission Equations
Table G-B-2. Input Parameter Units and Definitions for E30Wd
Symbol
asdm
Lc
veg
Z0
S
u+
PE
U
P
Fd
Units
mm
—
—
cm
w/w, %
m s-1
—
m s-1
dyr-1
1 mo-1
Definition
Mode of the aggregate size distribution
Ratio of the silhouette area of roughness elements too large to be included in sieving to
total base area
Fraction of surface covered with vegetation (fallow field)
Surface roughness height (fallow field)
Silt content of surface material
Observed or probable fastest mile of wind between disturbances
Thornthwaite Precipitation Evaporation Index
Mean annual wind speed
Mean number of days per year with >0.01 in precipitation
Frequency of disturbance per month where a disturbance is defined
exposes fresh surface material (fallow field)
as an action that
Table G-B-3. Active/Inactive Fields Assignments for veg', z'o, fd'
Symbol
veg'
z'o
fd'
Units
—
cm
Imo-1
Active Field
0.0
1.0
fd
Fallow Field
veg
Zo
0.0
Step 1: Calculate U*t
Calculate the threshold friction velocity, U*t (m s"1), the threshold wind speed for the
onset of wind erosion:
^= 0.650-cf-(asdmf
.425
Where
1.0
50.18Zc-647.89Zc2
(G-B-1)
Zc<2jd(T4
2jd(T4 2 x IQ'4)
increases the threshold friction velocity, which results in a relatively low or zero particulate
emission rate due to wind erosion. Low Lc (<2 x 10"4) is indicative of a bare surface with
homogeneous finely divided material (e.g., an agricultural field). Such surfaces have a relatively
low threshold friction velocity and increased particulate emissions. Equations (G-B-1 and G-B-2)
were derived from work from Cowherd et al. (1985, Figures 3-4 and 3-5).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-B-2
-------
Appendix G - Attachment B: Paniculate Emission Equations
Step 2: Calculate Ut
Ut (m s"1) is the threshold wind velocity at a height of 7.0 m (7.0 m is the typical weather
station anemometer height). It is calculated using an equation from Cowherd et al. (1985,
Equation, 4-3, with z=700 cm):
U = ^L In — z'0 < 700 (G-B-3)
0-4 I z\ ) °
where z'o is the roughness height in cm. Values for z'o for various surface conditions are
provided from Cowherd et al. (1985, Figure 3-6).
Step 3: Calculate ESOwd
E30wd is the annual average emission rate of particulate matter <30 um in diameter per
unit area of the contaminated surface. Note that the methodology developed by Cowherd et al.
(1985) estimates emission rates of particulate matter <10 um (or E10wd). E30wd can be
approximated from E10wd with knowledge of the ratio between PMso and PMio for wind erosion.
Cowherd (1998) advises that a good first approximation of this ratio is provided by the particle
size multiplier information presented from U.S. EPA (1995) for wind erosion from open fields
where PMso/PMio is equal to 2. Therefore, the HGSM incorporates a factor of 2 into the
Cowherd et al. (1985) equations for E10wdto allow estimation of E30wd.
For Sites with Limited Erosion Potential (U*t>0.75 m s-1)
The following equation was derived by using equations from Cowherd et al. (1985,
Equations 4-1 to 4-3), applying a factor of 2 as previously discussed, and converting the units to
gnr'd-1:
U.u(u+-Ut\\-veg')fd< 24
(PE/50)2 103 ' (G-B-4)
[0 U+
-------
Appendix G - Attachment B: Paniculate Emission Equations
where
x = 0.886^
u
(G-B-6)
1.91
2.2-0.6*
2.9-1.3*
0<*<0.5
0.5
-------
Appendix G - Attachment B: Paniculate Emission Equations
Table G-B-5. Parameter Units and Definitions for E30S,
Symbol
S
Nopc
fcult
Units
w/w, %
Id-1
—
Definition
Silt content of surface material3 (1.7-88) b
Number of tilling (or spreading and compacting) operations per day
Number of cultivations per application
a Silt is defined as particles <75 um in diameter. Silt content is determined by the percent of loose dry
surface material that passes through a 200-mesh screen using the ASTM-C-136 method (U.S. EPA,
1985).
Values in parentheses are the ranges of source conditions that were tested in developing the U.S. EPA
(1985) equation.
° Nop=(Nappl/365 x fcult)
Table G-B-6. Aerodynamic Particle Size Range Mass Fractions for E30S,
30-15 um
0.24
15-10 um
0.12
10-2.5 um
0.34
<2.5 jim
0.30
G-B.3 Particle Size Range Mass Fractions for Total PMao Emission Rate
Particle size range mass fractions characterizing the total annual average PMso emission
rate (E30; summed over all applicable mechanisms) are determined annually by applying the
mechanism-specific mass fractions to the E30; estimates to obtain size-specific emission rate
estimates Ey (g m"2 d"1) where subscript] identifies the particle size range (j=l indicates 30-15
um; 2 indicates 15-10 urn; 3 indicates 10-2.5 um; and 4 indicates <2.5 um). The total particle
size range mass fraction, pmfj, is calculated as
pmf] =
IX-
(G-B-9)
G-B.4 Annual Average Constituent Emission Rate (CE30) Equations
The amount of mass lost due to wind and mechanical disturbances, Mioss,wd (g m"2),
estimated using Equation G-92 and accumulated throughout the simulated year is used to
estimate CE30 (g m"2 d"l), the annual average, area-normalized emission rate of contaminant
mass adsorbed to particulate matter <30 um in diameter.
CE30 = -
365
(G-B-10)
Equation G-B-10 is directly applicable to the home garden during both the active and
fallow periods.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-B-5
-------
Appendix G - Attachment B: Paniculate Emission Equations
G-B.5 Estimation of First-Order Loss Rate (kwa)
An equation for kwd was derived by performing a mass balance on the surface layer of the
"soil" column to a depth of dz (the depth of the surface soil column cell) and considering losses
due to wind and mechanical activity only:
^L = -k C (G-B-11)
dt wd T
where
- (G-B-12)
^ }
dz KTL
G-B.6 References
Cowherd, CJ. 1998. Personal communication. Midwest Research Institute, Kansas City, MO,
February 27.
Cowherd, C.J., G.E. Muleski, PJ. Englehart, and D.A. Gillette. 1985. Rapid Assessment of
Exposure to Paniculate Emissions from Surface Contamination Sites. EPA/600/8-85/002.
U.S. Environmental Protection Agency, Office of Research and Development, Office of
Health and Environmental Assessment, Washington, DC. February. Available at
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=30001EPV.txt (accessed 12 December
2012).
U.S. DOC (Department of Commerce). 1968. Climatic Atlas of the United States. U.S.
Government Printing Office, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1985. Compilation of Air Pollutant Emission
Factors Volume 1: Stationary Point and Area Sources, 4th Edition. AP-42. EPA 420-R-
85-102. Office of Air and Radiation, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. September. Available at
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=2000KKEA.txt (accessed 13 December
2012).
U.S. EPA (Environmental Protection Agency). 1995. Compilation of Air Pollutant Emission
Factors Volume 1: Stationary Point and Area Sources, 5th Edition. AP-42. PB95-
196028INZ. U.S. Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC.
U.S. EPA (Environmental Protection Agency). 1999. Source Modules For Nonwastewater Waste
Management Units (Land Application Units, Wastepiles, And Landfills) Background And
Implementation For The Multimedia, Multipathway, AndMultireceptor Risk Assessment
(3mra) For Hwir99. U.S. Environmental Protection Agency, Office of Solid Waste and
Emergency Response, Office of Solid Waste, Washington, DC. October. Available at
http://www.epa.gov/osw/hazard/wastetypes/wasteid/hwirwste/pdf/risk/modules/s0056.pd
f (accessed 19 February 2014).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-B-6
-------
Appendix G - Attachment C: HGSMInput Parameters
Appendix G
Attachment G-C:
Home Garden Input Parameters Used for SFS Analysis
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment C: HGSMInput Parameters
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment C: HGSMInput Parameters
Attachment C:
Home Garden Source Model Input Parameters Used
for SFS Analysis
Table G-C-1 lists the HGSM input parameter values used to model particulate emissions,
erosion, and leaching from manufactured soil containing SFS applied to residential gardens.
Each variable entry also includes a parameter description, units, and a data source for each
variable. Variables are grouped by national constants, variables that are derived from other
variables, site-specific soil and land-use variables, and location and sites-specific variables.
Table G-C-1. Source Parameters
Parameter
Description
Value
Reference
Constants
AppDepth
asdm
CutOffYr
fwmu
mt
Nappl
nv
NyrMax
Depth of waste incorporation
(m)
Mode of the aggregate size
distribution (mm)
Operating life (years)
Fraction of waste in WMU
(Waste Management Unit)
(unitless)
Distance vehicle travels on (m)
Waste applications per year
(1 year1)
Vehicles per day on HGSM
(1 day1)
Maximum model simulation
time (years)
0.2
0.5
40
Set to 1, assuming that waste is
not mixed
0
1
0
200
Per EPA directive
Based on U.S. EPA, 1989
U.S. EPA (typical value
for manufactured soils)
Set to 0, assuming that no
regular vehicular activity
occurs on the agricultural
field
Per EPA directive
Set to 0, assuming that no
regular vehicular activity
occurs on the agricultural
field
Chosen to ensure that the
entire period in which
receptors may be exposed
was modeled. Value is
based on the fact that
exposure must begin
sometime during the
operation of the unit; the
maximum operation of the
unit is 40 years, and the
exposure maximum
exposure duration is 100
years.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-C-1
-------
Appendix G - Attachment C: HGSMInput Parameters
Table G-C-1. Source Parameters
Parameter
td
Vv
Description
Time period of deposition
(years)
Volatilization velocity (m day"1)
Value
200
0
Reference
Assumed time period for
modeling
Assumption that
degradation rates account
for volatilization
Derived
foc_soil
focW
LS
Rappl
X
Fraction organic carbon for soil
(unitless)
Fraction organic carbon of waste
solids (unitless)
USLE length-slope factor
(unitless)
Wet waste application rate (Mg
m"2 year1)
Flowlength for local watershed
(m)
Calculated using % organic
matter
set to native soil (foc_soil)
Calculated from X and Theta
Rappl = Nappl x application rate
x ha m"2
Default flow lengths by slope
(Theta)
Calculated based on U. S.
EPA EPACMTP, 1997b
set to native soil (foc_soil)
Calculated from length and
slope based on Williams
andBerndt, 1977
Process Design Manual,
U.S. EPA 1995
Lightle and Weesies,
1998
Distributions
Area_LAU
DRZ
effdust
Lc
veg
Area of the home garden (m2)
Root zone depth (cm)
Dust suppression control
efficiency for controlled areas
(unitless)
Roughness ratio (unitless)
Fraction vegetative cover
(fraction)
Set to 404.7 square meters (i.e.,
0.1 acre)
Uniform distribution
min=value for shallow-rooted
crops
max=value for deep-rooted
crops
see Table 3
normal distribution min=0 max=l
mean=0.5 stdev=0.3
Lognormal distribution
min 1E-04
max 1E-03
mean 3E-04
stdev 0.304
Normal distribution
min=0.8
max=1.0
mean=0.9
stdev=0.1
Home Garden Scenario
Dunne and Leopold, 1978
Based on U.S. EPA, 1989
U.S. EPA, 1989
Protective distribution for
screening purposes
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-C-2
-------
Appendix G - Attachment C: HGSMInput Parameters
Table G-C-1. Source Parameters
Parameter
Description
Value
Reference
Hydrologic Soil Group-Specific
CN
SMFC
SMWP
SoilHydGrp
SCS curve number (unitless)
Soil moisture field capacity
(Volume %)
Soil moisture wilting point
(Volume %)
Hydrologic soil group
HGSM: based on cover type and
hydrologic soil group all other
areas: uniform distribution
group A: min=39, max=72
group B : min=6 1 , max=8 1
group C: min=74, max=88
group D: min=80, max=91
Based on average hydrologic soil
group for each soil texture
Based on average hydrologic soil
group for each soil texture
Based on hydrologic soil properties
Based on Wanielista and
Yousef, 1993
Carsel and Parrish, 1988
Carsel and Parrish, 1988
USDA, 1994 (STATSGO)
Landu se-Specif ic
C
fcult
fd
P
PI
zruf
USLE cover management factor
(unitless)
Number of cultivations per
application (unitless)
Frequency of surface disturbance per
month on active HGSM (1 mo"1)
USLE supporting practice factor
(unitless)
Percent impervious (percent)
Roughness height (cm)
Set to 0.1
Set to 5
Calculated from cultivations per
application
Set to 1
HGSM: 0%
Set to 1
Based on Parameter
Guidance Document, U.S.
EPA, 1997a
Based on U.S. EPA, 1989
Based on U.S. EPA, 1989
Wanielista and Yousef,
1993
Center for Watershed
Protection, 1998
Based on information in
U.S. EPA, 1989
Regional
AirTemp
R
Twater
uw
Long-term average air temperature
(°C)
Meteorologic parameter - USLE
rainfall/erosivity factor (1 year"1)
Waterbody temperature (° K)
Meteorologic parameter - mean
annual wind speed (m sec"1)
Calculated from hourly ambient air
temperature data
Based on 22 -year station rainfall
records
Based on HUC region
Calculated from hourly windspeed
data
U.S. DOC and U.S. DOE,
1993
Wischmeier and Smith,
1978
van der Leeden et al., 1990
U.S. DOC and U.S. DOE,
1999
Scenario-Specific
Area buffer
Area of the buffer (m2)
10m x length of source - where
length of source is the total buffer
length
Buffer width based on 40
of 1993
Site-Specific
SiteLatitude
Site latitude (degrees)
Angular distance in degrees north
or south of the equator
U.S. DOC and U.S. DOE,
1993
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-C-3
-------
Appendix G - Attachment C: HGSMInput Parameters
Table G-C-1. Source Parameters
Parameter
Description
Value
Reference
Soil Texture-Specific
%Organic-
Matter
BD
BDw
K
Ksat
8Mb
SoilTexture
Ss
Theta
WCS
WSpH
Percent organic matter for
surface soil (percent)
Dry bulk density (g cm"3)
Dry bulk density for waste solids
(g cm"3)
USLE soil erodibility factor
(Kgm"2)
Saturated hydraulic conductivity
(cm h"1)
Soil moisture coefficient (unitless)
Texture of surface soil
Silt content of soil (mass percent)
Slope of the local watershed
(degrees)
Saturated volumetric water
content, porosity for soil (ml cm"3)
Watershed soil pH (pH units)
By predominant soil texture;
calculated based on area-
weighted average across all
map units for region (Appendix D)
Surface soil: calculated from
saturated water content (WCS);
see Appendix D
Set to bulk density for the
native soil
Area-weighted average for
each soil texture within
meteorological region (Appendix D)
Based on surface soil texture
Based on surface soil texture;
see Appendix D
Distribution of agricultural soil
textures within meteorological
region (Appendix D)
Area-weighted average silt content
for each soil texture within
meteorological region (Appendix D)
Area-weighted average slope for
each soil texture within
meteorological region (Appendix D)
Based on surface soil texture; see
Appendix D
Area-weighted average value for
each soil texture within
meteorological region (Appendix D)
USDA, 1994 (STATSGO)
Surface soil: Calculated
based on U.S. EPA
EPACMTP, 1997bfrom
saturated water content
(WCS)
Gunn et al., 2004
USDA, 1994 (STATSGO)
Carsel and Parrish, 1988
Clapp and Hornberger,
1978
USDA, 1994 (STATSGO)
USDA, 1994 (STATSGO)
USDA, 1994 (STATSGO)
Carsel and Parrish, 1988
USDA, 1994 (STATSGO)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-C-4
-------
Appendix G - Attachment C: HGSMInput Parameters
References
Carsel, R.F., and R.S. Parrish. 1988. Developing joint probability distributions of soil water
retention characteristics. Water Resources Research 2¥(5):755-769.
Center for Watershed Protection, Inc. 1998. Rapid Water shed Planning Handbook. Center for
Watershed Protection, Ellicott City, MD.
Clapp, R.B., and G.M. Hornberger. 1978. Empirical equations for some soil hydraulic properties.
Water Resources Research 14:601-604.
Dunne, T., and L.B. Leopold. 1978. Water in Environmental Planning. New York: W.H.
Freeman and Company.
Gunn, A.P, R.E. Dewhurst, A. Giorgetti, N.L. Gillott, S.J.W. Wishart, and S. Pedley. 2004. Use
of Sewage Sludge Products in Construction. London, CIRIA.
Hoppe, R.A., J. Johnson, I.E. Perry, P. Korb, I.E. Sommer, J.T. Ryan, R.C. Green, R. Durst, and
J. Monke. 2001. Structural and Financial Characteristics of U.S. Farms: 2001 Family
Farm Report. Agriculture Information Bulletin No. AIB768. U.S. Department of
Agriculture, Economic Research Service, Resource Economics Division, Washington,
DC. May.
Lightle, D.T., and G. Weesies. 1998. Default Slope Parameters. Memorandum submitted to S.
Guthrie, Research Triangle Institute, Research Triangle Park, NC, from D.T. Lightle and
G. Weesies, U.S. Department of Agriculture, Natural Resources conservation Service,
West Lafayette, IN. June.
Schroeder, E.D. 1977. Water and wastewater treatment. P. 156 m McGraw-Hill Series in Water
Resources and Environmental Engineering. Edited by V.T. Chow, R. Eliassen, and R.K.
Linsley. New York: McGraw-Hill, Inc.
U.S. DOC (Department of Commerce) and U.S. DOE (Department of Energy) National
Renewable Energy Laboratory. 1993. Solar and Meteorological Surface Observation
Network (SAMSON), 1961 1990. Version 1.0. National Climatic Data Center, Asheville,
NC.
U.S. EPA (Environmental Protection Agency). 1989. Hazardous Waste TSDF-Fugitive
Particulate Matter Air Emissions Guidance Document. EPA-450/3-89-019. U.S.
Environmental Protection Agency, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1995a. User's Guide for the Industrial Source
Complex (ISC 3) Dispersion Models. Volume II: Description of Model Algorithms. EPA-
454/B-95-003b. U.S. Environmental Protection Agency, Emissions, Monitoring, and
Analysis Division, Office of Air Quality Planning and Standards, Research Triangle Park,
NC. September.
U.S. EPA (Environmental Protection Agency). 1995b. Process Design Manual Land Application
of Sewage Sludge and Domestic Septage. U.S. Environmental Protection Agency, Office
of Research and Development, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1997a. The Parameter Guidance Document. A
Companion Document to the Methodology for Assessing Health Risks Associated with
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-C-5
-------
Appendix G - Attachment C: HGSMInput Parameters
Multiple Pathways Exposure to Combustor Emissions (InternalDraft). NCEA-0238. U.S.
Environmental Protection Agency, National Center for Environmental Assessment,
Cincinnati, OH, March.
U.S. EPA (Environmental Protection Agency). 1997b. EPA 's Composite Model for Leachate
Migration with Transformation Products. EPACMTP: User's Guide. U.S. Environmental
Protection Agency, Office of Solid Waste, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1998. Methodology for Assessing Health Risks
Associated with Multiple Pathways of Exposure to Combustor Emissions. Update to
EPA/600/6-90/003. Methodology for Assessing Health Risks Associated with Indirect
Exposure to Combustor Emissions. EPA 600/R-98/137. U.S. Environmental Protection
Agency, National Center for Environmental Assessment, Cincinnati, OH. December.
U.S. EPA (Environmental Protection Agency). 2001. FQPA Index Reservoir Screening Tool.
U.S. Environmental Protection Agency, Office of Pesticide Programs, Washington, DC.
Available at http://www.epa.gov/oppefedl/models/water/models4.htm#first.
U.S. EPA (Environmental Protection Agency). 2008. Technical Background Document: Updated
Biosolids Exposure and Hazard Assessment. U.S. Environmental Protection Agency,
Office of Water, Washington, DC. November 17.
USDA (U.S. Department of Agriculture). 1994. State Soil Geographic (STATSGO) DataBase.
Data use information. Miscellaneous Publication Number 1492. U.S. Department of
Agriculture, Natural Resources Conservation Service, Fort Worth, TX. December.
USDA (U.S. Department of Agriculture). 1997. Ponds—Planning, Design, Construction-
Revised. Agricultural Handbook No. 590. U.S. Department of Agriculture, Natural
Resources Conservation Service, Washington, DC. November.
van der Leeden, F., F.L. Troise, and D.K. Todd. 1990. The Water Encyclopedia. Chelsea, MI:
Lewis Publishers.
Wanielista M.P., and Y.A. Yousef. 1993. Stormwater Management. New York: John Wiley &
Sons, Inc.
Williams, J.R., and H.D. Berndt. 1976. Determining the universal soil loss equation's length-
slope factor for watersheds. Pp. 217-225 in Soil Erosion: Prediction and Control: The
Proceedings of a National Conference on Soil Erosion. Purdue University, West
Lafayette, IN, May 24-26. Ankeny, IA: Soil Conservation Society of America.
Wischmeier, W.H., and D.D. Smith. 1978. Predicting Rainfall Erosion Losses: A Guide to
Conservation Planning. Agricultural Handbook No. 537. U.S. Department of
Agriculture, Science and Education Administration, Washington, DC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-C-6
-------
Appendix G - Attachment D: Source Air Dispersion and Deposition Modeling
Appendix G
Attachment G-D:
Source Air Dispersion and Deposition Modeling
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment D: Source Air Dispersion and Deposition Modeling
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix G - Attachment D: Source Air Dispersion and Deposition Modeling
Attachment G-D:
Source Air Dispersion and Deposition Modeling
The constituent-specific emission rates (predicted by the source model) are combined
with air concentrations and deposition rates (supplied by the air dispersion model) to calculate
constituent-specific vapor- and particle-phase air concentrations and deposition rates. These
constituent-specific air concentrations and deposition rates are used in estimating aboveground
produce concentrations as shown in the equations presented in Appendix H. Rather than
performing new air dispersion modeling, the SFS evaluation used pre-existing dispersion and
deposition rates generated as part of EPA's evaluation of dioxins in biosolids applied to
agricultural land (U.S. EPA, 2003b), as well as ongoing biosolids-related risk assessment work.
The biosolids dataset reflects national-scale dispersion modeling for farms with areas that span a
range from 111 to 180 acres. The SFS evaluation used the portion of these data specific to
regions where SFS might be used (i.e. the "economic feasibility areas" described in Section
3.2.2). Although dispersion data from farm-size applications will likely overestimate impacts due
to residential gardening in a 404.7 sq. meter (i.e., 0.1 acre) area, the simplifying approach is
appropriate for this conservative SFS screening.
The remainder of Attachment G-D describes the biosolids modeling that was performed
to estimate the location-specific dispersion and deposition factors that were mapped to and
applied in modeling the SFS gardening scenario.
G -D.I Conceptual Air Model
Air dispersion and deposition modeling uses a computer-based set of calculations to
estimate ambient ground-level constituent concentrations and deposition rates associated with
constituent releases from land-use practices and wind erosion. The dispersion model uses
information on meteorology (e.g., wind speed, wind direction, temperature) to estimate the
movement of constituents through the atmosphere. Movement downwind is largely determined
by wind speed and wind direction. Dispersion around the centerline of the plume is estimated
using empirically derived dispersion coefficients that account for the movement of pollutants in
the horizontal and vertical directions. Pollutant movement from the atmosphere to the ground is
also modeled, to account for deposition processes driven by gravitational settling and removal by
precipitation.
The air model used in biosolids assessment (and, by extension, this SFS evaluation) is the
Industrial Source Complex-Short Term Model, version 3 (ISCST3),1 a steady-state Gaussian
plume model used for modeling concentration, dry deposition, and wet deposition from point,
area, volume, and open-pit sources. ISCST3 was designed primarily to support EPA's regulatory
modeling programs. The ISCST3 estimates annual average air concentration of dispersed
constituents and annual deposition rate estimates for vapors and particles at various locations in
and surrounding a source. The air concentrations and deposition rates developed by ISCST3
1 Modeling the deposition of particle-bound metals released from soil (e.g., via windblown emissions) onto plant
surfaces requires a model capable of estimating air concentrations and deposition rates. Although SCREENS (i.e..
the model used to conduct SFS Phase I screening of inhalation exposures) is an appropriate model for assessing
maximum inhalation exposures, it does not calculate wet or dry deposition. The ISCST3 model was therefore
chosen to support the refined probabilistic modeling of the SFS home garden scenario.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-D-1
-------
Appendix G - Attachment D: Source Air Dispersion and Deposition Modeling
were based on a unit emission rate (i.e., 1 jig s"1 m"2). The resulting air concentrations are called
unitized air concentrations (UACs) (i.e., jig m"3 per unit emission rate of 1 jig s"1 m"2), and these
are multiplied by the constituent-specific emission rates (predicted by the source models) and
appropriate conversion factors to calculate chemical-specific vapor- and particle-phase air
concentrations and deposition rates. Appendix H presents the equations used to develop the final
constituent-specific ambient air estimates and deposition rates.
G-D.2 Air Model Inputs
The key inputs to the air dispersion and deposition model include the following:
• Emission rates. The air concentrations and deposition rates developed by ISCST3 were
based on a unit emission rate (i.e., 1 jig s"1 m"2). The resulting air concentrations are
called UACs (i.e., jig m-3 per unit emission rate of 1 jig s"1 m"2).
• Surface area of the farm. As discussed previously, the size of the farm was varied
stochastically by sampling from a distribution using data from Hoppe et al. (2001)
reflecting lifestyle farms.
• Meteorological data for the site. Meteorological conditions at the site were modeled
using surface and upper air data obtained for the 41 climatic regions (See Appendix D).
• Locations of potential receptors. Receptors were placed uniformly over the modeling
domain. Outputs for these receptor points were averaged and used to estimate the mean
air concentrations and deposition rates.
• Particle diameter and mass fraction. Particle diameter and mass fraction are also
required inputs when modeling deposition. As input, a fixed distribution, consistent with
the Multimedia, Multipathway, Multireceptor Risk Assessment Modeling System
(3MRA) air modeling (U.S. EPA, 1999b), was used. The four size categories modeled
were 30-15 um, 15-10 um, 10-2.5 um and <2.5 um, with mass fractions of 0.4, 0.1, 0.3,
and 0.2, respectively.
G-D.3 Air Model Outputs
The air dispersion and deposition data were used to calculate environmental media
concentrations and food chain concentrations. The dispersion model outputs included annual
average air concentrations of the vapors and particles, wet deposition of the vapors and particles,
and dry deposition of the particles. Dry deposition of the vapors was also calculated, but outside
the dispersion model, based on an assumed dry deposition velocity of vapors of 1 cm s"1. These
outputs were produced for the grid of receptor points. These outputs were processed and
averaged in a GIS to produce areal averages for the following, based on the unit emission rate
approach:
• Air concentration of vapors and particles
• Wet deposition of vapors and particles
• Dry deposition of particles.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-D-2
-------
Appendix G - Attachment D: Source Air Dispersion and Deposition Modeling
References
Hoppe, R.A., J. Johnson, I.E. Perry, P. Korb, I.E. Sommer, J.T. Ryan, R.C. Green, R. Durst, and
J. Monke. 2001. Structural and Financial Characteristics of U.S. Farms: 2001 Family
Farm Report. Agriculture Information Bulletin No. AIB768. U.S. Department of
Agriculture, Economic Research Service, Resource Economics Division, Washington,
DC. May.
U.S. EPA (Environmental Protection Agency). 1995. User's Guide for the Industrial Source
Complex (ISC 3) Dispersion Models. Volume II: Description of Model Algorithms. EPA-
454/B-95-003b. Emissions, Monitoring, and Analysis Division, Office of Air Quality
Planning and Standards, Research Triangle Park, NC. September.
U.S. EPA (Environmental Protection Agency). 2003. Technical Background Document for the
Sewage Sludge Exposure and Hazard Screening Assessment. U.S. Environmental
Protection Agency, Office of Water, Washington, DC. 822-B-03-001. December.
Available at
http://water.epa.gov/scitech/wastetech/biosolids/upload/sewagesludge_background.pdf
(accessed 19 March 2012)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-D-3
-------
Appendix G - Attachment E: Arsenic Kd Evaluation
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-D-4
-------
Appendix G - Attachment E: Arsenic Kd Evaluation
Appendix G
Attachment G-E :
Arsenic Soil Partitioning (Kd) Evaluation
In environmental modeling, how a constituent partitions between soil components (i.e.,
how much adsorbs to soil solids, compared to how much transfers into soil water) is addressed
through the use of a soil/water partition coefficient, or Kd. The higher the Kd, the more
constituent adsorbs to soil solids, rather than transferring into soil water). Under the home
garden scenario, the properties and characteristics of the manufactured soil are assumed to mimic
those of natural soil in the area. Accordingly, the SFS-specific screening levels (generated as part
of Phase II national-scale modeling) were developed based on soil Kd values from U.S. EPA
(2005). Given the complexities of arsenic behavior in soil, an analysis was performed examining
the impact of Kd distributions on SFS arsenic screening levels. To better understand the
uncertainties and the sensitivity of these screening values to Kd, source modeling was also
performed for arsenic using a distribution of SFS waste-specific Kds. This distribution was
developed using the full set of whole waste/1 eachate pairs presented in Appendix B (i.e., the SFS
total waste concentration for each sample was divided by the corresponding leachate
concentration for that sample). It is important to note that the SFS waste-specific Kd distribution
reflects partitioning in pure SFS, and therefore would not accurately estimate partitioning in
soils. Modeling results using the SFS waste-specific Kd distribution can be seen as bounding
estimates.
Table G-E-1 compares the arsenic SFS-derived waste Kd distribution to the soil Kd
distribution from U.S. EPA (2005). The SFS-derived waste Kd distribution is relatively narrow
and the Kds are generally well below the soil-Kds. The minimum Kd values for the two
distributions are very similar, however, the mean waste Kd is approximately 6 times lower, and
the maximum waste Kd is about 10 times lower than corresponding soil-Kd values. Given that
Kd is a measure of sorption to solids, the SFS-waste Kd distribution would therefore tend to
estimate lower retention of arsenic in the soil and higher releases to groundwater than would the
soil Kd distribution. This is not surprising, as soils tend to have much higher levels of adsorbent
sources (e.g., Fe, Al, and Mn hydroxides and organic matter) compared to SFS, and would
therefore retain more arsenic in the solid phase. Figure G-E-1 provides a graphical comparison
of the cumulative distributions for the soil Kds and SFS waste-specific Kds.
Table G-E-1. Arsenic: Comparison of Soil Kds
with SFS Waste Kds (L kg-1)
Statistic
Minimum
Mean
Maximum
Soil Kd Distribution
(U.S. EPA, 2005)
Kd
2
1,585
19,952
logKd
0.3
3.2
4.3
SFS Waste-Specific Kd Distribution
(Derived based on Appendix B data)
Kd
5
241
1,960
logKd
0.7
2.4
3.3
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-E-1
-------
Appendix G - Attachment E: Arsenic Kd Evaluation
i.o
0,8
0.6
-Measured SFS Waste/Leachate
-3MRA soil Kd's
0.4
10
1,000
10,000
100,000
Kd (L/kg)
Figure G-E-1. Comparison of cumulative distributions for Arsenic Kd: SFS Waste Partition
Coefficients versus Soil Partition Coefficients (L kg-1)
Output from the analysis was a distribution of soil/produce and groundwater risks and
corresponding SFS-specific screening levels. Table G-E-2 presents the screening levels that
were developed using the two Kd distributions. As seen from this table, it is clear that application
of the SFS waste-specific Kd distribution results in a significantly lower screening level for the
groundwater pathway. However, it is important to note that this lower screening level is nearly
identical to the soil/produce screening level of 8.0 mg kg"1 obtained using the soil Kd
distribution. The similarity between the established SFS-specific screening level and the
bounding waste-specific estimate fosters a high level of confidence that the SFS screening level
will be protective of human health under a range of pathways and environmental conditions.
Table G-E-2. Home Gardening 90th Percentile Arsenic Screening Levels
for SFS in Manufactured Soil
Pathway
Soil/Produce
Groundwater
Arsenic SFS Screening Levels (mg kg'1)
Based on Soil Kd
Distribution
8.0
59
Based on SFS Waste-
Specific Kd Distribution
9.5
7.7
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
G-E-2
-------
Appendix G - Attachment E: Arsenic Kd Evaluation
References
U.S. EPA (Environmental Protection Agency). 2005. Partition Coefficients for Metals in Surface
Water, Soil, and Waste. EPA/600R-05/074. U.S. Environmental Protection Agency,
Office of Research and Development. July. Available at
http://www.epa.gov/athens/publications/reports/Ambrose600R05074PartitionCoefficients
.pdf (accessed 9 December 2013).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-E-3
-------
Appendix G - Attachment E: Arsenic Kd Evaluation
[This page intentionally left blank]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications G-E-4
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Appendix H:
Fate, Transport, Exposure, and Hazard Calculations
for Human Health and Ecological Effects
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table Hl-1. Total Concentration in Air (mg m3)
^air
Cflfr=exlFvxC^+(l-FjxeJx0.001
Name
Fv
Q
Cyp
*^-yv
0.001
Description
Fraction of air concentration in vapor phase (unitless)
Emission rate from source (g s^-nr2)
Normalized paniculate air concentration (ug-s-m2 g^-nr3)
Normalized vapor-phase air concentration (ug-s-m2 g^-nr3)
Conversion factor (mg ug"1)
Value
Fv=0; modeled constituents present
only in particle phase
Calculated from source model output
Calculated from dispersion modeling
Calculated from dispersion modeling
Source: Based on U.S.EPA IBM, 1998 and U.S.EPA HHRAP, 2005
Table H2-1. Participate Deposition Onto Plants
(mg m 2-yr*)
DP
Dp = (mOxQ)x(l-Fv)x(Dydp+(FwxD^))
Name
Q
Fv
Dydp
Fw
Dywp
1000
Description
Emission rate from source (g s^-nr2)
Fraction of air concentration in vapor phase (unitless)
Normalized annual average dry deposition from particle phase
(ug-s-m2 g-'-nr3)
Fraction of wet deposition adhering to plant surface (unitless)
Normalized annual average wet deposition from particle phase
(ug-s-m2 g-'-nr3)
Conversion factor (mg g"1)
Value
Calculated from source model output
Fv=0; modeled constituents present
only in particle phase
Calculated from dispersion modeling
Set to 0.6 (U.S. EPA HHRAP, 2005)
Calculated from dispersion modeling
Source: Based on U.S.EPA IBM, 1998 and U.S.EPA HHRAP, 2005
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-l
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table H3-1. Concentration in Aboveground Vegetation
Due to Deposition, Transfer, and Uptake
(mg kg -1 WW)
Name
Pa
Pv
Pr
MAP
100
Pag
100-M4
100
Description
Vegetative concentration due to particle deposition
(mgkgr'DW)
Vegetative concentration due to air-to-plant transfer
(mgkjr'DW)
Aboveground vegetation concentration due to root uptake
(mgkg^DW)
Plant tissue-specific moisture adjustment factor to convert
DW concentration into WW (percent)
Conversion factor to percent (unitless)
F
Value
Calculated; see Table H3-3
Pv=0; modeled constituents present
only in particle phase
Calculated; see Table H3-2
Exposed fruit: 85
Exposed vegetables: 91.77
Protected fruit: 89.59
Protected vegetables: 80.23
Source: Based on U.S.EPA IBM, 1998 and U.S.EPA HHRAP, 2005
Considered exposed and protected fruits and vegetables. Pv and Pd are always assumed to be zero for protected
fruit and vegetables.
Table H3-2. Aboveground Vegetation Concentration Due to Root Uptake
(mgkg'DW)
Pr
P,-c^B,
Name Description
Csoii Concentration of contaminant in soil (mg kg"1 )
R Soil-to -plant bioconcentration factor
((mg kg"1 DW plant) (mg kg"1 soil)"1)
Value
Output from Source Model
Chemical data; see Appendix F
Source: Based on U.S. EPA IBM, 1998 and U.S.EPA HHRAP, 2005
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-2
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table H3-3. Vegetative Concentration Due to Particle Deposition
(mgkg'DW)
Pd
Name
DP
RP
YP
KpPar
p DP*RP
*' YpxKpPar
Description
Particle deposition term for plants (mg nr^-yr1)
Interception fraction (unitless)
Crop yield (kg DW nr2)
Plant surface loss coefficient, paniculate (1 yr1)
Value
Calculated; see Table H2-1
Exposed fruit: 0.48
Exposed vegetables: 0.48
Exposed fruit: 1.17
Exposed vegetables: 1.17
Chemical data; see Appendix F
Source: Based on U.S. EPA IBM, 1998 and U.S. EPA HHRAP,
2005(Steady-state solution)
Table H3-4. Concentration in Belowground Vegetation Due to Root Uptake
(mgkg'DW)
Pbg
Name
For metals:
Pbg=CsollxBrmotxDWr
100-M4F,
DW - g
100
Description
Csoii Concentration of contaminant in soil (mg kg"1)
R Soil-to-plant bioconcentration factor for roots ((mg kg"1 DW
root plant) (mg kg"1 soil)"1)
DWr Dry weight fraction for root vegetables (unitless)
MAF Plant tissue-specific moisture adjustment factor for root
bg vegetables to convert DW concentration into WW (percent)
Value
Output from Source Model
Chemical data; see Appendix F
Calculated above
Below ground vegetables: 87.32
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-3
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Source: Based on U.S.EPA IBM, 1998 and U.S. EPA HHRAP, 2005
Table H4-1. Average Daily Dose from Total Ingestion
(mg kg -1 BW d-1)
AUUTotal Ingestion
ADDlotallnge^on = ^Aotf + ^D produce (soil pathways)
ADDTotalIngelion = ADDdw (groundwater pathway)
Name
ADDsoii Average daily dose
Ann Average daily dose
produce (mgkg^BWd-1)
Ann Average daily dose
AUUdw kg-1 BW d-1)
Description
from ingestion of soil (mg kg"1 BW d"1)
from consumption of produce
from ingestion of drinking water (mg
Value
Calculated; see Tables H4-2
Calculated; see Tables H4-3
Calculated; see Tables H4-3
Table H4-2. Average Daily Dose from Ingestion of Soil
(mg kg-1 BW d'1)
ADDsoil
Name
Csoil
CR,
F.OU
BW
0.000001
1 V 1 /? V h
. l_- ; /\ l_x J V /\ J 7 „ „ ,
Ar\r) — soil s soll^'00(
soil Tmtr -w-^
Description
Concentration of contaminant in soil (mg kg"1)
Soil ingestion rate (mg day'1)
Fraction of ingested soil that is contaminated (unitless)
Body weight (kg)
Conversion factor (kg mg-1)
)0001
Value
Output from Source Model
Human exposure data; see Appendix I
1 (i.e. 100%)
Human exposure data; see Appendix I
Source: Based on U.S. EPA IBM, 1998 and U.S. EPA HHRAP, 2005
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-4
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Risk Assessment of Spent Foundry Sands in Soil-Related Applications H-5
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table H4-3. Average Daily Dose from Consumption of Produce
(mg kg-1 BW d'1)
Name
Pi
CRPl
FH
LH
0.001
ALJLJ produce
1 V
' produce moL i- Pi- Pi'
Description
Concentration in vegetation as wet weight (g kg"1 WW)
Daily human consumption rate of produce
(gWWkg^BWday"1)
Fraction of vegetables grown in contaminated soil (unitless)
Food preparation loss (unitless)
Conversion factor (g kg"1)
Value
Calculated; see Tables H3-1 and H3-4
Human exposure data; see Appendix I
Human exposure data; see Appendix I
Human exposure data; see Appendix I
Source: Based on U.S.EPA IBM, 1998 and U.S. EPA HHRAP, 2005
Table H4-4. Average Daily Dose from Ingestion of Drinking Water
(mg kg-1 BW d'1)
ADDdw
^<,«c«,.^o«
Name
Cdw
CRdw
Fdw
0.001
Description
Concentration of contaminant in drinking water (mg L"1)
Drinking water ingestion rate (mL kg"1 d"1)
Fraction of ingested drinking water that is contaminated
(unitless)
Conversion factor (L ml"1)
Value
Output from EPACMTP Model
Human exposure data; see Appendix I
1 (i.e. 100%)
Source: Based on U.S. EPA IBM, 1998 and U.S. EPA HHRAP, 2005
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-6
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table H4-5. Lifetime Average Daily Dose
(mgkg'd1)
LADD
Name
ADD
ED
EF
AT
365
LlDD-^DXjEDXjEF
L^iJiLJLJ —
ATx365
Description
Average daily dose (mg kg"1 -day1)
Exposure duration (yr)
Exposure frequency (d yr1)
Averaging time (yr)
Conversion factor (days yr1)
Value
Calculated; see Tables H4-1 to H4-4
Human exposure data; see Appendix I
Human exposure data; see Appendix I
Human exposure data; see Appendix I
Source: Based on U.S.EPA IBM, 1998
Table H5-1. Unitized Human Dose Ratio Due to Ingestion
(unitless)
UDR
Name
ADD Average daily dose
ADD or LADD
HealthBenc hmark
Description
for Noncarcinogens (mg kg"1 day1)
LADD Lifetime average daily dose for Carcinogens (mg kg"1 day1)
Health RfD for noncancer or cancer risk level of 1E-05/CSF for
Benchmark cancer (mg kg"1 day1)
Value
Calculated; see Tables H4-1 to H4-
4
Calculated; see Table H4-5
Chemical data; see Appendix F
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-7
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
Table H6-1. Unitized Ecological Dose Ratio
(unitless)
UDR
Name
SoilConc
Eco-SSL
77D/? -
U LJI\ -
SoilConc
Eco - SSL
Description
Annual average SFS constituent-specific
(mgkg-1)
EPA's Ecological Soil Screening Levels
soil concentration
(mgkg-1)
Value
Calculated by home garden source
model; see Appendix G
Chemical data; see Appendix F
H.I References
U.S. EPA (Environmental Protection Agency). 1998. Methodology for Assessing Health Risks
Associated with Multiple Pathways of Exposure to Combustor Emissions. Update to
EPA/600/6-90/003 Methodology for Assessing Health Risks Associated With Indirect
Exposure to Combustor Emissions. EPA 600/R-98/137. U.S. Environmental Protection
Agency, National Center for Environmental Assessment, Cincinnati, OH. December.
U.S. EPA (Environmental Protection Agency). 2005. Human Health Risk Assessment Protocol
for Hazardous Waste Combustion Facilities. EPA 530-R-05-006. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response. September.
Available (with supporting documentation in a self-extracting file) at
http://www.epa.gov/osw/hazard/tsd/td/combust/risk.htm (accessed 19 March 2012).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
H-8
-------
Appendix H: Fate, Transport, Exposure, and Hazard Calculations for Human Health and Ecological Effects
[This page intentionally left blank]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications H-9
-------
Appendix I: Human Exposure Factors
Appendix I
Human Exposure Factors
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix I: Human Exposure Factors
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix I: Human Exposure Factors
Appendix I:
Human Exposure Factors
This appendix describes the collection or derivation of the human exposure factors that
were used in the SFS beneficial use in soils risk analysis. Exposure factors define the magnitude,
frequency, duration, and routes of exposure to SFS constituents that an individual may
experience.
The term "exposure," as defined by EPA's exposure guidelines (1992), is the condition
that occurs when a contaminant comes into contact with the outer boundary of the body. The
exposure of an individual to a contaminant completes an exposure pathway. After the body is
exposed, the constituent can cross the outer boundary and enter the body. The amount of
contaminant that crosses and is available for adsorption at internal exchange boundaries is
referred to as the "dose" (U.S. EPA, 1992).
Exposure factors are data that quantify human behavior patterns (e.g., ingestion rates of
soil and fruit) and characteristics (e.g., body weight) that affect human exposure to
environmental contaminants. These data can be used to construct realistic assumptions
concerning an individual's exposure to and subsequent intake of a constituent in the
environment. The exposure factors data also enable EPA to differentiate the exposures of
individuals of different ages (e.g., a child versus an adult). Section I.I presents an overview of
the receptors and selected exposure pathways considered for this analysis. The derivation and
values used for the human exposure factors in this risk assessment are described in Section 1.2
LI Receptors and Exposure Pathways
In the home gardening scenario, both adult and child members of a residential family are
exposed to chemicals through the use of SFS manufactured soil on their property. The adults are
20 years old or older when exposure begins, and the children begin exposure at 1 year of age.
As described in Section 5.3, Phase II refined probabilistic modeling was performed for
four constituents to evaluate potential exposures under the home garden soil/produce ingestion
pathway: arsenic, lead, manganese, and nickel. In addition, arsenic was also retained for more
refined evaluation under the groundwater pathway. The Phase II methodology as implemented
generates data to support the development of SFS concentrations based on cumulative exposure
across pathways or for individual pathways. Table 1-1 lists each receptor along with the specific
ingestion exposures that apply to that receptor for a given pathway. For the home gardening
groundwater pathway, receptors are exposed through the ingestion of groundwater used as a
drinking water source. For the soil/produce pathway, adult and child gardeners are exposed via
ingestion of soil, and homegrown above- and belowground produce.1
1 Although receptor exposures via the groundwater and soil/produce pathways were evaluated concurrently, separate
target SFS concentrations were developed for each pathway based on analyses discussed in Section 5.3.5 and
Appendix J that indicate that these exposures will not occur within the same timeframe.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-1
-------
Appendix I: Human Exposure Factors
Table 1-1. Receptors and Ingestion Exposure Pathways
Receptor
Adult Resident
Child Resident
Groundwater
Pathway
Drinking Water
A/
A/
Soil/Produce Pathway
Soil
A/
A/
Protected
Vegetables
A/
A/
Exposed
Vegetables
A/
A/
Root
Vegetables
A/
A/
Protected
Fruits
A/
A/
Exposed
Fruits
A/
A/
1.1.1 Childhood Exposure
Children are an important subpopulation to consider in a risk assessment because they are
likely to be more susceptible to exposures than adults. For example, children may eat more fruit
per unit of body weight than adults. This higher intake-rate-to-body-weight ratio can result in a
higher average daily dose (ADD) for children than for adults.
As children mature, their physical characteristics and behavior patterns change. To
capture these changes in the analysis, the life of a child was considered in stages represented by
the following cohorts: Cohort 1 (aged 1-5), Cohort 2 (aged 6-11), Cohort 3 (aged 12-19), and
Cohort 4 (aged 20-70). Each cohort is associated with distributions of exposure parameter values
that are required to calculate exposure to an individual. The exposure parameter distributions for
each cohort reflect the physical characteristics and behavior patterns of that age range. Data from
the 2011 Exposure Factors Handbook (EFH) and Child-Specific Exposure Factors Handbook
(CSEFH; U.S. EPA, 2008a) were used to derive distributions appropriate for each cohort. The
distributions for Cohort 4, the 20- to 70-year-olds, were the same as those used for adult
receptors.
The development of the child exposure parameters consisted of the following two steps:
1. Define the start age of the child
2. Select the exposure duration of the child.
To capture the higher intake-rate-to-body-weight ratio of children, a start age of 1 was
selected. The distribution of exposure durations for Cohort 1 (aged 1-5) was used to define
exposure duration for each of the Monte Carlo iterations in the probabilistic analysis.
1.1.2 Exposure Pathways
Human receptors may come into contact with chemicals present in environmental media
through a variety of pathways. In general, exposure pathways are either direct (e.g., ingestion of
groundwater) or indirect (e.g., food chain pathways). The exposure pathways considered in this
assessment were ingestion of soil, drinking water, and produce.
1.1.2.1 Ingestion of Soil
In the home gardening scenario, both adult and child receptors were exposed to soil based
on incidental ingestion, mostly as a result of hand-to-mouth behavior.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-2
-------
Appendix I: Human Exposure Factors
1.1.2.2 Ingestion of Drinking Water
In the home gardening scenario, both the adult and child receptors were assumed to
ingest groundwater contaminated by SFS constituents leaching from the manufactured soil used
in the garden.
1.1.2.3 Ingestion of Above- and Belowground Produce
The home gardening scenario included ingestion of the following categories of produce:
exposed fruit, protected fruit, exposed vegetables, protected vegetables, and root vegetables. For
aboveground produce, the term "exposed" indicates that the edible portion of the plant is
exposed to the atmosphere, and the term "protected" indicates that the edible portion of the plant
is protected from the atmosphere by an inedible skin. Home gardeners were assumed to grow
their fruits and vegetables in manufactured soil. The "aboveground" fruits and vegetables were
assumed to become contaminated via soil and air deposition. "Belowground produce" refers to
root crops grown by the gardener and were assumed to become contaminated via root uptake.
The evaluation used data developed by EPA on home gardeners, as well as data on the general
population, to define the amount of home grown produce consumed by adult and child receptors.
1.2 Exposure Parameters Used in Probabilistic Analysis
1.2.1 Introduction
The general methodology for collecting human exposure data for the probabilistic
assessment used the EFH (U.S. EPA, 2011) and Child-Specific Exposure Factors Handbook
(CSEFH; U.S. EPA, 2008) in one of the following three ways:
1. When data were adequate (most input variables), selected parametric distributions were
fit to the EFH or CSEFH data. The best distribution was then chosen using the chi-square
measure of goodness of fit. Parameter uncertainty information (e.g., averages, standard
deviations) was also derived.
2. If percentile data were not adequate for statistical model fitting, in most cases
distributions were selected based on the results for other age cohorts or, if no comparable
information was available, by assuming lognormal as a default distribution and
reasonable coefficients of variation (CVs).
3. Other variables for which data were not adequate for either approaches 1 or 2 above were
fixed at EFH-recommended mean values or according to established EPA policy.
Table 1-2 summarizes all of the parameters that were varied in the probabilistic
assessment. Fixed variables are presented later in Section 1.2.4.
Probabilistic risk assessments involve "sampling" values from probability distribution
functions (PDFs) and using the values to estimate risk. In some cases, distributions are infinite,
and there is a probability, although very small, that very large or very small values might be
selected from the distributions. Because selecting extremely large or extremely small values is
unrealistic (e.g., the range of adult body weights is not infinite), maximum and minimum values
were imposed on the distributions consistent with a methodology developed for the 3MRA
modeling system. For the probabilistic analyses, the maximum intake rates for most food items
were defined as 2 x (mean + 3 standard deviations). For exposed fruit (adult gardener) and
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-3
-------
Appendix I: Human Exposure Factors
exposed vegetable (children aged 12-19), twice the 99th percentile value was used as the
maximum intake rate. Minimum intake values for all food items were zero. The minimum and
maximum values are also included in Table 1-2.2
2 The 3MRA methodology for defining minimum and maximum values has been extensively peer reviewed and
reviewed by the Science Advisory Board. The defined minimum and maximum values preserve the shape and
scale of the distribution. For this reason, these values will typically not match the lower- and upper- most
percentiles (e.g., 1st and 99th) presented in the EFH (U.S. EPA, 2011).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-4
-------
O
i
K
I
•1
^j
^
a
^
OX
'a
o
c3
^
0
I
•a
%
£
£
t5
2
M
S3
0.
Si
M
3
Q
Q.
^
W
O
8?
S3
S
C/5
i
—
—
3
o
O
fl
1
•S
C£
3
"H
cS
S
£
3
'g
^
es
&f
OS
•a ui
+* •_
t/3 £
a
O
= &
!•
' '
B
.0
"S w
^2 ^^
"C ^"*
•£
a
IX
'3
^^
rameter
sS
a.
W
o
p
in
0
?
s
o
Tt
0
i
in
p
ts'
^H
?
W
•n
in
>— i
"c3
§
o
g
bO
0
h-1
bO
2
O
s — '
bo
'53
>~*
o
m
m
i
00
w
3
£
• — '
n
i-H
1
m
m
,_H*
i-H
?
W
^>
^
bO
2
y
^_j
1
i
0
&
W
00
1
JH
•§
^
o
^ — '
<
&
GO
ID
W
o
m'
0
?
S
o
o
0
S
CN
^H
in
o
?
W
oo
t^
ts'
"c3
§
o
g
bO
0
I-J
b
'bo
^>
^
bO
fN
2
y
^_j
1
i
o
&
W
W
oo
^'
0
o
S
o
o
0
o
w
"t
^t-
CN
o
?
w
Tt
in
^H
"c3
§
o
c
bO
^
b
'bo
>
^>
^
bO
2
y
^_,
1
i
0
&
W
W
OS
ts
^'
0
?
S
o
o
0
rn
CN
O
?
W
r--
in
^^
"c3
§
o
s
bO
0
h-1
b
bo
^>
^
bO
1
^_j
1
i
o
&
W
W
o
CN
0
?
S
o
o
0
S
CN
vq
CN
^H
9
W
0
t-
OS
g
3
G
o
b
bo
^>
^
bO
^^
2
y
bO
^
o
&
W
o
2
_u
•§
^
o
^ — '
<
&
GO
ID
W
o
r-
CN
0
O
S
o
o
0
o
w
in
OS
m'
O
?
W
•*t
vq
^H
"c3
§
o
c
bO
^
b
'bo
>
^>
^
bO
^^
ts
2
y
bO
u
1
0
&
W
W
o
^'
0
?
S
o
o
0
S
ON
i-H
^H*
i-H
9
W
0
^H
OS
a
3
G
0
b
bo
^>
^
bO
^.^
2
y
bO
^
o
&
W
W
0
^'
0
o
S
o
o
0
o
w
^0
t-;
o
?
W
f-
in
^H
^H
3
^O
*S
Kj
b
'bo
>
^>
^
bO
g~
1
bO
u
1
0
&
W
W
o
in
**'
0
?
S
o
o
1
OS
in
^H
9
W
r-
m
r~~
a
3
G
0
b
bo
^>
^
bO
2
y
M§
•8
ts
8
PH
OS
i
_u
•§
^
o
^ — '
<
&
GO
ID
W
o
ts'
0
o
S
o
o
0
o
w
in
^H
OO
^H
9
W
t-
m
r~~
a
3
G
o
b
'bo
>
^>
^
bO
ts
2
y
M§
1
8
PH
W
O
OO
m'
0
o
S
o
o
0
o
w
^0
in
m'
^H
9
W
^0
m
r~~
a
3
G
o
b
'bo
>
^>
^
bO
2
y
M§
1
8
PH
W
f-
t'
0
?
S
o
o
1
r-
•n
o
?
W
m
^0
^o
"c3
§
o
g
bO
0
I-J
b
bo
^>
^
bO
1
£
T3
«
[^
^
W)
^~^
2
y
1
c3
bo
>
1
8
PH
^H
m
_u
•§
^
o
^ — '
<
&
GO
ID
tO
w
o
0
oo
0
?
S
o
o
0
S
**
p
o
?
w
f-
0
^^
"c3
§
o
s
bO
0
h-1
b
bo
^>
^
bO
ts
2
y
1
brj
1
tj
8
PH
O>
W
O
0
V£>
0
O
S
o
o
p
w
o
OS
V£>
^H
9
w
0
t-
r~~
"c3
§
o
c
bO
^
b
'bo
>
^>
^
bO
^
2
y
1
gv
>
1
8
PH
O^
W
OS
V£>
0
?
S
o
o
0
S
OS
^H
_
o
?
w
^H
0
^^
"c3
§
o
g
bO
0
I-J
b
bo
^>
^
bO
^
I
3
brj
1
8
PH
W
O
t'
0
?
S
o
o
0
S
in
o
V£>
O
?
W
^H
m
ts
"c3
§
o
g
bO
0
h-1
b
bo
^>
^
bO
2
y
i-H
QJ
bO
>
^H
O
o
S
2
JH
•§
^
o
^ — '
<
&
GO
ID
W
o
in
^'
0
o
S
o
o
0
o
W
^0
p
^H
9
W
0
oo
VO
S
^o
>
b
'bo
>
^>
^
bO
ts
2
y
i-H
1
bO
^H
O
o
&
c>
W
o
0
OS
0
?
S
o
o
p
w
o
OS
^H
9
W
0
"sl-
oe
S
^O
%
b
bo
^>
^
bO
2
y
i-H
"§
bO
>
^H
O
o
o^
W
t-
r-'
0
o
S
o
o
0
o
W
ts
m
_
O
?
W
in
^H
S
^O
>
b
'bo
>
^>
^
bO
I
i-H
QJ
bO
-^H
O
o
&
£*~\
W
00
^'
m
p
W
r
>>
y
_0
2
1
8
0
&
W
^
^H
?
W
^
o
in
0
?
S
o
i-H
1
"*
t-;
O
?
W
•*
m
S
^O
^
^>
^
t
G
£5
1
8
o
&
W
cw
i-H
c3
OJ
>->
O
'oo
1
i
•-£
00
'ob
2
cS
1
.s
'o
p
CH
^O
1
i
V)
cS
g
S
,"S
^-4
S
>>
[^
73
i-s
o 3^
•s 3
^ X
x> o
11
rt fl
3 ^
'S o
^ B
•TS ^
R °
a M
cS g
1 ^
^ ^-.
Oo
in
tH ^J
(2 ^
11
s &
scale are pres
uration was c
.« "^
^ S
cd ^
ft 0
^ &
cw W
dxi
-------
Appendix I: Human Exposure Factors
1.2.2 Exposure Parameter Distribution Methodology
This section describes how stochastic or distributed input data for each exposure factor
were collected and processed. Most exposure factor distributions were developed by analyzing
data from the EFH or CSEFH to fit selected parametric distributions (i.e., gamma, lognormal,
Weibull). Development steps included preparing data, fitting distributions, assessing fit, and
preparing parameters to characterize uncertainty in the distribution inputs.
For many exposure factors, EFH and CSEFH data include sample sizes and estimates of
the following parameters for specific receptor types and age groups: mean, standard deviation,
standard error, and percentiles corresponding to a subset of the following probabilities: 0.01,
0.02, 0.05, 0.10, 0.15, 0.25, 0.50, 0.75, 0.85, 0.90, 0.95, 0.98, and 0.99. Where available, these
percentile data were used as the basis for fitting distributions. Although in no case were all of
these percentiles actually provided for a single factor, the EFH typically included seven or more.
Therefore, using the percentiles is a fuller use of the available information than simply fitting
data based on the method of moments (e.g., selecting models that agree with the data mean and
standard deviation). For some factors, sample sizes were too small to justify the use of certain
percentiles in the fitting process. Percentiles were used only if at least one data point was in the
tail of the distribution. If the exposure factor data repeated a value across several adjacent
percentiles, only one value (the most central or closest to the median) was used in most cases
(e.g., if both the 98th and 99th percentiles had the same value, only the 98th value was used).
The EFH and CSEFH do not use standardized age cohorts across exposure factors.
Different exposure factors have data reported for different age categories. Therefore, to obtain
the percentiles for fitting the four standardized age cohorts (i.e., aged 1-5, 6-11, 12-19, and 20-
70), each EFH or CSEFH cohort-specific value for a given exposure factor was assigned to one
of these four cohorts. When multiple cohorts fit into a single cohort, the percentiles were
averaged within each cohort (e.g., data on children aged 1-2 and 3-5 were averaged for Cohort 1
[aged 1-5]). If sample sizes were available, then weighted averages were used, with weights
proportional to sample sizes. If sample sizes were not available, then equal weights were
assumed (i.e., the percentiles were simply averaged).
Because the EFH and CSEFH data are always positive and almost always skewed to the
right (i.e., have a long right tail), three two-parameter probability distributions commonly used to
characterize such data (gamma, lognormal, and Weibull) were selected. The data were also fit to
a three-parameter distribution (generalized gamma) that unifies the two-parameter distributions
and allows for a likelihood ratio test of the fit of the two-parameter distributions. l This was a
considerable improvement over the common practice of using a lognormal model in which
adequate EFH data were available to support maximum likelihood estimation. However, in a few
cases (e.g., inhalation rate) the data were not adequate to fit a distribution, and the lognormal
distribution was assumed as the default.
1 The SFS evaluation ultimately only used the two-parameter distributions because the three-parameter distribution
did not significantly improve the goodness of fit over the two-parameter distributions.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-6
-------
Appendix I: Human Exposure Factors
1.2.3 Variable Parameters
1.2.3.1 Exposed Fruit Consumption
Table 1-3 presents exposed fruit consumption data. Data for consumption of homegrown
exposed fruit come from Table 13-58 of the EFH (U.S. EPA, 2011). Data (in g WW kg'1 d'1) are
presented by child age groups and for adults. For the age group of 1- to 5-year olds, data were
only available for those aged 3-5 years (not available for children aged 1-2); therefore, these
data were used for the entire 1- to 5-year-old age group.
Table 1-3. Exposed Fruit Consumption Data and Distributions
Age
Cohort
1-5
6-11
12-19
Adult
N
49
68
50
596
EFH Data (g WW kg * d'1)
Data
Mean
2.6
2.52
1.33
1.55
P01
0.042
P05
0.171
0.123
0.158
P10
0.373
0.373
0.258
0.258
P25
1
0.619
0.404
0.449
P50
1.82
1.11
0.609
0.878
P75
2.64
2.91
2.27
1.73
P90
5.41
6.98
3.41
3.41
P95
6.07
11.7
4.78
5
P99
32.5
15.7
5.9
12.9
N = number of samples, P01-P99 = percentiles
1.2.3.2 Protected Fruit Consumption
Data for consumption of homegrown protected fruit come from Table 13-59 of the EFH
(U.S. EPA, 2011) and are presented in Table 1-4. Data (in g WW kg d"1) were presented for the
following age cohorts: those aged 12-19, 20-39, 40-69, and all ages combined. No data for
adults or children aged 1-5 and 6-11 were available for homegrown protected fruit consumption.
However, per capita intake data for protected fruit (including store-bought products) were
available from the EFH for those aged 1-2, 3-5, and 6-11. Therefore, data for the general
population were used to calculate adjustment factors to develop distributions for the non-adult
age groups for consumption of homegrown protected fruit. The population estimated mean and
standard deviation for adults aged 20 and older (derived from the weighted average of means and
standard deviations of those aged 20-39 and those aged 40-69) were used to represent adults for
the analysis.
Table 1-4. Protected Fruit Consumption Data and Distributions
Source
EFH (gen)
EFH (gen)
EFH (gen)
EFH (gen)
EFH (gen)
Age
Cohort
All ages
1-5
6-11
12-19
20-69
EFH Data (g WW kg-1 d'1)
Data
Mean
1.9
5.45
2.7
1.8
1.4
P01
P05
P10
P25
P50
0.38
2.7
0.17
1.8*
0.93*
P75
2.6
7.7
3.8
2.6
2.1
P90
5.4
14.38
8.1
5.4
4.2
P95
8.1
20
11.4
8.4
5.8
P99
16.3
32.3
19.8
15.4
10.5
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-7
-------
Appendix I: Human Exposure Factors
HP
HP
EFH (HP)
EFH (HP)
EFH (HP)
EFH (HP)
1-5
6-11
12-19
20-69
All ages
Adult
2.960
5.1
5.740
5.9
0.12
0.13
0.15
0.12
0.16
0.3
0.266
0.265
0.283
0.39
0.335
0.335
0.393
0.94
0.933
1.116
1.23
2
2.34
2.42
2.84
6.9
7.45
7.46
7.44
15
16
16
11.4
19
19.7
19.1
19.1
36.59
47.3
47.3
gen = general population data, EFH = U.S. EPA (2011), HP = home-produced data, P05-P95 = percentiles,
* based on mean
The relative standard deviations (RSD) for consumption rates were assumed to be the
same for all age groups; the similarity of coefficients of variation (CVs) suggests that this is a
reasonable approximation for the general population. To develop consumption of homegrown
protected fruit distributions for the child age groups, it was also assumed that the mean intake
rates have the same fixed ratio for all the age groups of a given food type. That is, the ratio of the
mean amount consumed of homegrown protected fruit divided by the mean amount consumed of
protected fruit in the general population is the same for any two age groups. These two
assumptions (i.e., constant RSD and constant mean ratio) were used to infer the parameters of the
gamma distributions for the home-produced foods from those of the general population. Each
age-specific ratio (or adjustment factor) was multiplied by the "all ages" group data (e.g., mean,
standard deviation) to estimate each age-specific consumption rate.
1.2.3.3 Exposed Vegetable Consumption
Table 1-5 presents exposed vegetable consumption data and distributions. Data for
consumption of homegrown exposed vegetables come from Table 13-60 of the EFH (U.S. EPA,
2011). Data (in g WW kg'1 d'1) were presented for those aged 1-2, 3-5, 6-11, 12-19, 20-39, and
40-69, as well as for all adults. Weighted averages of percentiles, means, and standard deviations
were calculated for the age group of 1- to 5-year-olds (combining groups of children aged 1-2
years and 3-5).
Table 1-5. Exposed Vegetable Consumption Data and Distributions
Age
Cohort
1-5
6-11
12-19
Adult
N
105
134
143
1361
EFH Data (g WW kg-1 d'1)
Data
Mean
2.453
1.39
1.07
1.57
P05
0.102
0.044
0.029
0.089
P10
0.37
0.094
0.142
0.168
P25
0.833
0.312
0.304
0.413
P50
1.459
0.643
0.656
0.889
P75
3.226
1.6
1.46
1.97
P90
6.431
3.22
2.35
3.63
P95
8.587
5.47
3.78
5.45
P99
9.3
13.3
5.67
10.3
N = number of samples, P01-P99 = percentiles
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-8
-------
Appendix I: Human Exposure Factors
1.2.3.4 Root Vegetable Consumption
Table 1-6 presents root vegetable consumption rates and distributions. Homegrown root
vegetable consumption data come from Table 13-62 of the EFH (U.S. EPA, 2011). Data (in g
WW kg'1 d'1) were presented for those aged 1-2, 3-5, 6-11, 12-19, 20-39, and 40-69, and for
all adults. Weighted averages of percentiles, means, and standard deviations were calculated for
the Cohort 1 age group (combining groups of children aged 1-2 and 3-5).
Table 1-6. Root Vegetable Consumption Data and Distributions
Age
Cohort
1-5
6-11
12-19
Adult
N
45
67
76
682
EFH Data (g WW kg-1 d'1)
Data
Mean
1.886
1.32
0.937
1.15
P01
0.08
0.01
P05
0.081
0.014
0.008
0.036
P10
0.167
0.036
0.068
0.117
P25
0.291
0.232
0.269
0.258
P50
0.686
0.523
0.565
0.674
P75
2.653
1.63
1.37
1.5
P90
5.722
3.83
2.26
2.81
P95
7.502
5.59
3.32
3.64
P99
7.50
7.47
5.13
7.47
N = number of samples, P01-P99 = percentiles
1.2.3.5 Protected Vegetable Consumption
Homegrown protected vegetable consumption data come from Table 13-61 of the EFH
(U.S. EPA, 2011) and are presented in Table 1-7 below. Data (in g WW kg'1 d"1) were presented
for those aged 1-2, 3-5, 6-11, 12-19, 20-39, and 40-69 years, as well as for adults. Weighted
averages of percentiles, means, and standard deviations were calculated for Cohort 1 (children
aged 1-5), combining groups of children aged 1-2 and 3-5.
Table 1-7. Protected Vegetable Consumption Data and Distributions
Age
Cohort
1—5
6-11
12-19
Adults
N
53
63
51
602
EFH Data (g WW kg x d'1)
Data
Mean
1.76
1.1
0.776
1.01
P01
0.27
0.19
0.06
0.103
P05
0.265
0.208
0.161
0.153
P10
0.408
0.318
0.239
0.192
P25
0.829
0.387
0.354
0.336
P50
1.397
0.791
0.583
0.642
P75
2.066
1.31
0.824
1.21
P90
3.053
2.14
1.85
2.32
P95
6.812
3.12
2.2
3.05
P99
6.94
5.4
2.69
6.49
N = number of samples, P01-P99 = percentiles
1.2.3.6 Body Weight
Table 1-8 presents body weight data and distributions. Body weight data come from
Table 8-3 of the CSEFH (U.S.EPA, 2008) and Table 8-3 of the EFH (U.S. EPA, 2011). Data (in
kg) were presented by age and gender. Weighted averages of percentiles and means were
calculated for those aged 1-5, 6-11, 12-19, and adult age groups; male and female data were
weighted and combined for each age group. These percentile data were used as the basis for
fitting distributions.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-9
-------
Appendix I: Human Exposure Factors
Table 1-8. Body Weight Data and Distributions
Age
Cohort
1-5
6-11
12-19
20+
N
4,638
3,593
10,148
14,698
Body Weight Data (kg)
Data
Mean
15.6
31.8
63.9
81.63
P05
11.7
19.7
40.8
53.6
P10
12.4
21.3
44.3
57.7
P15
12.8
22.3
47.2
61.1
P25
13.6
24.4
51.4
67.0
P50
15.1
29.3
60.6
79.0
P75
17.0
36.8
72.5
92.9
P85
18.2
42.1
81.5
102.0
P90
19.2
45.6
88.1
108.5
P95
20.9
52.5
98.0
118.8
1.2.3.7 Drinking Water
Table 1-9 presents drinking water data and distributions. Drinking water data come from
Table 3-19 of the CSEFH (U.S.EPA, 2008) and Table 3-38 of the EFH (U.S. EPA, 2011). Data
(in mL kg"1 d"1) were presented by age and gender. Weighted averages of percentiles, and means
were calculated for those aged 1-5, 6-11, 12-19, and adult age groups; male and female data
were weighted and combined for each age group. These percentile data were used as the basis for
fitting distributions.
Table 1-9. Drinking Water and Distributions
Age
Cohort
1-5
6-11
12-19
20+
N
5462
1410
4143
7616
Drinking Water Data (mL kg-1 d'1)
Data
Mean
24.8
17
11
16
P10
3.3
2
1
2
P15
8.3
6
4
-
P25
19.5
13
8
6
P50
34
23
15
12
P75
34
23
15
22
P90
50.6
35
25
34
P95
66.1
47
34
42
P99
103
78
58
64
1.2.3.8 Exposure Duration
Table 1-10 presents exposure duration data and distributions. Exposure duration was
assumed to be equivalent to the average residence time for each receptor. Exposure durations for
adult and child residents were determined using data on residential occupancy from the EFH,
Table 16-109 (U.S. EPA, 2011). The data represent the total time a person is expected to live at a
single location, based on age. The table presented male and female data combined. Adult
residents aged 21-90 were pooled. Children aged 3 were used to represent those aged 1-5.
Table 1-10. Exposure Duration Data and Distributions
EFH Data
Age Cohort
Child (1-5)
Adult resident
Data Mean
(yr)
6.5
16.0
Distributions
Distribution
Weibull
Weibull
Pop-Estd Shape
(yr)
1.32
1.34
Pop-Estd Scale
(yr)
7.059
17.38
Pop-Estd = population-estimated
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-10
-------
Appendix I: Human Exposure Factors
1.2.4 Fixed Parameters
Certain exposure factors were fixed based on central tendency values from the best
available source (usually EFH recommendations), either because no variability was expected or
because the available data were not adequate to generate distributions. Fixed (constant) exposure
factors are shown in Table 1-11 along with the selected value and data source.
Table 1-11. Summary of Human Exposure Factor Data Used in Modeling: Constants
Description
Averaging time for carcinogens
Exposure frequency
Average
7.00E+01
3.50E+02
Units
yr
dy1
Source
U.S. EPA (1989)
U.S. EPA (1991)
Fraction food preparation loss
Exposed fruit
Exposed vegetables
Protected fruit
Protected vegetables
Root vegetables
2.10E-01
1.61E-01
2.90E-01
1.30E-01
5.30E-02
Fraction
Fraction
Fraction
Fraction
Fraction
U.S. EPA (2011); Table 13-69
U.S. EPA (2011); Table 13-69
U.S. EPA (2011); Table 13-69
U.S. EPA (2011); Table 13-69
U.S. EPA (2011); Table 13-69
Ingestion rate: soil
Children aged 1-5, 6-11, and 12-19
Adult
l.OOE+02
5.00E+01
mgd"1
mgd"1
U.S. EPA (2011); Table 5-1
U.S. EPA (2011); Table 5-1
• When evaluating carcinogens, total dose was averaged over the lifetime of the individual,
assumed to be 70 years.
• Exposure frequency was set to 350 days per year in accordance with EPA policy,
assuming that residents take an average of 2 weeks' vacation time away from their homes
each year.
• Mean soil ingestion rates were cited as 100 mg d"1 for children and 50 mg d"1 for adults
(U.S. EPA, 2011, Table 5-1). The EFH did not recommend any percentile data. The soil
ingestion rates were not varied for the probabilistic analysis.
Exposure Parameters Used for General Population
Consumption rate data for the general population were obtained directly from the EFH
based on per capita intake rates. Data for most parameters included 50th and 90th percentiles.
However, for exposed fruit (adults and 11-19 year olds) and protected vegetables (all cohorts),
mean data were used in the absence of 50th percentile data. Data for children aged 1-5 reflect a
weighted average for consumption rates reported for children aged 1-2 and 3-5. Data for adults
reflect a weighted average for consumption rates reported for adults aged 20-39 and 40-69.
Table 1-12 summarizes the parameters that were used in the analysis of the general population.
The fraction contaminated was assumed to be 0.5.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-11
-------
Appendix I: Human Exposure Factors
Table 1-12. Summary of Produce Consumption Rates (CR) for General Population
Parameters
General Population Estimates
grams (WW) kg"1 body weight day ~l
50th Percentile
90th Percentile
Source
Exposed Fruit
Children aged 1-5
Children aged 6-11
Children aged 12-19
Adults
3.9
2.2 a
0.87 a
0.646 a
21.24
6.3
2.9
2.12
U.S. EPA (2011); Table 9-18
U.S. EPA (2011); Table 9-18
U.S. EPA (2011); Table 9-18
U.S. EPA (2011); Table 9-18
Exposed Vegetables
Children aged 1-5
Children aged 6-11
Children aged 12-19
Adults
0.638
0.6
0.53
0.906
4.96
3.4
2.5
3.26
U.S. EPA (20 11); Table 9-20
U.S. EPA (20 11); Table 9-20
U.S. EPA (20 11); Table 9-20
U.S. EPA (20 11); Table 9-20
Protected Fruit
Children aged 1-5
Children aged 6-11
Children aged 12-19
Adults
2.7
0.17
1.8a
0.926 a
14.38
8.1
5.4
4.18
U.S. EPA (2011); Table 9-19
U.S. EPA (2011); Table 9-19
U.S. EPA (2011); Table 9-19
U.S. EPA (2011); Table 9-19
Protected Vegetables
Children aged 1-5
Children aged 6-11
Children aged 12-19
Adults
1.26a
0.78a
0.46 a
0.548 a
3.86
2.6
1.5
1.7
U.S. EPA (20 11); Table 9-21
U.S. EPA (20 11); Table 9-21
U.S. EPA (20 11); Table 9-21
U.S. EPA (20 11); Table 9-21
Root Vegetables
Children aged 1-5
Children aged 6-11
Children aged 12-19
Adults
1.44
1.0
0.82
0.7
6.02
4.2
3.0
2.58
U.S. EPA (20 11); Table 9-22
U.S. EPA (20 11); Table 9-22
U.S. EPA (20 11); Table 9-22
U.S. EPA (20 11); Table 9-22
Based on mean values.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
1-12
-------
Appendix I: Human Exposure Factors
1.3 References
USDA (U.S. Department of Agriculture). 1997. 1994-96 Continuing Survey of Food Intakes by
Individuals. CD-ROM. U.S. Department of Agriculture, Agricultural Research Service,
Washington, DC.
U.S. EPA (Environmental Protection Agency). 1989. Risk Assessment Guidance for Superfund.
Volume I: Human Health Evaluation Manual (Part A). Interim Final. EPA/540/1-89/002.
U.S. Environmental Protection Agency, Office of Emergency and Remedial Response,
Washington, DC.
U.S. EPA (Environmental Protection Agency). 1991. Risk Assessment Guidance for Superfund:
Volume 1—Human Health Evaluation Manual (Part B, Development of Risk-Based
Preliminary Goals). Interim Draft. EPA/540/R-92/003. U.S. Environmental Protection
Agency, Office of Emergency and Remedial Response, Washington, DC.
U.S. EPA (Environmental Protection Agency). 1992. Guidelines for Exposure Assessment.
EPA/600/Z-92/001. Risk Assessment Forum, Washington, DC. May 29. Available at
http://ofmpub.epa.gov/eims/eimscomm.getfile?p_download_id=429103 (accessed 17
December 2012).
U.S. EPA (Environmental Protection Agency). 2008. Child-Specific Exposure Factors
Handbook. EPA-600/R-06-096F. U.S. EPA, National Center for Environmental
Assessment, Cincinnati, OH. September. Available at http://cfpub.epa.gov/ncea/
cfm/recordisplay.cfm?deid=l99243 (accessed 31 December 2013).
U.S. EPA (Environmental Protection Agency). 2000. Options for Development of Parametric
Probability Distributions for Exposure Factors. EPA/600/R-00/058. U.S. Environmental
Protection Agency, National Center for Environmental Assessment, Office of Research
and Development, Washington, DC. July.
U.S. EPA (Environmental Protection Agency). 2011. Exposure Factors Handbook: 2011 Edition.
EPA/600/R-090/052F. U.S. Environmental Protection Agency, National Center for
Environmental Assessment, Office of Research and Development, Washington, DC.
September. Available online at http://cfpub.epa.gov/ncea/risk/recordisplay.cfm
?deid=236252 (accessed 31 December 2013)
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-13
-------
Appendix I: Human Exposure Factors
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications 1-14
-------
Appendix J: EPACMTP Groundwater Modeling
Appendix J
EPACMTP Groundwater Modeling
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix J: EPACMTP Groundwater Modeling
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix J: EPACMTP Groundwater Modeling
Appendix J:
EPACMTP Groundwater Modeling
National-scale probabilistic
groundwater modeling was performed for
arsenic using EPACMTP (U.S. EPA,
2003a,b,c; 1997). The EPACMTP model
addresses chemical reactions by adsorption
and decay processes. For the simulation of
metals, EPACMTP utilizes nonlinear
sorption isotherms which generally have a
linear range at lower leachate concentrations
and behave nonlinearly at higher leachate
Nonlinear Sorption Isotherms
A nonlinear sorption isotherm is an expression of the
equilibrium relationship between the sorbed
concentration of a metal (or other constituent) and the
aqueous concentration for a representative set of
subsurface system conditions. Nonlinear sorption
isotherms are important when modeling metals because
metal sorption coefficients (Kds), which influence
metal fate and transport, are significantly affected by
metal concentration in the aqueous phase. In general,
„ ,. metal mobility tends to be higher (and thus, Kds lower)
concentrations. The use of nonlinear metal as leachate concentrations increase. Therefore, as
leachate concentrations decrease during unsaturated
zone (soil) transport, metal mobility also tends to
decrease (and Kds tend to increase).
sorption isotherms enables EPACMTP to
model nonlinear behavior in the unsaturated
zone module for a wide array of subsurface
conditions. In the case of arsenic, the model
supports species-specific modeling of either arsenic III or V with arsenic III being the more
mobile of the two species. In this analysis, arsenic was modeled as arsenic III supporting the
development of conservative SFS Screening Levels for the groundwater pathway.
The leachate fluxes and annual average leachate infiltration rates estimated by the home
garden source model were used as input to EPACMTP, to estimate arsenic concentrations at the
receptor well. For both the child and adult receptors, the model generated distributions of
maximum time-average concentrations. These concentrations were calculated using receptor-
specific exposure durations and EPACMTP estimated peak well concentrations. The averaging
period for each iteration in the simulation was centered on the peak well concentration and
spanned the exposure duration for the receptor of interest (i.e., child or adult).
Under the SFS home garden scenario, the well was assumed to be placed 1 meter from
the edge of the garden in the centerline of the plume. The depth of the well was varied uniformly
throughout the aquifer thickness or throughout the upper 10m of the aquifer thickness,
whichever was less. That is, the well depth was never allowed to exceed 10m below the water
table. This limitation for well depth has been used in previous analyses primarily for two
reasons: (1) to be consistent with a residential well scenario (these wells are generally shallow
because of the higher cost of drilling a deeper well) and (2) to produce a conservative estimate of
exposure (because the infiltration rate is generally lower than the groundwater seepage velocity,
groundwater plumes tend to be relatively shallow).
The distributions of receptor drinking water concentrations were developed concurrently
with the soil pathway modeling using an initial soil concentration of 1 ppm for arsenic under the
"unitized" approach. As described in Chapter 5, the "unitized" approach scales the 90th
percentile unitized hazard quotient (HQ) to estimate a protective SFS-specific concentration
based on EPA's risk management criteria (e.g., HQ of 1). These SFS-specific concentrations
represent conservative estimates of the constituent concentration in SFS which, if the SFS were
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-l
-------
Appendix J: EPACMTP Groundwater Modeling
used as a component of manufactured soil in a home garden, would be protective of human
health and the environment. To ensure the appropriateness of applying the unitized approach to
the groundwater pathway, it was necessary to demonstrate that arsenic would behave linearly in
the subsurface under anticipated environmental conditions and at concentration levels found in
SFS. Section J.I describes the analysis that was performed to make the determination that the
linear approach was valid and defensible.
The remainder of this appendix discusses model inputs and outputs. Section J.2 discusses
and presents the EPACMTP input parameters used in the national-scale assessment. Section J.3
discusses key outputs including predicted arrival times for peak receptor well concentrations.
J.I Linear Behavior
EPACMTP simulates the migration of constituents from the source model through the
unsaturated and saturated zones to receptor drinking water wells. In the unsaturated zone,
EPACMTP simulates the effects of both linear and nonlinear sorption reactions. For metal
constituents such as arsenic with nonlinear sorption isotherms, the unsaturated zone module
simulates partitioning by using concentration-dependent partitioning coefficients. These
coefficients generally have a linear range at lower leachate concentrations and behave
nonlinearly at higher leachate concentrations, with Kd generally decreasing with increasing
leachate concentration. The saturated zone module uses a linearized isotherm, based upon the
maximum constituent concentration at the water table. The linear assumption applied in the
saturated zone reflects dilution of the leachate in the ambient groundwater (as the leachate enters
the saturated zone) to a range in which constituent isotherms generally are linear. In order to
apply the "unitized" approach to develop SFS-specific Screening Levels, each modeling
component along the exposure pathway, including the unsaturated zone, must be linear.
To ensure that a linear partitioning assumption is valid in the unsaturated zone,
consideration was given to the following. The assumption of linearity from emplacement to
exposure is dependent on the selection of Kd values from the linear range of the isotherms in the
unsaturated zone. Therefore, it was necessary to review the leachate concentrations generated by
the garden source model to ensure that the arsenic concentrations leaching from the garden
would not exceed the upper bound of an isotherm's linear range. In addition, it was necessary to
ensure that the predicted leachate concentrations associated with the estimated SFS-specific
Screening Levels would also fall within the linear range of the arsenic isotherms. For this reason,
the below analysis used the 95th percentile leachate concentrations derived from both SPLP and
ASTM testing methods as applied to pure SFS samples.
Analysis Overview
The analysis consisted of the following steps:
1. Establish a statistically representative leachate concentration from the population of SFS
leachate data compiled by EPA
2. Visually investigate the tabulated isotherms for arsenic III to identify if linear regions
exist and
3. Visually compare a conservatively representative leachate value for arsenic III to the
linear ranges of the isotherms to see if the assumption of linearity will hold for leachate
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-2
-------
Appendix J: EPACMTP Groundwater Modeling
values expected in the garden. When the representative leachate values are well within
the linear range, the defensible use of a unitized, scaling approach can be established.
4. Lastly, establish a bounding SFS concentration to benchmark that the calculated SFS-
specific Screening Levels will be associated with leachate concentrations that fall within
the linear range.
Establish a statistically representative leachate concentration
The USDA collected samples of SFS from U.S. foundries and conducted leaching
analyses of the materials in their raw form. Table J-l presents 95th percentile leachate
concentrations reflecting the USDA's leaching analyses, the synthetic precipitation leaching
procedure (SPLP), and the American Society for Testing and Materials (ASTM) International
method D 3897. The higher of the two values (i.e., ASTM value of 0.018 mg L"1) was used in the
linearity analysis for comparison to MINTEQA2-derived sorption isotherms.
Table J-l. USDA SPLP and ASTM Results
for Arsenic
Metal
Arsenic
SFS 95th %ile
SPLP
(mgL1)
0.017
ASTM
(mgL1)
0.018
Reference: Chapter 4, Table 4-2
Given that the home garden scenario assumes that SFS will be mixed in a 50:50 ratio
with native soils, the USDA 95th percentile leachate values serve as a conservatively high
estimate for the maximum likely SFS leachate concentration to be observed under the scenario.
Visual Inspection of Isotherms for Linearity
A visualization tool developed with the MATLAB (MathWorks, 2013) scientific
programming platform was used to plot individual MINTEQA2-derived tabulated sorption
isotherms of arsenic III. Figure J-l presents plots of two isotherms for a unique set of subsurface
conditions. The x- and y-axes represent aqueous dissolved concentration (mg L"1) and Kd (L kg"1)
on a base 10 logarithmic scale. The two isotherms are plotted in the main figure window, one for
the unsaturated (in blue) and one for the saturated (in red) regions of the subsurface. The plotted
curves representing the isotherms correspond to the same set of specific subsurface conditions as
specified by the selections shown on the left side of the figure:
• Groundwater compositional type (carbonate or non-carbonate)
• Dissolved concentration of representative anthropogenic (leachate) organic acids (LOM)
• pH of the receiving domain
• Concentration of adsorbents - ferric oxide [goethite] (FeOX) and particulate natural
organic matter (NOM).
Both curves display the same characteristic behavior: Kd is constant (i.e., linear) for
aqueous concentrations less than or equal to 0.3 mg L"1, above which Kd behaves nonlinearly,
decreasing with increasing aqueous concentration.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-3
-------
Appendix J: EPACMTP Groundwater Modeling
Figure J-l. Visualization of 2 (unsaturated and saturated) arsenic III nonlinear sorption
isotherms generated by MINTEQA2 in non-carbonate groundwater compositional
environment.
Visual Inspection and Comparison to SFS Concentrations
Figure J-2 is a duplicate of Figure J-l with the addition of a vertical line representing the
95th percentile ASTM leachate concentration of 0.018 mg L"1 for arsenic in SFS. This value is
over an order of magnitude less than the concentration at which the Kd begins to be dependent
on the dissolved concentration. If this behavior is consistent for all isotherms, then a linear
sorption assumption is reasonable.
Figure J-2. Visualization of 2 (unsaturated and saturated) Arsenic III nonlinear sorption
isotherms in non-carbonate groundwater compositional environment with SFS 95th
percentile ASTM leachate concentration of 0.018 mg L'1 superimposed.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-4
-------
Appendix J: EPACMTP Groundwater Modeling
Figure J-3 and Figure J-4 show all nonlinear sorption isotherms for arsenic III and the
95th percentile ASTM leachate concentration of 0.018 mg L"1 for arsenic in SFS for carbonate
(karst) and non-carbonate aquifer environments. In all cases, the benchmark leachate
concentration is comfortably less than the upper bound on the linear range of Kd.
GW Region GWType
Unsaturated f~\ ^^^^fT]
Saturated iNoncarbonate
•O' Log Kd
a Log Concentration
Unsaturated Zone O - Carbonate GW
± si1.; i fit PC I lone X -
10" 10" 10" 10'
Dissolved Concentration
Figure J-3. Visualization of all (unsaturated and saturated conditions) Arsenic III
nonlinear sorption isotherms in carbonate groundwater compositional environment with
SFS 95th percentile ASTM leachate concentration of 0.018 mg L"1 superimposed.
«. Log Kci
e Log Concentration
Unsaturated Zone 0 - Carbonate GW
Saturated Zone X - Noncarbonate
10 10 10 10 10 10 10 10
Dissolved Concentration
Figure J-4.Visualization of all (unsaturated and saturated conditions) Arsenic III
nonlinear sorption isotherms in non-carbonate groundwater compositional environment
with SFS 95th percentile ASTM leachate concentration of 0.018 mg L"1 superimposed.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-5
-------
Appendix J: EPACMTP Groundwater Modeling
Screening Level Leachate within Linear Range
To determine whether the resulting SFS Screening Levels would result in predicted
leachate concentrations that are within the linear range, a two-step approach was implemented:
Under the first step, a point of reference of 150 ppm was calculated and used for comparison to
the arsenic groundwater SFS screening level as an approximate breaking point indicator for
linear/non-linear behavior. This breaking point was estimated based on an initial unitized source
model runs where the 90th percentile maximum arsenic leachate concentration was identified to
be 0.004 mg L"1. Based on the above demonstration of linearity, the linearity/non-linearity
leachate concentration of 0.3 mg L"1 was used to back-calculate to a corresponding SFS
concentration of 150 ppm (corresponding to a manufactured soil concentration of 75 ppm).14
Under the second step, the manufactured soil concentration corresponding to the final 90th
percentile groundwater SFS Screening Level was used as input to the source model. The
resulting leachate distribution was reviewed and the 90th percentile maximum leachate
concentration was found to be 0.03 mg L"1 which is well below the established leachate
concentration of 0.3 mgL"1 discussed above.
Results from the analysis described above demonstrated that there is high confidence that
a linear assumption for the groundwater pathway modeling of arsenic III is reasonable and
defensible for calculating SFS Screening Levels. As shown above, the SFS 95th percentile
leachate concentration is, in all cases, comfortably within the linear range of all isotherms for
arsenic III.
J.2 Model Input Parameters
Attachment J-A identifies the key EPACMTP input parameters, values and distributions
used in evaluating the groundwater pathway.
Table J-A-1 presents all the EPACMTP input parameters organized by the primary
components of the groundwater modeling scenario:
• Aquifer (or saturated zone) parameters
• Chemical parameters associated with the leachate
• Exposure parameters associated with the receptor well
• Vadose (or unsaturated zone) parameters
• Waste Management Unit (garden) parameters.
For each input parameter, Table J-A-1 provides the EPACMTP variable ID, the
parameter description, units, distribution type, default values (if applicable), and data sources
(where appropriate).
14 It should be noted that this concentration was estimated only as a point of reference for this analysis and should
not be viewed as a definitive cut point between linear and non-linear behavior for any arsenic leaching scenario.
Rather, it is an approximation based on MINTEQ modeling that captures key controlling factors within the
subsurface environment. Therefore, the purpose of showing the approximate cut point should be recognized, and
the value should be used with caution.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-6
-------
Appendix J: EPACMTP Groundwater Modeling
Table J-A-2 presents detailed information on the four empirical correlated groundwater
pathway parameters: GRADNT, XKX, ZB, and DSOIL. Each record in the table represents a
correlated sampling of each of the four parameters from a single site. Table J-A-2 represents data
collected from 400 hazardous waste sites in the United States (Newell et al., 1990) grouped into
12 subsurface environments, identified by Aquifer Code. The value "-999" denotes that site data
were unavailable. This value (-999) triggers the model to estimate the value using methods
described in Section 5.5 of the EPACMTP Technical Background Document (U.S. EPA, 2003 a).
Details regarding data origins, organization, and use in EPACMTP are provided in Section 5.3.4
of U.S. EPA(2003b).
Table J-A-3 presents parameters represented by empirical distributions. For each
parameter, a set of paired values consisting of a parameter value and the associated cumulative
distribution function percentile (e.g., 0.25 denotes the 25th percentile) are shown. The derivation
of each distribution is discussed in detail in various sections of U.S. EPA (2003b).
The percolation of water through garden (infiltration) and soils surrounding the garden
(recharge) was estimated using modeling results from the Hydrogeologic Evaluation of Landfill
Performance (HELP) model (Schroeder et al., 1994). As shown in Table J-A-4, the rates are
correlated with the cover soil and the climate center nearest the garden. The compilation and
creation of these data are described in detail in Appendix A of U.S. EPA (2003b).
J.3 Model Outputs
This section discusses key outputs including arrival time predictions used to support the
development of separate SFS screening levels for the soil/produce and groundwater pathways.
The EPACMTP model outputs peak and average receptor well concentrations and the
estimated year when these concentrations are predicted to occur. The reported year is measured
from the time of initial contaminant release, and corresponds roughly to the middle of the
averaging period. To determine if surface and groundwater pathway exposures would occur
during the same or overlapping timeframes, the EPACMTP outputs were examined to
characterize arrival times. The timeframe estimates for arrival of plume at the receptor well are
presented in Table J-l. These estimates represent the year (after the SFS manufactured soil is
placed in the home garden) when the contaminant plume front would arrive at a well [Beginning]
and the year when the contaminant plume would pass the well [End]. Arrival of peak
concentration would only occur somewhere within this timeframe. The estimates shown are
based on EPACMTP outputs from the unsaturated zone transport simulation, including the first
arrival time of leachate at the water table and cessation time of leachate arrival at the water table.
Retardation effects in the aquifer due to sorption of arsenic onto soils were also accounted for in
these estimates using the following equation:
, -***
* »,
where
tAq = estimated travel time in the aquifer [yr]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-7
-------
Appendix J: EPACMTP Groundwater Modeling
R
= distance from source to well in X direction (along ground water flow
direction [m]
= porosity of aquifer [-]
= Retardation factor in aquifer [-]
= average groundwater velocity in X direction [m/yr]
Table J-2. EPACMTP Arrival Times
of Arsenic Plume at the Receptor Well
Percentile
90%
80%
70%
60%
50%
40%
30%
20%
10%
Arrival Time Zone (year)
Beginning
29
61
100
150
201
203
207
229
398
End
200
200
202
220
272
345
457
663
1112
The travel time in the aquifer was added to the water table information from the
unsaturated zone to estimate the windows shown in Table J-2. Based on these data, the front
edge of 90% of the simulated plumes would arrive at the receptor well no sooner than 29 years
after placed in the garden. Based on the end time of 200 years, the peak or maximum average
concentration would not occur until well beyond the initial introduction into the well. The
maximum exposures via the soil/produce pathway will occur during the first few years
immediately following the application of the manufactured soil. Given the predicted lag time
between the surface and groundwater pathway exposures, it is very unlikely that these exposures
would occur within the same timeframe. As a result, separate SFS Screening Levels were
developed for the soil/produce and the groundwater pathways.
J.4 References
MathWorks. 2013. MATLAB version R2013b. Natick, Massachusetts.
Newell, C.J., L.P. Hopkins, and P.B. Bedient. 1990. A hydrogeologic database for groundwater
modeling. Ground Water. September.
Schroeder, P.R., T.S. Dozier, P.A. Zappi, B.M. McEnroe, J.W. Sjostrom, andR.L. Peyton. 1994.
The Hydrologic Evaluation of Landfill Performance (HELP) Model: Engineering
Documentation for Version 3. EPA/600/R-94/168b. U.S. Environmental Protection
Agency, Office of Research and Development, Washington, DC.
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-8
-------
Appendix J: EPACMTP Groundwater Modeling
U.S. EPA. (Environmental Protection Agency). 1987. Process Coefficients andModelsfor
Simulating Toxic Organics and Heavy Metals in Surface Waters. U.S. EPA, Office of
Research and Development. Washington, DC: U.S. Government Printing Office.
U.S. EPA (Environmental Protection Agency). 1997. EPA 's Composite Model for Leachate
Migration with Transformation Products. EPACMTP: User's Guide. Office of Solid
Waste, Washington, DC. Available online at:
http://www.epa.gov/osw/nonhaz/industrial/tools/cmtp/index.htm
U.S. EPA (Environmental Protection Agency). 2001. WATER9. U.S. EPA, Office of Air Quality
Planning and Standards, Research Triangle Park, NC. Web site:
http://www.epa.gov/ttn/chief/ software/water/index.html
U.S. EPA (Environmental Protection Agency). 2003a. EPACMTP Technical Background
Document. Office of Solid Waste, Washington, DC. Available online at:
http://www.epa.gov/osw/nonhaz/industrial/tools/cmtp/index.htm
U.S. EPA (Environmental Protection Agency). 2003b. EPACMTP Parameters/Data Document.
Office of Solid Waste, Washington, DC. Available online at:
http://www.epa.gov/osw/nonhaz/industrial/tools/cmtp/index.htm
U.S. EPA (Environmental Protection Agency). 2005. Partition Coefficients for Metals in Surface
Water, Soil, and Waste. EPA/600R-05/074. U.S. Environmental Protection Agency,
Office of Research and Development. July. Available at
http://www.epa.gov/athens/publications/reports/Ambrose600R05074PartitionCoefficients
.pdf (accessed 9 December 2013).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-9
-------
Appendix J: EPACMTP Groundwater Modeling
[This page intentionally left blank]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications J-10
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
Attachment J-A
EPACMTP Input Parameters
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
[This page intentionally left blank]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
meters
g
Q
a.
4<
n
E^
!5
u
fe
•;
*S
K
r>^
S
g
s-^
^
^J
T^2
K
1
_S
3
.2
>•
"Z
•o
o
•_
^
"S
^
•o
a
s
o
O
5
—
H
0)
u
c
Si
*
PH
Comments
0)
>
1
X
'S
0)
Paramet
OJ
3
l«
^ HH
1
c-
01
^
_u
X
ctf
H
1
1
§
Q
on
"re
re
_H
5
^-H
1
"fl '-^
11
<2i
Empirical
1
5
^&
o
I
ob
Q
X
0
o
(N
S
W
oi
ID
is
Derived from particle diamete
h-<
^
Q
§
|
^
o
13
O
Derived
'g
o
"g
o
0
Effective p
e*
X
0
o
(N
S
W
VI
ID
Derived from porosity
h-<
^
Q
W
X
1
re
|
=3
o
1
'g
o
1*
Bulk densi
BULKD
(N
^
JJ
X
rt
H
•8
^2
O
re
t
re
(D
on
"B
O
•S
•g
^
Randomly selected based on i
(IGWR) using EPACMTP
o
'§
&
GO
1
3
^
Empirical
correlated
S
o
Aquifer thi
m
N
£
(N
L
^
JD
3
re
H
1
1
§
(D
on
"B
O
«!2
•g
^
Randomly selected based on i
(IGWR) using EPACMTP
o
'§
&
GO
1
1
Empirical
correlated
S
g
1
•5
Hydraulic j
1
X
0
o
(N
S
W
oi
^3
•3
00
T3
Derived from conductivity an
h-<
^
Q
§
1
^3
o
13
O
Derived
S
.£
_o
>
&D
Groundwat
KI
O
X
0
o
(N
S
W
oi
^3
^
'GO
2
OH
13
re
•T-T
Derived from bulk density, Ki
h-<
^
Q
w
1
re
|
=3
o
1
1
B
g
O
^H
Retardatior
RETARD
r*i
<^
JD
3
re
H
•8
.fl
O
=3
t
re
(D
C/3
_0
"3 M"
P
-H ^
fl 1|
'-3 o
§ "^
S ,
•j
•5
i
s
JZi
0
o
(N
PH
W
on
P
-P
'I
§
P ,
Derived from longitudinal dis
h-<
^
Q
w
1
re
1
3
O
1
g
.
I
Vertical dii
^
O
oo
s
•u1
^H
o
(D
re
>
re
o".
Groundwater temperature ma]
20- mi radius around location
o
I
Constant
1
(f0
(D
a
C3
h
er temps
Groundwa!
w
01
L
^
jj
X
rt
H
•8
^2
O
re
t
re
Q
on
.-H
fl
o
(N
S-H
^
&
Assumed equal to average soi
radius around location
§
"re*
g
s
^-H
1
•§1
C -fc2
Pi ^
Empirical
-a ' S
&o D
&
te
Groundwa!
ffi
re
'•S
1
z
<
»
I
§
X
ganic ca
Fraction or
1
•a
1
z
s
g
e?
^
"g
i
Radial dist
22
re
1
"S
<
8
1
(D
g
ta
15
Angle of w
centerline
ANGLE
o
S
§
^
s
1
g
o
a
VI
VI
"H
s,
bJ3
Q
O
JS
ffi
VI
"o
1
1
o
s
Constant
g
=
&
B
1
•3
Longitudin
XWELL
o
S
1
S
1
e
o
a
t£
on
(D
g
"On
'-+H
O
13
S
1
O
Constant
S
(D
S
fl
5 o
Distance fr
centerline t
YWELL
J3
0
O
(N
S
W
GO
ID
p —
O W
ii N_
O
2 o
J-H ^
s g
Uniform Distribution Parame
m; Max = minimum(3.0m,5.C
o
-g ^
*5
,O ^-H o
^ o ^
fl cL
g gH^,
^ fl A
"B g ffl
"Q -^ N
i s| >,
.§ jcT!
fl -S o &
S .~K S?
(2 1 S 2
Uniform
J
2
u
1
te
"S
^
o
13
X
"o
(t
ZWELL
O
O
o
&o
c
ed
&O
O
PH
o
fi
03
03
03
03
S
-------
CD
1 Variables (continu
CU
•o
o
^H
^i
•_
•Srf
«
•o
a
s
o
O
•^
CU
3
H
u
a
Si
O
M
Comments
0)
13
^
0)
s.
-2
'3
Parameter
—
la
13
Chemic
x
0
o
J>
CL<
W
GO
ID
s >,
S i?
* "S
S t3
o S3
o 53
. „ oo
o S
rt S3
£H ^
O O
£.2
Function of KOC and POM ft
tabulated empirical nonlinear
MINTEQ2A for metals
o
hemical-specii
O
|||
'bo
o
I.s
Freundlich isoth
coefficient (Kd)
aquifer
SFCOF
Not used in this analysis
rt
Constant
1
ID
1
1
Freundlich isoth
in aquifer
SFEXP
x
0
o
j>
CL<
W
GO
ID
V3
!>
"^
ed
3
1
3
t5
CL*
H
q
oj
alculated by E
O
Derived
^
1
3
o1
ctf
^
2
Chemical decay
CSLAM
Not used in this analysis
0
Constant
-
|
3
cr
rt
(D
S
Biodegradation i
Q
m
Not used in this analysis
o
'§
&
GO
i
1
6
Constant
oio
Q
2
1
Reference tempt
RTEMP
Not used in this analysis
o
'§
&
GO
i
1
6
Constant
'oo
g
S
'o
Organic carbon
distribution coei
(KOC)
§
0
o
W
P
^
oo
' — 1
J>
2i
W
GO
ID
o
hemical-specii
O
6
T
S
1
se
8
o
^0
Molecular diffus
DSTAR
3MRA LAU Module Output
*=3
j
c
g
<§
u
andomly seled
stribution
Pi '-3
Empirical
-
00
g
.§
"H
Leachate concen
CZERO
Not used in this analysis
13
8
c
g
'&
u
andomly seled
stribution
Pi '-3
Empirical
'oo
00
g
I
1
s
Leachable conce
in waste
CWASTE
^
0
O
J>
2<
W
on
P
&2
!>
"^
ed
3
1
3
ts
o
hemical-specii
O
lH
"C & "^
'bo
^S
o
1
g >
i.s
Freundlich isoth
coefficient (Kd)
zone
UFCOF
Not used in this analysis
rt
Constant
1
'B
§
I
1
Freundlich isoth
vadose zone
UFEXP
%
3
1
M
H
B
|
00
*n ^
Model Output used to represer
concentration for chronic expc
^
"3
<;
•a
§
2
O
istributions of
;posure times
Q 6
Empirical
^
Q
00
'oo
g
Groundwater av<
O
13
"5
x
0
o
SH
w
GO
ID
§
"QH
g
i
SFS Home Garden scenario as
w
o
S
onstant for An
O
Constant
1
'B
Q
*— \
METALE
PH
w
GO
ID
.0
_g
'-S ^~^
•5 o^
g ^1
•® *c3
3 X
Randomly selected value from
Min=0.0126;Max=1.115 (Lou
g
c
.0
^
3
g
'&
u
andomly seled
stribution
Pi '-3
Empirical
1
°?
i
fi
Iron Hydroxide i
1
xi
0
o
cs
J>
2<
W
on
P
.0
_g
i
•3
g
,o
3
Randomly selected value from
Min=0.001173;Max=0.00878
g
c
.0
^
3
g
'&
u
andomly seled
stribution
Pi '-3
Empirical
T
00
g
1
s
o
Leachate Organi
USLOM
X
0
o
J>
2<
W
on
P
_r
ed
PQ •g
M TT!
^
0 |
c O
s ^^
^ «-I
Randomly selected value from
distribution: Min=0.0;Max=8.
1988)
c
%
.m
^
|
£
u
andomly seled
3 distribution
Pi GO
: Empirical
§
^
1
ed
O
ed
1
c^
I
-------
meters
Q
V
Q
a.
^
^
n
t*H
Wj
^
1
K
1
O
^>
•J?
\
'"""J
tj
T~?^
K
1
/ v
-o
1>
C
1 Variables (cont
4*
•o
o
^
•
4*
1
•o
a
s
o
^
O
—
3
H
O
fl
E^
S
s
Comments
0)
>
OJ
&
H
K
a
Parameter
0)
^
.*
X
0
o
Q
•• — '
•8
O
M
O"1
ed
0
"8
GO
ed
JZi
1
"g
0
•a
1
O
1
&
1
1
1
15
o
cti
O
'En
fl
fS
g
GO
(D
O
£
(D
Unsaturated zon
on
Q
JH
0
o
^r
p (
W
&q
(D
O
|_g
"^
fi
O
N
Derived from unsaturated
fc
£
y
p (
w
JZi
Calculated
1
'S
0
g
^
1
fe
&H
Longitudinal dis
(vadose)
Q
X
0
o
(N
J
p
W
GO
1
S
"5b
ss
n9
^
_0
JZi
.2
13
S
£
Randomly selected value
texture (ISTYPEl)
o
Site-specif
KJ
_o
'Bn
g
fl
^p
Q
^
ed
S
O
cti
§
I
.0
0
J
p
W
GO
1
s
"5b
ss
n9
^
_0
JZi
.2
13
S
£
Randomly selected value
texture (ISTYPEl)
o
Site-specii
*«
o
'En
W
„
g
o
O
Soil bulk densit
1
1
g
"Hn
g
^
c§
,o
SFS Home Garden scenar
°
S
WMU width
Q
°
G
Infiltration rate
TUNIS
X
0
j
P
W
GO
^
X
T3
Kj
2
^
to
c§
g
^^
s
Calculated based on trans
EPACMTP
a.
S
!~s
P
w
X
Calculated
•8
,>
°
^
13
.2
OD
'^
Duration of leac
W
0
TSOUR
Q
Q
^
g
WMU depth
DEPTH
X
0
R
^r
PH
GO
X
Kj
^
o
_/
s
^
J5
S
(rt
Uniform Distribution Par;
g
S
^
1
1 5
>i '"^
*g ^
O ^3
T3 &o
1^
<& g
»|
P -g
^ 3
0
§ E
^ ^2
OP
1
3
.C
^
1
^-H
O
2
s
"o
>
*«
1 s
si
fi M
1
H- 1
O3
O
o
"B.
&
£
JS
f^
7^
O
C
cd
&O
>.
CT
c
3
o
PH
fi
U
O.
O
fi
S
03
03
03
03
03
S
-------
1
^}
S
Q
V
°^
1
iPACMTP
""H
K
1
-«
|
^
^
^S
K
1
*5*
CL>
c
1
O
J5
3
.2
"«
•o
O
5
4*
1
•o
=
s
o
O
•
i
^*
h^
—
3
H
0)
u
a
S
I
S
o
U
0)
13
4>
&.
H
-*2
'3
p
"8
fi
sS
h
PH
a
0)
Variabl
8
1
13
=5
1
1
g
§
.2
1
0
^
js
g
o
s
_o
15
I „
5 "g
o M
^2 "s
"^ S
Q &b
s
o
tfl
1
I
rt
_o
'Bn
s
fl
M
5
i
§
*«
s
'§>
ISTYPEl
GO
^3
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions
Hydro-
geologic
Environ-
ment
[IGWR]
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(m/yr)
-999
3.15
-999
-999
-999
946
1580
63.1
3470
28.4
126
15.8
315
-999
11000
94.6
-999
7570
6.31
6.31
31.5
31.5
-999
-8.52129
6.82319
1.07478
1.80348
-0.39418
3.15
11000
63.1
28.4
1890
5990
315
31.5
1580
315
22.1
284
9.46
221
3.15
Unsaturated
Zone Thickness
[DSOIL]
(m)
25.
16.8
15.2
610
5.79
4.57
3.05
4.88
6.1
2.04
6.1
3.81
21.3
6.1
3.05
1.83
1.22
1.52
0.914
1.83
6.1
0.305
9.14
2.8144
1.0747
0.800
0.5525
0.436
0.305
610
6.1
6.1
76.5
30.5
65.5
15.2
174
5.97
12.2
16.8
6.1
9.14
3.96
Aquifer
Thickness
[ZB]
(m)
.
152
15.2
-999
9.14
-999
-999
12.2
152
9.14
7.32
32.9
3.05
6.1
18.3
4.27
9.14
3.05
6.1
7.62
-999
6.1
152
3.76962
1.80348
0.55257
1.1956
0.17788
3.05
152
22.9
79.3
-999
183
45.7
21.3
30.5
3.6
10.7
3.05
152
-999
4.57
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.016
-
.
0.000
0.05
0.014
0.014
0.07
0.03
0.01
0.03
0.09
-
0.000007
0.02
0.04
0.01
0.000007
0.038
0.1
0.06
0.005
0.008
-3.97399
-0.39418
0.436
0.17788
0.81424
0.000007
0.81424
0.08
.
0.008
0.001
0.005
0.1
-
-
0.028
0.003
0.031
0.008
0.01
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-5
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
3.1
2210
11000
126
1330
31500
-999
1890
9780
6.31
3.15
12.6
22100000
34700
31500
3.15
315
315
-999
-999
63.1
189
22100000
-999
22.1
189
11000
-999
63.1
126
-999
-7.68877
12.3279
1.32509
0.47331
-1.46902
3.15
22100000
25500
946
1260
28.4
3780
Unsaturated
Zone Thickness
[DSOIL]
(m)
4.57
15.2
18.3
13.4
6.1
1.83
4.27
53.6
18.3
12.2
12.2
3.7
9.14
12.2
15.2
3.66
9.14
8.53
4.88
3.05
4.57
6.1
4.57
183
2.74
15.2
15.2
3.66
8.23
4.57
1.52
3.469
1.3250
0.5420
-0.01357
.
1.52
183
3.66
9.14
1.77
6.1
16.8
Aquifer
Thickness
[ZB]
(m)
91.4
30.5
91.4
7.62
21.3
3.05
89
6.1
30.5
24.4
12.2
30
1.52
4.57
6.1
9.14
21.3
19
-999
-999
19.8
61
1.83
12.2
3.05
61
22.9
18.3
518
107
91.4
4.2618
0.47331
-0.01357
1.61831
-0.39626
1.52
518
3.66
5.33
6.1
-999
1.52
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.001
0.033
.
0.004
0.005
-
-
0.043
0.012
0.015
0.025
0.01
1
0.008
0.05
0.04
0.005
0.025
-
0.024
0.04
0.023
1
0.000
-
0.012
0.000
.
0.007
0.03
.
-4.42479
-1.46902
-0.1757
-0.39626
1.75145
0.000
1
0.000
0.005
0.000000004
0.034
0.04
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-6
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
2680
31.5
-999
63.1
6620
126
31.5
8830
158
6.31
9.46
-7.81342
21.2765
2.78074
0.646
-1.30916
6.31
25500
50800
13900
-999
-999
1580
3.15
12.6
-999
2520
3150
9.46
94.6
-999
11600
12600
4100
-999
-999
3150
221
-999
3.15
631
-999
-999
Unsaturated
Zone Thickness
[DSOIL]
(m)
6.7
9.45
7.62
2.3
30.5
3.06
-999
5.33
0.91
1.37
2.56
2.7277
2.7807
1.0703
0.1746
0.2971
0.914
30.5
4.57
-999
6.1
12.2
2.13
19.8
4.57
0.91
1.52
2.44
1.83
0.61
6.98
15.2
7.62
2.13
10.7
0.61
0.30
1.52
4.57
3.05
2.44
50.8
15.2
Aquifer
Thickness
[ZB]
(m)
2.4
-999
-999
4.12
21.3
15.2
-999
45.7
4.57
3.66
2.74
2.93298
0.6463
0.17468
0.96341
-0.64536
1.52
45.7
9.14
33.5
-999
4.57
12.2
2.44
10.7
6.1
3.05
-999
6.04
3.96
53.3
76.2
6.4
32
8.53
7.62
9.14
7.62
27.4
3.05
7.62
145
6.1
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.009
0.05
0.01
0.007
0.02
0.01
0.01
0.000
0.003
0.027
0.042
-4.6888
-1.30916
0.29718
-0.64536
1.970
0.000000004
0.05
0.005
0.028
.
0.01
0.001
0.007
0.07
0.043
0.02
0.000002
0.055
0.006
.
0.004
0.049
0.003
0.000
0.001
0.003
0.004
0.015
0.02
0.005
0.092
0.0000001
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-7
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
31.
315
4420
631
-999
-999
7880
5360
-6.82634
9.60704
0.51036
1.46619
.
3.15
11600
5680
-999
946
-999
15800
63100
-999
15.6
12600
-999
7570
-999
1580
31500
-999
6.31
-999
23700
-999
1580
1260
3150
126
946
-999
-999
1390
-999
Unsaturated
Zone Thickness
[DSOIL]
(m)
33.5
9.14
1.52
2.21
1.22
9.14
22.9
3.05
2.6587
0.5103
1.522
-0.01024
0.093
0.305
50.8
3.05
0.91
-999
3.05
6.1
5.18
6.1
38.1
4.57
4.57
30.5
101
33.5
30.5
9.75
3.38
32.9
42.7
10.7
19.8
2.44
12.2
15.2
3.05
4.57
2.44
34.1
12.2
Aquifer
Thickness
[ZB]
(m)
_
3.05
19.8
0.33
-999
3.05
3.05
6.1
3.3063
1.46619
-0.01024
1.28413
-0.02391
0.33
145
21.3
3.96
15.2
6.1
3.05
1.52
3.05
1.52
4.57
22.9
-999
15.2
914
24.4
15.2
7.62
4.57
6.1
1.07
24.4
-999
3.81
4.57
3.05
-999
-999
91.4
85.3
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.023
0.002
0.002
0.001
.
0.005
0.02
0.001
-4.9212
-1.4956
0.093
-0.02391
1.83998
0.0000001
0.092
0.002
.
0.093
0.01
0.000
0.005
0.005
0.025
0.001
0.03
-
0.05
0.001
0.001
.
0.003
-
0.003
-
0.005
.
.
0.002
0.002
.
.
0.003
-
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-8
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
6
6
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
-999
-999
-999
94.6
2840
158
-999
1260
63.1
15800
3470
-999
126
2210
3.15
-999
-999
63700
3.15
-999
631
3190000
3150
3.15
946
3150
315
11000
-999
-999
12.6
2210
-999
22100
-5.61434
9.98295
0.28014
0.08839
2.96927
3.15
3190000
-999
-999
Unsaturated
Zone Thickness
[DSOIL]
(m)
3.6
27.4
15.9
7.01
42.7
1
18.3
7.32
82.3
36.6
7.62
12.2
1.83
15.2
3.66
12.2
36.6
6
6
7.01
14.6
9.14
10.7
4.72
13.7
7.62
4.88
2.44
2.44
3.96
2.13
9.14
3.05
6.1
3.4383
0.2801
0.839
0.5413
0.044
0.914
101
15.2
1.83
Aquifer
Thickness
[ZB]
(m)
_
-999
16.2
9.14
30.5
130
3.66
18.3
-999
-999
15.2
15.2
11
9.14
2.44
48.8
-999
-999
15.2
18.3
24.4
0.30
3.05
18.3
6.1
7.62
9.14
6.1
5.18
18.3
0.61
1.52
6.1
91.4
3.53678
0.08839
0.54136
2.05569
-0.71488
0.30
914
18.3
9.14
Hydraulic
Gradient
[GRADNT]
(ni/m)
_
0.006
0.000
0.000
0.002
0.001
0.01
0.000
-
0.001
0.02
0.001
0.002
-
0.005
0.01
0.068
.
0.015
.
0.003
0.000002
0.006
0.07
0.008
-
0.017
.
0.04
.
.
0.025
0.013
0.001
-5.61773
-2.96927
0.044
-0.71488
4.17328
0.000002
0.093
0.005
0.002
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-9
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
7
7
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
315
631
10700
1890
3.15
-999
4100
16700
11000
315
-999
11000
-999
-999
1580
33100
-999
252
14200
3150
5680
1890
315
31.5
3150
15500
5520
3150
158
22.1
-999
9.46
-999
-999
-
13.8058
1.67704
2.14642
-0.09303
3.15
10700
946
1260
Unsaturated
Zone Thickness
[DSOIL]
(m)
4.8
8.53
3.51
24.4
2.74
21.3
27.4
2.44
5.49
1.52
1.22
5.79
3.96
12.2
4.57
30.5
4.57
11.5
4.57
1.52
3.05
3.66
3.66
1.52
1.19
5.18
3.66
3.05
1.52
1.22
1.83
0.914
10.7
12.2
2.6584
1.6770
0.898
0.3495
-0.23716
0.914
30.5
2.44
2.13
Aquifer
Thickness
[ZB]
(m)
15.
9.14
7.32
36.6
3.66
7.62
3.05
6.4
13.1
3.05
1.83
-999
4.27
16.8
7.62
22.9
7.62
-999
18.3
1.52
6.1
6.1
0.61
-999
3.66
7.93
5.49
16.8
3.05
13.7
9.14
6.1
15.2
12.2
3.15814
2.14642
0.34951
0.86919
0.00252
0.61
36.6
8.23
305
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.001
0.01
0.005
0.001
0.003
0.001
0.001
0.004
0.002
0.002
0.008
0.000
0.017
0.002
0.04
0.01
0.1
0.005
0.000
0.0000004
0.001
0.002
0.000001
0.00000002
.
0.006
0.01
0.013
0.012
0.004
0.011
0.008
0.00008
0.000001
-5.6184
-0.09303
-0.23716
0.00252
1.23921
0.00000002
0.1
0.002
0.003
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-10
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
7
8
8
8
8
8
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
-999
6940
23300
4420
56100
55200
9460
-999
-999
946
9780
-999
4420
4420
1580
82000
946
11000
-999
6940
6310
23700
17700
1890
14500
12000
2520
12.6
315
31.5
-999
-5.22204
13.0649
-1.10808
0.50353
-0.73884
12.6
12000
6310
24000
30000
-999
2520
Unsaturated
Zone Thickness
[DSOIL]
(m)
35.
-999
15.2
1.83
3.05
3.05
57.9
9.14
12.2
3.05
3.05
5.18
3.66
24.4
1.52
14.9
12.2
3.05
4.57
2.13
7.01
4.88
5.79
4.57
1.52
2
1.52
5.79
0.61
0.457
45.7
2.8144
-1.10808
1.1384
0.049
0.2690
0.457
57.9
7.62
4.88
2.99
12.2
3.05
Aquifer
Thickness
[ZB]
(m)
_
22.9
36.6
38.1
10.1
61
9.14
9.14
9.14
3.05
3.05
12.2
15.2
21.3
24.4
8.53
18.3
4.57
13.7
7.99
5.18
18.3
42.7
10.7
18.3
-999
6.1
4.27
4.57
-999
3.05
3.78819
0.50353
0.0496
1.11517
-0.46202
3.05
305
61
22.9
18.9
6.71
21.3
Hydraulic
Gradient
[GRADNT]
(ni/m)
_
0.003
0.004
0.000
0.002
-
0.000001
0.000
0.002
0.008
0.013
0.002
0.005
0.01
0.01
0.003
0.000002
.
0.01
0.004
0.049
0.033
0.002
0.000004
0.012
0.01
0.011
0.021
0.006
0.001
.
-5.30668
-0.73884
0.26902
-0.46202
1.11713
0.000001
0.049
0.001
0.002
0.004
0.001
0.0000008
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-11
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
11000
13300
37800
1260
2210
9780
1890
34400
44200
15800
7250
13900
29000
99700
-999
14800
7880
-999
5680
18900
3880
-999
473
10400
22100
27800
27800
-999
11000
19200
631
19200
5050
-999
33100
-999
2210
60900
-3.59646
5.02
0.486
0.154
-
Unsaturated
Zone Thickness
[DSOIL]
(m)
9.1
5.49
4.57
10.7
3.05
3.35
48.8
7.62
4.88
2
9.14
12.2
2.74
2.13
4.57
1.83
2.44
15.2
2.44
4.57
3.66
2
6.1
7.62
9.14
7.62
7.62
6.1
12.2
5.33
0.91
18.3
0.61
7.62
15.2
4.57
2.13
2
2.9737
0.4862
0.8555
0.2696
0.0700
Aquifer
Thickness
[ZB]
(m)
21.
12.2
9.14
-999
22.9
15.2
32
26.2
18.6
24.4
39.6
122
10.1
7.01
6.1
61
3.05
76.2
6.1
7.62
7.62
18.3
4.57
30.5
7.62
24.4
24.4
4.57
3.05
12.2
10.7
10.7
12.2
30.5
30.5
22.9
3.66
30.5
3.92385
0.15471
0.26963
0.75329
-0.62236
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.004
0.006
0.003
0.008
0.000
0.000
0.03
0.006
0.002
0.001
0.000
0.002
.
0.000
0.003
0.001
0.03
0.000
0.001
0.005
0.004
0.000
0.017
0.001
0.005
0.002
0.002
0.00004
0.075
0.008
0.01
0.013
0.003
0.002
0.000
0.01
0.02
0.003
-5.86511
-0.8019
0.07004
-0.62236
1.62199
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-12
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
8
8
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
10
10
10
10
10
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(m/yr)
473
11000
946
315
18.9
21800
3470
3150
126
31.5
-999
31.5
315
63.1
915
-999
1890
3150
631
6310
-999
4100
126
126
-999
12.6
8830
315
284
9.46
1580
-7.67984
11.25
0.17085
0.72472
-0.72109
9.46
21800
-999
4420
284
19600
158
Unsaturated
Zone Thickness
[DSOIL]
(m)
0.6
48.8
2.1
13.7
3.66
6.1
39.6
21.3
1
7.62
3.05
5.18
3.96
4.57
2.44
7.32
1.83
7.62
3.66
2.44
2.13
1.52
3.05
3.05
0.61
1.83
1.52
1.52
1.74
18.3
3.35
2.4855
0.1708
0.8731
0.1347
-0.12094
0.61
39.6
3.35
11.6
4.57
39.6
4.57
Aquifer
Thickness
[ZB]
(m)
3.0
122
13.7
12.2
5.49
15.2
54.9
4.57
30
3.05
30.5
10.7
22.9
2.96
12.2
12.2
0.91
7.62
2.13
9.14
7.62
6.1
4.57
7.62
1.83
-999
18.3
6.1
9.14
2.44
6.1
3.22796
0.72472
0.13478
0.81983
-0.0043
0.91
54.9
14.6
54.9
7.62
21.4
3.05
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.0000008
0.075
0.05
0.001
0.008
0.004
0.017
0.01
.
0.009
0.0000005
0.03
0.007
0.022
0.000
-
0.005
-
.
0.00000004
0.009
0.01
0.05
0.02
-
0.04
0.004
-
0.01
0.003
0.000004
-4.68545
-0.72109
-0.12094
-0.0043
1.28625
0.00000004
0.05
0.03
0.005
0.01
0.000
0.000
Notes
Minimum
Maximum
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-13
-------
Appendix J- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
315
-999
126
315
31.5
126
-999
-999
631
3470
2210
-999
2840
-999
2210
126
-999
-999
3.15
25.2
4420
-999
-6.97635
4.99889
1.27993
0.51266
-1.74813
3.15
19600
946
63.1
7250
24300
-999
7570
12600
631
3150
1260
31.5
13900
-999
2520
Unsaturated
Zone Thickness
[DSOIL]
(m)
1.5
6.1
7.62
15.2
2.74
3.05
3.81
3.66
4.57
3.05
25.9
1.52
2.74
1.83
13.7
12.2
3.81
3.32
3.66
1.83
10.7
6.1
2.8094
1.2799
0.8603
0.4079
-0.71454
1.52
39.6
2.13
2.74
9.14
4.57
1.52
3.05
0.91
0.91
1.52
1.22
0.914
1.52
1.68
2
Aquifer
Thickness
[ZB]
(m)
6.1
3.66
2.29
10.7
6.86
4.12
6.1
15.2
0.91
3.05
7.62
15.2
4.57
2.44
7.62
12.2
16.8
1.83
11.6
4.57
9.14
42.7
3.15655
0.51266
0.40799
0.8467
0.03369
0.91
54.9
305
30.5
36.6
10.7
305
45.7
4.57
6.1
6.1
10.7
15.2
61
15.2
2
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.004
0.000001
0.005
0.01
0.017
0.003
0.00001
0.1
0.005
0.002
0.00001
0.002
.
0.008
0.01
0.025
0.002
0.06
0.01
0.009
0.014
0.00175
-5.57335
-1.74813
-0.71454
0.03369
3.61694
0.000001
0.1
0.01
0.03
0.000
0.006
0.001
0.006
0.005
0.01
.
0.002
0.005
0.002
0.002
0.002
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-14
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
1260
-999
315
1580
-999
315
284
946
-999
8170
-999
-999
-5.38023
3.48349
0.52513
-0.00429
-0.63963
31.5
24300
15800
-999
1580
-999
-999
1580
126
315
-999
-999
-999
15800
-999
221
315
24900
12300
-999
94.6
1260
2180
6310
-
12.0503
Unsaturated
Zone Thickness
[DSOIL]
(m)
1.2
0.91
1.52
2.74
3.35
3.05
1.07
2.13
2.74
7.01
-999
3.05
1.899
0.5251
0.4690
0.1806
.
0.914
9.14
3
5
50.8
15.2
3.05
45.7
3.05
12.2
30.5
320
5.33
29.3
18.3
-999
3.96
1.52
3.96
3.05
7.62
400
1.68
1.22
3.4776
1.4325
Aquifer
Thickness
[ZB]
(m)
3.0
7.62
1.52
4.57
4.27
24.4
30.5
1.68
21.3
6.1
6.71
42.7
3.7492
-0.00422
0.18069
2.02612
-0.08327
1.52
305
3
10
144
91.4
-999
-999
15.2
61
-999
-999
15.2
19.5
-999
39.6
3.05
-999
18.3
305
19.8
1
7.32
3.05
4.32063
0.53279
Hydraulic
Gradient
[GRADNT]
(ni/m)
0.017
-
0.05
0.023
0.019
0.001
0.003
0.000
0.00003
0.003
.
0.000
-5.61773
-0.63963
-0.2284
-0.08327
1.97797
0.00003
0.05
0.006
0.005
0.023
.
0.012
.
0.00005
0.033
0.02
0.009
0.001
.
.
0.002
0.018
0.002
0.009
0.001
0.01
0.000002
0.00042
.
-5.49537
0.79733
Notes
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
Mean
Covariance
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-15
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
Table J-A-2. Correlated Empirical Distributions (continued)
Hydro-
geologic
Environ-
ment
[IGWR]
12
12
12
12
12
13
13
13
13
13
13
13
13
Long.
Hydraulic
Conductivity
Aquifer
[XKX]
(ni/yr)
1.43257
0.53279
0.79733
94.6
15800
1890
.
12.0503
1.43257
0.53279
0.79733
3.15
22100000
Unsaturated
Zone Thickness
[DSOIL]
(m)
1.25667
0.9954
1.3551
1.22
400
5.18
3.4776
1.4325
1.2566
0.9954
1.3551
0.305
610
Aquifer
Thickness
[ZB]
(m)
0.99541
1.2437
0.81132
3.05
305
10.1
4.32063
0.53279
0.99541
1.2437
0.81132
0.30
914
Hydraulic
Gradient
[GRADNT]
(ni/m)
1.35511
0.81321
4.45451
0.000002
0.033
0.005
-5.49537
0.79733
1.35511
0.81321
4.45451
0.000000004
1
Notes
Covariance
Covariance
Covariance
Minimum
Maximum
Mean
Covariance
Covariance
Covariance
Covariance
Minimum
Maximum
References:
U.S.EPA, 2003b
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-16
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
Table J-A-3. Empirical Distributions of Selected Parameters
for Groundwater Modeling
VariablelD Parameter [Reference] Units
AL Longitudinal dispersivity (aquifer) m
DIAM Avg. particle diameter cm
PH Groundwater pH std. Units
USPH Unsaturated Zone pH std. Units
Value
0.1
1
10
100
0.00039
0.00078
0.0016
0.0031
0.0063
0.0125
0.025
0.05
0.1
0.2
0.4
0.8
3.2
3.6
4.5
5.2
6.07
6.8
7.4
7.9
8.2
8.95
9.7
3.2
3.6
4.5
5.2
6.07
6.8
7.4
7.9
8.2
8.95
9.7
CDF
0
0.1
0.7
1
0
0.038
0.104
0.171
0.262
0.371
0.56
0.792
0.904
0.944
0.976
1
0
0.01
0.05
0.1
0.25
0.5
0.75
0.9
0.95
0.99
1
0
0.01
0.05
0.1
0.25
0.5
0.75
0.9
0.95
0.99
1
CDF = Cumulative distribution function
References: U.S. EPA, 2003b
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-17
-------
Appendix J'- Attachment A: EPACMTP Input Parameters
Table J-A-4. HELP Infiltration Rates for Regional Recharge
ICLR
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
City
Fresno
Boise
Denver
Grand Junction
Pocatello
Glasgow
Bismarck
Pullman
Yakima
Cheyenne
Lander
Los Angeles
Sacramento
San Diego
Santa maria
Ely
Rapid City
Cedar City
Albuquerque
Las Vegas
Phoenix
Tucson
El Paso
Medford
Great Falls
Salt Lake City
Grand Island
Flagstaff
Dodge City
Midland
St. Cloud
E. Lansing
North Omaha
State
CA
ID
CO
CO
ID
KY
ND
WA
WA
WY
WY
CA
CA
CA
CA
NV
SD
UT
NM
NV
AZ
AZ
TX
OR
MT
UT
NE
AZ
KS
TX
MN
MI
NE
Silt Loam
-------
Appendix J - A ttachment A: EPACMTP Input Parameters
Table J-A-4. HELP Infiltration Rates
ICLR
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
City
Tulsa
Brownsville
Dallas
Oklahoma City
Concord
Pittsburg
Portland
Caribou
Chicago
Burlington
Bangor
Rutland
Seattle
Montpelier
Sault St. Marie
Put-in-Bay
Madison
Columbus
Cleveland
Des Moines
E. St. Louis
Columbia
Topeka
Tampa
San Antonio
Hartford
Syracuse
Worchester
Augusta
Providence
Portland
Nashua
Ithaca
Boston
State
OK
TX
TX
OK
NH
PA
OR
ME
IL
VT
ME
VT
WA
VT
MI
OH
WI
OH
OH
IA
IL
MO
KS
FL
TX
CT
NY
MA
ME
RI
ME
NH
NY
MA
for Regional Recharge (continued)
Silt Loam
(ISTYPE1 = 1)
0
0
0
0
0
0686
0549
0599
0612
1585
0.0894
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4171
1082
0798
1359
1471
1212
4384
1062
1651
0508
0912
0765
0780
1143
1435
1529
1049
0658
1095
1709
2545
2022
2116
2131
2294
2268
1684
2332
Sandy Loam
(ISTYPE1 = 2)
0
0
0
1006
1049
1067
0.0942
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2057
1313
4387
1491
1138
1781
2045
1598
4582
1483
2101
1003
1400
1158
1212
1641
1676
1989
1483
1031
1646
2228
3251
2591
2700
2863
2840
2812
0.2136
0
2383
Silty Clay Loam
(ISTYPE1 = 3)
0
0
0
0465
0384
0531
0.0389
0
0
0
0
0
0
0
0
0
1372
0792
3927
0886
0620
1166
1227
1008
4077
0.0879
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1435
0495
0686
0663
0823
1156
0704
1224
0762
0475
0820
1405
2118
1697
1674
1753
1872
1943
1392
1542
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-19
-------
Appendix K: Detailed Human Health Results
Table J-A-4. HELP Infiltration Rates for Regional Recharge (continued)
ICLR
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
City
Schenectady
Lynchburg
New York City
Philadelphia
Seabrook
Indianapolis
Cincinnati
Bridgeport
Orlando
Greensboro
Jacksonville
Watkinsville
Norfolk
Shreveport
Astoria
New Haven
Plainfield
Knoxville
Central Park
Lexington
Edison
Nashville
Little Rock
Tallahassee
New Orleans
Charleston
W. Palm Beach
Atlanta
Lake Charles
Miami
Annette
Bethel
Fairbanks
Honolulu
San Juan
State
NY
VA
NY
PA
NJ
IN
OH
CT
FL
NC
FL
GA
VA
LA
OR
CT
MA
TN
NY
KY
NJ
TN
AK
FL
LA
SC
FL
GA
LA
FL
AK
AK
AK
HI
PR
Silt Loam
(ISTYPE1 = 1)
0.1473
0.3081
0.2436
0.2007
0.1814
0.1300
0.1554
0.1953
0.1016
0.3256
0.1511
0.2891
0.3122
0.2296
1.0762
0.3520
0.1900
0.4107
0.3363
0.3294
0.3122
0.4674
0.3531
0.5913
0.5893
0.2609
0.2611
0.3416
0.3647
0.1450
1.6833
0.0564
0.0104
0.0523
0.1267
Sandy Loam
(ISTYPE1 = 2)
0.1928
0.3612
0.2944
0.2609
0.2428
0.1862
0.2210
0.2464
0.1697
0.3896
0.2106
0.3556
0.0000
0.2939
1.1494
0.4628
0.2540
0.4460
0.4171
0.3970
0.3914
0.5395
0.4336
0.7308
0.7445
0.3287
0.3490
0.3993
0.4641
0.2201
1.8354
0.0721
0.0234
0.0945
0.1923
Silty Clay Loam
(ISTYPE1 = 3)
0.1224
0.2570
0.1969
0.1641
0.1427
0.1064
0.1539
0.1615
0.0805
0.2705
0.1102
0.2332
0.2685
0.1842
0.9647
0.2855
0.1521
0.3543
0.2738
0.2700
0.2492
0.3769
0.2824
0.4564
0.4503
0.2123
0.1783
0.2822
0.2817
0.1019
1.4610
0.0554
0.0117
0.0366
0.0945
Reference: U.S.EPA, 2003b
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
J-A-20
-------
Appendix K: Detailed Human Health Results
Appendix K:
Detailed Human Health Results
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix K: Detailed Human Health Results
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix K: Detailed Human Health Results
Table K-l. Detailed Human Health Results
(Based on Home Gardener Consumption Rate Distributions for Produce)
RunID
Receptor
Type
%-tile
Exposure
Pathway
Soil/Produce
Concentration
(mgkg1)
ADD or
LADD
(mg kg -1 d'1)
Unitized
Dose Ratio
(unitless)
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
Arsenic - Cancer
228
8883
3686
4474
8971
6301
5114
430
8373
6864
590
2015
6176
4734
2638
2958
2631
5299
130
1697
4772
6628
11
6058
9680
6685
6301
7831
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
4.2E-01
3.9E-04
8.2E-04
2.0E-04
2.1E-04
5.7E-04
NA
4.5E-01
3.9E-04
8.4E-04
2.0E-04
3.0E-04
5.8E-04
NA
4.5E-01
3.7E-04
7.9E-04
2.1E-04
3.0E-04
5.5E-04
NA
5.0E-01
3.1E-04
8.0E-04
2.0E-04
2.9E-04
5.6E-04
NA
1.8E-07
3.1E-08
7.5E-08
6.5E-08
3.1E-08
3.8E-08
5.4E-07
3.2E-07
l.OE-07
3.1E-07
2.6E-07
1.1E-07
2.2E-07
1.1E-06
2.7E-08
2.6E-08
6.6E-08
4.4E-08
2.4E-08
3.8E-08
3.7E-07
8.2E-07
1.8E-07
5.9E-07
4.5E-07
1.8E-07
3.7E-07
1.8E-06
2.7E-02
4.7E-03
1.1E-02
9.7E-03
4.7E-03
5.7E-03
8.1E-02
4.8E-02
1.5E-02
4.6E-02
3.9E-02
1.6E-02
3.3E-02
1.7E-01
4.0E-03
3.9E-03
9.9E-03
6.6E-03
3.6E-03
5.6E-03
5.6E-02
1.2E-02
2.7E-02
8.9E-02
6.7E-02
2.7E-02
5.4E-02
2.8E-01
80
471
197
227
472
386
27
46
148
48
56
136
67
13
549
565
222
333
608
392
39
180
82
25
33
81
40
8
Arsenic- Noncancer
5759
1301
4323
529
7266
9022
6244
464
5066
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
50
50
50
50
50
50
50
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
7.4E-01
3.8E-04
8.1E-04
2.0E-04
3.1E-04
5.8E-04
NA
7.3E-01
3.9E-04
2.6E-06
4.1E-07
1.1E-06
9.2E-07
4.4E-07
5.2E-07
7.1E-06
3.9E-06
1.1E-06
1.5E-02
1.4E-03
3.6E-03
3.1E-03
1.5E-03
1.7E-03
2.4E-02
2.2E-02
3.8E-03
150
1,600
616
718
1,508
1,259
93
102
580
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-l
-------
Appendix K: Detailed Human Health Results
Table K-l. Detailed Human Health Results
(Based on Home Gardener Consumption Rate Distributions for Produce)
RunID
2672
1342
4818
1383
1587
254
197
4519
2590
5585
4446
6559
7850
6578
6391
2392
2095
4086
9631
Receptor
Type
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
%-tile
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Exposure
Pathway
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mgkg1)
8.0E-04
2.1E-04
3.0E-04
5.7E-04
NA
5.9E-01
3.9E-04
7.7E-04
1.9E-04
2.9E-04
5.4E-04
NA
9.7E-01
3.9E-04
7.8E-04
2.0E-04
2.9E-04
5.8E-04
NA
ADD or
LADD
(mg kg -1 d'1)
3.5E-06
3.1E-06
1.1E-06
2.4E-06
1.2E-05
2.3E-07
2.2E-07
6.6E-07
3.8E-07
2.1E-07
3.8E-07
3.0E-06
4.0E-07
7.4E-07
2.5E-06
2.2E-06
8.2E-07
1.5E-06
6.2E-06
Unitized
Dose Ratio
(unitless)
1.2E-02
l.OE-02
3.8E-03
8.1E-03
3.9E-02
1.3E-03
7.3E-04
2.2E-03
1.3E-03
6.9E-04
1.3E-03
l.OE-02
2.2E-03
2.5E-03
8.4E-03
7.3E-03
2.7E-03
5.0E-03
2.1E-02
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
186
213
576
272
56
1,694
3,022
1,002
1,758
3,174
1,718
219
986
897
263
302
802
440
106
Cobalt
1898
5384
7962
2528
7592
2253
495
2268
3048
1940
289
5040
622
5049
5203
6853
1048
390
8993
1736
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
5.2E-01
1.3E-03
1.1E-03
7.0E-04
l.OE-03
2.3E-03
NA
6.5E-01
1.4E-03
1.6E-03
7.2E-04
l.OE-03
2.1E-03
NA
6.0E-01
1.4E-03
1.6E-03
6.8E-04
2.3E-04
2.3E-03
3.6E-06
1.3E-06
1.9E-06
2.8E-06
1.3E-06
2.0E-06
1.7E-05
6.2E-06
3.7E-06
6.7E-06
l.OE-05
3.6E-06
9.8E-06
3.1E-05
3.1E-07
6.3E-07
1.1E-06
1.1E-06
5.8E-07
1.3E-06
1.2E-02
4.3E-03
6.3E-03
9.4E-03
4.4E-03
6.6E-03
5.5E-02
2.1E-02
1.2E-02
2.2E-02
3.4E-02
1.2E-02
3.3E-02
l.OE-01
l.OE-03
2.1E-03
3.6E-03
3.6E-03
1.9E-03
4.5E-03
1,038
283
143
99
267
111
38
106
178
98
65
184
67
22
2,099
1,045
608
612
1133
492
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-2
-------
Appendix K: Detailed Human Health Results
Table K-l. Detailed Human Health Results
(Based on Home Gardener Consumption Rate Distributions for Produce)
RunID
569
5751
9465
4894
1399
5914
6178
5661
Receptor
Type
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
%-tile
50
90
90
90
90
90
90
90
Exposure
Pathway
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mgkg1)
O.OE+00
8.1E-01
1.2E-03
1.5E-03
7.1E-04
l.OE-03
2.5E-03
NA
ADD or
LADD
(mg kg -1 d'1)
7.9E-06
6.4E-07
2.3E-06
4.6E-06
6.7E-06
2.5E-06
6.0E-06
1.8E-05
Unitized
Dose Ratio
(unitless)
2.6E-02
2.1E-03
7.8E-03
1.5E-02
2.2E-02
8.2E-03
2.0E-02
5.8E-02
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
83
1,038
283
143
99
267
111
38
Iron
1613
6612
506
5073
2396
5087
7672
1658
362
7137
4045
5016
806
3020
7194
5075
8713
3929
9918
3393
959
3159
6636
6766
6819
8211
3179
9766
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Child-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
Adult-HG
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
5.5E-01
1.9E-04
2.8E-04
9.4E-05
1.2E-04
4.8E-04
NA
9.0E-01
1.8E-04
4.0E-04
l.OE-04
1.6E-04
5.0E-04
NA
3.8E-01
1.6E-04
3.1E-04
9.4E-05
1.2E-04
2.5E-04
NA
8.1E-01
1.8E-04
3.1E-04
l.OE-04
1.8E-04
4.9E-04
NA
3.0E-06
1.9E-07
4.0E-07
4.3E-07
2.1E-07
4.2E-07
5.5E-06
5.9E-06
5.4E-07
1.4E-06
1.5E-06
5.7E-07
2.0E-06
9.6E-06
2.3E-07
9.2E-08
2.3E-07
1.6E-07
9.0E-08
2.8E-07
1.6E-06
5.9E-07
3.2E-07
8.8E-07
9.3E-07
3.8E-07
1.1E-06
3.2E-06
4.3E-06
2.7E-07
5.8E-07
6.1E-07
3.0E-07
5.9E-07
7.9E-06
8.5E-06
7.7E-07
2.0E-06
2.1E-06
8.1E-07
2.8E-06
1.4E-05
3.3E-07
1.3E-07
3.3E-07
2.3E-07
1.3E-07
4.0E-07
2.2E-06
8.4E-07
4.6E-07
1.3E-06
1.3E-06
5.4E-07
1.6E-06
4.6E-06
507,821
Capped
Capped
Capped
Capped
Capped
277,777
260,230
Capped
Capped
Capped
Capped
260,230
160,912
Capped
Capped
Capped
Capped
Capped
Capped
980,056
Capped
Capped
Capped
Capped
Capped
Capped
489,027
Capped = Modeling estimates indicated risks below levels of concern at concentrations above 1,000,000 mg kg"1
(i.e., SFS could be comprised entirely of this constituent and still not cause risk).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-3
-------
Appendix K: Detailed Human Health Results
Table K-2. Detailed Human Health Results
(Based on General Population Median Consumption Rates for Produce)
RunID
Receptor
Type
%-tile
Exposure
Pathway
Soil/Produce
Concentration
(mg kg-1)
ADD or
LADD
(mg kg -1 d'1)
Unitized
Dose Ratio
(unitless)
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
Arsenic - Cancer
228
9569
6290
3055
7051
9569
5208
430
631
5371
3314
8637
5148
2116
2638
1198
1198
1198
455
1198
8883
6628
3340
3410
3340
2136
3340
1770
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
4.2E-01
3.9E-04
8.3E-04
2.1E-04
3.2E-04
5.8E-04
NA
4.5E-01
3.8E-04
8.0E-04
2.0E-04
3.0E-04
5.6E-04
NA
4.5E-01
3.8E-04
7.9E-04
2.0E-04
2.9E-04
5.6E-04
NA
5.0E-01
3.6E-04
6.8E-04
1.9E-04
2.5E-04
5.4E-04
NA
1.8E-07
1.5E-08
1.5E-08
1.4E-08
3.4E-08
2.7E-08
2.9E-07
3.2E-07
2.7E-08
3.6E-08
1.8E-08
5.6E-08
5.4E-08
5.0E-07
2.7E-08
1.2E-08
4.1E-08
9.0E-09
l.OE-08
2.5E-08
1.2E-07
8.2E-08
5.3E-08
1.8E-07
3.9E-08
4.4E-08
1.1E-07
5.0E-07
2.7E-02
2.2E-03
2.3E-03
2.0E-03
5.1E-03
4.0E-03
4.4E-02
4.8E-02
4.1E-03
5.4E-03
2.7E-03
8.4E-03
8.1E-03
7.5E-02
4.0E-03
1.9E-03
6.2E-03
1.3E-03
1.5E-03
3.8E-03
1.9E-02
1.2E-02
8.0E-03
2.7E-02
5.8E-03
6.5E-03
1.6E-02
7.4E-02
80
1001
971
1092
435
546
50
46
541
409
823
263
271
30
549
1187
357
1635
1451
579
118
180
276
83
380
337
135
30
Arsenic- Noncancer
5759
34
9883
34
8356
34
2391
464
9993
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
50
50
50
50
50
50
50
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
7.4E-01
3.8E-04
8.1E-04
2.0E-04
3.2E-04
5.7E-04
NA
7.3E-01
3.9E-04
2.6E-06
2.1E-07
2.2E-07
1.9E-07
4.5E-07
3.9E-07
4.0E-06
3.9E-06
2.2E-07
8.8E-03
7.0E-04
7.2E-04
6.4E-04
1.5E-03
1.3E-03
1.3E-02
1.3E-02
7.2E-04
250
3142
3048
3413
1452
1712
166
169
3057
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-4
-------
Appendix K: Detailed Human Health Results
Table K-2. Detailed Human Health Results
(Based on General Population Median Consumption Rates for Produce)
RunID
9686
9993
8455
9993
2525
254
6349
4527
6349
5117
6349
70
7850
5770
8505
5770
4018
5770
4444
Receptor
Type
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
%-tile
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Exposure
Pathway
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mg kg-1)
8.4E-04
2.1E-04
3.4E-04
5.8E-04
NA
5.9E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
NA
9.7E-01
3.9E-04
8.4E-04
2.1E-04
3.5E-04
5.8E-04
NA
ADD or
LADD
(mg kg -1 d'1)
2.2E-07
2.0E-07
5.2E-07
4.0E-07
5.4E-06
2.3E-07
9.2E-08
3.1E-07
6.7E-08
7.6E-08
1.9E-07
9.7E-07
4.0E-07
9.4E-08
3.2E-07
6.8E-08
8.8E-08
1.9E-07
1.2E-06
Unitized
Dose Ratio
(unitless)
7.5E-04
6.6E-04
1.7E-03
1.3E-03
1.8E-02
7.8E-04
3.1E-04
l.OE-03
2.2E-04
2.5E-04
6.3E-04
3.2E-03
1.3E-03
3.1E-04
1.1E-03
2.3E-04
2.9E-04
6.4E-04
3.9E-03
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
2936
3321
1259
1666
121
2823
7174
2152
9882
8681
3498
682
1644
7024
2056
9676
7484
3425
570
Cobalt
1898
4798
6625
2100
6312
3001
3059
2268
9540
8465
9540
7152
9540
8674
5203
6413
2740
6413
2503
6413
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
5.2E-01
1.2E-03
1.6E-03
7.1E-04
1.1E-03
2.3E-03
NA
6.5E-01
1.4E-03
1.6E-03
7.2E-04
1.1E-03
2.5E-03
NA
6.0E-01
1.3E-03
1.6E-03
6.9E-04
l.OE-03
2.4E-03
3.6E-06
6.7E-07
4.2E-07
5.7E-07
1.4E-06
1.5E-06
8.2E-06
6.2E-06
7.5E-07
4.4E-07
6.9E-07
1.7E-06
1.7E-06
1.1E-05
3.1E-07
3.1E-07
5.9E-07
2.3E-07
2.6E-07
7.9E-07
1.2E-02
2.2E-03
1.4E-03
1.9E-03
4.8E-03
5.2E-03
2.7E-02
2.1E-02
2.5E-03
1.5E-03
2.3E-03
5.5E-03
5.7E-03
3.8E-02
l.OE-03
l.OE-03
2.0E-03
7.6E-04
8.5E-04
2.6E-03
181
986
1,570
1,152
460
427
80
106
876
1,495
951
399
384
58
2,099
2,115
1,112
2,913
2,586
830
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-5
-------
Appendix K: Detailed Human Health Results
Table K-2. Detailed Human Health Results
(Based on General Population Median Consumption Rates for Produce)
RunID
5410
5751
6328
7792
6328
9954
6328
5260
Receptor
Type
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
%-tile
50
90
90
90
90
90
90
90
Exposure
Pathway
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mg kg-1)
NA
8.1E-01
1.4E-03
1.6E-03
7.2E-04
1.1E-03
2.5E-03
NA
ADD or
LADD
(mg kg -1 d'1)
2.5E-06
6.4E-07
3.3E-07
6.3E-07
2.4E-07
2.7E-07
8.4E-07
2.9E-06
Unitized
Dose Ratio
(unitless)
8.3E-03
2.1E-03
1.1E-03
2.1E-03
7.9E-04
9.1E-04
2.8E-03
9.7E-03
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
265
1,038
2,012
1,053
2,772
2,405
790
226
Iron
1613
1020
3002
20
1262
7968
6883
1658
5217
9656
5217
6010
5217
4792
7194
1139
3977
1139
2167
1139
1301
3159
6286
9385
6286
9461
6286
5677
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Child-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
Adult-GP Med
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
5.5E-01
1.8E-04
3.1E-04
9.3E-05
1.4E-04
5.0E-04
NA
9.0E-01
1.9E-04
3.4E-04
l.OE-04
1.8E-04
5.0E-04
NA
3.8E-01
1.7E-04
2.9E-04
9.1E-05
1.4E-04
4.4E-04
NA
8.1E-01
1.9E-04
3.4E-04
l.OE-04
1.9E-04
5.0E-04
NA
3.0E-06
9.7E-08
8.2E-08
8.9E-08
2.2E-07
3.1E-07
3.8E-06
5.9E-06
1.1E-07
9.1E-08
9.8E-08
2.8E-07
3.4E-07
6.8E-06
2.3E-07
4.1E-08
1.1E-07
3.0E-08
3.5E-08
1.5E-07
5.9E-07
5.9E-07
4.6E-08
1.3E-07
3.4E-08
4.8E-08
1.7E-07
l.OE-06
4.3E-06
1.4E-07
1.2E-07
1.3E-07
3.1E-07
4.5E-07
5.4E-06
8.5E-06
1.5E-07
1.3E-07
1.4E-07
4.1E-07
4.8E-07
9.7E-06
3.3E-07
5.9E-08
1.6E-07
4.3E-08
5.0E-08
2.1E-07
8.5E-07
8.4E-07
6.6E-08
1.9E-07
4.8E-08
6.9E-08
2.4E-07
1.4E-06
507,821
Capped
Capped
Capped
Capped
Capped
404,714
260,230
Capped
Capped
Capped
Capped
Capped
226,140
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped
Capped = Modeling estimates indicated risks below levels of concern at concentrations above 1,000,000 mg kg'1
(i.e., SFS could be comprised entirely of this constituent and still not cause risk).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-6
-------
Appendix K: Detailed Human Health Results
Table K-3. Detailed Human Health Results
(Based on General Population High Consumption Rates for Produce)
RunID
Receptor Type
%-tile
Exposure
Pathway
Soil/Produce
Concentration
(mg kg-1)
ADD or
LADD
(mg kg -1 d'1)
Unitized
Dose Ratio
(unitless)
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
Arsenic - Cancer
228
9569
7457
5947
3555
9569
2701
430
1485
3921
631
5141
631
1692
2638
1198
1198
1198
455
1198
7041
6628
3340
3410
3340
2136
3340
3447
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
4.2E-01
3.9E-04
8.3E-04
2.1E-04
2.9E-04
5.8E-04
NA
4.5E-01
3.8E-04
8.0E-04
2.0E-04
2.9E-04
5.6E-04
NA
4.5E-01
3.8E-04
7.9E-04
2.0E-04
2.9E-04
5.6E-04
NA
5.0E-01
3.6E-04
6.8E-04
1.9E-04
2.5E-04
5.4E-04
NA
1.8E-07
4.5E-08
1.2E-07
7.2E-08
1.8E-07
1.1E-07
7.4E-07
3.2E-07
8.7E-08
2.3E-07
1.3E-07
2.5E-07
2.2E-07
1.3E-06
2.7E-08
3.9E-08
1.5E-07
4.1E-08
3.3E-08
9.4E-08
3.8E-07
8.2E-08
1.7E-07
6.4E-07
1.8E-07
1.4E-07
4.0E-07
1.6E-06
2.7E-02
6.7E-03
1.8E-02
1.1E-02
2.7E-02
1.7E-02
1.1E-01
4.8E-02
1.3E-02
3.4E-02
1.9E-02
3.7E-02
3.4E-02
1.9E-01
4.0E-03
5.8E-03
2.2E-02
6.1E-03
5.0E-03
1.4E-02
5.7E-02
1.2E-02
2.5E-02
9.5E-02
2.6E-02
2.1E-02
6.0E-02
2.4E-01
80
327
125
204
82
131
20
46
170
64
114
60
66
12
549
383
99
362
442
157
39
180
89
23
84
103
37
9
Arsenic- Noncancer
5759
34
217
34
9691
34
2614
464
9993
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
50
50
50
50
50
50
50
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
7.4E-01
3.8E-04
8.0E-04
2.0E-04
3.3E-04
5.7E-04
NA
7.3E-01
3.9E-04
2.6E-06
6.4E-07
1.7E-06
l.OE-06
2.5E-06
1.6E-06
9.9E-06
3.9E-06
6.6E-07
8.8E-03
2.1E-03
5.5E-03
3.4E-03
8.2E-03
5.4E-03
3.3E-02
1.3E-02
2.2E-03
250
1025
397
641
268
410
67
169
998
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-7
-------
Appendix K: Detailed Human Health Results
Table K-3. Detailed Human Health Results
(Based on General Population High Consumption Rates for Produce)
RunID
6242
9993
6864
9993
7163
254
6349
4527
6349
5117
6349
6954
7850
5770
8505
5770
4018
5770
5719
Receptor Type
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
%-tile
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Exposure
Pathway
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mg kg-1)
8.4E-04
2.1E-04
3.4E-04
5.8E-04
NA
5.9E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
NA
9.7E-01
3.9E-04
8.4E-04
2.1E-04
3.5E-04
5.8E-04
NA
ADD or
LADD
(mg kg -1 d'1)
1.7E-06
1.1E-06
2.9E-06
1.7E-06
1.2E-05
2.3E-07
2.9E-07
1.1E-06
3.0E-07
2.5E-07
7.0E-07
2.9E-06
4.0E-07
2.9E-07
1.2E-06
3.1E-07
2.9E-07
7.1E-07
3.1E-06
Unitized
Dose Ratio
(unitless)
5.8E-03
3.5E-03
9.5E-03
5.5E-03
3.9E-02
7.8E-04
9.5E-04
3.7E-03
l.OE-03
8.3E-04
2.3E-03
9.6E-03
1.3E-03
9.7E-04
3.9E-03
l.OE-03
9.6E-04
2.4E-03
l.OE-02
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
378
623
231
399
56
2823
2313
598
2189
2645
949
229
1644
2264
571
2144
2281
929
210
Cobalt
1898
2501
5502
8654
5236
994
9733
2268
9540
7830
9540
7152
9540
4005
5203
6413
2740
6413
2503
6413
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
5.2E-01
1.2E-03
1.6E-03
7.3E-04
l.OE-03
2.5E-03
NA
6.5E-01
1.4E-03
1.6E-03
7.2E-04
1.1E-03
2.5E-03
NA
6.0E-01
1.3E-03
1.6E-03
6.9E-04
l.OE-03
2.4E-03
1.2E-02
6.9E-03
l.OE-02
1.1E-02
2.5E-02
2.2E-02
8.6E-02
2.1E-02
7.7E-03
1.1E-02
1.2E-02
3.0E-02
2.4E-02
1.1E-01
l.OE-03
3.2E-03
7.1E-03
3.4E-03
2.8E-03
9.8E-03
1.2E-02
6.9E-03
l.OE-02
1.1E-02
2.5E-02
2.2E-02
8.6E-02
2.1E-02
7.7E-03
1.1E-02
1.2E-02
3.0E-02
2.4E-02
1.1E-01
l.OE-03
3.2E-03
7.1E-03
3.4E-03
2.8E-03
9.8E-03
181
318
215
203
87
102
25
106
286
192
179
73
92
21
2099
682
309
645
788
225
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-8
-------
Appendix K: Detailed Human Health Results
Table K-3. Detailed Human Health Results
(Based on General Population High Consumption Rates for Produce)
RunID
509
5751
6328
7792
6328
9954
6328
9534
Receptor Type
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
%-tile
50
90
90
90
90
90
90
90
Exposure
Pathway
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil/Produce
Concentration
(mg kg-1)
NA
8.1E-01
1.4E-03
1.6E-03
7.2E-04
1.1E-03
2.5E-03
NA
ADD or
LADD
(mg kg -1 d'1)
2.7E-02
2.1E-03
3.4E-03
7.5E-03
3.6E-03
3.0E-03
l.OE-02
3.0E-02
Unitized
Dose Ratio
(unitless)
2.7E-02
2.1E-03
3.4E-03
7.5E-03
3.6E-03
3.0E-03
l.OE-02
3.0E-02
Foundry Sand-
Specific Screening
Concentration
(mgkg'SFS)
80
1038
649
293
614
733
214
74
Iron
1613
3135
6911
2289
5124
2931
2508
1658
5217
8669
5217
6010
5217
7537
7194
1139
3977
1139
2167
1139
7952
3159
6286
9385
6286
9461
6286
4181
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Child-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
Adult-GP High
50
50
50
50
50
50
50
90
90
90
90
90
90
90
50
50
50
50
50
50
50
90
90
90
90
90
90
90
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil Ingestion
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
5.5E-01
1.8E-04
3.2E-04
9.4E-05
1.4E-04
4.6E-04
NA
9.0E-01
1.9E-04
3.4E-04
l.OE-04
1.8E-04
5.0E-04
NA
3.8E-01
1.7E-04
2.9E-04
9.1E-05
1.4E-04
4.4E-04
NA
8.1E-01
1.9E-04
3.4E-04
l.OE-04
1.9E-04
5.0E-04
NA
4.3E-06
4.3E-07
9.0E-07
6.8E-07
1.7E-06
1.9E-06
9.7E-06
8.5E-06
4.6E-07
l.OE-06
7.4E-07
2.2E-06
2.0E-06
1.5E-05
3.3E-07
1.8E-07
5.7E-07
1.9E-07
1.6E-07
7.8E-07
2.2E-06
8.4E-07
2.1E-07
6.7E-07
2.2E-07
2.3E-07
8.7E-07
3.0E-06
4.3E-06
4.3E-07
9.0E-07
6.8E-07
1.7E-06
1.9E-06
9.7E-06
8.5E-06
4.6E-07
l.OE-06
7.4E-07
2.2E-06
2.0E-06
1.5E-05
3.3E-07
1.8E-07
5.7E-07
1.9E-07
1.6E-07
7.8E-07
2.2E-06
8.4E-07
2.1E-07
6.7E-07
2.2E-07
2.3E-07
8.7E-07
3.0E-06
507,821
Capped
Capped
Capped
Capped
Capped
225,994
260,230
Capped
Capped
Capped
Capped
Capped
148,480
Capped
Capped
Capped
Capped
Capped
Capped
991,820
Capped
Capped
Capped
Capped
Capped
Capped
726,078
Capped = Modeling estimates indicated risks below levels of concern at concentrations above 1,000,000 mg kg'1
(i.e., SFS could be comprised entirely of this constituent and still not cause risk).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-9
-------
Appendix K: Detailed Human Health Results
Table K-4. Detailed Human Health Results: Groundwater Ingestion
RunID
Receptor Type
%-tile
Exposure
Pathway
Groundwater
Concentration
(mgL1)
ADD or
LADD
(mg kg -1 d'1)
Unitized
Dose Ratio
(unitless)
Foundry Sand-
Specific Screening
Concentration
(Groundwater
Pathway)
(mgkg1)
Arsenic - Cancer
NA
4302
NA
9716
Child
Child
Adult
Adult
50
90
50
90
Groundwater
Groundwater
Groundwater
Groundwater
Pathway
Incomplete
3.2E-04
Pathway
Incomplete
2.2E-04
NA
NA
NA
3.7E-02
NA
3.1E-02
NA
59
NA
71
Arsenic - Noncancer
NA
5146
NA
1578
Child
Child
Adult
Adult
50
90
50
90
Groundwater
Groundwater
Groundwater
Groundwater
Pathway
Incomplete
1.2E-04
Pathway
Incomplete
l.OE-03
NA
NA
NA
1.3E-02
NA
6.8E-03
NA
171
NA
321
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-10
-------
Appendix K: Detailed Human Health Results
Table K-5. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using Home Gardener Consumption Rate Distributions for Produce
RunID
%-tile
Consumption Rate
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19 | Adult
ED
(yr)"
EF
(d yr1)
AT
(yr)
Days
Year :
ADD""
(mg kg1
BW d'1)
LADD
(mg kg"1
BW d'1)
Benchmark
(RiD or
1E-5CSF1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFSConc"""
(mg kg^1 dry
weight)
Home Gardener Child - Cancer
5114
4734
50
90
100
2.87
1.03
19.48
1.84
0.21
25.44
100
0.99
1.44
7.83
4.64
1.01
15.91
100
0.54
0.12
0.00
0.83
2.86
100
2.60
0.10
18.76
1.24
0.81
23.51
NA
NA
NA
NA
NA
NA
100
1.30
2.85
0.80
8.80
0.01
13.77
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
(mg d-1)
(g WW kg -' BW d'1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(mg d"1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
7.0E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
8.3E-01
3.9E-04
8.2E-04
2.1E-04
3.0E-04
5.8E-04
6.5E-01
3.8E-04
7.9E-04
2.0E-04
2.9E-04
5.6E-04
5.5E-01
3.8E-04
8.0E-04
2.0E-04
2.9E-04
5.6E-04
NA
NA
NA
NA
NA
NA
3.6E-01
3.7E-04
7.7E-04
2.0E-04
2.8E-04
5.5E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
18
NA
NA
NA
NA
NA
14
NA
NA
NA
NA
NA
40
NA
NA
NA
NA
NA
26
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
65
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6
6
6
6
6
6
13
13
13
13
13
13
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
2.1E-06
8.4E-07
6.0E-07
2.4E-06
3.9E-07
3.5E-07
1.8E-06
5.7E-07
8.5E-07
1.7E-06
l.OE-06
3.6E-07
1.7E-07
6.9E-08
4.9E-08
1.9E-07
3.2E-08
2.9E-08
5.4E-07
3.2E-07
l.OE-07
1.6E-07
3.1E-07
1.8E-07
6.5E-08
1.1E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
2.5E-02
l.OE-02
7.4E-03
2.9E-02
4.8E-03
4.3E-03
8.1E-02
4.8E-02
1.6E-02
2.3E-02
4.6E-02
2.7E-02
9.8E-03
1.7E-01
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
27
13
Home Gardener Adult - Cancer
4772
7831
50
90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
2.33
1.00
2.94
2.37
0.34
8.97
50
0.39
1.83
1.52
0.64
2.11
6.50
(mg d"1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4.6E-01
3.8E-04
8.0E-04
2.0E-04
2.9E-04
5.7E-04
9.0E-01
3.9E-04
8.2E-04
2.1E-04
3.0E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
96
NA
NA
NA
NA
NA
97
NA
NA
NA
NA
NA
10
10
10
10
10
10
42
42
42
42
42
42
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
1.5E-07
7.8E-07
6.7E-07
4.2E-07
5.4E-07
1.8E-07
2.8E-07
1.3E-07
1.3E-06
2.2E-07
1.5E-07
1.2E-06
2.0E-08
1.1E-07
9.2E-08
5.8E-08
7.5E-08
2.5E-08
3.7E-07
1.6E-07
7.7E-08
7.2E-07
1.3E-07
8.7E-08
6.7E-07
1.8E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
3.0E-03
1.6E-02
1.4E-02
8.7E-03
1.1E-02
3.7E-03
5.6E-02
2.4E-02
1.2E-02
1.1E-01
1.9E-02
1.3E-02
l.OE-01
2.8E-01
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
39
8
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-ll
-------
Appendix K: Detailed Human Health Results
Table K-5. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using Home Gardener Consumption Rate Distributions for Produce
RunlD
%-tile
Consumpt on Rate
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19 | Adult
ED
(yr)"
EF
(d yr1)
AT
(yr)
Days
Year1
ADD""
(mg kg-1
BW d-1)
LADD
(mg kg-1
BW d-1)
Benchmark
(RfDor
1E-5CSF1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFSConc"""
(mg kjr1 dry
weight)
Home Gardener Child - Noncancer
6244
1587
50
90
100
1.85
0.77
17.28
2.13
0.67
22.70
100
2.26
0.04
10.67
5.54
8.16
26.66
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d"1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
8.7E-01
3.9E-04
8.4E-04
2.1E-04
3.5E-04
5.8E-04
8.8E-01
3.9E-04
8.4E-04
2.1E-04
3.4E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
21
NA
NA
NA
NA
NA
16
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4
4
4
4
4
4
3
3
3
3
3
3
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.5E-06
6.3E-07
5.4E-07
2.5E-06
5.8E-07
3.7E-07
7.1E-06
3.4E-06
7.7E-07
2.8E-08
1.6E-06
1.5E-06
4.5E-06
1.2E-05
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
8.2E-03
2.1E-03
1.8E-03
8.5E-03
1.9E-03
1.2E-03
2.4E-02
1.1E-02
2.6E-03
9.2E-05
5.2E-03
4.9E-03
1.5E-02
3.9E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
93
56
Home Gardener Adult- Noncancer
6559
9631
50
90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
2.35
0.76
4.15
0.53
0.86
8.65
50
0.64
7.68
1.15
0.41
0.70
10.58
(mg d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d'1)
(mg d-1)
(g WW kg -> BW d-1)
(gWWkg-'BWd-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
9.0E-01
3.9E-04
8.4E-04
2.1E-04
3.3E-04
5.8E-04
6.2E-01
3.8E-04
8.1E-04
2.0E-04
3.1E-04
5.6E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
61
NA
NA
NA
NA
NA
117
NA
NA
NA
NA
NA
4
4
4
4
4
4
11
11
11
11
11
11
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4.5E-07
8.0E-07
5.3E-07
6.1E-07
1.4E-07
4.7E-07
3.0E-06
1.6E-07
2.1E-07
5.2E-06
1.6E-07
l.OE-07
3.7E-07
6.2E-06
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
1.5E-03
2.7E-03
1.8E-03
2.0E-03
4.7E-04
1.6E-03
l.OE-02
5.3E-04
7.1E-04
1.7E-02
5.5E-04
3.4E-04
1.2E-03
2.1E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
219
106
* Model rounds exposure duration to whole number.
** Soil ingestion dose includes an adjustment to the arsenic soil concentration to reflect EPA's default relative bioavailability (RBA) value of 60%.
*** Includes conversion from wet to dry weight reflecting average modeled solids content of 90 percent (1 0 percent moisture).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-12
-------
Appendix K: Detailed Human Health Results
Table K-6. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using General Population Median Consumption Rates and Assuming 50% of Consumed Produce is Homegrown
RunlD
%-tile
Consumption Rate*
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19
Adult
ED
(yr)
EF
(d yr1)
AT
(yr)
Days
Year1
ADD""
(mg kg-1
BW d-1)
LADD
(mg kg-1
BW d-1)
Benchmark
(RfDor
1E-5 CSF -1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFSConc"""
(mg kjr1 dry
weight)
Gen. Pop. Child -Cancer
5208
2116
50
90
100
0.63
0.32
1.35
1.95
0.72
4.97
100
0.63
0.32
1.35
1.95
0.72
4.97
100
0.39
0.30
0.09
1.10
0.50
2.38
100
0.39
0.30
0.09
1.10
0.50
2.38
NA
NA
NA
NA
NA
NA
100
0.23
0.27
0.90
0.44
0.41
2.24
NA
NA
NA
NA
NA
NA
50
0.27
0.45
0.46
0.32
0.35
1.86
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d"1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
6.3E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
6.0E-01
4.0E-04
8.1E-04
2.0E-04
3.1E-04
5.7E-04
NA
NA
NA
NA
NA
NA
3.1E-01
3.7E-04
7.8E-04
2.0E-04
2.9E-04
5.5E-04
NA
NA
NA
NA
NA
NA
2.1E-01
3.7E-04
7.6E-04
1.9E-04
2.8E-04
5.4E-04
NA
NA
NA
NA
NA
NA
1.8E-01
3.6E-04
7.5E-04
1.9E-04
2.7E-04
5.3E-04
16
NA
NA
NA
NA
NA
17
NA
NA
NA
NA
NA
49
NA
NA
NA
NA
NA
26
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
76
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
52
NA
NA
NA
NA
NA
6
6
6
6
6
6
22
22
22
22
22
22
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
2.4E-06
2.1E-07
2.2E-07
2.0E-07
4.6E-07
3.9E-07
7.6E-07
1.2E-07
2.1E-07
l.OE-07
2.2E-07
2.6E-07
1.8E-07
1.6E-08
1.6E-08
1.5E-08
3.5E-08
3.0E-08
2.9E-07
2.2E-07
3.7E-08
6.1E-08
3.0E-08
6.6E-08
7.8E-08
5.0E-07
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
2.7E-02
2.4E-03
2.5E-03
2.2E-03
5.2E-03
4.5E-03
4.4E-02
3.4E-02
5.5E-03
9.2E-03
4.5E-03
9.9E-03
1.2E-02
7.5E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
50
30
Gen. Pop. Adult -Cancer
8883
1770
50
90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
0.27
0.45
0.46
0.32
0.35
1.86
50
0.27
0.45
0.46
0.32
0.35
1.86
(mg d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
7.2E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
3.3E-01
3.6E-04
7.6E-04
1.9E-04
2.8E-04
5.4E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
72
NA
NA
NA
NA
NA
71
NA
NA
NA
NA
NA
9
9
9
9
9
9
44
44
44
44
44
44
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
3.0E-07
9.3E-08
3.1E-07
6.7E-08
7.5E-08
1.9E-07
1.4E-07
8.7E-08
2.9E-07
6.3E-08
7. IE-OS
1.8E-07
3.6E-08
1. IE-OS
3.7E-08
8.1E-09
9.1E-09
2.3E-08
1.2E-07
8.5E-08
5.2E-08
1.7E-07
3.8E-08
4.2E-08
1.1E-07
5.0E-07
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
5.4E-03
1.7E-03
5.6E-03
1.2E-03
1.4E-03
3.4E-03
1.9E-02
1.3E-02
7.8E-03
2.6E-02
5.7E-03
6.4E-03
1.6E-02
7.4E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
118
30
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-13
-------
Appendix K: Detailed Human Health Results
Table K-6. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using General Population Median Consumption Rates and Assuming 50% of Consumed Produce is Homegrown
RunlD
%-tile
Consumption Rate*
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19
Adult
ED
(yrt
EF
(d yr1)
AT
(yr)
Days
Year1
ADD""
(mg kg"1
BW d'1)
LADD
(mg kg1
BW d'1)
Benchmark
(RiD or
1E-5 CSF"1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFS Cone """
(mg kg^1 dry
weight)
Gen. Pop. Child - Noncancer
2391
2525
50
90
100
0.63
0.32
1.35
1.95
0.72
4.97
100
0.63
0.32
1.35
1.95
0.72
4.97
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
(mg d"1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
7.0E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
9.4E-01
3.9E-04
8.3E-04
2.1E-04
3.1E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
16
NA
NA
NA
NA
NA
14
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3
3
3
3
3
3
1
1
1
1
1
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.6E-06
2.1E-07
2.2E-07
1.4E-07
4.6E-07
3.9E-07
4.0E-06
3.9E-06
2.2E-07
2.2E-07
2.0E-07
4.8E-07
4.0E-07
5.4E-06
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
8.5E-03
7.1E-04
7.2E-04
4.7E-04
1.5E-03
1.3E-03
1.3E-02
1.3E-02
7.2E-04
7.4E-04
6.6E-04
1.6E-03
1.3E-03
1.8E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
166
121
Gen. Pop. Adult - Noncancer
70
4444
50
90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
0.27
0.45
0.46
0.32
0.35
1.86
50
0.27
0.45
0.46
0.32
0.35
1.86
(mg d"1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d'1)
(gWWkg-'BWd-1)
(mg d"1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(gWWkg-'BWd-1)
(g WW kg -' BW d-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
8.2E-01
3.9E-04
8.2E-04
2.1E-04
3.1E-04
5.8E-04
8.5E-01
3.9E-04
8.2E-04
2.1E-04
3.0E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
109
NA
NA
NA
NA
NA
61
NA
NA
NA
NA
NA
1
1
1
1
1
1
2
2
2
2
2
2
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.2E-07
9.3E-08
3.1E-07
6.8E-08
7.9E-08
1.9E-07
9.7E-07
4.2E-07
9.4E-08
3.1E-07
6.8E-08
7.7E-08
1.9E-07
1.2E-06
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
7.5E-04
3.1E-04
l.OE-03
2.3E-04
2.6E-04
6.4E-04
3.2E-03
1.4E-03
3.1E-04
l.OE-03
2.3E-04
2.6E-04
6.4E-04
3.9E-03
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
682
570
* Produce consumption rates scaled to reflect 50% homegrown produce.
** Soil ingestion dose includes an adjustment to the arsenic soil concentration to reflect EPA's default relative bioavai ability (RBA) value of 60%.
*** Includes conversion from wet to dry weight reflecting average modeled solids content of 90 percent (10 percent moisture).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-14
-------
Appendix K: Detailed Human Health Results
Table K-7. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using General Population High Consumption Rates and Assuming 50% of Consumed Produce is Homegrown
RunID
%-tile
Consumption Rate*
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19
Adult
ED
(yr)
EF
(d yr1)
AT
(yr)
Days
Year :
ADD""
(mg kg"1
BW d'1)
LADD
(mg kg1
BW d'1)
Benchmark
(RiD or
1E-5 CSF"1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFS Cone"""
(mg kg : dry
weight)
Gen. Pop. Child - Cancer
2701
1692
30
90
100
1.93
2.48
7.19
10.62
3.01
25.23
100
1.93
2.48
7.19
10.62
3.01
25.23
100
1.30
1.70
4.05
3.15
2.10
100
1.30
1.70
4.05
3.15
2.10
12.30
NA
NA
NA
NA
NA
NA
100
0.75
1.25
2.70
1.45
1.50
7.65
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
6.4E-01
3.9E-04
8.0E-04
2.0E-04
3.0E-04
5.7E-04
3.8E-01
3.8E-04
7.9E-04
3.8E-04
2.9E-04
5.6E-04
4.1E-01
3.8E-04
7.8E-04
2.0E-04
2.9E-04
5.4E-04
1.6E-01
3.6E-04
7.5E-04
1.9E-04
2.7E-04
5.3E-04
NA
NA
NA
NA
NA
NA
l.OE-01
3.5E-04
7.2E-04
1.8E-04
2.6E-04
5.1E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
16
NA
NA
NA
NA
NA
15
NA
NA
NA
NA
NA
31
NA
NA
NA
NA
NA
29
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
60
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
6
6
6
6
6
6
19
19
19
19
19
19
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
2.0E-06
5.8E-07
1.5E-06
9.2E-07
2.1E-06
1.5E-06
5.6E-07
4.0E-07
1.1E-06
8.4E-07
l.OE-06
1.1E-06
1.7E-07
5.0E-08
1.3E-07
7.9E-08
1.8E-07
1.3E-07
7.4E-07
1.4E-07
l.OE-07
2.8E-07
2.2E-07
2.5E-07
2.8E-07
1.3E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
2.6E-02
7.5E-03
1.9E-02
1.2E-02
2.7E-02
1.9E-02
1.1E-01
2.1E-02
1.5E-02
4.3E-02
3.2E-02
3.8E-02
4.1E-02
1.9E-01
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
20
12
Gen. Pop. Adult -Cancer
7041
3447
50
90
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
0.85
1.63
2.09
1.06
1.29
6.92
50
0.85
1.63
2.09
1.06
1.29
6.92
(mg d"1)
(g WW kg -' BW d-1)
(g WW kg j BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(mg d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
7.8E-01
3.9E-04
8.3E-04
2.1E-04
3.2E-04
5.8E-04
2.2E-01
3.6E-04
7.5E-04
1.9E-04
2.7E-04
5.3E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
60
NA
NA
NA
NA
NA
64
NA
NA
NA
NA
NA
9
9
9
9
9
9
46
46
46
46
46
46
350
350
350
350
350
350
350
350
350
350
350
350
70
70
70
70
70
70
70
70
70
70
70
70
365
365
365
365
365
365
365
365
365
365
365
365
3.9E-07
2.9E-07
1.1E-06
3.1E-07
2.7E-07
7.0E-07
l.OE-07
2.7E-07
l.OE-06
2.8E-07
2.3E-07
6.5E-07
4.8E-08
3.6E-08
1.4E-07
3.8E-08
3.3E-08
8.7E-08
3.8E-07
6.6E-08
1.7E-07
6.5E-07
1.8E-07
1.5E-07
4.1E-07
1.6E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
6.7E-06
7.3E-03
5.3E-03
2.1E-02
5.6E-03
4.9E-03
1.3E-02
5.7E-02
9.9E-03
2.5E-02
9.8E-02
2.7E-02
2.2E-02
6.2E-02
2.4E-01
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
39
9
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-15
-------
Appendix K: Detailed Human Health Results
Table K-7. Example Calculations for Home Gardener Soil Pathway: Total Arsenic Ingestion Using General Population High Consumption Rates and Assuming 50% of Consumed Produce is Homegrown
RunlD
%-tile
Consumption Rate*
Child
1-5
Child
6-11
Child
2-19
Adult
Units
Soil/Food Cone
Child
1-5
Child
6-11
Child
12-19
Adult
Body Weight (kg)
Child
1-5
Child
6-11
Child
2-19
Adult
ED
(yr)
EF
(d yr1)
AT
(yr)
Days
Year1
ADD""
(mg kg"1
BW d-1)
LADD
(mg kg1
BW d-1)
Benchmark
(RfDor
1E-5 CSF -1)
Unitized
Dose
Ratio
(unitless)
Pathway
Allowable
SFSConc"""
(mg kg : dry
weight)
Gen. Pop. Child - Noncancer
2614
7163
50
90
100
1.93
2.48
7.19
10.62
3.01
25.23
100
1.93
2.48
7.19
10.62
3.01
25.23
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
(mg d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(mg d"1)
(g WW kg j BW d'1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
6.4E-01
3.9E-04
8.1E-04
2.0E-04
3.0E-04
5.7E-04
9.1E-01
3.9E-04
8.4E-04
2.1E-04
3.4E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
16
NA
NA
NA
NA
NA
14
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5
5
5
5
5
5
2
2
2
2
2
2
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.4E-06
6.5E-07
1.7E-06
l.OE-06
2.5E-06
1.6E-06
9.9E-06
3.8E-06
6.6E-07
1.8E-06
1.1E-06
2.9E-06
1.7E-06
1.2E-05
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
7.9E-03
2.2E-03
5.6E-03
3.5E-03
8.3E-03
5.4E-03
3.3E-02
1.3E-02
2.2E-03
5.8E-03
3.5E-03
9.5E-03
5.5E-03
3.9E-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
67
56
Gen. Pop. Adult - Noncancer
6954
5719
* Produce
** Soil in
""Inch
50
LJ
consumpt
'estion do*
des conv
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
50
0.85
1.63
2.09
1.06
1.29
6.92
50
0.85
1.63
2.09
1.06
1.29
6.92
(mg d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(mg d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
(g WW kg -> BW d-1)
(g WW kg -> BW d-1)
(g WW kg -' BW d-1)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5.2E-01
3.8E-04
7.9E-04
2.0E-04
2.9E-04
5.6E-04
9.6E-01
3.9E-04
8.5E-04
2.1E-04
3.5E-04
5.8E-04
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
55
NA
NA
NA
NA
NA
76
NA
NA
NA
NA
NA
25
25
25
25
25
25
1
1
1
1
1
1
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.8E-07
2.8E-07
1.1E-06
3.0E-07
2.5E-07
6.9E-07
2.9E-06
3.8E-07
2.9E-07
1.2E-06
3.1E-07
2.9E-07
7.1E-07
3.1E-06
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
3.0E-04
9.4E-04
9.4E-04
3.6E-03
9.9E-04
8.2E-04
2.3E-03
9.6E-03
1.3E-03
9.7E-04
3.9E-03
l.OE-03
9.7E-04
2.4E-03
l.OE-02
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
Soil
Protected Veg
Exposed Veg
Protected Fruit
Exposed Fruit
Root Veg
Total Ingestion
229
210
on rates scaled to reflect 50% homegrown produce.
e includes an adjustment to the arsenic soil concentration to reflect EPA's default relative bioavailability (RBA) value of 60%.
;rsion from wet to dry weight reflecting average modeled solids content of 90 percent (10 percent moisture).
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-16
-------
Appendix K: Detailed Human Health Results
Table K-8. Example Calculations for Home Gardener Groundwater Pathway for Arsenic
RunlD
%-tile
Consumption Rate
(mL kg1 BW dj)
Child
1-5
Child
6-11
Child
2-19
Adult
Groundwater Cone (mg L :)
Child
1-5
Child
6-11
Child
12-19
Adult
ED (yr)
EF (d yr-1)
AT
(yr)
Days
Year1
ADD
(mg kg1 BW
d1)
LADD
(mg kg -1 BW
d1)
Benchmark
(RfDor
1E-5 CSF -1)
Unitized
Dose Ratio
(unitless)
Pathway
Allowable SFS
Cone"
(mg kjr1 dry
weight)
Home Gardener Child - Cancer
4302
90
8.7128
12.233
NA
NA
3.2E-04
3.2E-04
NA
NA
6
350
70
365
2.8E-06
2.5E-07
6.7E-06
3.7E-02
Groundwater
59
Home Gardener Adult - Cancer
9716
90
NA
NA
NA
6.4184
NA
NA
NA
2.2E-04
11
350
70
365
1.4E-06
2.1E-07
6.7E-06
3.1E-02
Groundwater
71
Home Gardener Child - Noncancer
5146
90
33.096
9.6343
NA
NA
1.2E-04
NA
NA
NA
5
NA
NA
NA
3.9E-06
NA
3.0E-04
1.3E-02
Groundwater
171
Home Gardener Adult - Noncancer
1578
90
NA
NA
NA
1.9839
NA
NA
NA
0.001
13
NA
NA
NA
2.1E-06
NA
3.0E-04
6.8E-03
Groundwater
321
* Includes conversion from wet to dry weight reflecting average modeled solids content of 90 percent (10 percent moi. i
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
K-17
-------
-------
Appendix L: Detailed Ecological Results
Appendix L:
Detailed Ecological Results
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix L: Detailed Ecological Results
[This page intentionally left blank.]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
-------
Appendix L: Detailed Ecological Results
Table L-l. Detailed Ecological Results
Receptor
Name
Percentile
Soil
(mg kg -1)
Unitized
Dose
Ratio
(unitless)
Foundry Sand-
Specific Screening
Concentrations
(nig kg 'SFS)
Antimony
Soil Biota
Soil Biota
Mammals
Mammals
90
50
90
50
9.6E-01
8.2E-01
1.4E-01
1.2E-01
1 .2E-02
1 .OE-02
5.3E-01
4.5E-01
179
210
4.1
4.8
Chromium (III)
Mammals
Mammals
90
50
1.5E-01
1.4E-01
4.3E-03
4.1E-03
511
532
Copper
Plants
Plants
Soil Biota
Soil Biota
Mammals
Mammals
90
50
90
50
90
50
9.7E-01
9.0E-01
9.7E-01
9.0E-01
1.5E-01
1.3E-01
1.4E-02
1.3E-02
1.2E-02
1.1E-02
3.0E-03
2.7E-03
159
172
181
196
741
801
Manganese
Plant
Plants
Soil Biota
Soil Biota
Mammals
Mammals
90
50
90
50
90
50
9.7E-02
9.3E-02
9.7E-01
9.3E-01
9.7E-01
9.3E-01
4.4E-04
4.2E-04
2.2E-03
2.1E-03
2.4E-04
2.3E-04
4970
5212
1017
1066
9036
9477
Nickel
Plants
Plants
Soil Biota
Soil Biota
Mammals
Mammals
90
50
90
50
90
50
6.8E-02
6.4E-02
9.7E-01
9.2E-01
9.7E-01
9.2E-01
1.8E-03
1.7E-03
3.5E-02
3.3E-02
7.5E-03
7.1E-03
1230
1300
634
671
294
311
Risk Assessment of Spent Foundry Sands in Soil-Related Applications
L-l
-------
Appendix L: Detailed Ecological Results
[This page intentionally left blank]
Risk Assessment of Spent Foundry Sands in Soil-Related Applications L-2
------- |