kvEPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA.W6-90/008
August 1989
Development of Risk
Assessment
Methodology for
Municipal Sludge
Landfilling
-------
-------
Final
August 1989
DEVELOPMENT OF RISK ASSESSMENT METHODOLOGY
FOR MUNICIPAL SLUDGE LANDFILLING
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental
Protection Agency policy and approved for publication. Mention of trade
names or commercial products does not constitute endorsement or recommen-
dation for use.
-------
PREFACE
This 1s one of a series of reports that present methodologies for
assessing the potential risks to humans or other organisms from management
practices for the disposal or reuse of municipal sewage sludge. The manage-
ment practices addressed by this series Include land application practices,
distribution and marketing programs, landfUUng, Incineration and ocean
disposal. In particular, these reports deal with methods for evaluating
potential health and environmental risks from toxic chemicals that may be
present In sludge. This document addresses risks from chemicals associated
with municipal sludge landfUUng.
These proposed risk assessment procedures are designed as tools to
assist In the development of regulations for sludge management practices.
The procedures are structured to allow calculation of technical criteria for
sludge disposal/reuse options based on the potential for adverse health or
environmental Impacts. The criteria may address management practices (such
as site design or process control specifications), limits on sludge disposal
rates or limits on toxic chemical concentrations In the sludge.
The methods for criteria derivation presented In this report are
Intended to be used by the U.S. EPA Office of Water Regulations and Stan-
dards (OWRS) to develop technical criteria for toxic chemicals In sludge.
The present document focuses primarily on methods for the development of
nationally applicable criteria by OWRS.
This document was externally peer reviewed and completed 1n 1986.
Subsequent to further review by the U.S. EPA Science Advisory Board, a
revised draft incorporating review comments was produced in 1987. Various
scientific and editorial changes, which clarify but do not alter the overall
thrust of the document, have been made since that date.
111
-------
DOCUMENT DEVELOPMENT
Authors and Contributors
Larry Fradkin, Document Manager
Environmental Criteria and Assessment
Office
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Norman E. Kowal, Co-Document Manager
Health Effects Research Laboratory
Office of Health Research
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Gaynor Dawson, C. Joe English and
Rick W. Bond
ICF Northwest
601 Williams Blvd.
Richland, WA 99352
Randall J.F. Bruins
Environmental Criteria and
Assessment Office
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Cincinnati, OH 45268
William B. Peirano
Environmental Criteria and
Assessment Office
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Norma Whetzel
Office of Water Regulations
and Standards
U.S. Environmental Protection Agency
Washington, DC 20460
David Brown and Robert Swank
Environmental Research Laboratory
Office of Environmental Processes
and Effects Research
U.S. Environmental Protection Agency
Athens, GA 30613
Scientific Reviewers
Dr. Kirk Brown
Soil Science Department
Texas A&M
College Station, TX 75201
Dr. Tony Donigian
Aqua Terra Consultants
Mountain View, CA 94303
Dr. Wallace Fuller
Soil and Water Science Department
University of Arizona
Tucson, AZ 85721
Dr. James Geraghty
Geraghty and Miller, Inc.
Tampa, FL 33688
Dr. Dale Johnson
Dept. of Environmental Health
University of Cincinnati
Medical Center
Cincinnati, OH 45267
Dr. Fred Pohland
School of Civil Engineering
Georgia Institute of Technology
Atlanta, GA 30332
Dr. Martha Radike
Dept. of Environmental Health
University of Cincinnati
Medical Center
Cincinnati, OH 45267
IV
-------
Scientific Reviewers (cont.)
Dr. James Walsh
SCS Engineers, Inc.
Covington, KY 47017
Dr. Calvin H. Ward
Dept. of Environmental Science
and Engineering
Rice University
Houston, TX 77251
Document Preparation
Patricia A. Daunt, Bette L. Zwayer and Jacqueline Bohanon, Environmental
Criteria and Assessment Office, Cincinnati
Technical Publications Editor: Judith A. Olsen, Environmental Criteria and
Assessment Office, Cincinnati
-------
TABLE OF CONTENTS
Page
1. INTRODUCTION AND DESCRIPTION OF GENERAL METHODOLOGIC APPROACH . . 1-1
1.1. PURPOSE AND SCOPE 1-1
1.2. DEFINITION AND COMPONENTS OF RISK ASSESSMENT ....... 1-2
1.3. RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT PROCESS . . 1-3
1.3.1. Exposure Assessment 1-3
1.3.2. Hazard Identification and Dose-Response
Assessment 1-7
1.3.3. Risk Characterization 1-8
1.4. POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT . . . 1-10
1.5. LIMITATIONS OF THE METHODOLOGY 1-11
2. DEFINITION OF DISPOSAL PRACTICES 2-1
3. IDENTIFICATION OF KEY PATHWAYS 3-1
3.1. GROUNDWATER INFILTRATION 3-1
3.2. SURFACE RUNOFF .............. 3-5
3.3. PARTICULATE SUSPENSION 3-6
3.4. VOLATILIZATION 3-7
3.5. SUMMARY. 3-9
4. METHODOLOGY FOR GROUNDWATER CONTAMINATION PATHWAY ... 4-1
4.1. OVERVIEW OF THE METHOD . . 4-1
4.2. ASSUMPTIONS 4-6
4.3. CALCULATIONS 4-6
4.3.1. Source Term ....... 4-6
4.3.2. Unsaturated Zone Transport '. 4-15
4.3.3. Saturated Zone Transport . 4-29
4.3.4. Setting National Criteria . . 4-49
4.4. INPUT PARAMETER REQUIREMENTS ............... 4-52
4.4.1. Fate and Transport: Pathway Data . . 4-52
4.4.2. Fate and Transport: Chemical-Specific Data . . . 4-54
4.4.3. Health Effects Data 4-54
4.5. EXAMPLE CALCULATIONS 4-71
4.5.1. Site-Specific Application 4-72
4.5.2. National Criteria Site-Specific Application . . . 4-83
vi
-------
TABLE OF CONTENTS (cont.)
Page
5. METHODOLOGY FOR PREDICTING THE VAPOR CONTAMINANT PATHWAY 5-1
5.1. OVERVIEW OF THE METHOD 5-1
5.2. ASSUMPTIONS. . 5-3
5.2.1. Vapor Pressure 5-3
5.2.2. Loss Rate . . ... . . . . 5-5
5.2.3. Atmospheric Transport ..... 5-5
5.3. CALCULATIONS 5-6
5.3.1. Tier 1 5-6
5.3.2. Tier 2 5-8
5.3.3. TierS . . . . 5-T2
5,3.4. Procedure 5-15
5.4. INPUT PARAMETER REQUIREMENTS. 5-15
5.4.1. Fate and Transport: Pathway Data 5-15
5.4.2. Fate and Transport: Chemical-Specific Data . . . 5-16
5.4.3. Health Effects Data ........ 5-16
5.5. SITE-SPECIFIC APPLICATION. ................ 5-33
5.5.1. Tier! Calculation. . . . . . . . . 5-33
5.5.2. Tier 2 Calculation 5-38
5.6. NATIONAL CRITERIA SITE-SPECIFIC APPLICATION. . 5-42
6. REFERENCES 6-1
APPENDIX A: COLUMN METHOD FOR DETERMINING RETARDATION FACTOR (RF)
AND DISTRIBUTION COEFFICIENT (Kd) A-l
APPENDIX B: INPUT PARAMETERS FOR CONTAMINANTS OF INTEREST B-l
vii
-------
LIST OF TABLES
No. Title Page
3-1 Frequency of Threshold Wind Speeds for Windy Areas of
the United States 3-8
4-1 Assumptions for the Groundwater Pathway Methodology . .... 4-7
4-2 Relative Characteristics of Equilibrium Solutions and
Unsaturated Flow Modeling ........... 4-15
4-3 Background Inorganic Constituents for MINTEQ Model Runs . . . 4-34
4-4 Contaminant Concentrations Employed in Benchmark
MINTEQ Runs 4-35
4-5 Analytical Solutions of the Advective-Dispersive Equation . . 4-40
4-6 Required Parameters for Solution of the Advective-
Dispersive Equation , 4-45
4-7 Water Ingestion and Body Weight by Age-Sex Group in
the United States 4-58
4-8 Illustrative Categorization of Evidence Based on Animal
and Human Data. ....................... 4-65
4-9 Input Parameters for Example Calculations — Groundwater. . . 4-73
4-10 CHAIN Model Results for the National Criteria Calculation
for Benzene ...... 4-84
4-11 AT123D Model Results for the National Criteria
Calculation for Benzene 4-86
5-1 Assumptions for the Vapor Pathway Methodology ...*.... 5-4
5-2 Parameters Used to Calculate oz 5-13
5-3 Daily Respiratory Volumes for "Reference" Individuals
(Normal Individuals at Typical Activity Levels) and for
Adults with Higher-than-Normal Respiratory Volume or
Higher-than-Normal Activity Levels 5-20
5-4 Illustrative Categorization of Evidence Based on Animal
and Human Data 5-27
5-5 Input Parameters for Example Calculations: Vapor Loss. . . . 5-34
5-6 Supporting Sludge Landfill Characteristics 5-35
viii
-------
No.
LIST OF FIGURES
Title
Page
1-1 Relationship of Risk Assessment Methodology to Other
Components of Regulation Development for Sewage Sludge
Reuse/Disposal Options. ... .'•-. .............. 1-4
3-1 Contamination Migration Pathways for Pit or Wide Trench
Landfills 3-2
3-2 Contamination Migration Pathways for Narrow Trench
Landfills . . . 3-3
3-3 Possible Routes to Human Exposure from Landfilling Sludge . . 3-4
4-1 Logic Flow for Groundwater Pathway Evaluation of
Landfilled Sludges. ..;........... 4-4
4-2 Discretization Between Grid Points. . . . 4-22
4-3 Example MINTEQ Speciation Results for Entry of a
Contaminant into the Saturated Zone for Conditions of
pH = 7.0 and Eh = 1.50 mv . . . . . ... . . 4-37
4-4 Example Graph of the Family of Curves Obtained for the
National Criteria Case . 4-51
4-5 Graph of the Family of Curves for the Benzene National
Criteria Calculations 4-87
5-1 Logic Flow for Vapor Loss Pathway Evaluation of
Landfilled Sludges. . . ..... ... . . .... . . . . . 5-2
IX
-------
LIST OF ABBREVIATIONS
e
e
dt}i
3z~
dH
ax
A
ADI
AWQC
b
B
B/f
Standard deviation of the vertical concentration distance (m)
Density of sludge liquid (kg/8.)
Difference in total head
Elevation difference between grid points
Degradation rate constant (year*1)
Effective porosity (dimensionless)
Effective porosity
Degradation/decay rate parameter (day-*)
Saturated zone degradation rate constant
Unsaturated zone degradation rate constant
Average windspeed (m/sec)
Air entry matric potential
Pressure head at upper grid point
Pressure head at lower grid point
Matric potential
Hydraulic gradient in the vertical direction
Hydraulic gradient (dimensionless)
Landfill area
Acceptable daily intake (mg/kg bw/day)
Ambient water quality criteria
Slope of matric potential and moisture content plot
(dimensionless)
Bulk density of soil (g/cm3)
Soil-to-solution ratio
-------
LIST OF ABBREVIATIONS (cont.)
BI
bw
cdry
CET
CliET
Co
us
Cv
c(X)i
D*
D
ds
Os
°V
D
'w
EC
Background intake of pollutant from a given exposure route
(mg/day)
Bulk density saturated zone
Bulk density unsaturated zone material (kg/ma)
body weight (kg)
Concentration of contaminant in sludge/soil mixture (mg/kg)
Source concentration (mg/fi.)
Solution concentration (M/L3)
Dry weight concentration of contaminant in sludge (mg/kg)
Effects threshold concentration
Contaminant concentration in the liquid (mg/a)
Concentration of i in the solution (mol/ma)
Limiting sludge liquid concentration
Input concentration
Dry weight contaminant concentration
Contaminant concentration exiting the unsaturated zone (mg/9.)
Equilibrium vapor pressure
Concentration of i in air (mass/volume)
Atmospheric concentration (yg/m3)
Molecular diffusion coefficient of a solute in porous medium
Dispersion coefficient (cma/day)
Distance to property boundary (m)
Density of sludge (kg/m3)
Drainage volume (m3/m2-yr)
Density of water (kg/m3)
Environmental concentration
xi
-------
Eh
EP
erfc
ET
exp
f
*
fi
FH
foe
H
H1
HHAG
Hi
hy
l
*
Ki
K
Kd
LIST OF ABBREVIATIONS (cont.)
Oxidation reduction potential
Extraction procedure
Complementary error function
Evapotranspiration losses
Natural logarithm exponential
Soil moisture content (mVm3)
Average moisture content of the unsaturated zone
Harmonic mean moisture content between grid points
Fill height (m)
Fraction of organic carbon content (of soil or sludge)
(dimensionless)
Saturated soil moisture content (m3/m2)
Henry's Law Constant
Modified Henry's Law Constant (dimensionless)
Human Health Assessment Group
Henry's Law Constant for i (atm-m3/mol)
Depth to groundwater (m)
Air inhalation rate (ma/day)
Acceptable chronic pollutant intake rate (mg/day)
Total water ingest ion rate (8,/day)
Hydraulic conductivity of the medium
Harmonic mean hydraulic conductivity between grid points
Hydraulic conductivity as a function of matric potential
Distribution coefficient (8,/kg)
Organic carbon distribution coefficient for the contaminant
(a/kg)
xii
-------
LIST OF ABBREVIATIONS (cont.)
sat
L
L
L
M
MEI
ML
Ms
mv
MW-j
n
N
na
ne
OWRS
P
P
PB-PK
PH
Pi
q
Q
Q
Q
Octanol-water partition coefficient
Saturated soil hydraulic conductivity (m/yr)
Initial moisture content of sludge (kg/kg)
Soil layer
Mixing layer height (m)
Mass of contaminant
Most-exposed individual
Total Teachable mass (g/m2)
Weight of sludge solids (kg/ma)
millivolt ..
Molecular weight of contaminant i
Total porosity of cover soil (ma/cm3)
Dry weight concentration of contaminant in sludge (mg/kg)
Air-filled porosity of cover soil (cm3/cm3)
Effective porosity (cms/cms)
Office of Water Regulations and Standards
Total pressure in the system (atm)
Precipitation (m3/m2-yr)
Physiologically based pharmacokinetic
Acidity
Partial pressure of i above the solution (atm)
Steady-state flux
Allowable long-term average flux (g/mz-sec)
Source/sink strength
Volume of runoff
Flux during the active uncovered period for contaminant i
(g/ma-sec)
xiii
-------
Qf
-------
LIST OF ABBREVIATIONS (cont.)
TBI
TC
TC
tc
TCLP
TCVP
to
TP
TT
TU
TW
V
V
V
Vave
VOA
VW2
VZ
Ws
X
X
Total background intake rate (mg/day)
Travel time for a contaminant through unsaturated zone
Thickness of temporary soil cover
Thickness of cover (m)
Toxicity characteristic leaching procedure
Toxicity characteristic vapor procedure
Pulse duration (pulse input only) (days)
Pulse time (years)
Total travel time across all layers of unsaturated zone (years)
Steady-state travel time across an unsaturated zone soil layer
(years)
Travel time for water (years)
Interstitial pore-water velocity (cm/day)
Vertical term for transport (dimensionless)
Average interstitial pore-water velocity in the x direction
Average velocity across the unsaturated zone (m/year)
Volatile organic aromatics
Volume of water present in the fill (m3/m2)
Volume of water present in the fill after it drains
(ma/ma)
Seepage velocity in the vertical direction
Water content of sludge (kg/kg)
Leachate concentration (kg/ma)
Contaminant concentration in leachate (mg/9.)
Peak output concentration (at the property boundary)
Initial leachate concentration (below the landfill)
Initial concentration of X
xv
-------
ZHE
LIST OF ABBREVIATIONS (cont.)
Length or width of source (m)
Lateral virtual distance (m)
Mole fraction of i in the gas phase (dimensionless)
Zero-headspace extraction
xvi
-------
V. INTRODUCTION AND DESCRIPTION OF GENERAL METHODOLOGIC APPROACH
1.1. PURPOSE AND SCOPE
This is one of a series of reports that present methodologies for
assessing the potential risks to humans or other organisms from management
practices for the disposal or reuse of municipal sewage sludge. The
management practices addressed by this series include land application
practices, distribution and marketing programs, landfilling, incineration
and ocean disposal. In particular, these reports deal with methods for
evaluating potential health and environmental risks from toxic chemicals
that may be present in sludge. This document addresses risks from chemicals
associated with municipal sludge landfilling.
These proposed risk assessment procedures are designed as tools to
assist in the development of regulations for sludge management practices.
The procedures are structured to allow calculation of technical criteria for
sludge disposal/reuse options based on the potential for adverse health or
environmental impacts. The criteria may address management practices (such
as site design or process control specifications), limits on sludge disposal
rates or limits on toxic chemical concentrations in the sludge.
The methods for criteria derivation presented in this report are
intended to be used by the U.S. EPA Office of Water Regulations and Stand-
ards (OWRS) to develop technical criteria for toxic chemicals in sludge.
The present document focuses primarily on methods for the development of
nationally applicable criteria by OWRS. It is suggested that a user-
oriented manual based on these methods be developed for wider use in
deriving site-specific criteria for these sludge management practices.
Additional uses for the methodology may exist, such as developing guidelines
for local authorities for the selection of sludge management options,
1-1
-------
but these uses are not the focus of these documents and will not be
discussed.
These documents do not address health risks resulting from the presence
of pathogenic organisms in sludge. The U.S. EPA will examine pathogenic
risks in a separate risk assessment effort. These documents also do not
address potential risks associated with the treatment, handling or storage
of sludge; transportation to the point of reuse or disposal; or accidental
release.
1.2. DEFINITION AND COMPONENTS OF RISK ASSESSMENT
The National Research Council (NRC, 1983) defines risk, assessment as
"the characterization of the potential adverse health effects of human
exposures to environmental hazards." In this document, the MRC'.s definition
is expanded to include effects of exposures of other organisms as well. By
contrast, risk management is defined as "the process of evaluating
alternative regulatory actions and selecting among them," through
consideration of costs, available technology and other nonrisk factors.
The NRC further defines four components of risk assessment: (1) Hazard
identification is defined as "the process of determining whether exposure to
an agent can cause an increase in the incidence of a health condition." (2)
Dose-response assessment is "the process of characterizing the relation
between the dose of an agent ... and the incidence of [the] adverse health
effect " (3) Exposure assessment is "the process of measuring or
estimating the intensity, frequency and duration of ... exposures to an
agent currently present or of estimating hypothetical exposures that might
arise " (4) Risk characterization is "performed by combining the
exposure and dose-response assessments" to estimate the likelihood of an
effect (NRC, 1983). The U.S. EPA has broadened the definitions of hazard
1-2
-------
identification and dose-response assessment to include the nature and
severity of the toxic effect in addition to the incidence.
Figure 1-1 shows how these components are included in the development of
these risk assessment methodologies for sludge management practices. The
figure further shows how each methodology may be used to develop technical
criteria, and how these criteria could be used or modified by the risk
manager to develop regulations and permits.
1.3. RISK ASSESSMENT IN THE METHODOLOGY DEVELOPMENT PROCESS
As illustrated in Figure 1-1, the methodology development process begins
by defining the management practice. Even within a given reuse/disposal
option, real-world practices are '"highly variable, and so a tractable
definition must be given as a starting point. As a general rule, this
definition should include the types : of practices most frequently used. That
is, the definition shou-ld not be limited to ideal engineering practice, but
also need not include practices judged to be poor or substandard, unless the
latter are widely used. This definition, presented in Chapter 2 of this
document, helps to determine the limits of applicability of the methodology
and the exposure pathways that may be of concern. However, as also shown in
Figure 1-1 and as discussed in Section 1.4., this definition could be
modified as the methodology is applied because the methodology itself will
help to define acceptable practice.
1.3.1. Exposure Assessment. The exposure assessment step begins with the
identification of pathways of potential exposure. Exposure pathways are
migration routes of chemicals from, or within, the disposal/reuse site to a
target organism. For those pathways where humans are the target of concern,
special consideration is given to individual attributes that influence
exposure potential. Individuals will differ widely in consumption and
1-3
-------
FIGURE 1-1
Relationship of Risk Assessment Methodology to Other Components of
Regulation Development for Sewage Sludge Reuse/Disposal Options
1-4
-------
FIGURE 1-1 (cont.)
1-5
-------
contact patterns relative to contaminated media, and therefore will also
vary widely in their degree of exposure.
An ideal way to assess human exposure is to define the full spectrum of
potential levels of exposure and the number of individuals at each level,
thus quantifying the exposure distribution profile for a given exposure
pathway. The methodologies described in these reports will not attempt to
define exposure distributions in most cases, for the following reasons.
First, it is very difficult to estimate the total distribution of exposures,
because to do so requires knowledge of the distributions of each of the
numerous parameters involved in the exposure calculations and also requires
the modeling of actual or hypothetical population distributions and habits
in the vicinity of disposal sites. Such a task exceeds the scope of the
present methodology development effort.
Second, while knowledge of the total exposure distribution may be useful
for certain types of decision-making, it is not necessarily required for
establishing criteria to protect human health and the environment. If
criteria are set so as to be reasonably protective of all individuals,
including those at greatest risk, then as long as the risk assessment proce-
dures can reasonably estimate; the risk to these individuals, the quantifica-
tion of lesser risks experienced by other individuals is not required.
The drawback, however, of examining only a maximal-exposure situation is
that the true likelihood of such a situation occurring may be quite small.
The compounding of worst-case assumptions may lead to improbable results.
Therefore, the key to effective use of this methodology is a careful and
systematic examination of the effects of varying each of the input param-
eters, using estimates of central tendency and upper-limit values in order
to gain an appreciation for the variability of the result.
1-6
-------
Therefore, exposure will be determined for a most-exposed individual, or
MEI.* The definition of the MEI will vary with each human exposure
pathway. Chapter 3 of this document will enumerate the exposure pathways
and will define the MEI in qualitative terms; for example, for the
groundwater pathway, the MEI is a person receiving all of his or her
drinking water from an affected well at a landfill property boundary. The
MEI will not be quantitatively defined in this chapter, but relevant
information that allows the user to do so, such as available data on the
ranges of drinking-water consumption rates, will be provided in later
chapters. For exposure pathways concerning organisms other than humans, the
term MEI is not applied, but conservative assumptions are still made
regarding the degree of exposure. The remaining chapters (Chapters 4 and 5
in this document) explain the calculation methods and data requirements for
conducting the risk assessments for each pathway.
1.3.2. Hazard Identification and Dose-Response Assessment. To determine
the allowable exposure level for a given contaminant, the hazard identifica-
tion and dose-response assessment steps must be carried out. For human
health effects, these procedures already are fairly well established in the
Agency, although they still require improvement and specific assessments for
many chemicals remain problematic. Hazard identification in this case
consists first of all in determining whether or not a chemical should be
*The definition of the MEI does not include workers exposed in the produc-
tion, treatment, handling or transportation of sludge. This methodology is
geared toward the protection of the general public and the environment. It
is assumed that workers can be required to use special measures or equipment
to minimize their exposure to sludge-borne contaminants. Agricultural
workers, however, might best be considered members of the general public
since the use of sludge may not be integral to their occupation.
1-7
-------
treated as a human carcinogen. Procedures for weighing evidence of carcino-
genicity have been published in the U.S. EPA (1986a) and are further
discussed in later sections of this document. If treated as carcinogenic,
dose-response assessment would then consist of the use of Agency-accepted
potency values. If none are available, cancer risk estimation procedures
published by the Agency (U.S. EPA, 1986a) would be used to determine potency.
If not carcinogenic, hazard identification and dose-response assessment
normally consist of identifying the critical systemic effect, which is the
adverse effect occurring at the lowest dose, and the reference dose (RfD),
which is "the daily exposure ... that is likely to be without appreciable
risk of deleterious effects during a lifetime" (U.S. EPA, 1988). Further
description and procedures for deriving RfDs are found in U.S. EPA (1988).
For certain disposal options, effects on other organisms are of concern.
In these cases, existing Agency methodologies have been used where avail-
able. For example, existing guidelines for deriving ambient water quality
criteria (AWQC) (U.S. EPA, 1984d) are used to determine levels for aquatic
life protection. Where effects on terrestrial species are of concern, there
are no existing Agency guidelines, but suggested procedures for identifying
adverse effects (hazard identification) and threshold levels (dose-response
assessment) are provided.
1.3.3. Risk Characterization. Risk characterization consists of combin-
ing the exposure and dose-response assessment procedures to derive criteria.
Risk assessments ordinarily proceed from source to receptor. That is, the
source, or disposal/reuse practice, is first characterized and contaminant
movement away from the source is then modeled to estimate the degree of
exposure to the receptor, or MEI. Health effects for humans or other
1-8
-------
organisms are then predicted based on the estimated exposure. The calcula-
tion of criteria, however, involves a reversal of this process. That is, an
allowable exposure, or an exposure that is not necessarily allowable but
corresponds to a given level of risk, is defined based on health effects
data, as specified above. Based on this exposure level, the transport
calculations are either operated in reverse or performed iteratively to
determine the corresponding source definition. In this case, the resulting
source definition is a combination of management practices and sludge
characteristics, which together constitute the criteria. These steps are
carried out on a chemical-by-chemical basis, and criteria values are derived
for each chemical assessed and each exposure pathway. An example illustrat-
ing how these calculations may be carried out is provided in this document
for each pathway assessed. However, as indicated in Figure 1-1, the compi-
lation of data on specific chemicals to be used as inputs to the methodology
is a process separate from methodology development. Health effects data for
individual chemicals must be collected from the scientific literature. In
many cases, the U.S. EPA has already published approved values for cancer
potency or RfD. Data pertinent to a chemical's fate and transport
characteristics, such as solubility, partition coefficient, bioconcentration
factor or environmental half-life, must also be selected from the
literature. In some cases, data for particular health or fate parameters
were gathered for a variety of chemicals in the process of developing the
methodology. Where this was done, the information may appear as an
appendix. In most cases, however, such information does not appear in the
methodology documents and must be gathered as a separate effort.
Once these data have been selected, even on a preliminary basis, it may
be useful to carry out a rough screening exercise, using these data plus
1-9
-------
information on occurrence in sludges, to set priorities for risk character-
ization. Screening could reveal that certain pollutants are unlikely to
pose any risk, or that data gaps exist that preclude more detailed charac-
terization of risk. Methods for carrying out such a screening procedure
will not be discussed in this document.
Following chemical-specific data selection, risk characterization or
criteria derivation may be conducted. The values derived as limits on
sludge concentration or disposal rate, together with the management practice
definitions, will constitute the criteria. When calculating the numerical
limits, it is advisable to vary each of the input values used over its
typical or plausible range to determine the sensitivity of the result to the
value selected. Sensitivity analysis helps to give a more complete picture
of the potential variability surrounding the result.
1.4. POTENTIAL USES OF THE METHODOLOGY IN RISK MANAGEMENT
The results of the risk characterization step can then be used as inputs
for the risk management process, as shown in Part II of Figure 1-1.
Although this document does not specify how risk management should be con-
ducted, some potential further uses of the methodology in the risk manage-
ment process are briefly described here. These optional steps are shown as
dashed lines in Figure 1-1.
As suggested by NRC (1983), a risk manager may evalute the feasibility
of a set of criteria values based on consideration of costs, available
technology and other nonrisk factors. If it is felt that certain chemical
concentrations specified by the calculations would be too difficult or
costly to achieve, the management practice definition could be modified by
imposing controls or restrictions. For example, requirement of a greater
unsaturated zone thickness beneath a landfill could result in higher
1-10
-------
permissible sludge concentrations for some pollutants. The same degree of
protection would still be achieved.
Following promulgation of the criteria, it may also be possible to
evaluate sludge reuse or disposal practices on a site-specific basis, using
locally applicable data to rerun the criteria calculations. Criteria could
then be varied to reflect local conditions. Thus, the methodology can be
used as a tool for the risk manager to develop and fine-tune the criteria.
1.5. LIMITATIONS OF THE METHODOLOGY
Limitations of the calculation methods for each pathway are given in the
text and in tabular form in the chapters where calculation methods are
presented. However, certain limitations common to all of the methods are
stated here.
Municipal sludges are highly variable mixtures of residuals and
by-products of the wastewater treatment process. Chemical interactions
could affect the fate, transport and toxicity of individual components, and
risk from the whole mixture may be greater than that of any single compo-
nent. At present, these methodologies treat each chemical as though acting
in isolation from all the others. It should be noted that U.S. EPA's
mixture risk assessment guidelines (U.S. EPA, 1986b) caution that a great
deal of dose-response information is required before a risk assessment could
be quantitatively modified to account for toxic interactions. Future
revisions to these documents to include consideration of interactions will
most likely be limited to qualitative discussion of such interactions.
Transformation of chemicals occurring during the disposal practice,
including during combustion, or following release may result in exposure to
chemicals other than those originally found in the sludge. In many cases,
these assessment procedures may not adequately characterize risks from these
transformation products.
1-11
-------
In addition, these methodologies compartmentalize risks according to
separate exposure pathways. The use of an MEI approach, which focuses on
the most highly exposed individuals for each pathway, reduces the likelihood
that any single individual would receive such exposures by more than one
pathway simultaneously, and therefore the addition of doses or risks across
pathways is not usually recommended. However, it is possible that risk
could be underestimated in a small number of instances.
Finally, the methodologies look at exposed organisms in isolation.
Population-level or ecosystem-level effects that could result from a reuse
or disposal practice might not be predictable by this approach.
1-12
-------
2. DEFINITION OF DISPOSAL PRACTICES
In order to develop a risk assessment methodology for the landfilling of
municipal wastewater sludges, the following general management practices are
assumed.
For the present regulation, no codisposal practices will be considered.
The U.S. EPA's Office of Solid Waste has initiated an effort to develop a
risk assessment methodology for the codisposal of municipal sludge and
municipal solid waste. While Section 405(d) of the Clean Water Act requires
the U.S. EPA to regulate the disposal of sludge, including landfilling, the
Resource Conservation and Recovery Act (RCRA) requires the U.S. EPA to
regulate landfills, including those where nonhazardous material such as
sludge is codisposed with solid refuse (U.S. EPA, 1980a). Since 95% of
codisposed material is solid refuse on a weight basis, the Office of Solid
Waste will be regulating codisposed sludge under Subtitle D of RCRA. Thus,
this document assumes only monofills. The sludges may include digested or
undigested solids from primary, secondary or tertiary treatment processes.
Although undigested sludges are allowed to be landfilled, it is strongly
encouraged to use digested sludges because of esthetics and potential health
problems.
It is assumed that deposition occurs in a recessed trench or pit, or
that the working face is surrounded by surface drainage control ditches to
divert runon and capture any contained runoff. No flexible membrane liners
are assumed to be installed in the landfills. Clay liners may be present.
Although in some cases cover may be applied at greater intervals, cover is
assumed to be applied daily and to consist of excavated soils from the
trenches on site.
2-1
-------
The addition of bulk to the sludge is not assumed. Therefore, it is
assumed that sludges will contain 20-40% solids. This is necessary to
support a soil cover. Since a relatively flat site could pond, and an
excessively steep site could erode and create operational difficulties,
sludge landfilling is usually limited to areas that have slopes; >l/6 and <18#.
In addition to these assumed general management practices,, the following
practices are assumed for specific types of landfills.
In narrow trenches, sludge is ass.umed to be disposed in a single appli-
cation with a single layer of soil applied. Excavation is accomplished by
equipment based on solid ground adjacent to the trench, and the equipment
does not enter the excavation. Excavated material is usually immediately
i
applied as cover over an adjacent sludge-filled trench. Occasionally it is
stockpiled alongside the trench from which it was excavated for subsequent
application as cover over that trench.
Wide trenches are usually excavated by equipment operating inside the
trench. Excavated material is stockpiled on solid ground adjacent to the
trench from which it was excavated. Occasionally, however, it is imme-
diately applied as cover over an adjacent sludge-filled trench. Cover
thickness varies with the solids content and manner for covering. Cover is
applied either by equipment based on solid, undisturbed ground adjacent to
the trench or by equipment that is supported in the trench and that moves
over the sludge.
Area fills have an open face on one side that may be subject to surface
runoff. Drainage ditches are required in the downflow direction to contain
any runoff. The water collected in the drainage ditches will either
percolate into the soil or be routed to treatment. Treatment may consist of
a settling pond with subsequent discharge.
2-2
-------
When sludge is mixed with soil, it may be used as either a temporary or
a final cover, or both. When mixtures are applied as final cover, the
practice is most closely related to land application and should be evaluated
as such. A methodology has been developed to evaluate runoff from land
application of sludges.
While operating practices vary considerably, general practices can be
characterized by the following:
o While 72% of the states require or can require installation of
liners at landfill sites, most either do not have them or use
soil-based liners with a measurable permeability.
o Cover is applied daily (often twice daily) and consists of
excavated soils unless sludge/soil mixing is practiced. If the
final cover uses a sludge/soil mixture, the site should be
evaluated as a land application facility with respect to surface
runoff.
o Deposition occurs in a recessed trench or pit, or the working
face is surrounded by surface drainage control ditches to
capture any contaminated runoff.
o Most sludges will contain 20-40% solids.
2-3
-------
-------
3. IDENTIFICATION OF KEY PATHWAYS
Based on current design and operating practice considerations, the
potential routes of offsite migration can be summarized as those depicted in
Figures 3-1 and 3-2:
o Vapor loss from fill material migrating from the uncovered
working face and/or through the cover material and then being
dispersed in the atmosphere;
o Suspension of contaminated particles from the working face with
subsequent transport downwind;
o Dissolution of contaminants and/or carriage of contaminated
particles in surface runoff from the working face to nearby
surface waters (This pathway is not relevant to trench mono-
fills, since the sludge is emplaced below the surface in an
enclosed trench.);
o Infiltration of water and drainage of sludge moisture trans-
porting dissolved contaminants to the underlying aquifer.
The first two pathways threaten human health through intake of contami-
nated air either by onsite workers or downwind residents. The second two
pathways primarily affect human health through contamination of drinking
water, but may also be of concern as a result of use of contaminated water
on food crops and livestock with subsequent concentration in the food chain
as depicted in Figure 3-3.
3.1. 6ROUNDWATER INFILTRATION
Of the potential pathways, infiltration to groundwater and subsequent
uptake in drinking water is considered the most significant. That determi-
nation is based both on the likelihood and the consequences of occurrence.
All landfills receiving recharge will eventually allow for the generation of
leachate and subsequent downward migration to groundwater. Drinking-water
concentrations of pollutants established to protect human health are suffi-
ciently low that they will be breached before water would pose a threat
3-1
-------
CO
S "*
ro
i_
O>
3-2
-------
Vapor Loss
Leachate
•MHMMMM
Water Table
FIGURE 3-2
Contamination Migration Pathways for Narrow Trench Landfills
3-3
-------
Airborne
Pollution
Vapor
Particulates
Human
Exposure
Landfilled
Sludge
Dissolved
Particulates
Surface
Runoff
Dissolved in
Leachate
Unsaturated
Soil
Saturated
Groundwater
Recharge
Discharge
Dissolved and
Attached to
Suspended Matter
Surface
Water
Body
Drinking
Water
Irrigation/
Livestock
Water
Crop/Livestock
Consumption
Withdrawal
for Use
FIGURE 3-3
Possible Routes to Human Exposure from Landfming Sludge
3-4
-------
through uptake in food-chain crops. The groundwater might be used to
irrigate crops, but the sparse literature that exists on toxics uptake by
plants suggests that the threat is minimal (Kowal, 1985). Therefore, risk
evaluations based on drinking-water concerns will result in the most
restrictive sludge concentration criteria for the groundwater pathway.
Another possible exposure route is from edible aquatic organisms living
in surface water recharged by contaminated groundwater. This is considered
as a supplementary groundwater pathway.
3.2. SURFACE RUNOFF
Contaminant migration in surface runoff may result from dissolution into
the water or suspension of particulates to which the contaminant is
attached. In either case, transport requires physical contact between the
contaminant and the runoff. Therefore, contamination must be present at the
soil's surface for migration to proceed by this pathway. Since clean soils
are used for cover, except for the case of sludge/soil mixtures that
constitute land application, the working face is the only significant source
area for contaminated runoff. As noted earlier, operating procedures
require control of runon and runoff from the working face with drainage
ditches. In addition, since the working face is below grade for the
surrounding areas, all trench or pit fills will contain runoff by design.
This would not be the case for area or canyon fills, but these fills must
include provisions to contain drainage in the down-gradient direction.
Based on the assumption of good operating practices, runoff becomes a part
of the groundwater pathway or is eliminated. The precipitation that runs
off the working face will collect at the foot or in a drainage control ditch
where it will either percolate into the soil, be used for dust control or be
routed to treatment. In the first two cases, the runoff becomes a
3-5
-------
part of the groundwater pathway. In the third case, the pathway is
terminated. Therefore, the methodology does not consider an independent
surface runoff pathway.
Because good management practices will prevent the runoff pathway from
being a significant route by which toxic contaminants threaten human health
from sludge landfills, regulations to control this pathway are best focused
on requiring those practices rather than on establishing concentration
criteria. The necessary practices consist of two basic elements: (1)
diversion berms and/or ditches to redirect all runon from upflow areas away
from the fill area, and (2) berms and/or ditches at the foot of the fill .to
collect runoff from the fill area in general, and from the face in
particular. These berms/ditches should be capable of containing the
estimated volume of runoff from the 100-year, 24-hour design storm.
3.3. PARTICULATE SUSPENSION
The particulate suspension pathway is similar to that for surface runoff
in that it requires the contaminant-bearing particulates to be at the
surface where the wind and/or human activity will disturb it. The working
face is the only location where this will occur to any significant extent.
With daily application of cover, the face itself will not be exposed for
more than 8-12 hours in any given 24-hour period. In addition, suspension
will occur only when wind scour velocity exceeds a threshold value or with
mechanical agitation. For soils, the scour threshold has been reported as
6-13 m/sec (Gillette, 1973). Most sludges would be expected to be on the
high end of that scale or above the scale because of their moisture content
and tendency to mat as they dry. Composted or dried sludges, however, may
be very light and fine in texture and, therefore, easily resuspended.
3-6
-------
While each landfill site will have its own distinct characteristic wind
pattern and velocity distribution, a review of data from specific sites
gives some perspective on the frequency that the threshold windspeeds will
be exceeded. Table 3-1 provides data on the percent of the time that wind
velocities will exceed 12 m/sec at candidate wind-generation sites. Since
these sites were selected for wind-power potential, they represent the high
end of the scale. In no case did 12-m/sec winds occur for >5% of the time
at a 9-m height. Wind speeds diminish rapidly with proximity to the earth.
Therefore, 13-m/sec speeds at a landfill working face would occur even less
frequently. Thus, wind data coupled wit/) the operating times of 50% or less
?
without cover suggest that for windy sites, the winds will attain speeds
capable of suspending sludge from the working face for brief periods of
time. This will be augmented by mechanical agitation at times. In the
main, particulate suspension will be episodic rather than chronic with
regard to landfilled sludges.
Because particulate resuspension may occur under a limited set of
conditions, it is "best regulated through management practices rather than
concentration criteria. In particular, resuspension shall be controlled by
requiring placement of daily cover over landfilled sludges. Cover may
consist of clean soils or a mixture of sludge and soil at a depth of at
least 15 cm (6 In),,
3.4. VOLATILIZATION
Vapor loss from sludge may result from volatilization from the uncovered
working face, or release from within the fill and subsequent migration
through the sojj cover. The degree to which volatilization will occur
* '"
depends both of the physical properties of the contaminant (e.g., vapor
3-7
-------
TABLE 3-1
Frequency of Threshold Windspeeds for Windy Areas of the United States*
Location
Fraction of Time
Wind Exceeds 12 m/sec (%)
Amarillo, TX
Block Island, RI
Boardman, OR
Boone, NC
Clayton, NM
Cold Bay, AK
Culebra, PR
Holyoke, MA
Huron, SD
Kingsley Dam, NE
Ludington, MI
Montauk Point, NY
Point Arena, CA
Russell, KS
San Goronio, CA
1.27
0.11
0.45
4.98
1.73
4.56
1.54
0.24
0.27
0.29
1.24
3.51
0.34
0.84
3.31
*Source: Adapted from Sandusky and Renne, 1981
3-8
-------
pressure and solubility) and the nature of the sludge matrix. A strong
affinity between sludge and contaminant can bind otherwise volatile contami-
nants and reduce losses significantly. These effects are difficult to
predict a priori. Analytical methods are available to predict volatiliza-
tion from soils, but they do not account for the interactions that would
occur in a sludge. In general, most sludges will be subjected to thermal
and mechanical action that will facilitate volatilization prior to deposi-
tion in a landfill. Subsequent release of volatile residuals, however,
could occur if degradation of the sludge changed the matrix sufficiently to
alter sludge contaminant interactions. The uncertainty in the rate and end
products of sludge degradation in a landfill, thus, further frustrates
attempts to predict vapor losses.
While placement of daily cover over sludge will reduce flux rates,
preliminary calculations have revealed that vapor concentrations above
reference air concentrations (see Section 5.4.3.) can be observed with
sludges as a result of losses from the landfill working face prior to
application of cover. Therefore, proper management through application of
cover soils will not be adequate to control potential vapor problems, and
concentration criteria are also required. A methodology is provided to
predict vapor concentrations at the site boundary over extended periods of
time in order to determine concentration criteria for volatiles.
3.5. SUMMARY
From the above considerations, it is concluded that good management
practices through properly enforced regulations will control health problems
stemming from contaminant transport in runoff and resuspended particulates
in the atmosphere. Similar regulatory controls will not eliminate potential
contaminant losses through the groundwater and vapor pathways. Therefore,
contaminant concentration criteria are required to prevent infiltration and
3-9
-------
vapor losses from leading to conditions that exceed reference water and air
concentrations, respectively. The following chapters describe the method-
ologies developed to select those criteria and quantify concentrations
associated with placement of a given sludge in a designated landfill.
In all cases, use of the model and good management practices cannot
guarantee that environmentally significant releases will not occur. As a
consequence, a comprehensive monitoring program should be implemented with
any sludge disposal alternative. In the case of landfills, this would
consist of monitoring wells to detect groundwater contamination from
infiltration.
3-10
-------
4. METHODOLOGY FOR GROUNDWATER CONTAMINATION PATHWAY
4.1. OVERVIEW OF THE METHOD
As noted previously, the methodologies described herein are designed to
quantify risks associated with disposal of sludges in landfills. It has
been determined that the generation of leachate with subsequent migration to
and contamination of groundwater is a pathway of concern. It has also been
determined that the approach will be based on a risk assessment methodology
that can be applied directly to input data on a given site. The merits of
the proposed disposal activity will then be weighed on the basis of pre-
dicted risks to human health through drinking water. A tiered approach is
offered, beginning with simple comparisons to national criteria and going to
a more site-specific approach for contaminants in excess of the national
criteria. The second tier allows introduction of site-specific values to
reflect the conditions at the chosen site. Contaminants are considered
individually in sequence. If a contaminant is not present, it is deleted
and the analysis goes to the next contaminant. If a contaminant passes
through below criteria, it is dropped and the next contaminant considered.
To implement such a methodology, it is necessary to simulate the
movement of contaminants from the fill area through the unsaturated soil
column to the aquifer and then through the saturated zone laterally out from
the site. For compliance, the property boundary may be selected as the
point of compliance, since drinking-water wells could be constructed from
that point on and could then be affected by contamination with subsequent
public health implications. In no case is the compliance point allowed to
be set at a distance greater than 150 m from the landfill. Data on health
4-1
-------
effects, i.e., risk reference doses (RfDs) for noncarcinogens or potency
values for carcinogens, are used to evaluate allowable levels for
groundwater contamination. The premise is that a potable water supply must
be maintained at healthful levels for potential future uses even if there
are no current uses immediately off site.
The tiered approach begins with a comparison between measured chemical
concentrations in sludge and criteria generated for a reasonable worst case
landfill. Environmental setting parameters include six found to be
particularly influential on water quality. The set values for five of these
six parameters are as follows:
Depth to groundwater 1 m
Soil type Sand
Recharge 0.5 m/yr
Eh - oxidation potential +500 mv
pH - acidity 6.0
The value of the sixth parameter, partition coefficient, varies
according to the chemical evaluated. The criteria are calculated using the
model described herein operated to determine the sludge chemical
concentration that would raise groundwater concentrations to the reference
dose, but not above it.
If an operator determines that a given sludge contains chemicals
exceeding the criteria used for Tier I, the operator may calculate
site-specific criteria by inserting measured values for the parameters
listed above and rerunning the methodology.
4-2
-------
In either Tier I or Tier 2, the distance to the compliance point is set
on the basis of the classification of the groundwater at the disposal site.
If the landfill is underlain by a Class I groundwater, the compliance point
is set at the point of entry, i.e., no lateral movement in the aquifer is
considered. If the groundwater is Class II or III, the compliance point is
set at the smaller of the two options, the fenceline or 150 m. Hence, the
compliance point distance can never exceed 150 m. If the groundwater is
Class III on the basis of contamination with the toxic chemical of interest,
no groundwater degradation is allowed, and hence, the predicted leachate
concentration must be less than or equal to the current groundwater
concentration.
The tiered process is illustrated in Figure 4-1. To calculate Tier I
criteria, the methodology is applied assuming a^ given sludge concentration.
The resultant groundwater concentration is then compared with the
concentration that would produce a health-based reference dose. The ratio
of the two is used to recalculate until a sludge concentration is picked
that will just produce the limiting groundwater concentration.
The Tier 2 process is initiated by calculating a pulse or leach time
(the period of time required for all of the available contaminant to be
leached from the sludge) for each contaminant. For degradable contaminants,
a calculation is then made to determine how long it will take each contami-
nant to traverse the unsaturated zone and the amount of degradation that
will take place during that time. For nondegradable contaminants, the
concentration is unchanged through the unsaturated zone. For inorganic
contaminants, a speciation model (MINTEQ) is used to estimate the dissolved
concentration of contaminants in the saturated zone after accounting for
geochemical reactions.
4-3
-------
For Each Contaminant
Conduct Leachate
Extraction Test
Health
Criteria
Is
(xi)* > Health
Criteria?
No
Tierl
»«l*- End
Yes
Site
Data
Degradation
Rate
Distribution
Coefficient
Determine Pulse
Time
I
Determine Time of
Travel and Losses in
Unsaturated Zone
Inorganic
Contaminant
Geochemlcal
Considerations
I
Determine Initial
Concentration in
Aquifer
Model Input
Parameters
Determine Concentration
at Property Boundary
*(xl) = concentration of contaminant i
Tier3
Experimentally Determine
Attenuation Values
Yes
Tier 2
End
FIGURE 4-1
Structure of Tiered Methodology
4-4
-------
At this point, after accounting for degradation and geochemical
reactions, a comparison is made between predicted leachate concentration
entering the aquifer and the health criteria. The analysis is continued for
those contaminants exceeding the health criteria.
An analytical contaminant transport model, CHAIN, is used to predict
contaminant concentration at the base of the unsaturated zone. The
difference between the model and the unsaturated zone calculations discussed
above is that the model allows for dispersion as well as degradation. For
the metals (nondegradable), the output concentrations from the model are
adjusted based on geochemical reactions (MINTEQ). At this point, the model-
predicted or MINTEQ-adjusted contaminant concentrations at the base of the
unsaturated zone (point of entry into the aquifer) are compared to the
health criteria. The analysis is continued for those that do not pass.
The final step of the analysis is to use a saturated zone transport
model, AT123D, to predict contaminant concentration at the property
boundary. These final contaminant concentrations at the property boundary
are added to the background concentrations in the groundwater and again
compared to the health criteria. If they all pass the criteria, the
application would be accepted. If any one contaminant exceeds the criteria,
the application would be denied. If any contaminants exceed the criteria
and the analysis has been completed, then landfill disposal is not available
for the sludge unless the chemical levels in the sludge are reduced. The
procedures and details of each module in the methodology are described in
the following sections. The methodology for calculating the contamination
pathway from the groundwater to surface water to edible aquatic organisms, a
supplementary pathway, is also very briefly described (Section 4.3.3.5.).
4-5
-------
4.2. ASSUMPTIONS
In order to apply a methodology such as that presented here, it is
necessary to make simplifying assumptions. The assumptions, stated or
implied, required to implement the groundwater pathway analysis methodology
are outlined in Table 4-1.
4.3. CALCULATIONS
4.3.1. Source Term. The sludge itself is the starting point for a
contaminant's migration through the landfill and into the groundwater.
Therefore, the methodology must start with the sludge. This is most simply
done by assuming that the total mass of a contaminant is in dissolved form.
However, such an approach is extremely conservative. Chemical-physical
interactions between the organic matrix of sludge and contaminants are often
quite strong and may cause immobilization of pollutants. As a consequence,
not all of the contaminants in sludge are mobile. The availability of
organic contaminants is often related to their concentration, so that only a
fraction is mobilized at any one time. The latter is a partitioning
phenomenon that controls leachate levels to a discrete ratio between
concentrations in the sludge and the leachate. Current understanding of
these phenomena and the effects of various constituents on them is limited,
so it is not possible to accurately predict leachate levels through the use
of models at this time. Inorganic contaminants may have their leachate
concentrations dictated by solubility constraints.
If an applicant employs total sludge contaminant data and determines
that criteria will be exceeded, an estimate of leachate quality can be
derived.
4-6
-------
en
o
o
•o
o
x:
•P
cu
re
a.
4.
cu
-p
§
•a
c
3
O
CD
CO
4=
•P
4.
O
u-
on
c:
O
•p
a.
3
on
on
C
o
I *
re
u
*i
•r-
£
re
c
o
•p
ex
E
3
on
'i
|
/
/'
/
i
; re
CO
4-
'
^~
re
: c:
o
•r—
•P
U
C
3
U_
•
U
•r-
y-
u c
CO CO
cxx:
on 3
CO CO
cn-p
•o re
3 4-
i — 3
on o
u
co re
>}
-a 4-
r— CO
3 >
0
3 co
X)
on
•p -a
O l—
CU 3
u- o
u- x:
UJ CO
i
•r—
•H
C 4.
•r- (0
a
on
C >>
OXl
4J 13
re cu
4. C
^_) .(•».
d £
CU 4.
U CU
c -P
O CO
o -o
<— CO
re x)
u
•r™ ^—
£r-
cu •<-
J=. 3
u
co
0 -P
c x:
re o
en re
4- CO
O i—
CO
1—
CO
u
4.
3
O
00
•
•a
CO
f— »
ex
Q.
re
CO
4-
re
on
CO
3
r—
re
>
u
o
\s
•a
CO
4.
3
00
re
cu
e
CO
j= >>
•P X)
e -o
"~ XI
on >r-
T3 4.
»r~ =5
co "O re
.Q C >
re
en o
c= co o
•r— m\s
cz TO
O 3 CO
••- i — x:
•P +->
CO u-
x: o
-Pf—
<— >>
CU •'r- -P
o 3 "-
C. r—
•i— on t—
on r— XJ
re 3
co o i—
•P-r- 0
re E on
C CO
•r-x: co
EO -P
re
CO CO E
•P E "-
CO O -P
TO on r— •
C 3
•r-n-
O CU
oo .e
•r- CO -P
o
•p c: -p
u co u
CO on CO
U- ID U-
r lj_ i^ l}_
LU ex re
•o
CO
•r- re
c E E
•r- CO "r—
j^* 4«^
on u on
c co
O CO
•r-x: cu
-P-P XJ
re
4- 0 C
-p -P re
(Z U
CO T3
o co x:
c -p o
O-r- T-
o E x:
r~" ^~
(0
0 CO >>
CS •!•»
CU r— > —
f" r™ •!""•
O-r- XI
3 3
O I —
•i- CO O
c: -p on
re re
enx: on
4_ u -
o re i —
c: co re
(-<(— 0
•
4->
on
CO
CO
4^
C
•r-
U-
O
t—
to
U
•f—
E
CO
x:
u
CO
x:
•^
f
to
•p
re
T3
4_}
£Z
re
!~
ex
4«>
C
CO
E
4-*
(0
CU
&~
•4^
Sw.
CO
4"1
re
3
CO
4^
on
re
3
E
O
i_
<4-
>,
r—
XJ
•P -i^-
re on
x: 4_
•P CO
>
•P CU
C 4.
CU 4.
*P T^
X
cu cu
4_
co re
x:
^-» on
4-*
o c
-p re
c
on -i-
-P E
u re
•i- +j
•a c
CO O
4- u
ex
4- CO
CO £
> 0
O I/I
CO
x>
^^
p—
CO
•p
re
•r—
^_>
T—
3
r^
|»Mk
•(— '
3 '
on
-p
C
re
c •
•p- TO
E co
re NI
•P-I-
C i—
o •*—
U XI
3
r— r—
r— O
^ on
O
4*J
^n)
O
re
CO
4. .
r—
^) re
re u
£ »r~
' E
4. CO
o x:
u
on
TO 4-
•r— CO
r— JS
0 -P
on O
C
o re
•p
CO
•o E
e o
3 U
O CO
X) XI
i
C 1
O on
U CO •
4- i—
XI CX-r-
cn re
•i- on -p
x: -P
u c
<4- CO O
O r— •*••*
en-p
c co re
0 C 4_
•I- -p
re TO co
4- O 0
3 •!- C
TO 4- O
CO U
oo ex
•P 3
o c o
•r- O i—
T3 •!-
CO -P U-
4. re o
a. 4.
4- 4-> CO
co c u
> CO C
o u cu
on
re
•o
cu
X)
•r—
s_
u
on
CO
TO
cu
XI
c
re
u
c
4-
cu
-p
^~^ *
. re co
ex >
re
cu 3
•p
re co
x: 4-
u re
re 3
co cr
_l on
•
CO C
0 O
c: »r-
•i— on
on 4-
co
c ex
O on
•r— •!—
•P TO
re
•p "re
C -P
cu c
o o
c: si
0 -t-
0 4_
o
on x:
•p
o on
•i- CO
•0 4-
CO O
4- C
ex en
4— •!—
CO
> -P
O i-
j—
re
o
•r—
-P
s_
CO
^
CO
x:
c:
3
0
<4-
i —
re
c:
O
•i—-
on •
c c
co o
£.r-
•i- 4->
TO 0
1 CO
CO 4_
£= -i-
o -a
-p
4-
0
ex
on
C
re
4-
1—
CO
^z
o
f-J
X3
CO
re
4.
3
•P
re
on
c:
,
on
u
**^>
0 -P
•r— on
U- •r-
•i- 4_
U CO
CO -P
ex u
on re
1 4-
cu re
•p x:
•r- O
on
r—
CO -r-
X) O
on
-o
i — c:
3 O
O
3 en
on •!-
•P -0
u c:
CO CO
*i f~i
<4- CO
1 1 1 -Q
•
CO
re
•p
01
^>^
TO
re
cu
•P
to
on
•r™
3
o
14-
4-
co
•p
re
1
r—
re
u
£=
O
•T—
•P
re
4-
•P
C
CO
u
c:
0
0
•p
o
CO
u-
14-
re •
on
-P c:
O O
•P
on re
CO r-
O 3
0 0
*,]— „
o
E
O
4_>
+J
O
J3
•P
re
on
C
0
-p
•r- • .
T3 CU
c= c=
o o
U NI
CU TO
<— CO
XJ -P
re re
4-> 4-
3
4. +J
co re
-P on
re c:
3.3
. x:
u
•r—
x:
>> 3
XJ
4.
C CO
o > •
•i- O on
•P 4-
re cu 3
4- E 0
•p re o
(= 4_ O
co u-
u co >>
C E r—
O •<- i—
o -p re
3
on cn-p
•P c: o
o M- re
•r- -P
TO re co
co 4. en
4- CO 4_
exi— re
4- co x:
co u u
> O CO
o re 4_
• •
X
3
i —
*)
•P
<•"•
re
-p
<=
o
u
re
on
re
x:
^^
4_
re
c:
3 .
O
XI
4-
co
ex
ex
4-7
-------
CD
X: CD
•P O •
c cu
03 •(- -P
CD O in I
JZ • •!— S-
•P r— CU
V r— CL
O M-
-Q (O TJ
C i-
O >»-P CD
CU O -P •!- tO
+J .r- +J £ «-
re
O
re
(0 +J CD
•P on
CD re o
I/I -P S_
o re
ro
S_ on
.
E o
(/) O
>> c cu
•P CD i-
•r- 4-» Q.
U C
O O >>
r- o re
CD E
> o
•f* C.
•P C O
cz re s~ 3
(/I O O
•p -p «- I—
o ro CD r-
•r- x: s_ re
•a -p CD
cu x: -
s_ -p -p i-
CL C CD •
s- cu - 3 c
CD -P -O O O
> X C r— .1-
O CD ro on to
S_ -i- O
n3
4-> O
O)
o
T3 i
'5 2
CD
•r- 0)
s_ o
CL on CD C
CD S- CD E
O «- 3
-p- 5-
to
03 C
1- O I
01 "- to
'J= •*-> J=
3 u u
c a;
c 3 E
•fJ QJ a; o
CD -p s_
TO (O CU
4J S- i/l
O 3 T- •
Cl>
•r- CP
•a (O
d)
1- +->
CL to
CD CD t3 O) OJ
T3 - 3 •
O) C i— 4-
i_ O -r- CD
CL"- -O <4-
4_ +J T-
CD tO 5- 3
> -r- > s_ o cr
+•><*- (0
CD I
T3 r— O)
CO J=
O 3 4->
•i- CT
+->
I +-> ro
CD "- J=
-l-> -(-> O
•r- c: (O
oo ro a>
3 I —
CD cr
^a CD
CD -C
O C •(->
3 O O
.a •
ui Cn 4-
•(-> C U- CD
O •!- O <4-
0) TO .1-
<4- C >> 3
(4_ CD 4_> Cr
LU CL'l- tO
•H
U
CD
CL
U
o> si
i- a>
u
•r- >»
•P t=
i/) ro
a> o
ro -c
s- -P
ro CL
^Z CD
O XJ
•I- -p
•p I
3 i-
cu
to
•p
c o
tO <*-
S- O
u c:
•r- O •
i — ^"^
3 C I/I
(O T- CD
s- x:
-O T3 O
>> CD 03
•=!£
I— 01 CL
ro on CL
O to tO
S- O >
CD C T-
CD
•P
to
0) >>
S- r—
(/I O
1/1 O
«3 )
CD O
r— S_
tO O
> CL
CT-r-
0) O
on
•r- S- -P
x c:
•r- «- tO
S o c:
CD
en
cu
•p
"3
-o
a> •
•p >>
to i-
3 -P
T3 00
3 'i
O CD
i- x:
CD O
re
CD
•p
re
CD
re
cu
re
u.
a>
•P
re
•p -P
re c
I/) O
10
CD
re
re
CO
4-8
-------
01
c
0
ro
u
y-
•r-
CO
0£
C
O
•r—
4-*
&
E
01
in
^t
ro
CD
**
r—
CO
C
O
+J
o
c
3
U.
1
c: u- r—
O I-H CD
o -o
0
CD . E
E -o
CD CD CD
S- r- x:
4-* CD 4-*
X -0
CD O -
CD O
> CD i—
CO 01
J= O >
-a 4-» CD
i — >
3 TO
O C CD
0 O 4-
>> to
oo CD
CD XJ in
4-> CD
•r- oo 3
00 C 1
O CO
£? 1 ~*
O •!- 3C
00 -O Q.
oo y-
4-> O
CD
r— 4->
Q. CD
3 00
0
U CD
4-*
X: CO
LU 3
1 CT
3= CD
0.-0
to
4.
CD C
4-> to
JO
3 CD
•O T3
, C T- •
3 > 00
0 0 CO
CD Q. "i~
4-*
x -o to
•r- CD C
01 I— 4_
CD CD
CO T3 4->
J= O r—
r- E CO
0
O.
00
C
to
1—
CD
C
o
r-j
CD
4-> ^~^
CO •
4- 4->
3 c:
+•> O
ro u
00 >—
I
C
o
u
x:
LU
14—
•i—
•O
C
ro
.
u
•i—
•o
CD
0.
CD
T3
C
3
i —
'S
1
C
O
i — O
CD
•0 4-3
o c:
E to
cr
CD •<—
J= E
4-> ro
3 =
0 0
r— O
>4->
i- O
> -5
CD •
CO i- 00
4- 0. C
ro i- O
CD *r™
01 > 4->
C O ro
O 4_
4J r^ C
•r- •!- CD
•a 3 o
O •
•i- 13
g— Q^
to •»->
CD 0
4- CD
0 r-
CD
4» oo
O
«s
r^ CD
tO 4->
> 10
x:
JZ U
CD ro
•r- CD
CD CZ
r* »r—
** ^
CDr—
O CD
C >
OOi—
CD
4>J
to
x:
o
ro
CD
r—
^->
C
CD
oo
CD
(•
a.
CD
&_
u
•r-
C
to
CD
0 •
-o c
CD CD
+•> C
o o
CD Q.
CD O
OO 0
^ly
r—
3
O
00
00
.}_>
C
ro
X •
1— +J
O -i—
U JQ
O
ro
•r- -(->
4-> 0
CD T3
•(-> CD
O 4_
Q. Q.
CD CD
•!-> O
•
O
•r-
n
•r—
U
CD
o.
01
1
CD
U
O
00
CD
<")
T3
r—
3
i
00
0
CD
y-
LU
-M
O
c:
o
T3
01
C
o
•I—
4~3 •
0 >>
ro 4.
i_ -|rr*
4-> *r—
c E
•i- CD
x:
4-> 0
c o
CO CD
C CD
•I—
E -4-J
ro o
c: u-
o y-
«_j ro
c
O i—
•r- (O
4-* O
to •(-
4- CD
-H O
C r—
CD O
O •<—
c xi
o
U 4.
0
O i —
•i- CO
T3 0
CO T-
Q. CD
C f—
CD U
>
O CO
11
T3
C
CO
^t
r—
C
O
00
•i —
00
>5
r~
0
t__
•o
^y
f~
JD
•
oo 4-
•^ CO
£= 4_
0 O
•r—
4-^ 4-^
CO 00
•a 4_
CO T-
CD
CO oo
C5 «r-
4-
3
U
O
o
c
CO
u
o
•r—
^M>
ro
•O
ro
CD
CD
-a
^
-°
c y-
o o
•r-
4-> C
CO O
C CO
CD 3
U 1=
C CD
O 4->
0 4->
ro
4-> CD
O C
T3 4-> 00
CO CO U
Q-'i- C
4- E 10
CO -i- CD
> r— 4.
CD CO O
c
CD
14— CO
•i— 4.
3 10
cr
to oo
u
•r- C
ro
01 CD
CO *
•oo
00 ••
CD CD
4- i_
O
CD CD
4-> 4_ .
4-> CO -O
ro x: CD
E 4-> 4->
to
O -3
•i- -O C
C C CO
ro ro 4->
CD 4J
4- CO tO
O T-
•0 4->
O CD O
•z. E £=
CD
*l *f—
•r- 4-*
ro
C C
O CO
4-> •!— O
ro X)
4- 4-> ro
4-> C
C CD 4_
CD oo CD
O CD 4-^
C 4. CO
O O. 3
O T3 •
oo C H
00 -r- 3 i —
4-J O CD
0 4-» 4- >
•r- C CD CD
•o to <—
CO C 4-
4- -i- O T3
Q. E CO
4- CO (/> E
CD 4-» i — 3
•O C -r- 00
C O O 00
rD O on ro
C
•r- CD
CD
-t-> ro 4->
C 4- C
CO CO to
00 > C
CO to -i-
i_ £
Q. • to
00 4->
00 • C
•r- rs o
u
c • o
to x: «r-
•r- •(- CO •
E 3 CD oo
CO 4-.—
C 4- C >
O CO -r- CD
0 4-» r—
OO 4_
0 O -0
•r- O U- C
C 4-> 3
CO i — O
CD E T- 4_
4-. CD 03 CD
O 4-> > Jli
00 CO U
O >> 4_ ro
2: oo Q.X1
(
CD
>> to
•a 3
C CO
O 4_
•r— CO
4-> 3
ro Cr
4_ oo
C ro
CD
0 O
C 4-J
o
U CD
00
VI i —
4-> 3
o a.
•I—
•O CD
CD C
CLr —
4- to
CO 3
> cr
O CO
.
[ ^
c:
CO
4_J
00
C
O
o
oo
c:
o
ro
1 *
c:
CD
0
O
o
3
a.
c:
i— t
4-9
-------
For the purposes of this methodology, leachate concentrations are
estimated differently for organic contaminants and inorganic contaminants.
Organic concentrations in leachate are calculated using a partition
coefficient, K
oc
The value of K is determined as the ratio of the
oc
concentration of the chemical in the sludge to its concentration in the
water associated with the sludge, i.e.:
(4-1)
KOC
concentration of X in sludge organic carbon (mg/kg)
concentration of X in water (mg/8.)
a/kg
The value for K may be measured empirically or may be estimated from
oc
relations for solubility or octanol-water partition coefficients (Lyman et
a!., 1982), either of which may have been measured empirically. In any
case, the relation assumes that organic contaminants will adsorb onto solid
organic matter or organic coatings on solids as the primary mechanism of
retention.
Inorganic concentrations in leachate are assumed to be limited by
solubility constraints. In other words, it is assumed that inorganic
contaminants will desorb and/or dissolve from the sludge until they reach
their maximum solubility. Maximum solubility levels were estimated on the
basis of the maximum effluent or leachate levels reported for U.S. waste-
water treatment plants (U.S. EPA, 1985a).
In addition to determining the concentration of contaminants in leach-
ate, source term characterization is required to estimate the time over
which the contaminant will be present in the leachate, or pulse time.
Sludge does not act as an infinite source of contaminants. There is a
finite mass of contaminant present in the sludge that can be mobilized in
leachate. For some contaminants, that mass is less than the total mass in
4-10
-------
the sludge because of irreversible adsorption or other binding mechanisms.
In either event, the available mass will be released from the sludge over a
discrete pulse of time. It is for that period that concerns over leachate
effects on public health are real and measurable.
To calculate the pulse time, it is necessary to determine total
contaminant levels in the sludge, contaminant concentrations in the leach-
ate, sludge moisture content and recharge rate. These factors are relevant
to pulse time according to:
Q = M * X
(4-2)
Q = RT + (L - S)
(4-3)
where:
Q = volumetric water flow of leachate for the m2 unit area
required for contaminant to be completely leached (m3).
R = recharge or volume of infiltrate entering landfill/
mVyear -(in3/year). Can be calculated as R = P - ET "-
RO, where ET is evapotranspiration losses, RO is runoff and P
is precipitation. If runoff is retained for infiltration, it
should not be subtracted. RO refers only to runoff that is
routed to a treatment plant or otherwise allowed to leave the
site.
S = storage capacity for water in sludge defined as the "dry"
water content for the sludge under normal atmospheric condi-
tions/m2 (m3), i.e., the product of fill height after
drainage, sludge density and moisture content divided by 1000
kg/m3. .
L = water content, of sludge/m2 at time of disposal (m3),
i.e., the product of fill height, sludge density and moisture
content divided by 1000 kg/m3.
T = time of pulse over which all contaminant will be released
from the sludge (years).
M = mass of contaminant contained in a volume of sludge repre-
sented by the height of the sludge in the fill and a square
meter cross section (kg), i.e., M = (height of fill x 1.0
m3) x (density of sludge kg/m3) x (concentration [N]
of contaminant in sludge kg/kg) x (1 - moisture content).
4-11
-------
X = average concentration of contaminant in leachate (kg/ma).
Combining Equations 4-2 and 4-3:
T 0 M - X (L - S)
XR
(4-4)
For degradable contaminants, .the initial mass of contaminant (M) changes
with time. If a first-order decay mechanism is assumed at a degradation
rate \, Equation 4-4 becomes:
T = [In =
XR - \M
(4-5)
where:
T - pulse time (years)
X = degradation rate constant (year"1)
X » leachate concentration (kg/m3)
R = recharge or infiltration rate (ma/year)
H = mass of contaminant in sludge (kg)
Therefore, Equation 4-4 is applied for arsenic, copper, mercury, nickel,
bis(2-ethylhexyl)phthalate, trichloroethylene and any other contaminant
where x = 0. For all other contaminants, Equation 4-5 is applied. In
either case, the result describes the length of the pulse time (T) when
leachate is adding contaminant to the unsaturated zone at the concentration
calculated. Since anaerobic conditions are almost certainly likely to
prevail in both the unsaturated and saturated zones beneath a landfill,
anaerobic degradation rates should be used.
These formulations assume all contaminants in a sludge are ultimately
mobile and that contaminant concentrations remain relatively constant in
leachate until the total mass is virtually depleted. The first assumption
is conservative in that it does not subtract the nonmobile fractions of
4-12
-------
contaminant. The second assumption may not be conservative in that it
converts the actual contaminant concentration/time relation into a square
wave (i.e., a pulse of equal height throughout its duration). Because
actual leachate contaminant concentrations are likely to trail off with
time, the actual pulse time will be longer, but the concentrations will be
smaller. The degree of distortion arising from this assumption will depend
on the nature of the actual concentration/time relation. The square wave is
assumed at the landfill only. Dispersion and retardation are allowed to
transform the pulse once in transit.
In summary, source term characterization consists of two steps:
1. Derive a contaminant leachate concentration: by applying a
partition relation for organic contaminants and a maximum
solubility for inorganic contaminants. ,
2. Calculate a pulse time using Equation 4-4 or 4-5.
4.3.2. Unsaturated Zone Transport, As leachate is generated in the land-
fill, it moves vertically downward through the unsaturated zone to the
uppermost aquifer. To measure the risk to water quality by contaminants in
the leachate, it is necessary to determine the time of travel required to
reach the aquifer and subsequent effects on contaminant concentrations.
Factors affecting contaminant transport in the unsaturated zone include the
physical characteristics of the soil column, infiltration or recharges and
the distribution coefficient for the contaminant in that matrix.
4.3.2.1. SELECTION CRITERIA — No one method or model for calculating
time of travel in the unsaturated zone is appropriate for application to all
cases. In general, two criteria are used to select a procedure for use in
the sludge disposal risk methodology.
The first criterion is that the method should be generally applicable to
a wide range of problems or sites. Potential sites exist across the United
4-13
-------
States and, therefore, may be characterized by a range of values. It is
impossible to select any single method that is optimal for all sites. With
this in mind, methods should be selected that can be used for the wide range
of values potentially encountered without requiring different approaches for
each setting.
The second criterion is that the data required by the method are
generally available or can be estimated, i.e., the data are typically known
for most disposal sites, values can be obtained for most waste sites or
accurate data for a specific site can be obtained from the literature. This
criterion is important because it is not intended that expensive, specialized
site studies be conducted to support an application.
4.3.2.2. GENERAL APPROACHES TO CALCULATION OF TIME OF TRAVEL — Aside
from field observations, there are two basic approaches to determining
travel times in the unsaturated zone: equilibrium solutions and unsaturated
flow models. Both approaches are based on the same fundamental equations,
but differ in the simplifying assumptions made to solve the equations. As a
result of the simplifying assumptions, the approaches differ significantly
in the time necessary to obtain a solution, in computational difficulty and
in data requirements. Relative characteristics of the two approaches are
summarized in Table 4-2.
Based on the above information, and given the criteria discussed for
selecting methods of calculating travel time in the unsaturated zone, the
use of unsaturated flow models is ruled out. Therefore, time of travel will
be calculated through use of appropriate equilibrium solutions. Use of an
equilibrium model necessitates assumption of steady-state as opposed to
transient conditions. The effects of this requirement are minimized by the
use of output to compare with chronic exposure criteria. If acute exposures
4-14
-------
TABLE 4-2
Relative Characteristics of Equilibrium Solutions
and Unsaturated Flow Modeling
Characteristic
Equilibrium Solutions
Unsaturated Flow
Modeling
Computation time
Data requirements
Complexity of solution
Time dependency
Short
Low to medium
Simple
Steady state
Medium to long
Medium to large
Complex
Steady state or transient
4-15
-------
were to be evaluated, transient analyses utilizing unsaturated flow models
would be more appropriate.
Analytical solutions of travel time through the unsaturated zone are
based on Darcy's equation for one-dimensional flow:
dt|i
Vz =
(4-6)
where:
Vz = seepage velocity in the vertical direction
K(ip) = hydraulic conductivity as a function of matric potential
— - hydraulic gradient in the vertical direction
In unsaturated flow, both hydraulic conductivity and moisture content
are nonlinear functions of pressure head. Hydraulic conductivity, moisture
content and pressure head need not be constant throughout a soil column;
however, if they are not, a direct analytical solution of Darcy's equation
is not possible for unsaturated flow. In order to obtain a solution of
Darcy's equation for travel time in the unsaturated zone, the following
assumptions must be made:
o One-dimensional flow is in the vertical direction.
o Water flow is at steady state.
o Water-table conditions exist at the lower boundary (i.e., the
water table — the top of the saturated zone — lies at the
bottom of the unsaturated zone).
o The upper boundary has a constant flux.
o Soil characteristics (moisture content vs. matric potential and
hydraulic conductivity vs. matric potential) are constant with
depth.
o The hydraulic gradient is vertically down and equals unity.
(Drainage is due strictly to gravity, or a^/az = 1.)
4-16
-------
The steady-state assumption and that of a constant upper boundary flux
imply even infiltration over time rather than periodic storm events as is
typical. While these assumptions are essential for the analytical solution,
under most circumstances they overpredict velocity, thus underpredicting
time of travel and time-dependent attenuation such as degradation. The
water-table assumption has no effect on concentration calculations.
For nonhomogeneous soils, the constant property assumption can be
approximated by dividing the soil profile into a series of homogeneous
layers and performing the travel time calculation on each layer individually.
The unit gradient assumption greatly simplifies the analysis. This
assumption means that the matric potential and, therefore, moisture content
and hydraulic conductivity are constant with depth. Using this assumption,
it is possible to directly solve for moisture content in terms of the flux
through the system and saturated soil properties. Knowing the moisture con-
tent and flux, it is possible to calculate the pore-water velocity and the
time of travel through the unsaturated zone. The unit gradient assumption
is generally valid if gravitational forces dominate other forces (e.g.,
capillary forces). When invalid, this assumption overpredicts contaminant
concentrations by underpredicting travel time.
If the unit gradient assumption is not made, the analytical solution to
unsaturated flow becomes more complex. In this case, it is necessary to
employ an iterative solution for pressure head and moisture content. This
iterative solution is time consuming, but can be simplified through the use
of a computer.
All analytical solutions for travel time through the unsaturated zone
are one dimensional. This results in conservative estimates, since lateral
4-17
-------
dispersion would reduce the concentration of contaminant at any given point
of entry. When applying these solutions to specific sites, it is important
to consider the horizontal variability of soil characteristics. If soil
characteristics vary spatially, the solution should be applied to the, soil
profile having the highest hydraulic conductivity. The solution will then
yield the highest velocity and shortest travel time (e.g., worst case) for
the unsaturated flow system.
In summary, analytical solutions provide a means of quickly estimating
time of travel through the unsaturated zone. Several assumptions are
required to perform these solutions, and no single solution is appropriate
for all applications. Two analytical solutions that pan. be. used to obtain
an estimate of travel time through the unsaturated zone for many typical
problems are discussed in detail in the next section.
4.3.2.3. ANALYTICAL SOLUTIONS FOR ESTIMATING TIME OF TRAVEL THROUGH
THE UNSATURATED ZONE — Two analytical solutions for calculating time of
travel through the unsaturated zone are presented. The first solution
assumes a unit gradient condition exists and is, therefore, the simplest.
The unit gradient assumption is not made in the second solution, which
allows for a variable moisture content, and, therefore, it is a little more
complex. The applicability of these methods is limited owing to the
simplifying assumptions used (see the previous section); however, the
methods can be used in a wide range of applications to calculate estimated
travel times.
The two methods provided here for estimating time of travel in the
unsaturated zone are consistent with the methods recommended in the Time of
Travel Manual (U.S. EPA, 1985b) developed for reviewing applications for
hazardous waste landfills.
4-18
-------
The data required by these analytical solutions for calculating travel
time through the unsaturated zone are stratigraphy of the site, thickness of
geologic units or soils, soil moisture characteristics for each unit or soil
and steady-state flux of water/moisture in the unsaturated zone.
Stratigraphic information is necessary for determining the types of
soils that are present in the unsaturated zone and for establishing the
layering sequence of these soils. Stratigraphy is most often determined
from logs of borings drilled at the site.
The thickness of the unsaturated zone, or layers within the unsaturated
zone, establishes the distance that water/moisture must travel before it
reaches the water table. This information would most likely be obtained
from borings.
The soil characteristics refer to the relationship between soil moisture
content (f) and matric potential (<|»), and the relationship between
hydraulic conductivity (K) and matric potential (1^) • Ideally, these
relationships should be measured in the laboratory using soil samples
obtained from the site. If laboratory measurements are not possible, the
following simple analytical relationships between pressure head and water
content, and between conductivity and matric potential (Campbell, 1974), can
be used:
< Vf>
(4-7)
where:
K = K
sat
(4-8)
Hie = air entry matric potential
fs = saturated water content
f = field water content
Ksat = saturated hydraulic conductivity
4-19
-------
b = negative one times the slope of the log-log plot of i|»m vs. f
n = 2 + 3/b .
Using these relationships, it is necessary to know only the slope of the
log-log plot of \|» vs. f, the saturated hydraulic conductivity and the
field moisture content. If experimentally derived data are not available, b
can be estimated from values provided in Appendix B. The saturated
hydraulic conductivity can be determined in the field or measured in the
laboratory. Appendix B lists representative values of saturated hydraulic
conductivity for a variety of materials.
The saturated moisture content (f ) can also be obtained from labora-
tory measurements. If measurements are not possible, f can be estimated
from the total (actual) porosity. Representative values of total porosity
are given in Appendix B.
Analytical solutions of travel time assume steady-state flow of moisture
through the unsaturated zone. A simple approximation of steady-state flux
is to assume that it is equal to the net infiltration at the site. Net
Infiltration is equal to the net precipitation minus actual evapotranspira-
tion. This information should be obtainable from weather stations or
agricultural research stations.
The following solution assumes steady-state flow and a unit hydraulic
gradient and employs the analytical soil moisture, pressure and conductivity
relationships described earlier (Campbell, 1974). Utilizing Darcy's
equation and the soil characteristic relationships described by Campbell
(1974), it is possible to derive the following expression for moisture
content as a function of steady-state flux (Heller et al., 198Ji):
f =
'Ksat
m
) fs
(4-9)
4-20
-------
where:
q = steady-state flux
Ksat = saturated hydraulic conductivity
fs = saturated moisture content
m = l/(2b 4- 3), where b is negative 1 times the slope of the
log-log plot of tm vs. f, as described earlier
Using Equation 4-9, it is possible to directly calculate the steady-state
moisture content of the soil. Pore-water velocity (the velocity of a water
particle) is defined as:
V - q/f (4-10)
Therefore, travel time for water (TW) can be calculated as the thickness of
the soil layer (L) divided by the pore-water velocity:
TW = L/V <= Lf/q (4-11)
The above solution of travel time can be applied to single- or multiple-
layered systems. For multiple layers, the above calculations are performed
for each layer. The total travel time through the unsaturated zone is then
equal to the sum of the travel times for each layer.
Solution of the variable moisture content case is more complex and
requires division of the soil profile into a number of discrete nodes or
grid points as shown in Figure 4-2. The nodes do not have to be evenly
spaced, but can be variably spaced to best represent different material
types (layers) if they are present. The analytical solution for this case
is as follows (Jacobson et al., 1985):
. , + Az (q/K* - 1)
i — I i
(4-12)
where:
Vi
= pressure head at the upper grid point
= pressure head at the lower grid point
4-21
-------
Grid i + 1
Node or
1 Grid Point
Grid i
Hydraulic
Conductivity
of Grid
Grid i-1
Grid i-2
AZj AY:
I, I
FIGURE 4-2
Discretization Between Grid Points
4-22
-------
K*
= flux through the soil column
= elevation difference between grid points
= harmonic mean hydraulic conductivity between grid
points
K* =
AZi
6 zi/Ki
(4-13)
KT, K^-i = hydraulic conductivity at the upper and lower grid
point, respectively
The solution begins with the grid point located at the lower boundary
(water table), where t. is known to be zero, and K is known from
I— 1 1— 1
r.
and the soil characteristic curve. The solution proceeds itera-
tively by assuming a value for y., determining K* and then solving for
\l».. A new value is assumed for »|». and the process repeated until
there is convergence on a solution. The calculated value of »ji. is then
used as *|i. for the next pair of grid points, and the process is
repeated.
Once the solution has determined the pressure head at every grid point,
the moisture content and hydraulic conductivity at every grid point can be
obtained from soil characteristic curves. Knowing the moisture content and
hydraulic conductivity at two grid points, the travel time between the grid
points is given by:
(a
where:
*
f
Ti
Ah.
i
*
Ki
Kf A h-j
harmonic mean moisture content between grid points
difference in total head
harmonic mean hydraulic conductivity between grid points
elevation difference between grid points
(4-14)
4-23
-------
The above equation is used to determine the travel time between every pair
of grid points. These travel time segments are then summed to obtain the
total travel time through the soil column.
It is possible to perform the above solutions manually for very simple
systems. However, as with all iterative solutions, the process can be very
time consuming. Therefore, the use of a computer is recommended. A computer
code to perform the above solutions for pressure head and travel times has
been developed by Oacobson et al. (1985).
It is emphasized that soil systems are often heterogeneous. They con-
tain a variety of materials that may create preferred routes of migration.
Therefore, if field values for time of travel are available, they should be
employed. Otherwise, data input to time of travel should be selected as
that for the most conductive media at the site, such as the coarse sands and
gravels. If the landfill includes a clay liner, the clay layer should be
the upper sequence evaluated in the unsaturated zone. If a membrane liner
is present, the methodology cannot be applied as described:.
4.3.2.4. ESTIMATION OF CONTAMINANT TRAVEL TIME — The analytical
solutions discussed above provide an estimate of the time for leachate to
travel through the unsaturated zone as a fluid. Contaminants will either
travel with the leachate or at a slower velocity, depending on the degree to
which they are adsorbed onto soil particles. The retardation factor is a
measure of how much more slowly a contaminant moves than the bulk leachate
and is a function of the contaminant/soil matrix distribution coefficient.
The retardation factor (RF) for a particular contaminant can be calcu-
lated by the formula:
RF = 1 + (B/f)(Kd) (4-15)
4-24
-------
where:
B/f = soil-to-solution ratio (bulk density of the soil divided by
its moisture content)
Kd = distribution coefficient
The Kd can be either measured in the laboratory or obtained from the
literature for a wide range of soil types and contaminants. Literature
values for sludge contaminants of concern are provided in Appendix B.
For organic contaminants, the Kd has been estimated using an organic
carbon distribution coefficient and data on the organic carbon content of
the soil. This underpredicts attenuation, since some surface adsorption
also occurs on soil particles. Use of a distribution coefficient treats all
adsorption as reversible. This, too, is conservative, since some sorption
is irreversible and, therefore, removes contaminants permanently.
The travel time for a contaminant through the unsaturated zone (TC) can
be estimated as the water travel time (TW) times the retardation factor (RF)
for that particular contaminant:
TC - TW x RF (4-16)
4.3.2.5. ESTIMATION OF CONTAMINANT CONCENTRATION — As the leachate
travels through the unsaturated zone, contaminant concentrations will be
reduced through chemical and biological processes. Reductions due to
precipitation of inorganic contaminants in excess of solubility limits may
occur in both the unsaturated and saturated zones, but can be most easily
predicted upon entry into the aquifer. As a consequence, these geochemical
considerations are left to the saturated zone transport segment. Reductions
due to degradation, such as hydrolysis and biochemical oxidation, will occur
in the unsaturated zone. These mechanisms are characterized by a degradation
rate constant (X). This can be related to environmental half-life repre-
sented by t . , the time required for the contaminant concentration to be
4-25
-------
reduced to one-half its initial value. If a first-order decay mechanism is
assumed, the concentration X at any time can be defined as:
X-X0e-" , (4-,7)
where X is the initial concentration of X and t is the time. Therefore,
the half-life, t.. .„, can be derived as:
or
t 1/2 =
In2 0.693
(4-18a)
(4-18b)
K \
The methodology assumes all degradation by-products are less toxic than
their parent compound. This assumption will not be true for all contami-
nants, but it is difficult to eliminate because degradation products from a
complex sludge environment have not been well characterized.
Equation 4-17 can also be employed to determine the degree of concentra-
tion reduction (degradation) that will occur as the leachate moves through
the unsatiirated zone. In this way, the average leachate concentration from
the source (X = X ) can be converted to the predicted value upon entry
into the aquifer. This is done by inserting the contaminant travel time in
the unsaturated zone, which is TC. From Equations 4-16 and 4-17:
X = X0e-VTC
or
In2 TC
X =
(4-19a)
(4-19b)
The resultant X is the contaminant concentration that should be applied
to all subsequent saturated zone transport calculations. Values for the
rate constant x. should be obtained from the scientific literature. If
4-26
-------
more than one degradation, mechanism is applicable, a composite value,
\ , should be derived. When all mechanisms are of the same order, the
composite value is derived as the sum of the individual rate constants. The
calculated values for concentration are compared to health criteria or
effects thresholds. If they do not exceed the thresholds, the contaminant
is dropped from further evaluation.
It is assumed that nitrogen leaches from the landfill mainly as the
ammonium ion, and that negligible amounts of nitrate will be found because
of the anaerobic conditions prevailing in the unsaturated and saturated
zones beneath the landfill.
Use of the time of travel approach gives the reduction in contaminant
concentration in the unsaturated zone due only to degradation. That is, it
is conservatively assumed that the contaminant plume moves as a pulse
through the unsaturated zone with no dispersion. In actuality, dispersion
will cause the contaminant pulse to elongate as it moves through the unsatu-
rated zone, with a resultant decrease in concentration. If dispersion in
the unsaturated zone is expected to be significant, the applicant may wish
to apply an analytical transport model to predict the concentration reduc-
tion due to dispersion.
The one-dimensional CHAIN code can be used (Van Genuchten, 1985) to
estimate the effects of dispersion of contaminants in the unsaturated
zone.* This code solves the convective-dispersive transport equation for a
zone-dimensional case and accounts for retardation and degradation. For
input data, the model requires the average pore-water velocity, the
*A more data-intensive alternative to CHAIN is the PRZM model (Carsel et
al., 1984), which can produce more detailed short-term predictions; these
would normally be unnecessary in the present context of chronic exposure.
4-27
-------
dispersion coefficient, the water content, the pulse time, the retardation
factor, the decay rate and several coefficients describing the source term.
The pore-water velocity can be obtained from Equation 4-10. The
dispersion coefficient may be highly site specific. Where the coefficient
is not known, Donigian et al. (1983) recommend a 2-fold approach for
modeling dispersion in the unsaturated zone. The dispersion coefficient is
at first set close to zero (0.01 cm2/day = 3.65xlO"4 m2/year) to
represent the situation where pollutant transport is convection dominated
and dispersion is relatively unimportant. Next, a reasonable value is
used, such as by setting the dispersion coefficient equal to one-tenth the
depth to groundwater (hy) times the pore-water velocity (V). Results from
these procedures are then compared to determine the potential influence of
unsaturated zone dispersion. Where the latter is important, site-specific
estimates may be advisable. The water content can be obtained using
Equation 4-9, the pulse time from Equation 4-4 or 4-5, and the retardation
factor from Equation 4-15. The source term is specified as a pulse of
constant concentration as calculated by partitioning (organic contaminants)
and solubility (inorganic contaminants), for a duration equal to the pulse
time. Degradation rates for chemicals to be analyzed should be taken from
the available literature. Care must be taken to ensure that values used are
appropriate to the subsoil environment. In particular, since anaerobic
conditions will prevail in the leachate-influenced unsaturated zone (as well
as in the saturated zone), rates should be representative of anaerobic,
rather than aerobic, systems.
The CHAIN code can be run to determine the unsaturated contaminant con-
centration at a depth equal to the depth to groundwater for a period equal
4-28
-------
to several contaminant travel times. The maximum resulting concentrations
are compared to the effects threshold. If the maximum concentration is
below the effects threshold, the analysis can be concluded without analysis
of the saturated zone transport.
4.3.3. Saturated Zone Transport.
4.3.3.1. INITIAL CONCENTRATION SELECTION — For inorganic contami-
nants, attenuation in soils may result from solution chemistry effects, as
well as interactions with the soil matrix. In the former case, the presence
of other chemical species in solution leads to the formation of insoluble
salts that precipitate out as solids. At this point in the analysis,
.geochemistry has been accounted for only to the extent that it affects the
composition of the leachate as simulated in the extraction test. While
further geochemical reactions may occur as the leachate travels through the
unsaturated zone, they are difficult to predict and often overshadowed by
adsorption and/or exchange on particle surfaces. Therefore, they are not
considered in this analysis until the leachate enters the saturated zone.
The adsorption and exchange phenomena have been accounted for in selection
of the unsaturated zone distribution coefficient.
It is assumed that when the leachate enters the saturated zone, it has
little or no effect on the hydrology of the aquifer, i.e., leachate produc-
tion is small compared to the volumetric flow of the aquifer. This will be
true whenever the area of the disposal facility comprises a small fraction
of the total recharge zone (a common reality). Under these circumstances,
it is possible to predict solution reactions in the groundwater and subse-
quent solution levels for contaminants of interest. These predictions are
4-29
-------
made through the application of a geochemical model that utilizes thermo-
dynamic data to predict equilibrium of total dissolved-phase (mobile) metal
concentrations. By applying the model at this point in the analysis, it is
possible to account for dissolution of salts and reduce the inventory of
contaminant in leachate to those levels likely to be encountered in the
aquifer. Since the final solution concentration is dictated by the
solubility of product salts, considering the geochemistry in the saturated
zone only and not throughout the system is not likely to have a significant
impact on the results.
Geochemical models involve complex codes and massive amounts of data.
The analyst must interact with the program during the analysis and, there-
fore, must be trained in the application of the code utilized. As a conse-
quence, it is not the intent of this methodology to require each applicant
to apply a geochemical model to site-specific conditions. Rather, a series
of model runs have been made across a spectrum of conditions. Output of
these runs is provided here for the applicant to utilize in selecting
contaminant concentrations based on matching groundwater conditions of a
given run to those of the site of interest.
The MINTEQ geochemical code was applied to generate predicted contami-
nant concentrations under selected groundwater conditions. HINTEQ is one of
the more advanced computer codes that the U.S. EPA employs to characterize
the chemical processes that may control the concentrations of constituents
dissolved in leachate and natural waters. MINTEQ is a hybrid code that
combines an efficient mathematical structure with,a large, well-documented
thermodynamic data base. Functionally, the code models the mass distribu-
tion of a dissolved element between various uncomplex and complex aqueous
4-30
-------
species; it also calculates the degree to which the water is saturated with
respect to the solids in the thermodynamic data base. Adsorption,
precipitation and dissolution reactions can be included in calculations.
Only the latter two were applied here, since a Kd is used in the unsaturated
model to address adsorption. Detailed documentation of the MINTED, code and
data can be found in Felmy et al. (1983, 1984) Morrey (1985) and Deutsch and
Krupka (1985).
Each element will exist in the subsurface environment as a relatively
complex distribution of different species, each having a specific set of
properties. Relative concentrations of individual species within the
distribution are controlled by equilibrium constants governing the
individual reactions, and by the chemical environment in which this
speciation process occurs. Some elements exist in several oxidation states
simultaneously and form a number of individual species of widely differing
chemical characteristics. Because such chemical and physical
characteristics determine the ability of the species to be transported, the
MINTEQ code is used in this methodology.
To apply the MINTEQ code, it was necessary to specify key solution
parameters, including organic constituents, background ionic species, pH and
Eh. Organic ligands that could solubilize metals were selected on the basis
of data from studies of municipal landfill leachate. A total organic level
of 15,000 mg/a. in raw leachate was utilized on the basis of the maximum
organic levels measured in sludge landfill leachate (U.S. EPA, 1978). The
total organic loading modeled consisted of six representative compounds for
which required data on thermodynamics were available:
4-31
-------
Acetate-1 (mol. wt. 59.05)
Glutamate-2 (mol. wt. 145.13)
Glycine-1 (mol. wt. 74.07)
Phthalate-2 (mol. wt. 164.13)
Salicylate-2 (mol. wt. 136.12)
Tartrate-2 (mol. wt. 148.09)
Since it was assumed that the leachate will not control groundwater
chemistry, it was necessary to simulate dilution arising from mixing of the
leachate into the aquifer. It was also assumed that only a portion of the
total organic load in leachate represented ionic species capable of solu-
bilizing metals. In recognition of both dilution and the identity of
individual fractions of the total organic loading, it was estimated that
0.01 of the maximum level of organics observed in sludge monofill leachate
(15,000 mg/fi,) would be present as ionic organic ligands capable of
solubilizing metals.
Given the 0.01 fraction selected, the organic ligands for the MINTEQ
runs were entered in the following concentrations:
Acetate 11.99 mg/a, .
Glutamate 29.46 mg/a.
Glycine 15.04 mg/a .
Phthalate 33.32 mg/B. , ,
Salicylate 27.63 mg/a,
Tartrate 30.06 mg/a.
Total 147.5 mg/fi. ,
4-32
-------
Inorganic species and concentrations for the MINTEQ runs were selected
on the basis of median values of national groundwater data included in the
STORE! system as presented in Table 4-3.
Six combinations of pH and Eh were selected for model runs: pH = 6.0
and 7.0; Eh = -200, +150 and +500 mv. The Eh values bracket those reported
as typical for groundwater (Baas-Becking et a!., 1960). The pH values
address the lower half of the 6-8.5 range reported for groundwaters. These
lower values are considered conservative because they are more likely to
mobilize metals than pH levels of 7.5-8.5. Low pH and Eh values represent
water affected by Teachate where acid has been formed and oxygen is depleted.
For each contaminant of interest, a series of model runs were made
introducing the contaminant at the concentrations listed in Table 4-4.
Results of the model runs were then plotted showing the output
concentrations as a function of input levels for each pH-Eh combination.
These results are presented in Appendix B (Figures B-2 through B-6). Since
each contaminant was modeled separately, no provision was made for
interactions arising from multiple contaminants entering simultaneously.
To account for the geochemistry, the applicant need only determine the
pH and Eh conditions of local groundwater and the level of contaminants in
leachate. It is difficult to obtain accurate Eh measurements because the
sample rapidly goes to a high oxidation level upon contact with air.
However, because the more oxidized state yields higher metal mobility, the
error introduced by poor sampling will essentially add a greater degree of
safety. The appropriate graphs are selected from Appendix B (Figures B-2
through B-6) on the basis of having similar pH and Eh values. The leachate
contaminant concentrations are then entered as inputs and the resulting
4-33
-------
TABLE 4-3
Background Inorganic Constituents for MINTEQ Model Runs
(temperature: 14°C)
Chemical
Concentration
Aluminum
Arsenic
Barium
Bicarbonate
Bromide
Cadmium
Calcium
Carbonate
Chloride
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Nitrate
Phosphate
Selenium
Silver
Sulfate
Sulfide
Thallium
0.200
0.010
o.;?oo
190.000
0.300
0.005
48.000
0.000
15.000
0.202
0.020
0.2100
0.010
14.000
0.040
0.0005
1.000
0.090
0.005
0.010
25.000
0.200
0.040
4-34
-------
TABLE 4-4
Contaminant Concentrations Employed in Benchmark MINTEQ Runs
Contaminant
(mg/SL)
Model Input Concentrations
Arsenic
Copper
Lead
Mercury
Nickel
0.06/1.25/2.5/5.0
1.32/32.5/65.4/130
0.03/0.5/1.0/2.0/10.0
0.0035/0.075/0.15/0.30
0.16/3.75/7.5/15
4-35
-------
groundwater concentration identified from the curve as the starting point
for subsequent saturated zone modeling.
To illustrate, assume an applicant has a predicted leachate concentra-
tion at the base of the unsaturated zone of 6 mg/a, and a site with ground-
water at pH 7.2 and Eh 200 mv. The contaminant concentration in the
groundwater (aquifer) would be determined by selecting the MINTED, figure for
this contaminant with the closest match for the pH and Eh conditions (pH of
7.0 and Eh of 150 mv) and reading the appropriate value. As illustrated in
Figure 4-3, an input concentration of 6 mg/fi, (abscissa value) yields an
aquifer concentration of 1.0 mg/fi. (ordinate value). This result would
then be used for the input concentration (C ) to the saturated zone model.
o
Nitrate is included in the MINTEQ data base. However, it was found that
nitrate, like chlorides, was not solubility limited in any of the runs;
therefore, no graph was constructed, and input levels to the saturated zone
should be equated with output from the unsaturated zone.
The predicted inorganic contaminant concentrations in the saturated zone
are compared to health criteria or effects thresholds. If they are less
than the threshold, the contaminant is dropped from further consideration.
If they exceed the threshold, they are input to the saturated zone transport
model.
4.3.3.2. TRANSPORT CODE SELECTION CRITERIA — The same criteria used
to select a method for calculating travel time in the unsaturated zone are
relevant for the saturated zone, namely the following: (1) the method
should be appropriate for a wide range of applications, and (2) the data
required by the method should be generally available. A more detailed
discussion of these criteria is presented in Section 4.3.2.1.
4-36
-------
O) "*•*.
> g>
o E
w •*-
.2 c
Q
CD O
§§
NO
"O ^
£ i
II
•+•* ^
1.2-
1.0--
0.8-
0.6-
0.4-
0.2-
0.0
Unsaturated Zone Input Concentration (mg/.t)
FIGURE 4-3
Example MINTEQ Spedatlon Results for Entry of a Contaminant Into the Saturated Zone
for Conditions of pH = 7.0 and Eh = 1.50 mv
4-37
-------
4.3.3.3. TECHNICAL APPROACHES FOR DETERMINING TIME OF TRAVEL AND CON-
TAMINANT CONCENTRATION IN THE SATURATED GROUNDWATER FLOW SYSTEM — There
are two basic approaches to estimating contaminant travel time (velocity)
and concentration in the saturated groundwater flow system: analytical
solutions and numerical modeling. Analytical solutions are relatively quick
and simple to use. However, they are based on a variety of simplifying
assumptions related to contaminant characteristics and the subsurface
environment. Consequently, the methods provide order-of-magnitude estimates
of contaminant travel time and concentration. Numerical models, on the
other hand, are far less restricted with regard to simplifying assumptions,
but they typically require more data, are time consuming to set up and run,
and require expensive and/or specialized equipment and expertise. Based on
the above information and the selection criteria discussed earlier, the use
of numerical models is not required. However, if the applicant believes
that a more accurate portrayal of transport is worth the added costs, he or
she may opt to employ such a model.
Numerous analytical methods/models for predicting contaminant transport/
concentrations in the groundwater flow system are available (Lapidus and
Amundson, 1952; Davidson et al., 1968; Lindstrom and Boersma, 1971; Lai and
3urinak, 1972; Warrick et al., 1971; Cleary et al., 1973; Lindstrom and
Stone, 1974; Marino, 1974; Kuo, 1976; Yeh and Tsai, 1976; Van Genuchten and
Wierenga, 1976; Selim and Mansell, 1976; Wang et al., 1977; Yeh, 1981;
Donigian et al., 1983). Most of these solutions/models are based on the
advection-dispersion equation for predicting solute movement through porous
media. The basic difference among them is their simplifying assumptions
that make them specific to a particular problem.
4-38
-------
4.3.3.4. ANALYTICAL METHODS/MODELS FOR ESTIMATING CONTAMINANT CONCEN-
TRATIONS IN THE GROUNDWATER FLOW SYSTEM -- A number of analytical models
are available for predicting saturated transport as presented in Table 4-5.
Two of these analytical methods for estimating contaminant concentration in
the saturated groundwater flow system are discussed here for illustrative
purposes. In both cases, the methods are described in fairly general terms.
References are provided for a detailed discussion of the methods and their
application.
The first method is the analytical solution to the advective-dispersive
equation. The second method describes the use of the AT123D analytical
model (Yeh, 1981), which solves the advective-dispersive equation. It is
coded such that it can be run on a personal computer, and it has the
capability to handle many different types of boundary conditions.
The advective-dispersive equation forms the basis of all solution
algorithms for predicting solute movement through saturated porous media.
This equation assumes constant groundwater velocity (steady flow) in the
longitudinal direction. It was developed for solving the limiting case of
unidirectional advective transport with three-dimensional dispersion in a
homogeneous and isotropic aquifer. Contaminant decay and retardation can be
described by a first-order degradation rate and an equilibrium (partitioning
or distribution) coefficient, respectively.
In three dimensionals with the average flow along the x axis, the
advective-dispersive equation can be written as:
3C _,_ n 3C
— + v — =
3T
3X
3X2
3aC
—
aya
(4-20)
4-39
-------
TABLE 4-5
Analytical Solutions of the Advective-Dispersive Equation
Author
Title
Boutwell, S.H
S.M. Brown,
B.R. Roberts and
A.D. Atwood
Cleary, R.W. and
M.J. Ungs
Codell, R.B.
Codell, R.B.
Donigian, A.S.
et al.
Van Genuchten, M.T.
and W.J. Alves
Modeling Remedial Actions at Uncontrolled Hazardous
Waste Sites. EPA/540/2-85-001. U.S. EPA, Cincinnati,
OH. 1985.
Analytical Models for Groundwater Pollution and
Hydrology. Report 78-WR-15. Water Resources Program,
Department of Civil Engineering, Princeton University,
Princeton, NJ. 1978.
Collection of Mathematical Models for Dispersion in
Surface Water and Groundwater. NUREG-0868. Nuclear
Regulatory Commission, Bethesda, MD. 1982.
Simplified
NUREG-1054.
MD. 1984.
Analysis for Liquid Pathway Studies.
Nuclear Regulatory Commission, Bethesda,
Rapid Assessment of Potential Groundwater Contamination
Under Emergency Response Conditions. Anderson-Nichols
&Co., Inc., Palo Alto, CA. EPA-68-3116. 1983.
Analytical Solutions of the One-Dimensional Convective-
Dispersive Solute Transport Equation. U.S. Dept. of
Agriculture, Tech. Bull. No. 1661. 1982.
Yeh, G.T.
AT123D: Analytical Transient One-, Two-, and Three-
Dimensional Simulation of Waste Transport in the
Aquifer System. ORNL-5602. Environmental Sciences
Div., Pub. No. 1439. Oak Ridge National Laboratory,
Oak Ridge, TN. 1981.
4-40
-------
where:
= solution concentration (M/L )
°2.» DT!l» DTv = longitudinal, lateral transverse and vertical
transverse hydrodynamic dispersion coefficients
(La/T)
V
= average interstitial pore-water velocity in the x
direction (L/T)
T = time (T)
x, y, z = Cartesian coordinates
X = degradation/decay rate (T"1)
RF = retardation factor
Because the flow is unidirectional along the layering and is almost
horizontal, the Cartesian coordinate axes are oriented in directions
parallel to and normal (perpendicular) to the mean flow direction. The
coefficients of hydrodynamic dispersion appearing in Equation 4-20 include
both the effects of mechanical dispersion and molecular diffusion (D*).
They are of the form:
DTJt =
V * D*
DTv - "TV V
(4-21a)
(4-21b)
(4-21c)
If the amount of spreading due to molecular diffusion is insignificant rela-
tive to the mixing caused by mechanical dispersion, then D*, the molecular
diffusion coefficient of a solute in porous medium, is generally ignored.
Since this analysis focuses on a compliance point directly downflow from the
source, dispersion will always exceed diffusion effects unless velocity
4-41
-------
approaches zero. In those cases, the dispersion coefficient is selected to
encompass both effects. Unless site-specific information on dispersion is
available, dispersion in the direction of flow may be assumed equal to
one-tenth the distance to the point of compliance times the pore-water
velocity (Donigian et a!., 1983).
For a contaminant that travels with the groundwater, the average linear
pore-water velocity can be calculated as:
V
e ax
(4-22)
where:
K
aH
ax
e
hydraulic conductivity of the medium (L/T)
hydraulic gradient (dimensionless)
effective porosity (dimensionless)
For contaminants that adsorb onto the soil matrix, the retardation
factor must be estimated. The retardation factor (RF) is a measure of the
mobility of the contaminant in the porous media. It represents the ratio of
i -
the mean pore-water velocity to the mean contaminant migration velocity and
can be calculated as:
RF = 1 + (B/e)(Kd) (4-23)
where:
B ~ bulk density of the soil (kg/m3)
6 = effective porosity (dimensionless)
Kd = distribution coefficient (8,/kg)
Values for effective porosity are provided in Appendix B.
4-42
-------
The pollutant source is applied as continuous step functions of initial
tions:
C (x,o) = 0 (4-24a)
concentration (C ) and duration (T ) with the following boundary conditions:
C (o,t) = C
(4-24b)
jp (»,t) =0 (4-24c)
The analytical solution of the advective-dispersive equation (4-20), as given
by Cho (1971), Misra et al. (1974), Van Genuchten and Alves (1982) and Rao
(1982), can be expressed as:
C(x,t) = C C*(x,t) for o < t < T (4-25)
= C C*(x,t) - C*(x, t-T ) for t > T
o p P
where C*(x,t) is given by
C*(x,t) = 1
* exp
20
erf c
and
w = (v2 + 4 OR \)
(exp denotes the natural logarithm exponential and erfc the complementary
error function.)
erfc(z) = J exp(-s2)ds
z
4-43
-------
as found tabulated in standard reference texts (i.e., Abramowitz and Steguin,
1972).
The boundary conditions shown in Equation 4-24(a,b,c) indicate that:
o No contaminant is present in the soil prior to input from the
source.
o The input concentration at the surface is constant at C0.
o A semi-infinite column is assumed with a zero-concentration
gradient at the bottom. This last boundary condition is often
assumed to allow development of the analytical solution; Van
Genuchten and Alves (1982) indicate that this assumption has a
relatively small influence on the accuracy of the solution in
most circumstances when applied to well-defined finite systems.
The parameters required to solve the advective-dispersive equation,
along with their symbols and recommended units, are listed in Table 4-6.
The second method for predicting the spatio-temporal distribution of
contaminants in an aquifer is the AT123D code developed by Yeh (1981). This
method is based on the basic advective-dispersive equation just discussed;
however, it is coded in a format that makes it easy to use, and that allows
for implementation of numerous options (450 in all). These options provide
the code with the capability to simulate a wide variety of configurations
and situations of source release and types of boundary conditions.
The data required to run the AT123D code are as follows:
o Geometry of the region of interest (x, y and z dimensions);
o Geometry of the source of contamination (xs, ys and zs
dimensions);
o Dispersion coefficients in the x, y and z directions (Dx, Dy
and Dz);
o Soil properties of effective porosity and bulk density (0, B);
o Hydraulic conductivity (K);
o Source/sink strength (Q);
4-44
-------
TABLE 4-6
Required Parameters for Solution of the Advective-Dispersive Equation*
V
Parameter/Boundary Condition
Source concentration
Interstitial pore-water velocity
Dispersion coefficient
Degradation/decay rate parameter
Retardation factor (function of
following characteristics)
Symbol
C
V
D
X.
RF - 1 + BKd
6
Recommended Unit
mg/fc
cm/day
cm2 /day
day-i
dimensionless
Distribution coefficient
Soil bulk density
Effective porosity
Pulse duration (pulse input only)
*Source: Donigian et a!., 1983
Kd
B
6
mSL/g
g/cm3
dimensionless
day
4-45
-------
o Distribution coefficient (soil-waste interaction parameter) (Kd);
o Flow field (groundwater velocity) (V , V and V );
o Decay constant (X); and
o Background concentrations of contaminants of interest;
A complete description of this code and its application is contained in Yeh
(1981).
4.3.3.5. CALCULATING STREAM CONCENTRATIONS RESULTING FROM GROUNDWATER
SEEPAGE — A supplementary groundwater pathway is the exposure from edible
aquatic organisms living in surface water recharged by contaminated ground-
water. This pathway is assumed to be represented by the consumption of fish
caught in a stream near a landfill. The concentration of the contaminant in
the stream is first calculated under the conservative assumption that all of
the contaminants exiting the unsaturated zone beneath the landfill seep into
the stream:
R x A
Stream Concentration = Cus x
Qf
(4-25a)
where:
CKS = contaminant concentration exiting the unsaturated zone
R = net recharge
A = landfill area
Qf = stream flow
and the units are internally consistent.
The resulting stream concentration is then compared with the reference
concentration in surface water (RWC), as described in the surface runoff
chapter of the companion methodology document on land application and
4-46
-------
distribution and marketing of municipal sludge (U.S. EPA, 1989). The
expressions in that document take into consideration bioconcentration
factors, fish ingestion rate, water ingestion rate, etc.
As an example, if the values of:
R =0.5 m/year
A = 100 m x 10 m = 1000 m2
Qf = 1 m3/sec
are used to calculate the factor to convert leachate concentration to stream
concentration, a dilution factor of 0.0000158 is obtained.
4.3.3.6. ACCOUNTING FOR BACKGROUND CONTAMINANT CONCENTRATIONS IN THE
GROUNDWATER— If background groundwater concentration levels for the
contaminants of interest are measurable, these should be incorporated into
the procedure by adding the background concentration to the final
model-predicted concentration at the property boundary and comparing this
total concentration to the health criteria.
4.3.4. Setting National Criteria. The methodology presented herein has
been devised to evaluate municipal sludge landfill disposal on a site-
specific basis. As mentioned previously, it can also be employed to estab-
lish sludge contaminant concentration 'criteria to be administered on a
national or regional scale. For this purpose, the mode of operation is the
reverse of that presented in earlier portions of this chapter. That is, the
analysis begins with, the selected health effects criteria and works back-
ward to determine how high a concentration in the sludge could be before
environmental levels would exceed that threshold. Hence, the analysis
begins with the point of compliance, moves back through the aquifer to the
4-47
-------
point below the disposal facility and then moves up through the unsaturated
zone to the disposal cell and the sludge itself.
Functionally, the reverse operation is not so straightforward. Some of
the models and constructs employed in the methodology cannot be operated in
the reverse mode, i.e., the analytical model is not closed form, so one can-
not set the final concentration and solve for the initial concentration.
Given constant dispersion and degradation, outflow concentration is a func-
tion of two factors — leachate concentration and leach duration. Therefore,
specifying an outflow concentration does not specify what the input para-
meters must be because many combinations of the input parameters can produce
the same outflow concentration. Thus, it is necessary to make a series of
forward calculations for a range of values and then work backward to obtain
the national criteria levels.
Because the methodology is not reversible, the results of the national
criteria calculation are not unique values of acceptable contaminant concen-
trations in the sludge. Instead, the result presented is a graph of a
family of curves that relate the input leachate concentration (out of the
landfill) to the groundwater concentration at the facility boundary. Each
curve on the graph represents a certain pulse or release time. Given the
health effects criteria concentration at the facility boundary and the
initial pulse or release duration, the maximum acceptable input concentration
can be obtained from the graph. From the input concentration, the total
mass of contaminant in the sludge can then be calculated. This entire
methodology will be discussed in more detail in the following text, and an
example will be provided.
4-48
-------
To develop the graph of the family curves, it is necessary to run the
modeling sequence (unsaturated and saturated zone models) discussed previ-
ously for a range of pulse durations and contaminant input concentrations.
The first step is to run the unsaturated zone transport model (CHAIN) using
a range of pulse durations and contaminant (leachate) concentrations. Pulse
durations of 1, 10, 100 and 1000 years and contaminant concentrations of 10,
100 and 1000 times the health effects criteria are recommended. All other
input parameters to the model should be representative of the region being
simulated, or in the case of national criteria, representative of the
reasonable or probable worst case.
The results of the CHAIN model runs provide a series of release
durations and peak concentrations for input to the saturated transport code
(AT1230). The release duration obtained from the CHAIN results is defined
as the period of time when the output concentration first reaches 1% of the
peak concentration until the time after the peak declines to <1% of the
peak. The peak concentration is used as the contaminant concentration
during the entire release period and is a conservative assumption.
For the organics, the peak output concentration from CHAIN becomes the
input concentration to ATI230.. For the metals, the peak output concentra-
tion from CHAIN is adjusted using the appropriate MINTEQ curve in Appendix 8
(Figures B-2 through B-6) to predict the resulting concentration after
geochemical reaction. This resulting concentration is used as the input
concentration to AT123D.
The AT123D code requires a contaminant input flux rate that is the input
concentration obtained from CHAIN or MINTEQ (Appendix B) times the surface
recharge rate. The release times from CHAIN become the input pulse times to
4-49
-------
AT123D. All other input parameters to AT1230 shall be representative of the
region or national case being simulated. The AT123D code is then run for
this range of input concentrations and pulse times to predict a range of
peak output concentrations.
At this point in the methodology the output consists of pairs of initial
leachate concentration (X.) (below the landfill) and peak output concen-
tration (Xf) (at the property boundary) for each initial pulse duration
simulated. These pairs can all be plotted on the same graph (Xf on the
horizontal axis and X.. on the vertical axis) to yield a family of curves
(one curve for each pulse duration). An example graph is shown in
Figure 4-4.
The family of curves can be used as follows in working backward to
determine a national criteria. Locate the point on a specific pulse time
curve whose abscissa is equal to the health effects criteria value for the
desired contaminant. The ordinate of this point is the maximum allowable
leachate concentration in the landfill for pulse times equal to or less than
the pulse time of the corresponding curve. A maximum allowable leachate
concentration can be obtained from the graph for each pulse time curve.
The last step of the methodology is to calculate the total Teachable
mass (M.) and the total mass (M) of contaminant in the landfill. For non-
degradable contaminants, the total Teachable mass can be calculated as the
product of the initial leachate concentration (X.) (as determined from the
family of curves), times the recharge rate (R), times the pulse time (T ):
- X. R Tp
(4-26)
4-50
-------
Acceptable Leachate Concentration
o
O
0)
Effects Threshold Value
Peak Output Concentration
FIGURE 4-4 ......
Example Graph of the Family of Curves Obtained
for the National Criteria Case
4-51
-------
For degradable contaminants, the degradation rate constant (X) enters
the calculations and the equation is:
ML =
(4-27)
The total mass of contaminant in the sludge consists of that sorbecl on
the sludge solids and that dissolved in the sludge water. Therefore, the
total mass of contaminant in the landfill for both degradable and nondegrad-
able contaminants can be calculated as the total Teachable mass plus the
product of the initial leachate concentration (as determined from the family
of curves) and the volume of water that drains from the sludge (D ):
M = ML + Xi DV (4-28)
There may exist a whole range of solutions for a particular simulation,
so it is not possible to choose a unique result. However, from experience
it appears that for many cases this methodology can be simplified and a
somewhat unique solution can be obtained. This simplification is usually
the result of the peak output concentration asymptotically reaching a
maximum value as the pulse time increases, or as a result of solubility
limits being realized in the saturated flow system. Examples of these
processes are provided in the site-specific applications; discussed in
Section 4.5.
4.4. INPUT PARAMETER REQUIREMENTS
A number of inputs are required to apply the landfill alternative
groundwater pathway review methodology to a specific site or proposed site.
This section summarizes these inputs and provides information on where data
may be obtained.
4.4.1. Fate and Transport: Pathway Data.
4-52
-------
4.4.1.1. SOURCE TERM —
1, Sludge Moisture Content (L) -- Derived directly from gravi- metric
analysis of sludge [ASTM Method G 51-77 (1984)].
2. Storage Capacity of Sludge (S) — Derived gravimetrically as the
moisture content of the sludge when completely drained by gravity
[ASTM Method 6 51-77 (1984)].
3. Net Recharge (R) — Obtained from local weather station data or
agricultural extension offices.
4.4.1.2. UNSATURATED ZONE —
1. Depth to Groundwater (hy)— Determined from site plan and borings.
2. Distance from Landfill to Property Boundary (ds) — Determined from
site plan. Should equal the buffer strip width between the fill
area and the property fence. Cannot exceed 150 m.
3. Stratigraphy — Taken from site borings and/or local geological
maps to determine the soil types and sequencing of the types.
4. Stratigraphic Layer Thickness — Estimated from borings and/or
local geological maps.
5. Saturated Soil Hydraulic Conductivity (Ksat) — Measured in the
field or laboratory [ASTM Method 02434-68 (1974)].
6. Slope of the Log-Log Plot of Air Entry Matrix Potential (ye) and
Field Moisture Content (f) for Soils (b) — Derived experi-
mentally or estimated from data presented in Appendix B (ASTM
Method D2216-80).
7. Saturated Soil Moisture Content (fs) — Derived experimentally or
estimated from data presented in Appendix B (ASTM Method D2216-80).
8. Bulk Density of Soil (B) — Derived experimentally or taken from
the literature. A common value applied here is 1600 kg/m3
(ASTM Method D2937-83).
4.4.1.3. SATURATED ZONE —
1. Groundwater pH — Determined by direct measurement [ASTM Method
G 51-77 (1984)].
2. Groundwater Eh — Determined by direct measurement.
3. Hydraulic Conductivity in the Aquifer (K) — Determined from
field tests or taken from data presented in Appendix B.
4-53
-------
4.
5.
6.
7.
Effective Porosity (ee) — Determined directly or taken
from data presented in Appendix B (ASTM Method D 4404--84).
Hydraulic Gradient (aH/ax) — Determined from field data on
potentiometric head (water table) or estimated from topography.
Bulk Density of Aquifer Media (B) — Determined experimentally
or taken from the literature. Typical value for soils is 1600
kg/ma (ASTM Method D 2937-83).
Dispersion Coefficient (Kd) — Derived from the following:
3H K
Dx-0.1 < ()ds
(4-29a)
aH K
g£) (g)ds
(4-29b)
Dy,Dz = 0.01
8. Geometry of the Site — Taken from site maps.
4.4.2. Fate and Transport: Chemical-Specific Data.
4.4.2.1. SOURCE TERM —
1. Contaminant Concentration in Sludge (N) — Derived directly for
each contaminant by analyzing a sample of the sludge using an
approved digestion technique.
2. Contaminant Concentration in Leachate (X) — Derived directly
for each contaminant by applying the partition coefficient to
the total sludge concentration for organic contaminants and
maximum solubility levels for inorganic constituents.
4.4.2.2. UNSATURATED ZONE —
1. Unsaturated Zone Distribution Coefficient (Kd) — In Tier 2,
the Kd value is derived experimentally.
2. Unsaturated Zone Degradation Rate Constant (\u) — Selected
values from the literature are provided.
4.4.2.3. SATURATED ZONE —
1. Saturated Zone Distribution Coefficient (Kd) — Selected from
Appendix B.
2. Saturated Zone Degradation Constant (xs) — Values provided
as taken from the scientific literature.
4.4.3. Health Effects Data. A reference water concentration (RWC, in
mg/2.) will be defined as a groundwater concentration used to evaluate the
4-54
-------
potential for adverse effects on human health as a result of sludge land-
filling. That is, for a given landfill site, and given the practice defini-
tions and assumptions stated previously in this methodology, the criterion
for a given sludge contaminant is the concentration in the sludge that
cannot be exceeded, and is calculated to result in groundwater concen-
trations below the RWC at the well site. Exceeding the RWC would be a basis
for concern that adverse health effects may occur in a human population in
the site vicinity.
RWC is determined based upon contaminant toxicity and water ingestion
rate, from the following general equation:
Reference Water Concentration: RWC =1/1
P w
(4-30)
where I is the acceptable chronic pollutant intake rate (in mg/day) based
on the potential for health effects, and I is the water ingestion rate
(in JL/day). This simplified equation assumes that the ingested contami-
nant is absorbed into the body via the gastrointestinal tract at the same
rate in humans as in the experimental species tested, or between routes of
exposure (e.g., oral and inhalation). Also, this equation assumes that
there are no other exposures of the contaminant from other sources, natural
or manmade. I varies according to the pollutant evaluated and to whether
the pollutant acts according to a threshold or nonthreshold mechanism of
toxicity.
4.4.3.1. THRESHOLD-ACTING TOXICANTS — Threshold effects are those
for which a safe (i.e., subthreshold) level of toxicant exposure can be
estimated. For these toxicants, RWC is derived as follows:
4-55
-------
Reference Water Concentration: RWC =
/RfD x bw\
I RE J~
TBI
* Iw
(4-31)
where:
RfD = reference dose (mg/kg/day)
bw = human body weight (kg) ''
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
Iw = water ingest ion rate (8,/day)
RE = relative effectiveness of exposure (unitless) |
The definition and derivation of each of the parameters used to estimate RWC
for threshold-acting toxicants are further discussed below.
4.4.3.1.1. Reference Dose (RfD) — When toxicant exposure is by
ingestion, the threshold assumption has traditionally been used to establish
an acceptable daily intake, or ADI. The Food and Agricultural Organiza-
tion and the World Health Organization have defined ADI as "the daily intake
of a chemical which, during an entire lifetime, appears to be without
appreciable risk on the basis of all the known facts at the time. It is
expressed in milligrams of the chemical per kilogram of body weight (mg/kg)"
(Lu, 1983). Procedures for estimating the ADI from various types of toxico-
logical data were outlined by the U.S. EPA in 1980 (Federal Register,
1980). More recently the Agency has preferred the use of a new term, the
reference dose, or RfD, to avoid the connotation of acceptability, which is
often controversial.
The RfD is an estimate (with uncertainty spanning perhaps an order of
magnitude) of the daily exposure to the human population (including
sensitive subgroups) that is likely to be without appreciable risk of
deleterious effects during a lifetime. The RfD is expressed in units of
mg/kg bw/day. The RfD is estimated from observations in humans whenever
4-56
-------
possible. When human data are lacking, observations in animals are used,
employing uncertainty factors as specified by existing Agency methodology.
RfD values for noncarcinogenic (or systemic) toxicity have been derived
by several groups within the Agency. An Intra-Agency Work Group verifies
each RfD, which is then loaded onto the Agency's publically available
Integrated Risk Information System (IRIS) database. Most of the
noncarcinogenic chemicals that are presently candidates for sludge criteria
for the landfill pathway are included on the Agency's RfD list, and thus no
new effort will be required to establish RfDs for deriving sludge criteria.
For any chemicals not so listed, RfD values should be derived according to
established Agency procedures (U.S. EPA, 1988).
4.4.3.1.2. Human Body Weight (bw) and Water Ingestion Rate (I ) —
w
Both bw and I vary widely among individuals according to age and sex.
w
Variations of mean drinking-water intake and body weight with age and sex
for the U.S. population are illustrated in Table 4-7. The choice of values
for use in risk assessment depends on the definition of the individual at
risk, which in turn depends on exposure and susceptibility to adverse
effects. The RfD (or ADI) was defined before as the dose on a body-weight
basis that could be safely tolerated over a lifetime. As shown in Table
4-7, water consumption on a body-weight basis is substantially higher for
infants and toddlers than for teenagers or adults. Therefore, infants and
toddlers would be at greater risk of exceeding an RfD when exposure is by
drinking water. However, the effects on which the RfD is based may occur
after a long cumulative exposure period, in some instances approaching the
4-57
-------
TABLE 4-7
Water Ingestion and Body Weight by Age-Sex Group in the United States
Age-Sex Group
6-11 months
2 years
14-16 years, female
14-16 years, male
25-30 years, female
25-30 years, male
60-65 years, female
60-65 years, male
Mean Water
Ingestiona
(mfc/day)
308
436
587
732
896
1050
1157
1232
Median
Body Weight
.(kg)
8.8b
13. 5b
51 .3b
54. 2b
58. 5C
67. 6C
67. 6c
73.9°
Water Ingestion
per Unit
Body Weightd
(mfc/kg/day)
35.1
32.2
11.4
13.5
15.3
15.5
17.1
16.7
aSource: Pennington, 1983. From the revised FDA Total Diet Study.
Includes categories 193, 195-197, 201-203.
bSource: Nelson, 1969, as cited in Bogert et a!., 1973.
Calculated by averaging several age or sex groups.
cSource: Society of Actuaries, 1959, as cited in Bogert et all., 1973.
Average body weights for median heights of 156 cm (5 ft, 5 in) and 173 cm
(5 ft, 8 in) for females and males, respectively.
water ingesti on/body-weight ratios have been derived from the
referenced values for illustrative purposes only.
4-58
-------
human lifespan. In these cases it may be reasonable to base the derivation
of criteria upon adult values of bw and I . In cases where effects have a
w
shorter latency (e.g., <10 years) and where children are known to be at
special risk, it may be more appropriate to use values for toddlers or
infants.
The approach presently employed in the derivation of recommended maximum
contaminant levels (RMCLs) by the U.S. EPA Office of Drinking Water is to
assume a bw and I of 70 kg and 2.0 i/day, respectively (Federal
w
Register, 1985), for adults and a bw and IM of 10 kg and 1.0 a/day,
respectively, for a child.
4.4.3.1.3. Total Background Intake Rate of Pollutant (TBI) — It is
important to recognize that sources of exposure other than sludge disposal
practices may exist, and that the total exposure should be maintained below
the RfD. Other sources of exposure include background levels (whether
natural or anthropogenic) in drinking water (other than groundwater), food
or air. Other types of exposure, due to occupation or habits such as
smoking, might also be included depending on data availability and regula-
tory policy. These exposures are summed to estimate TBI.
Data for estimating background exposure usually are derived from
analytical surveys of surface, ground or tap water, from FDA market-basket
surveys and from air-monitoring surveys. These surveys may report means,
medians, percentiles or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to
estimate TBI should be consistent with the value of bw. Where background
data are reported in terms of a concentration in air or water, ingestion or
4-59
-------
inhalation rates applicable to adults or children can be used to estimate
the proper daily background intake value. Where data are reported as total
daily dietary intake for adults and similar values for children are unavail-
able, conversion to an intake for children may be required. Such a conver-
sion could be estimated on the basis of relative total food intake or
relative total caloric intake between adults and children. ' •.
As stated in the beginning of this subsection, the TBI is the summed
estimate of all possible background exposures, except exposures resulting
from a sludge disposal practice. To be more exact, the TBI should be a
summed total of all toxicologically effective intakes from all nonsludge
exposures. To determine the effective TBI, background intake values (BI)
for each exposure route must be divided by that route's particular relative
effectiveness (RE) factor. Thus, the TBI can be mathematically derived
after all the background exposures have been determined, using the following
equation:
TBI
BI (nonsludge-
BI (food) derived water) BI (air)
RE (food) * RE (water) + RE (air)
RE (n)
(4-32)
where:
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
BI = background intake of pollutant from a given . exposure
route, indicated by subscript (mg/day)
RE = relative effectiveness of the exposure route indicated by
subscript (unitless)
4.4.3.1.4. Fraction of Ingested Water From Contaminated Source-- It
is recognized that an individual exposed to contaminated groundwater from a
landfill may not necessarily remain in the landfill proximity for 24 hours/
4-60
-------
day. However, if it is assumed that residential areas may be contaminated,
it is likely that less mobile individuals will include those at greatest
risk. Therefore, it is reasonable to assume that 100% of the water ingested
by the MEIs will be from the area of the landfill,.
4.4.3.1.5. Relative Effectiveness of Exposure (RE) — RE is a
unitless factor that shows the relative toxicological effectiveness of an
exposure by a given route when compared to another route. The value of RE
may reflect observed or estimated differences in absorption between the
inhalation and ingestion routes, which can then significantly influence the
quantity of a chemical that reaches a particular target tissue, the length
of time it takes to get there, and the degree and duration of the effect.
The RE factor may also reflect differences in the occurrence of critical
toxicological effects at the portal of entry. For example, carbon
tetrachloride and chloroform were estimated .to be 40% and 65% as effective,
respectively, by inhalation as by ingestion based on high-dose absorption
differences (U.S. EPA, 1984b,c). In addition to route differences, RE can
also reflect differences in bioavailability due to the exposure matrix. For
example, absorption of nickel ingested in water has been estimated to be 5
times that of nickel ingested in the diet (U.S. EPA, 1985d). The presence
of food in the gastrointestinal tract may delay absorption and reduce the
availability of orally administered compounds, as demonstrated for
halocarbons (NRC, 1986).
Physiologically based pharmacokinetic (PB-PK) models have evolved into
particularly useful tools for predicting disposition differences due to
exposure route differences. Their use is predicated on the premise that an
effective (target-tissue) dose achieved by one route in a particular species
4-61
-------
is expected to be equally effective when achieved by another exposure route
or in some other species. For example, the proper measure of target-tissue
dose for a chemical with pharrnacologic activity would be the tissue concen-
tration divided by some measure of the receptor binding constant for that
chemical. Such models account for fundamental physiologic and biochemical
parameters such as blood flows, ventilatory parameters, metabolic capacities
and renal clearance, tailored by the physicochemical and biochemical prop-
erties of the agent in question. The behavior of a substance administered
by a different exposure route can be determined by adding equations that
describe the nature of the new input function. Similarly, since known
physiologic parameters are used, different species (e.g., humans vs. test
species) can be modeled by replacing the appropriate constants. It should
be emphasized that PB-PK models must be used in conjunction with toxicity
and mechanistic studies in order to relate the effective dose associated
with a certain level of risk for the test species and conditions to other
scenarios. A detailed approach for the application of PB-PK models for
derivation of the RE factor is beyond the scope of this document, but the
reader is referred to the comprehensive discussion in NRC (1986). Other
useful discussions on considerations necessary when extrapolating route to
route are found in Pepelko and Withey (1985) and Clewell and Andersen (1985).
Since exposure for the groundwater pathway is by drinking water, the RE
factors applied are all with respect to the drinking-water route.
Therefore, the value of RE in Equation 4-31 gives the relative effectiveness
of the exposure route and matrix on which the RfD was based when compared to
drinking water. Similarly, the RE factors in Equation 4-32 show the
relative effectiveness, with respect to the drinking-water route, of each
background exposure route and matrix.
4-62
-------
An RE factor should only be applied where well-documented, referenced
information is available on the contaminant's observed relative
effectiveness or its pharmacokinetics. When such information is not
available, RE is equal to 1.
4.4.3.2. CARCINOGENS — For carcinogenic chemicals, the Agency con-
siders the excess risk of cancer to be linearly related to dose, except at
high-dose levels (U.S. EPA, 1986a). The threshold assumption, therefore,
does not hold, as risk diminishes with dose but does not become zero or
background until dose becomes zero.
The decision whether to treat a chemical as a threshold- or nonthresh-
old-acting (i.e., carcinogenic) agent depends on the weight of the evidence
that it may be carcinogenic to humans. Methods for classifying chemicals as
to their weight of evidence have been described by U.S. EPA (1986a), and
most of the chemicals that presently are candidates for sludge criteria have
recently been classified in Health Assessment Documents or other reports
prepared by the U.S. EPA's Office of Health and Environmental Assessment
(OHEA), or in connection with the development of RMCLs for drinking-water
contaminants (U.S. EPA, 1985e). To derive values of RWC, a decision must be
made as to which classifications constitute sufficient evidence for basing a
quantitative risk assessment on a presumption of carcinogenicity. Chemicals
in classifications A, and B, "human carcinogen" and "probable human
carcinogen," respectively, have usually been assessed as carcinogens,
whereas those in classifications D and E, "not classifiable as to human
carcinogenicity because of inadequate human and animal data" and "evidence
of noncarcinogenicity for humans," respectively, have usually been assessed
according to threshold effects. Chemicals classified as C, "possible human
4-63
-------
carcinogen," have received varying treatment. For example, lindane,
classified by the Human Health Assessment Group (HHA6) of the U.S. EPA as
B2-C, or between the lower range of the B category and category C, has been
assessed both using the linear model for tumorigenic effects (U.S. EPA,
1980b) and based on threshold effects (U.S. EPA, 1985e). Table 4-8 gives an
illustration of these U.S. EPA classifications based on the available weight
of evidence.
The use of the weight-of-evidence classification, without noting the
explanatory material for a specific chemical, may lead to a flawed conclu-
sion because some of the classifications are exposure-route dependent.
Certain compounds (e.g., nickel) have been shown to be carcinogenic by the
inhalation route but not by ingestion. The issue of whether or not to treat
an agent as carcinogenic by ingestion remains controversial for several
chemicals.
If a pollutant is to be assessed according to nonthresholld, carcinogenic
effects, the RWC is derived as follows:
Reference Water Concentration: RWC =
(RL x bw \ _
U * x RE/
TBJ:
w
(4-33)
where:
q-|* = human cancer potency [(mg/kg/day)"1]
RL = risk level (unitless) (e.g., 10~5, lO"6, etc.)
bw = human body weight (kg)
RE = relative effectiveness of exposure (unitless)
lw
water ingestion rate (8,/day)
TBI = total background intake rate of pollutant (mg/day); from
all other sources of exposure
4-64
-------
TABLE 4-8
,•;•'. ' f'
Illustrative Categorization of Evidence Based on Animal and Human Data*
Animal Evidence
Human
Evidence
Sufficient
Limited
Inadequate
No data
No evidence
Sufficient
A
81
B?
82
B2
Limited
A
81
C
C
C
Inadequate
A
81
0
D
D
No Data
A
81
D
D
D
No
Evidence
A
81
D
E
E
*The above assignments are presented for illustrative purposes. There may
be nuances in'the classification of both animal and human data indicating
that different categorizations than those given in the table should be
assigned. Furthermore, these assignments are tentative and may be modified
by ancillary ^vidence. In this regard all relevant information should be
evaluated to jetermine if the designation of the overall weight of evidence
needs to be mpdified. Relevant factors to be included along with the tumor
data from human and animal studies include structure-activity relationships;
short-term test findings; results of appropriate physiological, biochemical
and toxicologifal observations; and comparative metabolism and pharmaco-
kinetic studies. The nature of these findings may cause an adjustment of
the overall categorization of the weight of evidence.
4-65
-------
The RWC, in the case of carcinogens, is thought to be protective because the
q* is typically an upper-limit value (i.e., the true potency is
considered unlikely to be greater and may be less). The definition and
derivation of each of the parameters used to estimate RWC for carcinogens
are further discussed in the following subsections.
4.4.3.2.1. Human Cancer Potency (q *) — For most carcinogenic
chemicals, the linearized multistage model is recommended for estimating
human cancer potency from animal data (U.S. EPA, 1986a). When
epidemiological data are available, potency is estimated based on the
observed relative risk in exposed vs. nonexposed individuals, and on the
magnitude of exposure. Guidelines for use of these procedures have been
presented in the U.S. EPA (1980c, 1985e) and in each of a series of Health
Assessment Documents prepared by OHEA (for example, U.S. EPA, 1985c). The
true potency value is considered unlikely to be above the upper-bound
estimate of the slope of the dose-response curve in the low-dose range, and
it is expressed in terms of risk/dose, where dose is in units of mg/kg/day.
Thus, q.|* has units of (mg/kg/day)"1. OHEA has derived potency
estimates for each of the potentially carcinogenic chemicals that are
presently candidates for sludge criteria. Therefore, no new effort will be
required to develop potency estimates to derive sludge criteria.
4.4.3.2.2. Risk Level (RL) — Since by definition no "safe" level
exists for exposure to nonthreshold agents, values of RWC are calculated to
reflect various levels of cancer risk. If RL is set at zero, then RWC will
be zero. If RL is set at 10 6, RWC will be the concentration which, for
lifetime exposure, is calculated to have an upper-bound cancer risk of one
case in one million individuals exposed. This risk level refers to excess
4-66
-------
cancer risk, i.e., over and above the background cancer risk in unexposed
individuals. By varying RL, RWC may be calculated for any level of risk in
the low-dose region, i.e., RL <10~2. Specification of a given risk
level on which to base regulations is a matter of policy. Therefore, it is
common practice to derive criteria representing several levels of risk
without specifying any level as "acceptable."
4.4.3.2.3. Human Body Weight (bw) and Water Ingestion Rate (I ) —
w
As with toxicants, it is important to gear the selection of bw and I to
w
the nature of the effect of concern. Carcinogenesis normally has a long
latency period and, therefore, adult values of bw and I have usually been
w
applied. For example, the HHAG assumes 70 kg and 2 ,8,/day, respectively,
to derive unit risk estimates for drinking water, which are potency
estimates transformed to units of (vg/8,)"1. In addition, although
exposure is somewhat higher in children when expressed on a body-weight
basis (see Table 4-8), water ingestion occurs lifelong and groundwater
concentrations tend to change only very slowly.
4.4.3.2.4. Relative Effectiveness of Exposure (RE)— RE is a unit-
less factor that shows the relative toxicological effectiveness of an expo-
sure by a given route compared to another route. The value of RE may
reflect observed or estimated differences in absorption between the
inhalation and ingestion routes, which can significantly influence the
quantity of a chemical that reaches a particular target tissue, the length
of time it takes to get there, and the degree and duration of the effect.
The RE factor may also reflect differences in the occurrence of critical
toxicological effects at the portal of entry. For example, carbon
tetrachloride and chloroform were estimated to be 40% and 65% as effective,
4-67
-------
respectively, by Inhalation as by ingestion based on high-dose absorption
differences (U.S. EPA, 1984b,c). In addition to route differences, RE can
also reflect differences in bioavailability due to the exposure matrix. For
example, absorption of nickel ingested in water has been estimated to be 5
times that of nickel ingested in food (U.S. EPA, 1985d). The presence of
food 1n the gastrointestinal tract may delay absorption and reduce the
availability of orally administered compounds, as demonstrated for
halocarbons (NRC, 1986).
PB-PK models have evolved into particularly useful tools for predicting
disposition differences due to exposure route differences. Their use is
predicated on the premise that an effective (target-tissue) dose achieved by
one route in a particular species is expected to be equally effective when
achieved by another exposure route or in some other species. For example,
the proper measure of target-tissue dose for a chemical with pharmacologic
activity would be the tissue concentration divided by some measure of the
receptor binding constant for that chemical. .Such models account for
fundamental physiologic and biochemical parameters such as blood flows,
ventilatory parameters, metabolic capacities and renal clearance, tailored
by the physicochemical and biochemical properties of the agent in question.
The behavior of a substance administered by a different exposure route can
be determined by adding equations that describe the nature of the new input
function. Similarly, since known physiologic parameters are used, different
species (e.g., humans vs. test species) can be modeled by replacing the
appropriate constants. It should be emphasized that PB-PK models must be
used in conjunction with toxicity and mechanistic studies in order to relate
the effective dose associated with a certain level of risk for the test
4-68
-------
species and conditions to other scenarios. A detailed approach for the
application of PB-PK models for derivation of the RE factor is beyond the
scope of this document, but the reader is referred to the comprehensive
discussion in NRC (1986). Other useful discussions on considerations
necessary when extrapolating route to route are found in Pepelko and Withey
(1985) and Clewell and Andersen (1985).
Since exposure for the groundwater pathway is by drinking water, the RE
factors applied are all with respect to the drinking-water route.
Therefore, the value of RE in Equation 4-33 gives the relative effectiveness
of the exposure route and the matrix on which the q * was based when
compared to drinking water. Similarly, the RE factors in Equation 4-32 show
the relative effectiveness, with respect to .the drinking-water route, of
each background exposure route and matrix.
An RE factor should only be applied where well-documented, referenced
information is available on the contaminant's relative effectiveness or its
pharmacokinetics. When such information is not available, RE is equal to 1.
4.4.3.2.5. Total Background Intake Rate of Pollutant (TBI) — It is
important to recognize that sources of exposure other than sludge disposal
practices may exist, and that the total exposure should be maintained below
the determined cancer risk-specific exposure level. Other sources of
exposure include background levels (whether natural or anthropogenic) in
drinking water (other than groundwater), food or air. Other types of
exposure, due to occupation or habits such as smoking, might also be
included depending on data availability and regulatory policy. These
exposures are summed to estimate TBI.
4-69
-------
Data for estimating background exposure usually are derived from
analytical surveys of surface, ground or tap water, from FDA market-basket
surveys and from air-monitoring surveys. These surveys may report means,
medians, percentiles or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to esti-
mate TBI should be consistent with the value of bw. Where background data
are reported in terms of a concentration in air or water, ingestion or
inhalation rates applicable to adults can be used to estimate the proper
daily background intake value.
As stated in the beginning of this subsection, the TBI is the summed
estimate of all possible background exposures, except exposures resulting
from a sludge disposal practice. To be more exact, the TBI should be a
summed total of all toxicologically effective intakes from all nonsludge
exposures. To determine the effective TBI, background intake values (61)
for each exposure route must be divided by that route's particular relative
effectiveness (RE) factor. Thus, the TBI can be mathematically derived
after all the background exposures have been determined, using the following
equation:
BI (nonsludge-
TBI _
TBI ~
BI (food)
RE (food)
derived water) BI (air) BI (n)
+ RE (air) + ''' . RE (n)
RE (water)
(4-32)
where:
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
BI = background intake of pollutant from a given exposure route,
indicated by subscript (mg/day)
4-70
-------
RE = relative effectiveness, with respect to drinking-water
exposure, of the exposure route indicated by subscript
(unitless)
4.4.3.2.6. Fraction of Ingested Water From Contaminated Source — It
is recognized that an individual exposed to contaminated groundwater from a
landfill may not necessarily remain in the landfill proximity for 24 hours/
day. However, if it is assumed that residential areas may be contaminated,
it is likely that less mobile individuals will include those at greatest
risk. Therefore, it is reasonable to assume that 100% of the water ingested
by the MEIs will be from the area of the landfill.
For volatile contaminants, the estimated intake from inhalation is
converted to an equivalent ingestion dose in drinking water, which is added
to the background concentration in groundwater. This accounts for intake of
these contaminants via both routes simultaneously. Atmospheric
concentrations are calculated in vig/m3 and converted to equivalent
drinking water concentrations in mg/9. by assuming an individual breathes
20m3 of air per day and drinks 2 8, of water. Therefore,
1 yg/m3 of air results in a total intake of 20 yg from the air,
being equivalent to drinking 2 a, of TO yg/8. or 0.01 mg/9, water.
Therefore, atmospheric concentrations in micrograms/cubic meter are
multiplied by 0.01 to convert them to equivalent liquid concentrations
before adding them to the composite aquifer concentration (leachate from the
landfill plus background).
4.5. EXAMPLE CALCULATIONS
The methodology presented in the previous section can best be illus-
strated with an example calculation. In the following, calculations are
first made for an individual site as would be the case with a site-specific
4-71
-------
application. Then an example is given for development of criteria for maxi-
mum allowable contaminant levels in sludge. To illustrate the methodology,
the example considers an organic contaminant. Input parameters for the
example calculations are provided in Table 4-9.
4.5.1. Site-Specific Application. A step-by-step discussion of a site-
specific application follows. The application uses benzene as the constitu-
ent of interest. The input data are reiterated on the first pages of the
application.
The reference water concentration (RWC) for the carcinogen benzene is
derived using Equation 4-33:
RL x bw
RWC =
/RL x bw \_
\qf x RE/
TBI
•f I w
(4-33)
The risk level (RL), the body weight (bw) and the daily ingestion rate
(I ) are set for this example at 10
W
-6
70 kg and 2 it, respectively.
The relative effectiveness factor (RE) is set at 1. The human cancer
potency for benzene has been determined by the U.S. EPA to be 5.2x10
(mg/kg/day)
—3.
Current total background intake (TBI) of benzene from
all other sources, except from landfilling of sludges, has not been
determined for 1986, but for illustrative purposes a TBI of zero is used
here to derive an example RWC. Determination of an RWC for a specific
landfill site should be based on a current local assessment of TBI.
' 10 ~& x 70 kg
RWC =
,5.2 x 10-2 (mg/kg/day)-'
= 6.73 x 10~4 mg/5.
=0.673
- 0
* (2 a)
4-72
-------
TABLE 4-9
Input Parameters for Example Calculations -- Groundwater
Fate and Transport: Pathway Data
Source Term
1. Water content of sludge
2. Storage capacity of sludge
3. Density of sludge
4. Net recharge
Unsaturated Zone
5. Depth to groundwater
6. Distance to property boundary
7. Material
8. Thickness
9. Saturated soil hydraulic conductivity
10. Slope of matric potential and
moisture content plot
11. Saturated soil moisture content
12. Bulk density
Saturated Zone
13. Groundwater
14. Groundwater
15. Hydraulic conductivity
16. Effective porosity
17. Hydraulic gradient
WS = 0.95 kg/kg
S = 0.90 kg/kg
Ds = 1012 kg/m3
R = 0.5 m/year
hy = 1 m
ds = 100 IB
Sandy loam
m = 1 m
ksat =10* m/year
b = 4.0
fs = 0.39 ma/m3
Bu = 1400 kg/m3
pH = 6
Eh = 150 mv
K = 1.5x10s m/yr
ee = 0.10
(3H/3X) =0.003
4-73
-------
TABLE 4-9 (cont.)
Fate and Transport: Pathway Data
18. Bulk density
19. Dispersion coefficient
20. Site geometry
Landfill width
Landfill length
Fill height
Saturated Zone (cont.)
Bs = 2390 kg/m3
Dx = 4.5x10* mVyear
Dy = 4.5x103 ma/year
Dz = 4.5x103 ma/year
SW = 10 m
SL = 100 m
FH = 3.46 m
Chemical-Specific Data
Fate and Transport:
Source Term
21. Benzene
22. Benzene
Unsaturated Zone
23. Benzene
24. Benzene
Saturated Zone
25. Benzene
26. Benzene
Chemical-Specific Data — Health or Environmental Effects
Concentration in leachate (X) = 0.05 mg/8,
Concentration in sludge (N) =3 mg/kg
Distribution coefficient (Kd) = 7.4xlO"3 ft/kg
Degradation rate constant (\) = 3.9 year"1
Distribution coefficient (Kd) = 0.0 8,/kg
Degradation rate constant (\) = 0.0 year"1
27. Benzene
Reference water concentration (RWC)
0.000673 mg/a
4-74
-------
4.5.1.1, TIER 1 — Compare the concentration of each contaminant in
the leachate (X) to the RWC.
X (mg/a) RWC (tng/a.)
Benzene 0.05 0.000673
Continue with Tier 2 only if X exceeds RWC, which, in this case, is true for
benzene.
4.5.1.2. TIER 2 — Since benzene did not pass Tier 1, a Tier 2 analy-
sis is required. The procedure that would be followed in a Tier 2 analysis
is presented below in a step-by-step fashion.
4.5.1.2.1. Sludge/Landfill Calculations —
A. Determine the weight of sludge solids/m2 of fill as:
MS = (FH) (Os) (1 - W$)
where:
Ms = weight of sludge solids (kg/m2)
FH = fill height (m)
Ds = density of sludge (kg/ma)
Ws - water content of sludge (kg/kg)
Ms = (3.46 m) (1012 kg/m3) (1 - 0.95)
= 175.08 kg/m2
B. Calculate the mass of each contaminant/m2 of fill material as:
M = (Ms) (N)
where:
M = mass of contaminant in fill (g/m2)
N = dry weight contaminant concentration in sludge (mg/kg)
Benzene
M = (175.08 kg/m2) (3.0 mg/kg) (0.001 g/mg)
=0.53 g/m2
4-75
-------
C. Calculate the volume of water/m present in the fill at the time
of disposal as:
= C(WS)
- ws)]
where:
VW1 * volume of water/m2 present in the fill (m3/m2)
DW =* density of water (kg/m3)
VW1 = [(0.95) (175.08 kg/ma)]/[(1000 kg/m3) (1 - 0.95)]
= 3.33 mVm2
D. Calculate the volume of water/m2 present in the fill after the
sludge drains as:
VW2 = C(S)
where:
2 .
VW2 = Vo1ume of water/m in the fill after it drains (m /m )
S = storage capacity of sludge (kg/kg)
Vw? - [(0.9) (175.08 kg/m2)]/[(!000 kg/m3) (1 - 0.9)]
=1.58 ma/m2
E. Calculate the volume of water/m that will drain from the sludge
as:
where:
D = volume of water/m that drains from the sludge (m /m )
= 3.33 m3/m2 - 1.58 m8/ma
= 1.75 m3/m2
F. Calculate the Teachable mass/m for each contaminant as:
ML = M - (X) (Dv)
4-76
-------
where:
ML = Teachable mass/m*
X = contaminant concentration in leachate (mg/5t)
Benzene
ML = 0.53 g/m2 - (0.05,g/m3) (1.75 m3/m2)
= 0.44 g/m
6. Calculate the period of time needed sto leach the leachable mass from
the landfill. For nondegradable contaminants the formula is:
Tp = ML/(R) (X)
For degradable contaminants the formula is: . ;
Tp = In {[(R) (X)]/[(R) (X) + (\) (ML)]
where:
Tp = leach time (pulse time) (years)
R = net recharge (m/year)
X = leachate concentration (g/ma = mg/S.)
\ = degradation rate constant (year"1)
Benzene
Tp -
1
In
(0.5 m/year) (0.05 g/m3)
3.9 year-i (0.5 m/year) (0.05 g/ms) + (3.9 year-i) (0.44 g/m*)
=1.09 year
4.5.1.2.2. Unsaturated Zone Calculations —
A. Calculate the steady-state moisture content of the soil for each
layer in the unsaturated zone as:
/ R \ [l/(2b + 3)]
= s IK1 '
where:
f = steady-state moisture content (m3/m3)
4-77
-------
fs = saturated moisture content for the unsaturated zone soil
Ksat - saturated hydraulic conductivity of the unsaturated zone
soil (m/year)
b = negative 1 times the slope of the log-log plot of matric
potential and saturated moisture content (dimensionless)
Layer 1
f = 0.39
= 0.16 m3/m3
0.5 m/year 1/C(2)(4) + 3]
10* m/year
(If multiple layers are present in the unsaturated zone, solve for each
layer.)
B. Calculate the steady-state travel time of the water across each
unsaturated zone soil layer as:
TU = (hy) (e)/R
where:
Tu - steady-state travel time across an unsaturated zone soil
layer (years)
hy = depth to groundwater or thickness of the unsaturated zone
beneath the landfill (m)
Layer 1
3 , 3,
Tul = (1.0 m) (0.16 m /m )/(0.5 m/year)
= 0.32 year
(If multiple layers are present in the unsaturated zone, solve for each
layer.)
C. Calculate the total travel time of the water across the unsaturated
zone as:
TT=Tul +Tu2+-
4-78
-------
where:
T = total travel time across all layers of the unsaturated zone (years)
T = 0.32 year
0. Calculate the average velocity of the water across the unsaturated
zone as:
where:
vave = average velocity across the unsaturated zone (m/year)
hy# = thickness of each unsaturated zone layer where #=1,2, ...
V =1.0 m/0.32 year
ave
=3.14 m/year
E. Calculate the average moisture content of the unsaturated zone as:
f = R/V
ave ave
where:
fgve = average moisture content of the unsaturated zone (m3/m3)
rave =
0.5 m/vear
3.14 m/year
= 0.16 mVm3
F. Obtain the unsaturated zone distribution coefficient, Kd, and
unsaturated zone degradation rate for each contaminant from the
scientific literature. Select Kd values based on soil type. If
there is more than one type of soil (i.e., more than one layer) in
the unsaturated zone, use a weighted average for Kd. Calculate the
weighted average by summing the products of the thickness of each
layer and the Kd for each soil type (layer) and dividing by the
total thickness of all layers.
4-79
-------
Contaminant Kd (it/kg)
Degradation Rate (year"1)
Benzene 7.4x10~3 3.9
G. Calculate the retardation factor for the contaminant in the unsatu-
rated zone as:
RF = 1 + (B Kd/f )
v u ave'
where:
RF = retardation factor (dimensionless)
Bu = bulk density of unsaturated zone material (kg/m3)
Kd = distribution coefficient (8,/kg)
If multiple layers are present in the unsaturated zone, calculate a
weighted average retardation factor using weighted average values for
bulk density and saturated moisture content (weight according to layer
thickness).
Benzene
7.4xlO~3 Si/kg) (0.001 m3/9,) (1400 kg/m3)
RF » 1 +
0.39 ma/m3
= 1.03
H. For degradable contaminants, calculate the concentration of contami-
nant leaving the unsatura^ed zone accounting for degradation as:
X exp [(-1) (\) (TT) (RF)]
where:
C = contaminant concentration exiting the unsaturated zone (mg/8.)
Benzene
l
C - 0.05 mg/8. exp [(-1) (3.9 year ) (0.32 year) (1.03)]
us
= 0.014 mg/9,
4.5.1.2.3. HINTEQ Adjustment for Geochemical Reactions — For
metallic contaminants, determine the amount of concentration reduction that
4-80
-------
will occur due to geochemical reactions in the saturated flow system
(aquifer) using the MINTEQ graphs given in Appendix B (Figures B-2 through
B-6).
4.5.1.2.4. Tier 2 Intermediate Comparison to Reference Water
Concentrations — Compare the Tier 2 concentrations at the base of the
unsaturated zone, as calculated without allowing for dispersion, with the
reference water concentration. The final Tier 2 concentrations for benzene
are given in H on the preceding page.
Benzene
Concentration after allowing for decay = 0.014 mg/H
Reference Water Concentration^ 0.000673 mg/5.
For benzene, the Tier 2 concentrations at the base of the unsaturated
zone, without accounting for dispersion, are greater than the reference
water concentration. Therefore, the Tier 2 analysis needs to be continued
using the CHAIN and AT123D models.
4.5.1.2.5. CHAIN Model —Since the Tier 2 result presented above is
not below the reference water concentration, it is necessary to run the
CHAIN analytical transport model to predict the contamination concentration
at the base of the unsaturated zone. The input data required by the CHAIN
model are the leachate concentration (X), the net recharge (R) as the flux
rate, the leach time out of the landfill (T ) as the input pulse time, the
retardation factor (RF) and the degradation rate (X). The dispersion
coefficient used in CHAIN is calculated as one-tenth the depth to ground-
water (unsaturated zone thickness, hy) times the average groundwater veloc-
ity in the unsaturated zone (V ).
ave
For organic contaminants, compare the maximum model-predicted, concentra-
tion to the reference water concentration values. If the model-predicted
4-81
-------
concentration is less than the reference water concentration, no further
Tier 2 analysis is required. If not, the saturated transport model ATI230
should be run.
For inorganic contaminants, take the maximum model predicted concentra-
tions and enter the appropriate curves in Appendix B (Figures B-2 through
B-6) to predict the resulting concentration after geochemical reaction. If
the resulting concentration is less than the threshold, no further Tier 2
analysis is required. If not, the resulting concentrations should be
entered into the saturated transport model AT123D.
The CHAIN model results for this example problem are as follows:
Dependable Organics
Benzene
Unsaturated Model
Results (mg/g,)
Reference Water Concentration
(mg/il)
0.015 0.000673
The unsaturated model results for benzene are above the reference concen-
tration value; therefore, the saturated transport code AT123D needs to be
run for benzene.
4.5.1.2.6. AT123D Model -- Since the CHAIN model results are not
below the reference concentration for benzene, it is necessary to run the
AT123D saturated zone transport model to predict the contaminant
concentration at the facility boundary. The peak contaminant concentrations
and the pulse time as predicted by the CHAIN model are used as inputs to
AT123D. The other input data required by ATI230 are the degradation rates
(X), the retardation factors (calculated for the saturated zone using the
saturated distribution coefficient, bulk density and porosity), the
groundwater velocity (calculated as the saturated hydraulic conductivity
times the hydraulic gradient divided by the effective porosity), the
longitudinal dispersion coefficient (calculated as one-tenth the distance to
4-82
-------
the landfill boundary times the groundwater velocity), and the transverse
and vertical dispersion coefficients (calculated as one-tenth the
longitudinal dispersion coefficient).
The AT123D model results for the example problem are as follows:
Saturated Model Reference Water Concentration
Degradable Organics Results (mg/g.) (mg/a.)
Benzene 9.8xlO~8 0.000673
The saturated model results for benzene are well below the reference
water concentration. Therefore, this application would be acceptable. If
the predicted output concentrations were very close to the criteria values,
the permit writer may require a characterization of input parameter uncer-
tainty and additional runs to determine sensitivity to that uncertainty.
If background concentrations of benzene are present in the groundwater,
these concentrations would be added to the saturated model results and this
total concentration would be compared to the reference water concentration.
4.5.2. National Criteria Site-Specific Application. In order to set
national criteria, the methodology is applied in reverse order with the same
site- and chemical-specific inputs. In this case the starting point is the
environmental concentration (EC) criteria and the endpoint is the acceptable
leachate concentration or the acceptable amount of total contaminant in the
landfill. Example calculations for the trial scenario follow for benzene.
The input data for this application are provided in Table 4-10.
Benzene
The first step of the methodology was to run the CHAIN code to predict
the peak benzene concentration and the release duration of benzene into the
saturated flow system. Pulse times of 0.01, 0.1, 1 and 10 years and contam-
inant concentrations of 0.00079 and 0.0079 mg/9. were used as input to
4-83
-------
TABLE 4-10
CHAIN Model Results for the National
Criteria Calculation for Benzene
Pulse Time
(years)
0.01
0.01
0.1
0.1
1.0
1.0
10.0
10.0
Input
Concentration
(mg/8.)
0 . 00079
0.0079
0.00079
0.0079
0.00079
. 0.0079 t
0.00079
0.0079
Peak Output
Concentration
(mg/9.)
0.11x10-*
O.llxlO-3
0.98x10-*
0.98xlO~3
0.24xlO~3
0.24xlO~2
0.24xlO-g
0.24x10-2
Release Duration
(years)
0.7
0.7
0.8
0.8
1.6
1.6
10.0
10.0
4-84
-------
CHAIN. Pulse times of 0.01 and 0.1 years are not realistic, but they were
simulated for illustrative purposes since, as will be discussed, all pulse
times >1 year yield the same peak concentration at the facility boundary.
The CHAIN model results are listed in Table 4-11. The peak concentra-
tion for pulse times of 1 and 10 years are identical and would be the same
for all pulse times >10 years. The reason for this is that the benzene
travel time through the unsaturated system is short (0.3 years) compared to
pulse times >1 year; therefore, there is little dispersive effect and an
equilibrium concentration is reached in the flow system.
j. '
The second step is to use the peak concentration and release time from
.CHAIN, as the input concentration and pulse time to AT123D. Since AT123D
requires an input flux rate, the actual input to AT1230 is the peak output
concentration from CHAIN multiplied by the recharge rate (0.5 m/year).
The output from AT123D is a series of peak concentrations (X) for
each of the cases simulated as shown in Table 4-11. Table 4-11 also lists
the pulse times and input concentrations (X.) used in the CHAIN model.
The third step of the national criteria methodology is to plot the pairs
of points (X. vs. X ) for identical pulse times in Table 4-11. The data
points produce three curves as shown in Figure 4-5; the curves for pulse
times of 1 and 10 years overlay each other. The curves for 1 and 10 years
also represent all pulse times >10 years.
The maximum benzene concentration for all pulse times >1 year that would
not exceed the health effects criteria of 0.000673 mg/a. can be determined
as follows. Locate the point on the 1-year pulse time curve whose abscissa
is equal to the health effects criteria for benzene. The ordinate of this
point is the maximum allowable leachate concentration for pulse times >1
4-85
-------
TABLE 4-11
AT123D Model Results for the National
Criteria Calculation for Benzene
Pulse Time
(years)
0.01
0.01
0.1
0.1
1.0
1.0
10.0
10.0
CHAIN Model Input
Concentration (X^)
(mg/a,)
0.00079
0.0079
0.00079
0.0079
0.00079
0.0079
0.00079
0.0079
ATI 230 Model Peak Output
Concentration (Xf)
(mg/!l)
6.7xlO~12
6.7x10-"
6.0X1Q-11
6.0xlO"10
1.5xlO~10
1.5xlO~9
l.SxlQ-10
l.SxlO"9
4-86
-------
Benzene
Outflow Concentration (mg//)
FIGURE 4-5
Graph of the Family of Curves for the Benzene
National Criteria Calculations
4-87
-------
year (see dotted lines in Figure 4-5). The maximum allowable benzene
concentration in the landfill is 4000 mg/a,. Pulse times of <1 year are
probably not realistic and are not considered.
To check this result, the CHAIN and ATI230 models were run with a pulse
length of 1 year and a benzene input concentration of 4000 'mg/a,. The peak
concentration at the property boundary was calculated to be 0.00075 mg/a.,
just above the health effects criteria limit.
The final step of the calculation was to determine the total mass (M) of
benzene that could be present in the landfill and still meet the health
effects criteria at the property boundary. The total Teachable mass (M )
of benzene was calculated as:
ML =
R (eKTP-l)
(4000 mg/il) (0.5 m/vear) Fe <3-9 Vear ^ (] year)- TJ
s
_ 3.9 year-*
= 24,800 g/m
The total mass of benzene in the landfill was calculated as:
M = M, + X. D
L i v
- 24,800 g/m2 + (4000 mg/fc) (1.75 m3/m2)
- 31,800 g/m2
The total mass of benzene that can be present in the landfill and still
be close to the health effects criteria at the boundary is quite large due
to the rapid decay time for benzene, which has a half-life of approximately
65 days.
It should be noted that benzene solubility in water is reported as 820
mg/a. Therefore, if other sludge constituents do not significantly
increase benzene's solubility, these criteria will not restrict the land-
filling of any benzene-containing sludges for this set of site conditions.
4-88
-------
5. METHODOLOGY FOR PREDICTING THE VAPOR CONTAMINANT PATHWAY
5.1. OVERVIEW OF THE METHOD
Vapor loss from the landfill has been identified as a possible pathway
of concern for migration of certain volatile toxic chemicals (e.g., benzene,
cyanide, dimethylnitrosamine and trichloroethylene) from sludge disposal/
reuse. In concert with the risk assessment framework provided here, the
tiered approach and concern for chronic exposure, three levels of analysis
are outlined. As in the case of groundwater contamination, the initial tier
is a simple comparative structure that can be implemented to quickly screen
a chemical for the landfilling option. Contaminants failing Tier 1 can be
evaluated at Tier 2 to consider site-specific conditions.
Regardless of the level of assessment (Tiers 1 through 3), the basic
approach requires some degree of simulation of the movement of vapors up
from the waste into the atmosphere and downwind to the property boundary.
The property boundary has been selected as the point of compliance, since
that would be the first area of unrestricted access where continuous expo-
sure is likely to occur. Workers would be exposed potentially to higher
concentrations on the landfill property, but that contact,is regulated under
OSHA, would be limited to a 40-hour work week, and should be controlled
through use of respirators as appropriate. Chronic risk at the property
boundary is measured against the selected health effects threshold value.
This approach parallels that taken to evaluate landfill bans and exemption
requests for hazardous wastes.
The tiered approach and sequencing of the overall methodology is illus-
trated in Figure 5-1. Tier 1 involves a simple partial pressure calculation
using Henry's Law to predict maximum potential vapor levels above the
5-1
-------
Health Criteria
For Each Contaminant i
Calculate Equilibrium
Vapor Concentration
Yes
Calculate Boundary
Line Concentration
Yes
Experimentally Derive
Vapor Flux
Yes
Calculate Boundary
Line Concentration
Yes
Develop New
Option
No
No
No
No
»• End
» End
-»» End
"*»• End
*(Xi) = Concentration of Contaminant i in Atmosphere
FIGURE 5-1
Logic Flow for Vapor Loss Pathway Evaluation of Landfilled Sludges
5-2
-------
sludge. This is a very conservative estimate of concentration, since it
does not account for air/sludge matrix partitioning or dispersion in the
atmosphere. If the Tier 1 concentrations are lower than the threshold
value, no further evaluation of the contaminant is necessary. If the pre-
dicted concentration exceeds the threshold value, the applicant may opt to
proceed to Tier 2 where transport through the soil cover and atmospheric
dispersion are taken into consideration.
The Tier 2 analysis employs an analytical model developed to evaluate
vapor loss and dispersion from hazardous waste sites as a part of the land-
fill ban analysis (Environmental Science and Engineering, 1985). Elements
of the model consider those periods when the cell face is open, the period
of temporary cover and the postclosure period. Degradation and deposition
are not accounted for since travel times will be relatively short.
The procedures and details of each tier in the methodology are described
in the following sections.
5.2. ASSUMPTIONS
In order to apply a methodology such as that presented here, it is nec-
essary to make simplifying assumptions. The assumptions, stated or implied,
required to implement the vapor pathway analysis are outlined in Table 5-1
and discussed in the following sections.
5.2.1. Vapor Pressure. The Tier 1 and Tier 2 methodologies require the
vapor pressure (vapor concentration) of the contaminant to be specified.
Because vapor pressures are not routinely measured, the methodologies use
Henry's Law to specify vapor concentration as a function of liquid concen-
tration. Henry's Law is most appropriate for low aqueous concentrations and
low solids content sludges. As the concentrations and solids contents in-
crease, Henry's Law will tend to overpredict vapor pressure as a result of
5-3
-------
CD
to
to
r— Q.
in
LU
CQ
re
re
O
to
a.
re
cu
re
TT r— •
C r— to |
co in -I- c re
c 3 o i-
10 O •!— OJ
e .r- -rs -p Q.
o -p o re £
•r- CO JC i- CD
+J S_ +J +J +J
re -P cu c
&- c £ cu
-a
c
to
re
o>
cu
o o
co - c i-
cu c cu o
O O t/1 O
CD
CU
o re
Q. o
X S-
O> Q.
o i/i &- o •!— cu
re
i— • o .c
to to •(-> to o
i_ +J +J o I—
o c e •!- re cu
u- CD cu -o s-
•»-> 4-> CD i— 3
CD C C 5- i— to
•4-> O O Q.T- to
re o o s- 3 cu
•i- CD S-
i- t/» tO > tO CL
CLT3 T3 O C
O T- -r- O T3
i_ r- r- O -I- C
o. o o +J +-> re
to
re
CD i.
i- CU
CL-I-
i- K-
CU
O •)-»
Q. on to
(O
3 -o <= s- s-
o c QJ re rj
< r— re ••->>+->
i- en
o. c
re Q.-I-
'i— cu re r—
V) >,
re JD
•a c:
C
C O
re o
O =J 3
•i- cr re
+->•!- _J
rer—
S- V)
•»-> y —
c o >>
CD i-
u c c
C O CD
O 3
Q.U- CD
re i/>
> re 3
S- O
re o
re
1/1 a> -i—
V) > O
o o to
CD
re -
er>
3
o o
to S_
I/I <4- 4-»
CD O
to CD to
re e 3
CD U-
c\i o y—
re T-
cu 'o.
l— cu JQ
i- a
cu
re
cu
•*->
re
OJ
JQ
o
n.
0} i-
X
•r- OJ
T3 0
a; a>
O CU •r- CD
i- +•> -P i-
cn to ex 3
•r- £ to
r— to 3 O
r— C in O.
•r- O VI X
SO
re •
"O O a>
o. c:
+j cu -i—
c o i—
re CD s~
4-> 1- CO
o re cu
en
TS c
co
re i —
re o -o
•i— C
CM 4-» -I—
o 3
S- CU C
CD I- 3
x w
re c:
E o
-)-> re
o s-
••- •)->
•a c:
C
CL'i-
E 3
=9 C
oo 3
cu
re
CO
re
cu
re o
CM -4->
cu c
•r— •<— •<— «r- O -r- O
a.
re
5-4
-------
activity effects and partitioning between solid and liquid phases. As such,
the use of Henry's Law represents a conservative approach.
5.2.2. Loss Rate. The Tier 2 methodology assumes that the loss rate of
contaminants following emplacement of the soil cover is controlled by
diffusion of contaminants through the soil. The loss rate is then
independent of wind speed. This assumption is not appropriate for
describing direct volatilization from solid or liquid surfaces and, there-
fore, is not used to describe losses during the active period of disposal.
The assumption is appropriate for describing volatile losses when contami-
nants must first diffuse to the atmosphere and is, therefore, appropriate
for describing losses from a landfill. It is, however, recognized that some
absorption and degradation of organic vapors within the soil cover would
occur, thereby decreasing the concentration of air emissions.
Unfortunately, little is known of these processes, so the conservative
approach is taken here.
The Tier 2 methodology assumes that the final soil cover applied to a
landfill cell has the same permeability as the temporary soil cover. In
practice, the final soil cover should not be more permeable than the tempo-
rary cover and will usually be less permeable. This assumption, therefore,
will lead to an overprediction of loss rates.
5.2.3. Atmospheric Transport. The Tier T methodology assumes no atmos-
pheric dilution of contaminants. The result of this assumption will be to
grossly overpredict atmospheric contaminant concentrations. This approach
is clearly conservative and is consistent with the Tier 1 approach.
To simplify use of the atmospheric transport model in Tier 2, it is
assumed that the wind speed and direction are constant and that the receptor
5-5
-------
of concern is always located downwind along the center line of the plume.
The effect of these assumptions is to predict the maximum possible downwind
atmospheric concentrations and, therefore, the maximum possible exposure.
This conservative approach is consistent with the other exposure
methodologies.
In applying the atmospheric model, it is assumed that stable atmospheric
conditions will always be encountered. The effect of this assumption will
be to maximize the resulting downwind concentrations, thereby predicting the
maximum possible exposure levels. Again, this conservative approach is con-
sistent with other exposure methodologies.
5.3. CALCULATIONS
5.3.1. Tier 1. The first tier embodies a simple comparison of source
term vapor concentrations to the threshold value. Source term vapor concen-
trations are predicted on the basis of sludge contaminant concentrations and
Henry's Law. This does not account for any dispersion in the atmosphere
and, thereby, overpredicts concentrations. Henry's Law describes vapor
compositions over dilute solutions. The relation is given as::
P. - H.C1.
(5-1)
where:
P-J ~ partial pressure of i above the solution (atm)
Hj = Henry's Law Constant for i (atm-ma/mol)
Cl-j » concentration of i in the solution (mol/ma)
Assuming the vapor phases act as ideal gases, the partial pressure can be
translated into an atmospheric concentration using Oalton's Law:
P1 = VjP (5-2)
where:
y-j = mole fraction of i in the gas phase (dimensionless)
P = total pressure in the system (atm)
5-6
-------
For the landfill environment of interest here, P can be set at 1 atm. Then
combining Equations 5-1 and 5-2, atmospheric concentration (y.) can be
calculated as:
Py. = H.C1. (5-3)
With the molecular weight of the contaminant and air and the molar volume of
air, this can be converted to an atmospheric concentration in terms of
weight fraction or mass per volume.
An alternate approach is to use a dimensionless modified Henry's Law
Constant (H1) defined as:
H1 = Cv./Cl.
(5-4)
where:
Cv-j = concentration of i in air (mass/volume)
Cl-j = concentration of i in water (mass/volume)
This eliminates the need for conversions to obtain the atmospheric concen-
tration of the contaminant in the desired units. H can be converted to H'
by using the Universal Gas Law to calculate atmospheric concentration
(mol/volume) from partial pressure.
The use of the Henry's Law approach assumes ideal gas behavior and
dilute solutions. Both assumptions are appropriate for the levels of vola-
tile contaminants expected in municipal sludge, since handling and treatment
prior to disposal are likely to have allowed high concentrations to diminish
through the vaporization process. In empirical work with municipal sludge,
English et al. (1980) found Henry's Law to be useful in predicting atmos-
pheric concentrations of ammonia. Values for the Henry's Law Constant can
be obtained from the literature or calculated from physical properties.
Henry's Law Constants and modified Henry's Law Constants for contaminants of
interest are provided in Appendix B.
5-7
-------
The Henry's Law approach is likely to overpredict vapor concentrations
because the organic solids in sludge will bind some of the contaminants,
making them less available in the water solution for volatilization.
Because the prediction is conservative, it provides a greater margin of
safety. Underprediction would occur if the contaminant were present in the
liquid phase only and the solids comprised a major fraction of the overall
solution. This is not the case for the sludges anticipated. The estimated
vapor concentrations are compared to the appropriate reference air
concentration (RAC). If the vapor concentrations do not exceed the RAC, no
further analysis is required. If the vapor concentrations do exceed the
RAC, the applicant can decide if the greater accuracy of Tier 2 or 3 is
likely to be advantageous and, therefore, worth the added cost.
5.3.2. Tier 2. The first-tier methodology treats the landfilled sludge
as though it were a solution in a surface impoundment. The vapor concentra-
tions of contaminant are a result of direct volatilization from the surface
and subsequent diffusion into the air column. In reality, the sludge will
reside in a landfill cell. The active face of the cell may remain uncovered
for short periods of time (<8 hours), but will soon receive a layer of soil
overburden that will remain intact throughout the postclosure period. In
some cases, a tighter capping material will be added and vegetation estab-
lished as a part of the final cover. In either event, a finite layer of
soil will reside between the sludge's surface and the air column. All
vapors lost from the sludge must migrate through the cover to reach the
atmosphere and be available for downwind transport.
In considering vapor loss from hazardous waste landfills,, Environmental
Science and Engineering (1985) depicted three discrete phases for which pre-
dictive constructs were devised: operating period with uncovered wastes,
5-8
-------
operating period with shallow temporary cover, and postclosure period with
permanent cover. Since regulations do not require impermeable caps for
sludge or municipal refuse landfills, the permanent cover may not differ
significantly from the temporary cover with respect to vapor losses. The
permanent cover will likely include revegetation and a greater total
thickness. The analytical methodology used to predict vapor migration
through the temporary cover is the same as that used for the permanent
cover. Therefore, two analytical procedures are presented for the Tier 2
evaluation, one for predicting vapor loss from an open landfill cell and one
for predicting vapor migration through a soil cover.
Two types of exposure are evaluated during the Tier 2 analysis. The
first evaluation addresses the exposure due to losses during the active
operating period for a landfill cell. This exposure will be characterized
by relatively high loss rates (primarily from the uncovered waste) and rela-
tively small surface areas (i.e., the area of the active cell). The second
evaluation will be of the exposure due to losses from the covered wastes
during the postclosure period. This exposure is characterized by relative-
ly low loss rates from the covered wastes and relatively large surface areas
(i.e., the surface area of the entire closed landfill).
According to Environmental Science and Engineering (1985), the loss rate
from uncovered wastes during the active life of the landfill can be calcu-
lated as:
0.17 v (0.994)T"20 Cv-j
(5-5)
where:
= emission rate during active uncovered period for contami
nant i (g/mz-sec)
5-9
-------
v - windspeed (m/sec)
T = temperature (°C)
T-20 « a temperature correction factor derived empirically
Cv-j = vapor concentration of contaminant i (g/m3)
MWi = molecular weight of contaminant i
From Equation 5-3, Cv.. can be determined from the liquid concentration of
contaminant i, CK and Henry's Law Constant for contaminant i, H..
Equation 5-5 can then be expressed as:
0.17 v (0.994)1"-20 H.C1.
qai =
(5-6)
The emission rate for vapors emanating through cover materials during
the active and postclosure period can be calculated as (Environmental
Science and Engineering, 1985):
where:
qpi =
9.2 x IP"5 na10/3 CviQ.006)7 20
tc n2
(5-7)
Ipi
Demission rate through landfill cover to contaminant i
(g/mz-sec)
na = air filled porosity of cover soil (cma/cm3)
Cvi = vapor concentration
T = temperature (°C)
T-20 = a temperature correction factor derived empirically
tc = thickness of cover (m)
n = total porosity of cover soil (cma/cm3)
MW^ = molecular weight of contaminant i
5-10
-------
As with Equation 5-6, the vapor concentration can be expressed in terms of
the liquid concentration and Henry's Law Constant to yield:
9.2 x 10~s nj°/3 (1.006)1'20 H.C1.
a I!
qpi =
tc n2
(5-8)
The contaminant emission rate for the active and postclosure periods
calculated using either Equation 5-6 or 5-8 is used to calculate
atmospheric concentrations at the compliance point using a source-receptor
ratio (SRR).*
C(X)i = q x SRR (5-9)
where C(X)i is the atmospheric concentration (yg/m3), X is the down-
stream distance from the source to the receptor (m) and SRR is calculated as
(Environmental Science and Engineering, 1985):
SRR = 2.032 x 106 [•
where:
Xn
(r1 + Xy) v Oz)
(5-10)
the characteristic length of the landfill assumed to be a square
(m)
V = vertical term which is a function of source height and z
r1 =
X =
oz
distance from the landfill center to the receptor or point of
compliance (m)
lateral virtual distance (m)
mean wind speed (m/s)
standard deviation of the vertical concentration distance (m)
*If the landfill is vent, the total emission Qpc.j = qpc-| (1-fv) + qvent-
qvent is derived as qvent = [0.082U/ MW H'C^ (0.994) T-20, and fv is the
fraction of the fill that is vented.
5-11
-------
The vertical term for gases, V, is set at
M 1.6
2L
and
V = e
-1/2
n=l
2e
(5-11)
(5-12)
where L = mixing layer height (m).
The lateral virtual distance, Xy, is calculated as:
x o \Cot
~
(5-13)
where A0' = sector width (radians) for 22.5° 0' = 0.393.
The standard deviation in the vertical distance, a , can be taken from
tables for various distances and stability classes or calculated as indi-
cated in Table 5-2.
5.3.3. Procedure. In order' to establish sludge concentration criteria
for volatile contaminants, it is once again necessary to operate the method-
ology provided here in a reverse mode just as described for the groundwater
pathway in Chapter 4. That is, the RAC must be taken as input to determine
the maximum allowable concentration in the source sludge. From Equation
5-9, the compliance point concentration is defined as:
C = Q x SRR (5-14)
Since SRR is characteristic of a site and not concentration dependent, a
single value can be calculated for a representative site. When C is set at
C__, the effects threshold concentration, the allowable flux, Q, is
11
defined as:
Q = C£T/SRR
(5-15)
5-12
-------
TABLE 5-2
Parameters Used to Calculate az a
Pasquill
= a x (mb)*3
<_ruuu i i i ujr ouv^^vi y
Very unstable^
Unstable13
Slightly unstable13
Neutral
Slightly stable
f\ y iMiiy
0.10 - 0.15
0.16 - 0.20
0.21 - 0.25
0.26 - 0.30
0.31 - 0.40
0.41 - 0.50
0.51 - 3.11
3.11
0.10 - 0.20
0.21 - 0.40
0.40
0.10
0.10 - 0.30
0.31 - 1.00
1.01 - 3.00
3.01 - 10.00
10.01 - 30.00
30.00
0.10 - 0.30
0.31 - 1.00
1.01 - 2.00
2.01 - 4.00
4.01 - 10.00
10.01 - 20.00
20.01 - 40.00
40.00
a
158.080
170.222
179.520
217.410
258.890
346.750
453.850
+
90.673
98.483
109.300
62.141
34.459
32.093
32.093
33.504
36.650
44.053
23.331
21.628
21.628
22.534
24.703
26.970
35.420
47.618
b
1.04520
1.09320
1.12620
1.26440
1.40940
1.72830
2.11660
+
0.93198
0.98332
1.09710
0.91465
0.86974
0.81066
0.64403
0.60586
0.56589
0.51179
0.81956
0.75660
0.63077
0.57154
0.50527
0.46713
0.37615
0.29592
5-13
-------
TABLE 5-2 (cont.)
Pasquill
= a x (mb)')
Stable 0.10
0.21
0.71
1.01
2.01
3.01
7.01
15.01
30.01
60
- 0.20
- 0.70
- 1.00
- 2.00
- 3.00
- 7.00
- 15.00
- 30.00
- 60.00
.00
a
15.209
14.457
13.953
13.953
14.823
16.187
17.836
22.651
27.084
34.219
b
0.81558
0.78407
0.68465
0.63227
0.54503
0.46490
0.41507
0.32681
0.27436
0.21716
aSource: Environmental Science and Engineering, 1985
&If the calculated value of oz exceeds 5000 m,
-------
It has been found that the flux during the active life of the facility is
greater than that during postclosure, and will, therefore, be the limiting
factor. From Equation 5-6:
0.17 v (0.994)
T-20
(5-16)
where:
qai - allowable flux during the active period for contaminant i
(g/ma-sec)
v = windspeed (m/sec)
T = temperature (°C)
H^ = Henry's Law Constant (dimensionless)
Cli = concentration of contaminant in the sludge liquid (mg/Sl)
MWi = molecular weight of contaminant
Combining Equations 5-6 and 5-15 for the criteria case yields:
0.17 v (0.994)1"-20 H^Cli
CET/SRR =
(5-17)
which can be solved for Cl.. to give Cli£T, the limiting sludge liquid
concentration:
_
C1iET ~ SRR (0.17 v)(0.994)T-20 H-J
(5-18)
5.4. INPUT PARAMETER REQUIREMENTS
5.4.1. Fate and Transport: Pathway Data.
1. Vertical Term for Transport (V) — It is conservatively assumed
that atmospheric conditions are stable. Therefore, V will
always be equal to 1.
2. Lateral Virtual Distance (Xy) — Equal to X0 [Cot
(A0'/2)]//iT, where A01 is the sector width in
radians. It is assumed that the sector width is 0.393 (22.5°);
therefore, Xy = 2.84X0.
3. Average Wind Speed (v) — Obtained from local weather station.
4. Average Air Temperature (T) — Obtained from local weather
station.
5-15
-------
5. Air-Filled Porosity of Cover Soil (na) — It is assumed that
cover soils will be drained to field capacity. Therefore, the
air-filled porosity is assumed to be equal to the effective
porosity (ne). Values for effective porosity can be obtained
from Table 4-5.
6. Porosity of Cover Soil (n) — Can be measured in the laboratory
or obtained from Table 4-5.
7. Cover Thickness (tc) — Obtained from site design or operating
procedures.
8. Length or Width of Source (X0) — Obtained from site map or
plans. It is assumed that source areas are square. For active,
uncovered case, Xo is equal to the square root of the area of
an individual landfill cell. For postclosure, covered case,
X0 is equal to the square root of the area of the overall
landfill.
9. Distance from Center of Source to Receptor (r1) — Obtained
from site plans, r1 is taken as the sum of one-half the width
of the total landfill area (X0/2) plus the width of the
buffer area from the landfill area to the property boundary.
10. Standard Deviation of the Vertical Concentration Distance
(oz) — Atmospheric conditions are assumed to be stable.
5.4.2. Fate and Transport: Chemical-Specific Data.
1. Contaminant Concentration in Sludge Liquid (C-j) — Derived
directly for a contaminant by applying the TCLP. Alternately,
C-j can be calculated from the dry weight contaminant
concentration, Cs, the organic carbon distribution
coefficient for the contaminant, K0ct the fraction of organic
carbon in the sludge solids, foc, and the solids content of
the sludge, S.
2. Henry's Law Constant (H'j -- Obtained from Appendix B or
derived directly.
3. Molecular Weight of Contaminant (MW) — Obtained from the
literature.
5.4.3. Health Effects Data. A reference air concentration (RAC, in
3
mg/m ), will be defined as an ambient air concentration used to evaluate
the potential for adverse effects on human health as a result of sludge
landfilling. That is, for a given landfill site, and given the practice
5-16
-------
definitions and assumptions stated previously in this methodology, the cri-
terion for a given sludge contaminant is that concentration in the sludge
that cannot be exceeded, and is calculated to result in air concentrations
below the RAC at a compliance point downwind from the site. Exceeding the
RAC would be a basis for concern that adverse health effects may occur in a
human population in the site vicinity.
RAC is determined, based upon contaminant toxicity and air inhalation
rate, from the following general equation:
Reference Air Concentration: RAC = I /I (5-19)
pa
where I is the acceptable chronic pollutant intake rate (in nig/day) based
on the potential for health effects and I is the air inhalation rate (in
a
mVday). This simplified equation assumes that the inhaled contaminant
is absorbed into the body via the lungs at the same rate in humans as in the
experimental species tested, or between routes of exposure (e.g., oral and
inhalation). Also, this equation assumes that there are no other exposures
of the contaminant from other sources, natural or manmade. I varies
according to the pollutant evaluated and according to whether the pollutant
acts according to a threshold or nonthreshold mechanism of toxicity.
5.4.3.1. THRESHOLD-ACTING TOXICANTS — Threshold effects are those
for which a safe (i.e., subthreshold) level of toxicant exposure can be
estimated. For these toxicants, RAC is derived as follows:
/RfD x bw
Reference Air concentration: KAU =
where:
RfD = reference dose (mg/kg/day)
bw = human body weight (kg)
/RfD x bw\
\ RE /
TBI
(5-20)
5-17
-------
TBI = total background Intake rate of pollutant from all other sources
of exposure (mg/day)
Ia = air inhalation rate (ma/day)
RE « Relative effectiveness of inhalation exposure (unitless)
The definition and derivation of each of the parameters used to estimate RAC
for threshold-acting toxicants are further discussed below.
5.4.3.1.1. Reference Dose (RfD) —When toxicant exposure is by
ingestion, the threshold assumption has traditionally been used.to establish
an acceptable daily intake, or ADI. The Food and Agricultural Organization
and the World Health Organization have defined ADI as "the daily intake of a
chemical which, during an entire lifetime, appears to be without appreciable
risk on the basis of all the known facts at the time. It is expressed in
milligrams of the chemical per kilogram of body weight (mg/k;g)" (Lu, 1983).
Procedures for estimating the ADI from various types of toxicological data
were outlined by the U.S. EPA in 1980 (U.S. EPA, 1980c). More recently the
Agency has preferred the use of a new term, the "reference dose," or RfD, to
avoid the connotation of acceptability, which is often controversial.
The RfD is an estimate (with uncertainty spanning perhaps an order of
magnitude) of the daily exposure to the human population (including
sensitive subgroups) that is likely to be without appreciable risk of
deleterious effects during a lifetime. The RfD is expressed in units of
mg/kg bw/day. The RfD is estimated from observations in humans whenever
possible. When human data are lacking, observations in animals are used,
employing uncertainty factors as specified by existing Agency methodology.
RfD values for noncarcinogenic (or systemic) toxicity have been derived
by several groups within the Agency. An Intra-Agency Work Group verifies
each RfD, which is then loaded onto the Agency's publically available
5-18
-------
Integrated Risk Information System (IRIS) database. Most of the
noncarcinogenic chemicals that are presently candidates for sludge criteria
for the landfill pathway are included on the Agency's RfD list, and thus no
new effort will be required to establish RfDs for deriving sludge criteria.
For any chemicals not so listed, RfO values should be derived according to
established Agency procedures (U.S. EPA, 1988),
5.4.3.1.2. Human Body Weight (bw) and Air Inhalation Rate (I ) —
a
An assumption of 20 m3 inhalation/day by a 70-kg adult has been widely
used in Agency risk assessments and will be used in this methodology for
adults. Table 5-3 shows values of I for a typical man, woman, child and
a
infant with a typical activity schedule, as defined by the International
Commission on Radiological Protection (ICRP, 1975). Additional values have
been derived for an adult with the same activity schedule, but using upper
limit rather than average assumptions about respiration rates for each
activity, and for an adult with normal respiration rates, but whose work is
moderately active and who practices 1 hour of heavy activity (i.e.,
strenuous exercise) per day (Fruhman, 1964, as cited Diem and Lentner, 1970;
Astrand and Rodahl, 1977,, as cited in Fiserova-Bergerova, 1983).
Representative body weights have been assigned to each of these individuals
to calculate a respiratory volume-to-body weight ratio. (These ratios have
been derived for illustrative purposes only.) The resulting ratio values
range from 0.33 to 0.47 m3/kg/day, all of which exceed the ratio value
of 0.29 m3/kg/day estimated from the 70-kg adult inhaling 20 ma/day,
as used currently by the Agency. Therefore, the typically assumed values
for adults may underestimate actual exposure. In cases where children or
infants are known to be at specific risk, it may be more appropriate to use
values of bw and I for children or infants.
a
5-19
-------
in
O.O
- —
o 8
o\
•8
I rtn,
o +-
— CL-
8.
5-20
-------
5.4.3.1.3. Total Background Intake Rate of Pollutant (TBI) — It is
important to recognize that sources of exposure other than sludge disposal
practices may exist, and that the total exposure should be maintained below
the RfD. Other sources of exposure include background levels (whether
natural or anthropogenic) in drinking water, food or air. Other types of
exposure, due to occupation or habits such as smoking, might also be
included depending on data availability and regulatory policy. These expo-
sures are summed to estimate TBI.
Data for estimating background exposure usually are derived from
analytical surveys of surface, ground or tap water, from FDA market-basket
surveys and from air-monitoring surveys. These surveys may report means,
medians, percentiles or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to esti-
mate TBI should be consistent with the value of bw. Where background data
are reported in terms of a concentration in air or water, ingestion or
inhalation rates applicable to adults or children can be used to estimate
the proper daily background intake value. Where data are reported as total
daily dietary intake for adults and similar values for children are unavail-
able, conversion to an intake for children may be required. Such a conver-
sion could be estimated on the basis of relative total food intake or rela-
tive total caloric intake between adults and children.
For example, in deriving the National Emission Standard for mercury, the
average dietary contribution of 10 pg Hg/70 kg/day was subtracted from the
assumed threshold of 30 jjg/70 kg/day to give an allowable increment from
inhalation exposure of 20 yg/70 kg/day. An assumed inhalation volume of
5-21
-------
20 m /day for a 70-kg man was then applied to derive an allowable
ambient air concentration of 1 yg Hg/m3 (U.S. EPA, 1984a). For the
purposes of this methodology, however, TBI should be an estimate of
background exposure from all sources, including inhalation.
As stated in the beginning of this subsection, the TBI is the summed
estimate of all possible background exposures, except exposures resulting
from a sludge disposal practice. To be more exact, the TBI should be a sum-
med total of all toxicologically effective intakes from all nonsludge
exposures. To determine the effective TBI, background intake values (131)
for each exposure route must be divided by that route's particular relative
effectiveness (RE) factor. Thus, the TBI can be mathematically derived,
after all the background exposures have been determined, using the following
equation:
BI (food) BI (water) BI (air) BI (nV
TBI - RE (food) + RE (water) + RE (air) + •" + RE (n)
(5-21)
where:
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
BI = background intake of pollutant from a given exposure route,
indicated by subscript (mg/day)
RE = relative effectiveness, with respect to inhalation exposure,
of the exposure route indicated by subscript (unitless)
5.4.3.1.4. Fraction of Inhaled Air from Contaminated Area — It is
recognized that an individual exposed to air emissions from a landfill may
not necessarily remain in the landfill proximity for 24 hours/day. However,
if it is assumed that residential areas may be contaminated, it is likely
that less mobile individuals will include those at greatest risk. Therefore,
it is reasonable to assume that 100% of the air breathed by the ME Is will be
from the area of the landfill.
5-22
-------
5.4.3.1.5. Relative Effectiveness of Exposure (RE) — RE is a
unitless factor that shows the relative toxicological effectiveness of an
exposure by a given route when compared to another route. The value of RE
may reflect observed or estimated differences in absorption between the
inhalation and ingestion routes, which can then significantly influence the
quantity of a chemical that reaches a particular target tissue, the length
of time it takes to get there, and the degree and duration of the effect*
The RE factor may also reflect differences in the occurrence of the critical
toxicological effects at the portal of entry. For example, carbon tetra-
chloride and chloroform were estimated to be 40% and 65% as effective,
respectively, by inhalation as by ingestion based on high-dose absorption
differences (U.S. EPA, 1984b,c). In addition to route differences, RE can
also reflect differences in the exposure matrix. For example, absorption of
nickel ingested in water has been estimated to be 5 times that of nickel
ingested in the diet (U.S. EPA, 1985d). The presence of food in the
gastrointestinal tract may delay absorption and reduce the availability of
orally administered compounds, as demonstrated for halocarbons (NRC, 1986).
Physiologically based pharmacokinetic (PB-PK) models have evolved into
particularly useful tools for predicting disposition differences due to
exposure route differences. Their use is predicated on the premise that an
effective (target-tissue) dose achieved by one route in a particular species
is expected to be equally effective when achieved by another exposure route
or in some other species. For example, the proper measure of target-tissue
dose for a chemical with pharmacologic activity would be the tissue concen-
tration divided by some measure of the receptor binding constant for that
chemical. Such models account for fundamental physiologic and biochemical
5-23
-------
parameters such as blood flows, ventilatory parameters, metabolic capacities
and renal clearance, tailored by the physicochemical and biochemical prop-
erties of the agent in question. The behavior of a substance administered
by a different exposure route can be determined by adding equations that
describe the nature of the new input function. Similarly, since known
physiologic parameters are used, different species (e.g., humans vs. test
species) can be modeled by replacing the appropriate constants. It should
be emphasized that PB-PK models must be used in conjunction with toxicity
and mechanistic studies in order to relate the effective dose associated
with a certain level of risk for the test species and conditions to other
scenarios. A detailed approach for the application of PB-PK models for
derivation of the RE factor is beyond the scope of this document, but the
reader is referred to the comprehensive discussion in NRC (1986). Other
useful discussions on considerations necessary when extrapolating route to
route are found in Pepelko and Withey (1985) and Clewell and Andersen (1985).
Since exposure for the vapor pathway is by inhalation, the RE factors
applied are all with respect to the inhalation route. Therefore, the value
of RE in Equation 5-20 gives the relative effectiveness of the exposure
route and matrix on which the RfD was based when compared to. inhalation of
contaminated air. Similarly, the RE factors in Equation 5-21 show the
relative effectiveness, with respect to the inhalation route, of each back-
ground exposure route and matrix.
An RE factor should only be applied where well-documented, referenced
information is available on the contaminant's observed relative effective-
ness or its pharmacokinetics. When such information is not available, RE is
equal to 1.
5-24
-------
5.4.3.2. CARCINOGENS — For carcinogenic chemicals, the Agency con-
siders the excess risk of cancer to be linearly related to dose, except at
high-dose levels (U.S. EPA, 1986a). The threshold assumption, therefore,
does not hold, as risk diminishes with dose but does not become zero or
background until dose becomes zero.
The decision whether to treat a chemical as a threshold- or nonthreshold-
acting (i.e., carcinogenic) agent depends on the weight of the evidence that
it may be carcinogenic to humans. Methods for classifying chemicals as to
their weight of evidence have been described by U.S. EPA (U.S. EPA, 1986a),
and most of the chemicals that presently are candidates for sludge criteria
have recently been classified in Health Assessment Documents or other
reports prepared by the U.S. EPA's Office of Health and Environmental
Assessment (OHEA), or in connection with the development of recommended
maximum contaminant levels (RMCLs) for drinking-water contaminants (U.S.
EPA, 1985e). To derive values of the reference air concentration (RAC), a
decision must be made as to which classifications constitute sufficient
evidence for basing a quantitative risk assessment on a presumption of
carcinogenicity. Chemicals in classifications A and B, "human carcinogen"
and "probable human carcinogen," respectively, have usually been assessed as
carcinogens, whereas those in classifications D and E, "not classifiable as
to human carcinogenicity because of inadequate human and animal data" and
"evidence of noncarcinogenicity for humans," respectively, have usually been
assessed according to threshold effects. Chemicals classified as C,
"possible human carcinogen," have received varying treatment. For example,
lindane, classified by the Human Health Assessment Group (HHAG) of the U.S.
EPA as B2~C, or between the lower range of the B category and category C,
5-25
-------
has been assessed using both the linear model for tumorigenic effects (U.S.
EPA, 1980b) and based on threshold effects (U.S. EPA, 1985e). Table 5-4
gives an illustration of these U.S. EPA classifications based on the
available weight of evidence.
Using the weight-of-evidence classification without noting the
explanatory material for a specific chemical may lead to a flawed conclu-
sion, since some of the classifications are exposure-route dependent.
Certain compounds (e.g., nickel) have been shown to be carcinogenic by the
inhalation route, but not by ingestion. The issue of whether or not to
treat an agent as carcinogenic by ingestion remains controversial for
several chemicals.
If a pollutant is to be assessed according to nonthreshold, carcinogenic
effects, the RAC is derived as follows:
Reference Air Concentration: RAC =
(5-22)
where:
q-j* = human cancer potency [(mg/kg/day) 1]
RL = risk level (unitless) (e.g., l(r5, 1CT6, etc.)
bw = human body weight (kg)
RE - relative effectiveness of inhalation exposure (unitless)
Ia - air inhalation rate (m3/day)
TBI = total background intake rate of pollutant (mg/day), from
all other sources of exposure
The RAC, in the case of carcinogens, is thought to be protective since the
q * is typically an upper-limit value (i.e., the true potency is consid-
ered unlikely to be greater and may be less). The definition and derivation
5-26
-------
TABLE 5-4
Illustrative Categorization of Evidence Based on Animal and Human Data*
Animal Evidence
Human
Evidence
Sufficient
Limited
Inadequate
No data
No evidence
Sufficient
A
Bl
82
B2
B2
Limited
A
Bl
C
C
C
Inadequate
A
Bl
D
0
D
No Data
A
Bl
D
D
D
No
Evidence
A
Bl
D
E
E
*The above assignments are presented for illustrative purposes. There may
be nuances in the classification of both animal and human data indicating
that different categorizations than those given in the table should be
assigned. Furthermore, these assignments are tentative and may be modified
by ancillary evidence. In this regard, all relevant information should be
evaluated to determine if the designation of the overall weight of evidence
needs to be modified. Relevant factors to be included along with the tumor
data from human and animal studies include structure-activity relationships;
short-term test findings; results of appropriate physiological, biochemical
and toxicological observations; and comparative metabolism and pharmaco-
kinetic studies. The nature of these findings may cause an adjustment of
the overall categorization of the weight of evidence.
5-27
-------
Cancer Potency (q*) — For most carcinogenic
of each of the parameters used to estimate RAC for carcinogens are further
discussed in the following subsections.
5.4.3.2.1. Human
chemicals, the linearized multistage model is recommended for estimating
human cancer potency from animal data (U.S. EPA, 1986a). When
epidemiological data are available, potency is estimated based on the
observed relative risk in exposed vs. nonexposed individuals, and on the
magnitude of exposure. Guidelines for use of these procedures have been
presented in the U.S. EPA (1980c, 1985e) and in each of a series of Health
Assessment Documents prepared by OHEA (e.g., U.S. EPA, 1985d). The true
potency value is considered unlikely to be above the upper-bound estimate of
the slope of the dose-response curve in the low-dose range, and it is
expressed in terms of risk/dose, where dose is in units of mg/kg/day. Thus,
q,* has units of (mg/kg/day) 1. OHEA has derived potency estimates
for each of the potentially carcinogenic chemicals that are presently
candidates for sludge criteria. Therefore, no new effort will be required
to develop potency estimates to derive sludge criteria.
5.4.3.2.2. Risk Level (RL) — Since by definition no "safe" level
exists for exposure to nonthreshold agents, values of RAC are calculated to
reflect various levels of cancer risk. If RL is set at zero, then RAC will
be zero. If RL is set at 10 6, RAC will be the concentration which, for
lifetime exposure, is calculated to have an upper-bound cancer risk of one
case in one million individuals exposed. This risk level refers to excess
cancer risk, i.e., over and above the background cancer risk in unexposed
individuals. By varying RL, RAC may be calculated for any level of risk in
the low-dose region, i.e., RL <10~2. Specification of a given risk
5-28
-------
level on which to base regulations is a matter of policy. Therefore, it is
common practice to derive criteria representing several levels of risk
without specifying any level as "acceptable."
5.4.3.2.3. Human Body Weight (bw) and Air Inhalation Rate (I ) —
9
Considerations for defining bw and I are similar to those stated in Sec-
3
tion 5.4.3.1.2. The HHAG assumes respective values of 70 kg and 20
mVday to derive unit risk estimates for air, which are potency
estimates transformed to units of (yg/m )
3.-1
As illustrated in
Table 5-3, exposures may be higher in children than in adults when the
ratios of inhalation volumes to body weights are compared. However, because
exposure is lifelong, values of bw and I are usually chosen to be
a
representative of adults.
5.4.3.2.4. Relative Effectiveness of Exposure (RE) — RE is a unit-
less factor that shows the relative toxicological effectiveness of an expo-
sure by a given route when compared to another route. The value of RE may
reflect observed or estimated differences in absorption between the inhala-
tion and ingestion routes, which can significantly influence the quantity of
a chemical that reaches a particular target tissue, the length of time it
takes to get there, and the degree and duration of the effect. The RE
factor may also reflect differences in the occurrence of critical toxico-
:logical effects at the portal of entry. For example, carbon tetrachloride
and chloroform were estimated to be 40% and 65% as effective, respectively,
by inhalation as by ingestion based on high-dose absorption differences
(U.S. EPA, 1984b,c). In addition to route differences, RE can also reflect
differences in the exposure matrix. For example, absorption of nickel
ingested in water has been estimated to be 5 times that of nickel ingested
5-29
-------
in food (U.S. EPA, 1985d). The presence of food in the gastrointestinal
tract may delay absorption and reduce the availability of orally
administered compounds, as demonstrated for halocarbons (NRC, 1986).
PB-PK models have evolved into particularly useful tools for predicting
disposition differences due to exposure route differences. Their use is
predicated on the premise that an effective (target-tissue) dose achieved by
one route in a particular species is expected to be equally effective when
achieved by another exposure route or in some other species. For example,
the proper measure of target-tissue dose for a chemical with pharmacologic
activity would be the tissue concentration divided by some measure of the
receptor binding constant for that chemical. Such models account for
fundamental physiologic and biochemical parameters such as blood flows,
ventilatory parameters, metabolic capacities and renal clearance, tailored
by the physicochemical and biochemical properties of the agent in question.
The behavior of a substance administered by a different exposure route can
be determined by adding equations that describe the nature of the new input
function. Similarly, since known physiologic parameters are used, different
species (e.g., humans vs. test species) can be modeled by replacing the
appropriate constants. It should be emphasized that PB-PK models must be
used in conjunction with toxicity and mechanistic studies in order to relate
the effective dose associated with a certain level of risk for the test
species and conditions to other scenarios. A detailed approach for the
application of PB-PK models for derivation of the RE factor is beyond the
scope of this document, but the reader is referred to the comprehensive
discussion in NRC (1986). Other useful discussions on considerations
necessary when extrapolating route to route are found in Pepelko and Withey
(1985) and Clewell and Andersen (1985).
5-30
-------
Since exposure for the vapor pathway is by inhalation, the RE factors
applied are all with respect to the inhalation route. Therefore, the value
of RE in Equation 5-22 gives the relative effectiveness of the exposure
route and matrix on which the q * was based when compared to inhalation of
contaminated air. Similarly, the RE factors in Equation 5-21 show the
relative effectiveness, with respect to the inhalation route, of each back-
ground exposure route and matrix.
An RE factor should only be applied where well-documented, referenced
information is available on the contaminant's observed relative effective-
ness or its pharmacokinetics. When such information is not available, RE is
equal to 1.
5.4.3.2.5. Total Background Intake Rate of Pollutant (TBI) — It is
important to recognize that sources of exposure other than sludge disposal
practices may exist, and that the total exposure should be maintained below
the determined cancer risk-specific exposure level. Other sources of
exposure include background levels (whether natural or anthropogenic) in
drinking water, food or air. Other types of exposure, due to occupation or
habits such as smoking, might also be included depending on data availabil-
ity and regulatory policy. These exposures are summed to estimate TBI.
Data for estimating background exposure usually are derived from analyt-
ical surveys of surface, ground or tap water, from FDA market-basket sur-
veys, and from air-monitoring surveys. These surveys may report means,
medians, percentiles or ranges, as well as detection limits. Estimates of
TBI may be based on values representing central tendency or on upper-bound
exposure situations, depending on regulatory policy. Data chosen to esti-
mate TBI should be consistent with the value of bw. Where background data
5-31
-------
are reported in terms of a concentration in air or water, ingestion or
inhalation rates applicable to adults can be used to estimate the proper
daily background intake, value. For certain compounds (e.g., nickel) that
have been shown to be carcinogenic by the inhalation route, but not by the
ingestion route, the TBI should not include background exposure from the
ingestion route. Thus, in such a case only background exposures from other
air emission sources should be included in the TBI.
As stated in the beginning of this subsection, the TBI is the summed
estimate of all possible background exposures, except exposures resulting
from a sludge disposal practice. To be more exact, the TBI should be a sum-
med total of all toxicologically effective intakes from all nonsludge expo-
sures. To determine the effective TBI, background intake values (BI) for
each exposure route, must be divided by that route's particular relative
effectiveness (RE) factor. Thus, the, TBI can be mathematically derived,
after all the background exposures have been determined, using the following
equation:
_ BI (food) BI (water) BI (air) BI (n)
TBI ~ RE (food) + RE (water) + RE (air) + •" "''RE (n)
(5-23)
where:
TBI = total background intake rate of pollutant from all other
sources of exposure (mg/day)
£
BI = background intake of pollutant from a given exposure route,
indicated by subscript (mg/day)
RE ='relative effectiveness, with respect to inhalation exposure,
of the exposure route indicated by subscript (unitless)
5.4.3.2.6. Fraction of Inhaled Air From Contaminated Area — It is
recognized that an individual exposed to air emissions from a landfill may
not necessarily remain in the landfill proximity for 24 hours/day. However,
5-32
-------
if it is assumed that residential areas may be contaminated, it is likely
that less mobile individuals will include those at greatest risk. There-
fore, it is reasonable to assume that 100% of the air breathed by the MEIs
will be from the area of landfill.
5.5. SITE-SPECIFIC APPLICATION
This section presents sample calculations for determining the vapor
exposure resulting from landfilling of sludge. In the following, calcula-
tions are first made for a particular landfill on a site-specific applica-
tion and then an example is given for calculating maximum allowable contami-
nant levels in sludge. Benzene, because it is a volatile contaminant of
concern, is used for the example calculations. For the examples, data de-
scribing the occurrence and concentration of benzene in sludge are taken
from U.S. EPA (1985a). The pathway and chemical parameters used in the
calculations are summarized in Table 5-5. , Data describing waste sites are
values assumed to represent reasonable cases. In actual practice, the data
used in the calculations would be those measured or collected by the
applicant.
Assume operating procedures include excavation of a 4- by 16-m trench,
disposal of three daily 0.5-m lifts in each trench, application of a daily
cover of 0.3 m soil and application of a final cover of 1.0 m soil. Assume
that 67% of the total disposal site area is available for trenches (Table
5-6).
5.5.1. Tier 1 Calculation. The Tier 1 calculation involves comparing the
equilibrium vapor concentration of the constituent with the reference air
concentration (RAC). This approach represents the worst possible case with
no allowance made for atmospheric dilution, dispersion or degradation. The
5-33
-------
TABLE 5-5
Input Parameters for Example Calculations: Vapor Loss
Fate and Transport: Pathway Data
1. Vertical term for transport
2. Lateral virtual distance
3. Average windspeed
4. Average air temperature
V = 1
Xy = 2.84xo
xo = 22.64 m active
1361.46 m postclosure
v = 2 m/sec
T = 15°C
5. Air-filled porosity of cover soil na = 0.1
6. Porosity of cover soil
7. Cover thickness
8. Length of source
9. Distance from center of source
to receptor
10. Standard deviation of the
vertical concentration distance
n = 0.4
Tc
0.3 m active,; 1.0 m post-
closure
X0 = 8 m active, 480 m post-
closure
r1 = 340 m
az =6.2 m
Fate and Transport: Chemical-Specific Data (Benzene)
11. Contaminant concentration in
sludge liquid
12. Henry's Law Constant
13. Molecular weight of contaminant
Health Data (Benzene)
14. Reference concentration in air
X = 0.0083 mg/9.
H1 = 0.24
MW = 78
RAC = 6.73xlO~a vg/m3
5-34
-------
TABLE 5-6
Supporting Sludge landfill Characteristics
Daily disposal rate
Trench dimensions
Depth of daily fill
Life of facility
Total trench area
Total disposal site area
10 dry metric tons/day
4 m by 16 m
0.5 m
20 years
156,000 m2
234,000 m2
5-35
-------
equilibrium vapor concentration is taken as the product of the Henry's Law
Constant of the constituent and the liquid phase concentration of the
constituent.
The liquid phase constituent concentration can be obtained in several
ways. If the leachate extraction procedure is used, the liquid concentra-
tion will be determined directly from the procedure. If the analytical
results are expressed in terms of dry weight, it will be necessary to con-
vert the dry weight results to an equivalent liquid phase concentration
accounting for partitioning between the liquid and solid phases. This is
accomplished with Equation 5-24:
cdrv s
Ci =
(5-24)
where:
Koc foe
C-|
~ concentration of contaminant in sludge liquid (mg/9.)
= dry weight concentration of contaminant in sludge (mg/kg)
S = solids content of sludge (kg dry solids/kg total wet
sludge)
Koc = organic carbon distribution coefficient
mg contaminant/kg organic carbon
mg contaminant/5, sludge liquid
foc = organic carbon content of sludge (kg organic carbon/
kg sludge solids)
Y8, = density of sludge liquid (kg sludge liquid/8. sludge liquid)
For the example calculation, the mean dry weight concentration of benzene
in sludge, 0.326 mg/kg, reported in U.S. EPA (1985a) is used. The organic
carbon distribution coefficient for benzene is 74.2 a./kg (U.S. EPA, 1985a).
Assuming a solids content of 30% for dewatered sludge, an organic carbon
content of 50% for the sludge solids and a density of 1 kg/a, for the sludge
liquid, the equivalent liquid concentration is the following:
5-36
-------
Cl =
(0.326 mg/kgH0.30 kg/kg)
(74.2 !l/kg)(0.5 kg/kg)(0.30 kg/kg) + (1-0-30) kq/kq
1.0 kg/I,
= 0.0083 mg/8.
The Henry's Law Constant for benzene is then used to calculate the vapor
concentration in equilibrium with the liquid concentration:
Cv = H C
1
(5-25)
From Appendix B, the dimensionless Henry's Law Constant for benzene is 0.24.
The equilibrium vapor pressure is:
Cv = (0.24) (0.0083 mg/fc) = 0.0020 mg/!l = 2.0 mg/m3
The RAC for the carcinogen benzene is derived using Equation 5-22:
RAC =
/RL x bw \
\q!* x REJ
- TBI
*' la
The risk level (RL), the body weight (bw), and the daily inhalation volume
(Ia) are set for this example at 10~6, 70 kg and 20 m3, respec-
tively. The relative effectiveness factor (RE) is set at 1. The human
cancer potency for benzene has been determined by the U.S. EPA to be
5.2xlO~2 (mg/kg/day)"1. Current total background intake (TBI) of
benzene from all other sources (i.e., except from landfilling of sludges)
Ms
has not been determined for 1986, but for illustrative purposes a TBI of
zero is used here to derive an example RAC. Determination of an RAC for a
specific landfill site should be based on a current local assessment of TBI.
RAC
fc
10~6 x 70 kg
.2x10-2 (mg/kg/day)
= 6.73 x 10~s mg/m3
= 6.73 x 10~2 yg/m3
= 0.0673 yg/m3
) -o
-1 X I/
* 20 ma
5-37
-------
The above vapor concentration is then compared to the reference value for
benzene, RAC = 6.73xlO~2 yg/m3. Since 2 mg/ma » 6.73xlO~2 vg/m3 ,
it is necessary to proceed to Tier 2.
5.5.2. Tier 2 Calculation. The Tier 2 methodology involves estimating
the flux of contaminant out of the landfill and using an atmospheric
dispersion model to predict the atmospheric concentration of contaminant
downwind of the site. The long-term average downwind concentration is then
compared with the RAC. .
The first step of the Tier 2 methodology is to calculate the flux rate
of contaminants during the active life of the trench (i.e., before emplace-
ment of final cover). The flux rate for the active period of the cell is
taken as the time-weighted average of the flux rate with no cover and the
flux rate with temporary daily cover. Under the assumed operating condi-
tions, each trench will be active for a 3-day period. On each of the 3
days, a lift of sludge will be applied followed by a temporary soil cover.
It is assumed that the sludge will be uncovered for 4 hours each day.
Therefore, the fraction of time that the sludge is uncovered is
(3x4)/(3x24) = 0.17. The fraction of time that the sludge is covered by
temporary cover is .( 3x20) /( 3x24) = 0.83.
The flux rate during the portion of time that wastes are uncovered is
calculated using Equation 5-6:
T-20 /"""""
q . = [0.17 v (0.994) H.C,.]/YMWI
CM III
Based on the parameter values provided in Tables 5-4 and 5-5, the resulting
contaminant flux rate is:
q • =
31
m/sec)(0.994)15~2°(0.24)(0.0083
= 7.9 x 10~s g/m2-sec
5-38
-------
The flux rate during the portion of time that wastes are covered by
temporary cover is calculated using Equation 5-8:
2
q = [9.2 x 10~5 n!0/3 (1 .006)T~20
pi
a
TC n
As shown in Table 5-5, the thickness of the temporary soil cover, T , is
\f
0.3m. The air-filled porosity of the soil cover, n , and the total
9
porosity of the soil cover, n, are assumed to be 0.1 and 0.4, respectively.
The resulting contaminant flux rate is:
qp. = [(9.2 x 10~5)(0.1)10/3 (1.006)15"20 (0.24) (0.0083) ]/|VMWI (0.3)(0.4):
qp1 = 2.0 x 10~10 g/m2-sec
The time-weighted flux for the active period is then:
(7.9 x 10~5 g/m2-sec)(0.17) + (2.0 x 10"10 g/m2-sec)(0.83)
= 1.3 x 10~5 g/m2-sec
The second step of the Tier 2 methodology is to calculate the flux rate
of contaminants during the postclosure period. The flux during the post-
closure period is calculated using Equation 5-8:
'Pi
[9.2xlO-5na10/3(1.006)T-2°
c n2
The values of variables used will be the same as those used to calculate flux
through the temporary cover, except for the cover thickness, T . For post-
closure, the cover thickness is 1.0 m. The resulting postclosure flux is:
qp. = [9.2 x 10
5
(1.006)15~20 (0.24)(0.0083) ]/|
|V78~(1)(0.4)2|
= 5.8 x 10 g/m -sec
The long-term average exposure level is based on exposure to contami-
nants over a 70-year period. The fluxes for the active and postclosure
cases, therefore, must be adjusted to reflect the period of exposure. For
the active case, one landfill cell will be open at any time during the
entire 20-year active period. The long-term average flux associated with
the active portion of the fill is then 20/70 times the active flux or:
5-39
-------
(20/70)(1.3 x 10~s g/m2-sec) = 3.71 x 10~6 g/m2-sec
The long-term average flux associated with postclosure will be based on the
average postclosure life of the landfill cells. That is, all cells will be
closed at least 50 years and the maximum postclosure period wiII be 70 years.
Because the postclosure period varies linearly from 50 to 70 years, the
average period of 60 years was used. The long-term average Flux associated
With postclosure will then be 60/70 of the postclosure flux or:
(60/70)(5.8 x 10"11 g/m2-sec) = 5.0 x 10"11 g/m2-sec '
Because the active and postclosure fluxes involve different source areas,
the fluxes could not be summed to obtain a single long-term average flux to
calculate exposure. That is, the active flux is associated with the area of
one landfill cell, while the postclosure flux is associated with the area of
the total disposal site. Therefore, the downwind concentrations associated
with each flux were calculated.and summed to obtain the average exposure.
Downwind exposure concentrations were calculated using Equation 5-9:
X02Q V
C(r,0) - (2.032 x 10*) [—
Xy) v oz
The above equation is for the concentration along the center line of the
plume, which represents a worst case.
For the active-period exposure, the side of the source area, X , is
108 m. In determining the distance from the source center to the receptor,
r1, it was assumed that the active cell was always located in the center of
the disposal area. The distance was then taken as one-half the square root
of the area of the disposal site plus a buffer zone distance, assumed to be
100 m.
r1 = (0.5) (234,000 m2) + 100 m = 340 m
5-40
-------
The lateral virtual distance, X. was calculated using Equation 5-13:
= 8(2.84) = 22.69 m
a was calculated for a worst-case condition corresponding to stable
atmospheric conditions. For a downwind distance corresponding to the dis-
tance r1 (0.34 km), a was calculated using the data presented in
Table 5-1.
oz = 14.456 (0.34)0'78407 = 6.2 m
For stable atmospheric conditions and a contaminant release height of 0 m, L
is infinite and therefore the vertical term, V, is equal to 1 (Equation
5-11). For the assumed wind speed of 2 m/sec, the resulting downwind con-
centration is:
C (r,0) = (2.032 x
= (2.032 X 10s)(8)2 (3.71
(340 + 22.69)(2)(6.2)
,' 3
0.11 vg/m
For the postclosure period, the side of the source area, X , is 480
o
m. The distance from the source center to the receptor, r1, is the same as
for the active case. As for the active case, the lateral virtual distance,
X , was calculated using Equation 5-13:
Of
= 1361.46 m
5-41
-------
As with the active case, <*z was selected for a worst-case condition cor-
responding to stable atmospheric conditions. Because the downwind distance
is the same as for the active case, o will be 6.2 m. The vertical
term, V, will also be equal to 1 , as in the active case. Using the above
data and the assumed wind velocity of 2 m/sec, the downwind contaminant
concentration is:
C(rt0) • 2.032 X10*
2.032 x IP6 (480)2 (5.0 x 10~11)(1)
(340 + 1361.46)(2)(6.2)
= 1 .1 x 10~3 yg/m3
The above results show that the exposure due to the postclosure release
is negligible compared to the exposure due to active release. The total
exposure concentration for comparison to the reference level will then be
the active exposure, or 0.11 ng/m3. This is compared to the reference
air concentration, RAC = 6.73xlO~2 yg/m3, for benzene.
5.6. NATIONAL CRITERIA SITE-SPECIFIC APPLICATION
To establish sludge concentration criteria for volatile contaminants, it
is necessary to operate the methodology provided here in a reverse mode.
That is, the RAC must be taken as input to determine the maximum allowable
concentration in the source sludge. From Equation 5-14, the compliance
point concentration is defined as:
C = Q x SRR
Since SRR is characteristic of a site and not concentration dependent, a
single value can be calculated for a representative site. When C is set at
C , the long-term effects threshold concentration, the allowable
long-term average flux Q is defined as:
Q = CET/SRR
5-42
(5-26)
-------
The flux during the active life of the facility was shown in Section 5.5.2.
to be far greater than that during postclosure, and therefore the latter may
be set equal to zero in the calculation of criteria. From Equation 5-6:
0.17 v (0.994)T-20 HjClj
Qai = /-——
y MWi
where:
qai = flux during the uncovered period (g/m2-sec)
v = windspeed (m/sec)
T = temperature (°C)
Hi = Henry's Law Constant (dimensionless)
Cli = concentration of contaminant in the sludge liquid (mg/s.)
MWi = molecular weight of contaminant
The average flux during the human lifetime is determined by adjusting the
uncovered period flux, q ., for the fraction of time that the sludge is
uncovered (0.17) during the facility active life, and for the assumed total
active life of the facility (20 years) during the human lifetime (70 years),
as described in Section 5.5.2. The resulting relationship is as follows:
(5-27)
Combining Equations 5-6, 5-26 and 5-27 for the criteria case yields:
0.17 v (0.994)1"-20
Q = q . x 0.17 x (20/70)
31
CET/SRR
x 0.17 x
(5-28)
The latter can be solved for Cl to give Cl , the limiting sludge
liquid concentration:
CET
CliET =
70
SRR (0.17 v)(0.994)T-20Hi 0-17 20
(5-29)
For benzene, the MW
6.73xlO~2 pg/m3. Therefore:
6.73 x IP"2 C/78)
is 78 g/mol, the H. is 0.24 and the RAC is
CliET =
(3xlO*)(0.17)(2)(0.994)
= 4.9 x 10~3 mg/8,
ic_
ID
1 70
x - x— (5-30)
(0.24) 0.17 20
5-43
-------
-3
The national criteria would, therefore, be set at 4.9x10 " mg/9. in
leachate. Based on Equation 5-24 and assuming a solids content of 30% for
dewatered sludge, an organic content of 50% for the sludge solids and a
density of 1.0 kg/8, for the sludge liquid, the corresponding dry weight
concentration of benzene in the landfilled sludge would be 0.19 mg/kg.
5-44
-------
6. REFERENCES
Abramowitz, M. and I.A. Stegun. 1972. Handbook of Mathematical Functions.
Dover Publications, New York. p. 1045.
Astrand, P.-O. and K. Rodahl. 1977. Textbook of Work Physiology, 2nd ed.
McGraw-Hill, New York. (Cited in Fiserova-Bergerova, 1983)
Baas-Becking, L.G.M., I.R. Kaplan and 0. Moore. 1960. Limits of the natu-
ral environment in terms of pH and oxidation-reduction potentials. J. Geol.
68(3): 243-284.
Bogert, L.O., G.M. Briggs and D.H. Galloway. 1973. Nutrition and Physical
Fitness, 9th ed. W.B. Saunders Co., Philadelphia, PA. 578 p.
Boutwell, S.H., S.M. Brown, B.R. Roberts and D. Atwood. 1985. Modeling
Remedial Action at Uncontrolled Hazardous Waste Sites. U.S. EPA,
Cincinnati, OH. EPA/540/2-85-001.
Campbell, G.S. 1974. A simple method for determining unsaturated conduc-
tivity from moisture retention data. Soil Sci. 117: 311-314.
Carsel, R.F., C.N. Smith, L.A. Mulkey, J.D. Dean and P. Jowise. 1984.
User's manual for the pesticide root zone model (PRZM) release 1, EPA-600/
3-84-109. Prepared by Environmental Research Laboratory, Athens, GA. U.S.
EPA, Washington, DC. 216 p.
6-1
-------
Cho, D.H. 1971. Convective transport of ammonium with nitrification in
soil. Can. J. Soil Sci. 51: 339-350. As referenced in Van Genuchten, M.T.
1985. Convective-dispersive transport of solutes involved in sequential
first-order decay reactions. Comput. Geosci. 11:129-147.
Cleary, R.W., T.J. McAvary and W.L. Short. 1973. Unsteady-State Three-
Dimensional Model of Thermal Pollution in Rivers. Water-1972. Am. Inst.
Chem. Engineers, Symp. Ser. 129, Vol. 69. p. 422-431. As referenced in
Yeh, G.T. 1981. AT123D: Analytical Transient One-, Two-, and
Three-Dimensional Simulation of Waste Transport in the Aquifer System.
ORNL-5602, Environmental Sciences Division, Pub. No. 1439.
Cleary, R.W. and M.J. Ungs. 1978. Analytical Models for Groundwater
Pollution and Hydrology. Water Resources Program, Department of Civil
Engineering, Princeton University, Princeton, NJ. Report 78-WR-15.
Clewell, H.J. and M.E. Andersen. 1985. Risk assessment extrapolation and
physiological modeling. Toxicol. Ind. Health. 1: 111-134.
Codell, R.B. 1982. Collection of Mathematical Models for Dispersion in
Surface Water and Groundwater. NUREG-0868. Nuclear Regulatory Commission,
Bethesda, MD.
Codell, R.B. 1984. Simplified Analysis for Liquid Pathway Studies.
NUREG-1054. Nuclear Regulatory Commission, Bethesda, MD.
6-2
-------
Davidson, J.M., C.E. Rick and P.W, Santelmann. 1968. Influence of water
flux and porous material on the movement of selected herbicides. Soil Sci.
Soc. Am. Proc. 32: 629-633. As referenced in Yen, G.T. 1981. AT123D:
Analytical Transient One-, Two-, and Three-Dimensional Simulation of Waste
Transport in the Aquifer System. ORNL-5602, Environmental Sciences
Division, Pub. No. 1439.
Diem, K. and C. Lentner, Ed. 1970. Scientific Tables. Ciba-Giegy, Ltd.,
Basle, Switzerland.
Deutsch, W.J. and K.M. Krupka. 1985. MINTEQ Geochemical Code: Compilation
of Thermodynamic Data-Base for the Aqueous Species, Gases, and Solids Con-
taining Chromium, Mercury, Selenium, and Thallium. Draft. Battelle
Memorial Institute, Richland, WA.
Donigian, A.S., T.Y.R. Lo and E.W. Shanahan. 1983. Rapid Assessment of
Potential Ground-Water Contamination Under Emergency Response Conditions,
EPA-68-03-3116. Prepared by Anderson-Nichols & Co., Inc., Palo Alto, CA.
U.S. EPA, Washington, DC.
English, C.J., J.R. Miner and J.K. Koelliker. 1980. Volatile ammonia
losses from surface-applied sludge. J. Water Pollut. Control Fed. 52:
2340-2350.
Environmental Science and Engineering. 1985. Exposure to Airborne Contami-
nants Released from Land Disposal Facilities — A Proposed Methodology.
Prepared by Environmental Science and Engineering, Gainesville, FL, for
Office of Solid Wastes, U.S. EPA, Washington, DC.
6-3
-------
Felmy, A.R., S.M. Brown, Y. Onishi, S.B. Yabusaki and R.S. Argo. 1983.
HEXAMS — the Metals Exposure Analysis Modeling System, Prepared by
Environmental Research Laboratory, Athens, 6A. U.S. EPA, Washington, DC.
Felmy, A.R., D.C. Girvin and E.A. Jenne. 1984. MINTEQ — A Computer Pro-
gram for Calculating Aqueous Geochemical Equilibria, EPA-600/3-84-032,
Prepared by Environmental Research Laboratory, Athens, GA. U.S. EPA,
Washington, DC.
Fiserova-Bergerova, V., Ed. 1983. Modeling of Inhalation Exposure to
Vapors: Uptake, Distribution and Elimination, Vol. II. CRC Press, Inc.,
Boca Raton, FL.
Fruhman, G. 1964. Title not given. Z. Exp. Med. 138: 1. (Cited in Diem
and Lentner, 1970)
Gillette, D.A. 1973. On the Production of Soil Wind Erosion Aerosols
Having the Potential for Long-Range Transport. Special Issue of Journal de
Recherches Atmospherique on the Nice Symposium on the Chemistry of Sea-Air
Particulate Exchange Processes. Nice, France. As referenced in Dynamac,
Corp. 1983. Methods for Assessing Exposure to Windblown Particulates,
EPA-600/4-83-007. Springfield, VA. NTIS PB 83-177659.
ICRP (International Commission on Radiological Protection). 1975. Report
of the Task Group on Reference Man (No. 23). Pergamon Press, Ltd., London.
6-4
-------
Jacobson, E.A., M.D. Freshley and F.H. Dove. 1985. Investigations of Sen-
sitivity and Uncertainty in Some Hydrologic Models of Yucca Mountain and
Vicinity. Draft. PNL-5306, SAND84-7212. U.S. .Dept. of Energy, Richland,
WA.
Kowal, N.E. 1985. Health Effects of Land Application of Municipal Sludge,
EPA/600/T-85/015. Prepared by Health Effects Research Laboratory, Research
Triangle Park, NC. U.S. EPA, RTP, NC.
Kuo, E.Y.T. 1976. Analytical solution for 3-D diffusion mode]. J.
Environ. Eng. Div. ASCE. 102: 805-820. As referenced in Yeh, G.T. 1981.
AT123D: Analytical Transient One-, Two-, and Three-Dimensional Simulation
of Waste Transport in the Aquifer System. ORNL-5602, Environmental Sciences
Division, Pub. No. 1439.
Lai, S.H. and J.J. Jurinak. 1972. The transport of cations in soil columns
at different pore velocities. Soil Sci. Soc. Am. Proc. 36: 730-733. As
referenced in Yeh, G.T. 1981. AT123D: Analytical Transient One-, Two-,
and Three-Dimensional Simulation of Waste Transport in the Aquifer System.
ORNL-5602, Environmental Sciences Division, Pub. No. 1439.
Lapidus, L. and N.R. Amundson. 1952. Mathematics of adsorption in beds.
VI. The effects of longitudinal diffusion in ion exchange and chromato-
graphic columns. J. Phys. Chem. 56: 984-988. As referenced in Yeh, G.T.
1981. AT123D: Analytical Transient One-, Two-, and Three-Dimensional
Simulation of Waste Transport in the Aquifer System. ORNL-5602,
Environmental Sciences Division, Pub. No. 1439.
6-5
-------
Lindstrom, F.T. and L. Boersma. 1971. A theory on the mass transport of
previously distributed chemicals in a water saturated sorbing-porous medium.
Soil Sci. Ill: 192-199. As referenced in Yeh, G.T. 1981. AT123D:
Analytical Transient One-, Two-, and Three-Dimensional Simulation of Waste
Transport in the Aquifer System. ORNL-5602, Environmental Sciences
Division, Pub. No. 1439.
Lindstrom, F.T. and W.M. Stone. 1974. On the start up or initial phase of
linear mass transport of chemicals in a water saturated sorbing porous
medium. I. SIAM. J. Appl. Math. 26: 578-591. As referenced in Yeh, G.T.
1981. AT123D: Analytical Transient One-, Two-, and Three-Dimensional
Simulation of Waste Transport in the Aquifer System. ORNL-5602,
Environmental Sciences Division, Pub. No. 1439.
Lu, F.C. 1983. Toxicological evaluations of carcinogens and noncarcino-
gens: Pros and cons of different approaches. Reg. Toxicol. Pharmacol. 3:
121-132. .
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt. 1982. Handbook of Chemical
Property Estimation Methods. McGraw-Hill, San Francisco, CA.
Marino, A.M. 1974. Distribution of contaminants in porous media flow.
Water Resour. Res. 10: 1013-1018. As referenced in Yeh, G.T. 1981.
AT123D: Analytical Transient One-, Two-, and Three-Dimensional Simulation
of Waste Transport in the Aquifer System. ORNL-5602, Environmental Sciences
Division, Pub. No. 1439.
6-6
-------
Misra, C., D.R. Nielsen and J.W; Biggar. 1974. Nitrogen transformation in
soil during leaching. I. Theoretical considerations. Soil Sci. Soc. Am.
Proc. 38: 289-293. As referenced in Van Genuchten, M.T. 1985.
Convective-dispersive transport of solutes involved in sequential
first-order decay reactions. Comput. Geosci. 11: 129-147.
Morrey, J.R. 1985. PROOEF: A Code to Facilitate the Use of the Geochemical
Code MINTEQ. Draft. Battelle Memorial Institute, Richland, WA.
Nelson, W.E., Ed. 1969. Textbook of Pediatrics, 9th ed. W.B. Saunders
Co., Philadelphia, PA. (Cited in Bogert et al., 1973)
NRC (National Research Council). 1983. Risk Assessment in the Federal
Government: Managing the Process. National Academy Press, Washington, DC.
NRC (National Research Council). 1986. Dose-route extrapolations: Using
inhalation toxicity data to set drinking water limits. In: Drinking Water
and Health, Vol. 6. National Academy Press, Washington, DC.
Pennington, J.A.T. 1983. Revision of the total diet study food list and
diets. J. Am. Diet. Assoc. 82: 166-173.
Pepelko, W.E. and J.R. Withey. 1985. Methods for route-to-route extrapola-
tion of dose. Toxicol. Ind. Health. 1: 153-175.
6-7
-------
Rao, P.S.C. 1982. Unpublished report prepared for Anderson-Nichols & Co.,
Inc. Univ. of Florida, Gainesville, FL. As referenced in Donigian, A.S.,
T.Y.R. Lo and E.W. Shanahan. 1983. Rapid Assessment of Potential
Ground-Water Contamination Under Emergency Response Conditions,
EPA-68-03-3116. Prepared by Anderson-Nichols & Co., Inc. Palo Alto, CA.
U.S. EPA, Washington, DC.
Sandusky, W.F. and D.S. Renne. 1981. Candidate Wind Turbine Generator Site
Annual Data Summary for January 1979 through December 1979. PNL-3703,
U.8-60. Prepared by Pacific Northwest Laboratory, Richland, WA. U.S. Dept.
of Energy, Washington, DC.
Selim, H.M. and R.S. Mansell. 1976. Analytical solution of the equation
for transport of reactive solutes through soil. Water Resour. Res. 12:
528-532. As referenced in Yeh, G.T. 1981. AT123D: Analytical Transient
One-, Two-, and Three-Dimensional Simulation of Waste Transport in the
Aquifer System. ORNL-5602, Environmental Sciences Division, Pub. No. 1439.
Society of Actuaries. 1959. Build and blood pressure study. (Cited in
Bogert et al., 1973)
U.S. EPA. 1980a. Background Document, Resource Conservation and Recovery
Act, Subtitle C, Section 3001, Section 261.24. Office of Solid Waste,
Washington, DC.
6-8
-------
U.S. EPA. 1980b. Ambient Water Quality Criteria Document for Hexachloro-
cyclohexane. Prepared by the Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH, for the Office
of Water Regulations and Standards, Washington, DC. EPA 440/5-80-054. NTIS
PB 81-117657. r:
U.S. EPA. 1980c. Water Quality Criteria Documents; Availability. Federal
Register. 45(231): 79318-79379. : '
U.S. EPA. 1984a. Mercury Health ••'••Effects'' Update. Health Issue Assessment.
Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park, NC. Washington, DC. EPA
600/8-84-019F. NTIS PB 85-123925.
U.S. EPA. 1984b. Health Assessment Document for Carbon Tetrachloride.
Office of Health and Environmental 'Assessment, Environmental Criteria and
Assessment Office, Cincinnati, OH. EPA 600/8-82-001F. NTIS PB 85-124196.
U.S. EPA. 1984c. Health Assessment Document for Chloroform. Office of
Health and Environmental Assessment, Environmental Criteria and Assessment
Office, Research Triangle Park, NC. EPA 600/8-84-004A. NTIS PB 84-195163.
U.S. EPA. 1984d. Guidelines for Deriving Numerical Aquatic Site-Specific
Water Quality Criteria by Modifying National Criteria. Environmental
Research Lab., Duluth, MN. EPA/600/3-84/099. NTIS PB 85-121101/REB.
6-9
-------
U.S. EPA. 1985a. Environmental Profiles and Hazard Indices for Constitu-
ents of Municipal Sludges: Office of Water Regulations and Standards,
Washington, DC.
U.S. EPA. 1985b. Technical Support for Development of Guidance on Hydro-
geologic Criterion for Hazardous Waste Management Facility Location. Office
of Solid Waste, Washington, DC.
U.S. EPA. 1985c. Health Assessment Document for Polychlorlnated D1benzo-j>-
dloxln. Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH. EPA 600/8-84/014F. NTIS PB
86-122546.
U.S. EPA. 1985d. Drinking Water Criteria Document for Nickel,, Prepared by
the Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office, Cincinnati, OH, for the Office of Drinking Water,
Washington, DC. EPA/600/X-84/193. NTIS PB86-117801.
U.S. EPA. 1985e. National Primary Drinking Water Regulations; Synthetic
Organic Chemicals, Inorganic Chemicals and Microorganisms; Pro- posed Rule.
(40 CFR Part 141) Federal Register 50(219): 46936-47022.
U.S. EPA. 1986a. Guidelines for Carcinogen Risk Assessment. Federal
Register 51(185): 33992-34003.
U.S. EPA. 1986b. Guidelines for Health Risk Assessment of Chemical
Mixtures. Federal Register 51(185): 34014-34025.
6-10
-------
U.S. EPA. 1988. Reference Dose (RfD): Description and Use 1n Health Risk
Assessments. Integrated Risk Information System (IRIS). Online.
Intra-Agency Reference Dose (RfD) Work Group, Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office,
Cincinnati, OH. February.
U.S. EPA. 1989. Development of Risk Assessment Methodology for Land
Application and Distribution and Marketing of Municipal Sludge. Prepared by
the Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office, Cincinnati, OH, for the Office of Water Regulations
and Standards, Washington, DC. EPA/600/6-89/001. NTIS PB90-135740/AS.
Van Genuchten, M.T. 1985. Convective-dlspersive transport of solutes
Involved 1n sequential first-order decay reactions. Comput. Geosd.
11: 129-147.
Van Genuchten, M.T. and W.J. Alves. 1982. Analytical Solutions-of the One-
Dimensional Convective-D1spers1ve Solute Transport Equation. U.S. Dept.
Agrlc. Tech. Bull. No. 1661. 151 p. As referenced in Van Genuchten, M.T.
1985. Convective-d1spers1ve transport of solutes involved 1n sequential
first-order decay reactions. Comput. Geosci. 11:129-147.
Van Genuchten, M.T. and P.J. Wierenga. 1976. Mass transfer studies In
sorb- 1ng porous media. I. Analytical solutions. Soil Sci. Am. 0. 40:
473-480. AT123D: Analytical Transient One-, Two-, and Three-D1mens1onal
Simulation of Waste Transport in the Aquifer Systems. ORNL-5602,
Environmental Sciences Division, Pub. No. 1439.
6-11
-------
Wang, S.T., A.F. McMillan and B.H. Chen. 1977. Analytical model of disper-
sion in tidal fjords. J. Hydraulic Div., ASCE. 103: 737-751. As
referenced in Yen, G.T. 1981. AT123D: Analytical Transient One-, Two-,
and Three-Dimensional Simulation of Waste Transport in the Aquifer System.
ORNL-5602, Environmental Sciences Division, Pub. No. 1439. Supplied at
earlier date.
Warrick, S.T., J.W. Biggar and D.R. Nielsen. 1971. Simultaneous solute and
water transfer for an unsaturated soil. Water Resour. Res. 7: 1216-1225.
As referenced in Yeh, G.T. 1981. AT123D: Analytical Transient One-, Two-,
and Three-Dimensional Simulation of Waste Transport in the Aquifer System.
ORNL-5602, Environmental Sciences Division, Pub. No. 1439. Supplied at
earlier date.
Yeh, G.T. 1981. AT123D: Analytical Transient One-, Two-, and Three-Dimen-
sional Simulation of Waste Transport in the Aquifer System. ORNL-5602.
Environmental Sciences Div., Pub. No. 1439. Oak Ridge National Laboratory,
Oak Ridge, TN.
Yeh, G.T. and Y.J. Tsai. 1976. Analytical transient three-dimensional
modeling of effluent discharges. Water Resour. Res. 12: 533-540. As
referenced in Yeh, G.T. 1981. AT123D: Analytical Transient One-, Two-,
and Three-Dimensional Simulation of Waste Transport in the Aquifer System.
ORNL-5602, Environmental Sciences Division, Pub. No. 1439.
6-12
-------
APPENDIX A
COLUMN METHOD FOR DETERMINING RETARDATION FACTOR (RF)
AND DISTRIBUTION COEFFICIENT (Kd)
A-l
-------
A.I. SCOPE AND APPLICATION
The column methods described herein can be used to experimentally
determine the velocity of a contaminant through a column of porous
soil/rock. The method is directed to measurement of a retardation factor
(the ratio of water velocity to contaminant velocity). A distribution
coefficient can subsequently be derived based on the porosity and density of
the soil/rock matrix. The method is applicable to any porous; media through
which water-borne contaminants may flow. Water is passed through a column
of the porous media on a once-through or recirculating basis. Contaminant
is introduced continuously or as a spike. The time of travel for the
contaminant is determined by measuring contaminant in effluent volumes. The
result is compared to the velocity of water through the column. The ratio
of the two values is defined as the retardation factor.
A. 2. THEORY
RF = V /V
gw c
The column method, which measures the migration velocity of a
contaminant (V ) relative to the groundwater velocity (V ), provides a
c gw
retardation factor (RF) according to the following equation:
(A-l)
However, when a measurement is made to determine the value of a particular
contaminant retardation factor in a rock/groundwater system, the solution's
chemical composition (including pH, Eh, cations and anions), the rock's
characteristics (chemical composition, mineralogy, surface area, cation and
anion exchange capacities) and the equilibrium between rock and groundwater
should also be considered. These parameters are important because they can
greatly affect the measured value of RF.
A-2
-------
Two common expressions used to describe equilibrium adsorption reactions
are:
S =
abC
and
1 -i- aC
J/n
(A-2)
(A-3)
S = KC
where
S = contaminant concentration sorbed on the rock (yg/g)
C = contaminant concentration in solution (yg/ma.)
and a, b, K and n are constants.
These equations (Equation A-2 after Langmuir; Equation A-3 after
Freundlich) may describe the relationship between S and C for a given solid
and solution composition at a constant temperature (often called adsorption
isotherms). Both equations are commonly used for an empirical description
of experimental adsorption data.
When contaminant concentrations are small, such that aC is <1 in
Equation A-2, the isotherm equation reduces to:
S = abC (A-4)
The exponential constant 1/n in Equation A-3 is usually close to unity,
and that equation, too, can be approximated by:
S = KC (A-5)
Both Equations, A-2 and A-3, can then be approximated by:
S = KdC (A-6)
where the constants ab and K can be taken as the Kd. It is important to
remember that Equation A-6 is usually an approximation and that it holds
only under the conditions mentioned above.
A-3
-------
When a linear isotherm, such as that given by S = KdC, can be used to
describe the adsorption reaction, the transport equation for a contaminant
in equilibrium with both rock and water in a one-dimensional porous medium
flow path is:
where
D
x
e
!£
at
= D
3x2
(A-7)
ax
gw
dispersion coefficient (cm /sec)
the distance along the flow path
groundwater velocity (average pore-water velocity, cm/sec)
JW
3
b = bulk density (g/cm )
e = porosity of the porous medium
The expression that relates the Kd to the retardation factor,
RF = [1 + (b/6) Kd] (A-8)
can be substituted into Equation A-7 to obtain a simpler form for the
transport equation:
!£ = D i!£
at 3x2
+ V,
gw
ac
ax
(A-9)
If a spike of contaminant is added to the groundwater as it enters the
column, adsorption delays the elution of the peak until RF pore volumes have
been eluted. The pore volume or void volume of a column is given by the
porosity of the porous media (e) times the total column volume (CV). The
retardation factor can then be calculated from the ratio of the volume
required to elute the contaminant's peak (or maximum concentration) to the
pore volume of that column.
If the porosity is unknown, it can be calculated from:
6 = 1 - b/p (A-10)
A-4
-------
where p is the average density of the individual particles used to pack
the column.. An experimental check on the calculated pore volume can be
obtained by the elution of a nonsorbing element, which will have a maximum
concentration at exactly one pore volume.
In summary, the transport equation for contaminant migration used in
most safety assessment models utilizes a linear adsorption isotherm
(Equation A-6). Adsorption of the contaminant results in a lower migration
velocity for the contaminant than that of the groundwater: Vr = V /RF.
gw
Generally, this is true only when the groundwater composition, rock chemical
composition and temperature do not vary (i.e., they are at equilibrium).
A.3. INTERFERENCES
Interferences of two types may occur in the column method:
(1) interference in the analysis of eluent for the contaminant of interest,
and (2) interaction of the contaminant with the apparatus or column
material. In the former case, interferences are identified in the methods
prescribed for conducting the analysis required to monitor water for the
contaminant. In the latter case, tubing, pumps and column materials must be
selected that are compatible with the contaminant of interest. If
compatibility cannot be determined from analytical laboratory or materials
handling handbooks, a simple laboratory test should be conducted as a blank
run. Results of the blank run will indicate if the apparatus itself is
retarding or removing contaminant.
A-5
-------
A.4. APPARATUS AND MATERIALS
Equipment requirements vary with the selection of high- or low-pressure
systems in a single-pass or recirculating mode.
A.4.1. CONTACT COLUMN
A contact column is required to hold the soil/rock matrix during the
contact period. A typical low-pressure configuration is depicted in Figure
A-l. The column must be constructed of material that will withstand the
intended operating pressures and not interact with the groundwater, the
contaminant or the soil/rock matrix. For low-pressure experiments, a clear,
inert plastic is desirable because it permits direct observation of the
column, which will help identify problems with changes in the porous media
or bubble entrapment. The upflow configuration is preferred to facilitate
bubble migration out of the column. A double layer of screen (inert
material such as plastic) should be placed at the ends of the column to
disperse flow and reduce the end-cap volume while holding the matrix in
place.
The column diameter should be at least 30 times the average particle
size of the porous media. The column length should be at least 4 times the
column diameter. The column volume should also be selected such that
uncertainty about the volume of end-caps and tubing does not greatly affect
the estimate of pore volume.
A.4.2. SYSTEM LAYOUT -- LOW-PRESSURE METHODS
Low-pressure column studies require the use of a fluid reservoir, a
fluid delivery system, a column and an effluent collection system. Contact
with groundwater may be accomplished in a single pass or through use of a
recirculating system.
A-6
-------
Effluent Solution
Tubing
Tubing Connector Nipple
O Ring
End Cap
Screen
Column Body
Influent
FIGURE A-l
A Detailed View of the Column Used for Low-Pressure
Column Retardation Studies (Single-Pass or Recirculating)
A-7
-------
A.4.2.1. Single-Pass Column Method. A schematic of the apparatus needed
for a single-pass, low-pressure column method is illustrated in Figure A-2.
The reservoir can be constructed of any suitable, nonreacting material for
maintaining influent solution. If volatile contaminants are to be studied,
an open reservoir will not be suitable unless contaminants are injected in
line as a spike. If steady feed methods are employed, a diaphragm system
may be required to prevent volatile losses to the atmosphere.
The groundwater velocity through the column is controlled by the
hydraulic head gradient [pressure difference between the column's inlet and
outlet (AH) divided by the column length (L)] and the hydraulic
conductivity of the porous media (K) according tot'
AH
L
Vgw - K
(A-ll)
A pump is not required for/nonvolatil.e. systems if the reservoir is elevated
above the column outlet: Such a gravity feed system is practical for heads
of up to 50 cm of water. At greater heads, the physical dimensions of the
apparatus become limiting, and a pump is more desirable.
The hydraulic conductivity of the soil/rock matrix may also constrain
the size/configuration of the apparatus. If small columns (~5 cm) are
employed at a head of H = 50 cm water, the practical upper limit to the
hydraulic head gradient for a gravity feed system is +H/L = 10 cm water/cm
of column. The minimum velocity (Equation 4-10) should be 3x10
cm/sec, which limits the system to samples having values of K>3xlO
cm/sec. Less permeable media (K<10~5 cm/sec) will require a pump.
Low-pressure syringe and peristaltic pumps are available that will maintain
flow rates over a range suitable, for :controlling velocities in experiments
on relatively permeable'columns.' / ..,•..• •> , :
A-8
-------
Groundwater
Reservoir
Column
Apparatus
Spike
Injection
Valve
nnnnn
FIGURE A-2
Apparatus Needed for a Low-Pressure, Single-Pass
Column Retardation Study
A-9
-------
The effluent fraction collector can be obtained commercially or may
consist of a test-tube rack with tubes that are changed manually. If an
automated collector is employed, it should be adjusted to receive
small-volume increments (i.e., v
-------
Groundwater
Reservoir
Pump
88
I
Sampling
Port
Column
FIGURE A-3
Apparatus Needed for a Low-Pressure Recirculating
Column Retardation Study
©Registered Trademark of E.I. duPont deNemours and Co., Wilmington, DE.
A-l 1
-------
Groundwater
Reservoir
High-
Pressure
Pump
Spike
Injection
Valve
Confining
Pressure
Pump
"Pressure*
Transducer
Compressed
Air
Pressure
Transducer
Throttle
Valve
FIGURE A-4
Apparatus Needed for a High-Pressure
Column Retardation Study
®Registered Trademark of E.I. duPont deNemours and Co., Wilmington, DE.
A-12
-------
The influent pump must maintain high pressures to force liquid through
low-permeability samples at a relatively constant velocity. The maintenance
of a constant velocity is complicated by fluctuations in permeability over
time. Constant-flow-rate pumps can accommodate decreases in permeability by
increasing the pressure gradient along the column. However, a maximum
pressure-setting control is necessary for safety considerations.- When that
pressure is reached, further declines in permeability will result in
decreased groundwater velocity.
High-pressure systems are often applied for rock systems of low
permeability. When rock cores are sufficiently impermeable, the groundwater
may flow around the core down the edges of the column rather than through
the sample. To prevent such short ciWcuiting, the core can be cast in an
epoxy jacket that bonds to the rock surface(and forms a column wall. Spike
injections of contaminant are most commonly employed in high-pressure
systems.
A.5. REAGENTS
A.5.1. GROUNDWATER
To the extent possible, groundwater representative of the site of
interest should be utilized. If natural groundwaters are not available,
they can be synthesized based on key parameters such as total dissolved
solids, conductivity, ionic strength, pH, Eh and total organic carbon.
Barring the availability of good data, distilled water can be employed to
represent meteoric water. Regardless of the source, the water should be
analyzed to determine the presence or absence of the contaminant of interest.
A-13
-------
Special attention must be directed to maintaining the redox or Eh status
of the leaching solution. The solubility of metals is greatly affected by
changes in redox potential because of the presence of species couples, such
as S /S04 2, which can produce low-solubility metal salts (i.e., the
sulfides). The dissolution and/or evolution of gasses, especially
atmospheric oxygen, can greatly .affect redox potential. As a consequence,
measures should be taken to maintain leaching solutions at the desired redox
potential values. Common measures include:
o Purge oxygen from the air space above leaching solutions by
maintaining a nitrogen blanket.
o Employ a redox buffer in the leaching solution. One such buffer
is the pyrogallol-fe+2 complex. The concentration of the
two species is. selected on the basis of the desired Eh level.
o Prepare the leaching solution fresh daily and monitor Eh before
and after use of each batch.
A.5.2. CONTAMINANT
A clean source of the contaminant of interest is required to prepare
spikes or continuous-feed solutions. Certified materials should be
utilized. Spikes should be prepared as aqueous solutions prior to injection
to eliminate problems with solution kinetics. A purity check is advised
here. For organics, shelf-life is limited and, therefore, purity checks
should be conducted periodically.
A.6. SAMPLE COLLECTION, PRESERVATION AND HOLDING
Samples should be collected serially with a fraction collector or by
manual replacement of sample vials at the effluent port. Change-out time
should be selected to accumulate a sample volume <1/20 of a pore volume.
A-14
-------
Sample preservation should be done as normally prescribed for the
contaminant of interest. If preservatives are indicated, the proper amount
should be added to the sample vial and, where necessary, calculations made
to account for the added volume of fluid.
Special precautions are required for collection/preservation of volatile
contaminants. In the case of cyanides, an alkaline receiving solution in
the sample vial can be used to prevent vapor loss. For organic volatiles,
direct feed to the analytical instrument or provisions for collection in a
closed container are necessary. Holding times should be minimized.
A.7. PROCEDURE
Select the system configuration on the basis of the materials of
interest and the availability of apparatus. High-pressure systems are
required if low-permeability matrices such as rock cores are to be
evaluated.
Assemble the system sizing the column so that diameter is >30 times the
maximum particle diameter and column length is >4 times column diameter. In
all cases, the column volume should be greater than the dead volume (sum of
tubing, end-caps, sample-holding screens, etc.).
If an intact core is to be evaluated, the column must be fitted to the
core in such a manner that side flow is minimized. For low-permeability
cores, an epoxy jacket may be cast around the core. For loose aggregates,
the material must be added to the column and packed to a density
representative of natural conditions. This can be accomplished mechanically
or by repeated pulses with uncontaminated groundwater.
A-15
-------
The height of the groundwater reservoir or the pump size/speed should be
selected to accomplish the desired groundwater velocity. To reduce the
effects of diffusion, select conditions such that:
Vgw > 1.6 x 10 /L cm/sec
where
V
gw
groundwater velocity in cm/sec
L = length of the column in cm
Calculate the number of mass transfer units (n) according to:
(b) (Kd) (L) (Sk)
n =
(e) (Vgw)
(A-12)
where
b
Kd
bulk density of the soil/rock matrix (g/cm3)
distribution coefficient in
contaminant/ms. groundwater)
contaminant/g soil)/(jjg
gW
length of the column in cm
porosity of the soil matrix (dimensionless)
groundwater velocity in cm/sec
Sfc = sorption rate constant (sec"1)
to determine if equilibrium is to be expected. In general, 90% of
equilibrium is attained when n = 20, while only 50% is reached when n = 3.
If a single-pass system is employed, make up a spike solution such that
the concentration approximates contaminant levels of interest and has a
total spike volume <10% of the total pore volume.
Activate the flow system and observe until flow conditions are steady.
Activate the sample collection system. Inject the spike and note the time
of injection. Analyze effluent samples and determine the time of passage
for the centroid of the peak. Calculate the retardation factor (RF) as:
= vn .-/effective pore volume
(J. b
(A-13)
A-16
-------
where VQ 5 is the volume eluted when 50% of the total spike has passed
(the centroid of the spike).
If a constant-feed system is employed, the feed water should be brought
to the desired contaminant concentration and allowed to equilibrate. The
system is then activated with the contaminated groundwater feed and the
effluent analyzed until the effluent concentration is one-half the influent
concentration (C = 1/2 C ). The volume of eluent at the time C = 1/2 C
o o
is defined as V_ _ and can be used to calculate RF according to:
RF = VQ 5/effective pore volume (A-14)
If a recirculating system is employed, the feed groundwater is brought
to the desired contaminant concentration and flow initiated. The effluent
is monitored until effluent concentrations are equivalent to the influent.
At that time, the volume and concentration of eluent are measured to
determine the total mass of contaminant adsorbed on the column. This is
used to calculate the distribution coefficient (Kd) according to
Kd = S/Cf (A-15)
where
S = concentration of contaminant on soil/rock (ng/g) or
determined by mass of contaminant removed over mass
of the soil/rock core
C = concentration of feed water
RF is then calculated from Equation A-8.
A.8 CALCULATIONS
Methods for determining the distribution coefficient from column
adsorption studies depend on the contact system employed. If a spike feed is
utilized, it is necessary to determine when half of the mass of contaminant
A-17
-------
has passed through the column. This is accomplished through analysis of
effluent concentration data. Each sample of effluent is analyzed for the
contaminant and results plotted in terms of concentration (vertical axis)
and column or pore volumes of effluent (horizontal axis). The spike will
appear as a peak in the effluent with width and height determined by the
column dimensions, water velocity and attenuation. The area under the peak
represents the total mass of contaminant in the effluent. If the peak is
symmetrical, the centroid lies at a point directly below the maximum
concentration. The cumulative pore volume at that point is defined as
V , or the volume required for half of the spike to pass from the
0.5
column. If the peak is not symmetrical, the centroid must be located. The
centroid is defined as the vertical line dividing the area under the curve
into two equal portions. Once again, the intersection of the vertical line
with the horizontal axis defines V . The two cases are illustrated in
\j • 3
Figure A-5. The retardation factor (RF) is calculated from Equation A-8.
When a constant-feed system is utilized, the plot of concentration and
effluent volume represents a breakthrough curve, as illustrated in
Figure A-6, rather than a peak. For this system, VQ 5 is selected as the
volume at which effluent concentrations are half of the feed concentrations,
or C/C = 0.5. Once again, Equation A-8 is applied to determine the
o
retardation factor.
Once the value for RF has been determined, the distribution coefficient
(Kd) can be calculated from the conversion of Equation A-8:
Kd = (RF - l)/e/b (A-16)
where
e = porosity of the soil column (dimensionless)
b = bulk density of the soil in the column (g/cm3)
A-18
-------
O
.80-
.70-
.60-
.50-
.40-
.30-
.20-
.10-
B. Symmetrical Spike
Effluent Volume
FIGURE A-5
Selection of VQ from Spike Elution Data
A-19
-------
o
§
10
Effluent Volume
FIGURE A-6
Selection of V. _ from Continuous-Feed Data
U. D
A-20
-------
If a recirculating system is employed, Kd can be determined directly.
The influent and effluent lines are analyzed continuously for the
contaminant of interest, and the eluent volume is monitored. The
concentration and volume are recorded at the time when influent and effluent
concentrations are equivalent. The total mass of contaminant (M ) in the
system is defined as:
M = V C
T wo
(A-16)
where
VM = total volume of ?solution in the apparatus (9.)
CQ = initial concentration of contaminant in the solution (mg/8,)
The mass of dissolved contaminant (M_) at the end of the procedure is
defined as:
Mc = V C.
S w f
(A-18)
where Cf = final concentration of contaminant in solution (mg/a,).
Therefore, the absorbed mass of contaminant (Ma) is defined as:
Ma = M - M
I O
- Vw
-------
Combining Equations A-18, A-19b, A-20 and A-21, the distribution coefficient
(Kd) is calculated as:
Kd = (V /V )[C - CJ/C_]/(b/e)
W S 0 ft
(A-22)
A.9. REFERENCES
Material in this appendix is derived from the following references:
Relyea, J.F. 1981. Status report: Column method for determining
retardation factors. U.S. Dept. of Energy, Richland, WA. PNL--4031, UC-70.
Relyea, J.F. 1982. Theoretical and experimental considerations for the use
of the column method for determining retardation factors. Radioactive Waste
Management and the Nuclear Fuel Cycle. 3: 151-156. (Modified)
A-22
-------
APPENDIX B
INPUT PARAMETERS FOR CONTAMINANTS OF INTEREST
B-l
-------
B.I. DISTRIBUTION COEFFICIENTS
Distribution coefficients are required to determine how a contaminant
will partition itself between the soil particles and the soil water. The
distribution coefficient (Kd) is defined as:
Kd = S/C (B-l)
where
S = concentration of contaminant on soil (mg/kg)
C = concentration of contaminant in water (mg/S,)
The concept of Kd is a gross simplification of attenuation of inorganic
contaminants in soil. Precipitation chemistry is an important factor in
attenuation over and above adsorption and exchange. Precipitation does not
yield a solution concentration in proportion to the mass of contaminant in
the system. As a consequence, the use of a Kd is most valid at low
contaminant concentration levels where contaminants do not exceed solubility
thresholds.
For organics, the Kd concept is more broadly useful because adsorption
accounts for most soil attenuation. In the case of organics, Kd is
calculated from the distribution as a function of organic carbon content of
the soil (K ) and the fraction of soil (f ) consisting of organic
oc oc
matter as follows:
Kd = (K )(f ) (B-2)
oc oc
If values for K have not been determined experimentally, equations are
available that relate K to octanol/water partition coefficient data —
oc
(K ) or solubility.
ow
Table B-l is provided to assist the analyst in selecting Kd values for
contaminants of interest. Values for, inorganic contaminants were derived
B-2
-------
1
CD
1
VI
4-
c
10
c
I
J
T3
4-
O
01
O
CO
^
o
^
(/)
4-
c
o
°u
M-
1
c
tributio
o
5
•N.
i
V.
1
"t
CD
O
H-
CO
C4-
CO
TJ
•H i
IB O
1. -J
3
4- >
10 4-
C •"•
=5 CO
0) S
4- 10
2 3
2
4- >
10 T3
M c
C IB
1
CO
a
£
'o
CO
•
J
£
C
c
1
1
o
d
CD
d
PA
•"•
d
o
o
d
d
o
d
o c
— o
O £
u_
•v >•
*-f o
"o o
^ IB
"^v D)
O} 1»
S3
«i
* S 2 2 *° S "**
O IB 10 C 10
3 • OO 03 tSi 03 b • (/3 CD
IB— • OS «Ov 5*-« 'Cv
OIB-3 — 3 — —110 3 —
O .
10 CL
U ID 10 10
C — C *•* vs
— c co o o
i- O N N N
T3 « C C C
— l_ CD 4) CO
«c < as oa en
<<<< <^— 3
o o o o in . oo CM o
d FA d do vo 03 CM d
*M _ . ^f} j^
oooo ococAoa
o *t o o d \o — dp
en
P
. X „
r-j
^f ^ • \O
O\ CO fV hO
IT\ CO
CNJ \O O CN
• " • r"* • ' -
•!•» O — O
in in — CM
fACMO^" «CO «O
VOCM -in »VO »CM
fA^-pACO — inCOCNO
in co
CM Ov \O CM 00 »» Ov
r*» -o o-ov-o
d — d — d — — 5d
in
o
X
c\i«cin -co<:«c«s<
f-Z\O — CMZZZZ
CO
— ~
x o
a> —
jz j=
- §
-H ® ffl E
T3 ' O — •+- ^ "O
*0 E jQ O O O *0 CL C
W.JCT3 i_ — — t_ ^j CL *D
— 4- (0 •
CQ a. c_> u o o o o o o
B-3
-------
ci
1
1
r-i
*•*,
ot
"S.
I
•s.
'c
*x
I
UJ
4-
c
C
"c
H-
1
o
8
1
"u
ti
5
JO
•gi
's.~->
3"O
•fei
toco
TO
i-g
I5
s
•o o
0_J
u &
£U
CO
{A ^
C/)
o
d
o
d
— -
o
"is
"6
U_J
lo-?
i/i-
sc^
•o
<0 O
O "O
II
8
d
in
S
d
•o
o
4-
10
UTO
3 C
4- «J
§
1
—
*O
1X1 (0 ••• 3 C — C — C — » £ i r i 1 1 1 1 1 1 1 1 1
10 O O 10 E~OX!— MO O O — E flj 0) H) H)
3^ Ofl OIO 3— LU XI M O in O«J o"lT3 O^ OOO Z>— SJ — 3 — d —
C3COC3OO OOO OO
ddddd odd r-ddo
O\ 00
OOOOO OOO OO
ddddd dddinjffddcM
n
» i
0 0
.
X X
vO
in CM CM o *r m co
CM O\ CM O •
voo — ino xim— — CM
H
I
n I 2 f
OOO
— — X —
XX— X 0 C3CO
UN M » * d rO^CO --
O «»
~ k •
in — in x o ^.^foo
H
I
no*
i — i
0 0
— X —
co
SX — XO *"l«^00 «3- — — CM
• • •••inco««cM
incMov^fo OOO^CMOOCO
VO
o v
in CM iy o 21 co co r^ ^£ z *•• ~~ z
i
£ S § ; '
O •— fl} o
a> 1 § -5
•e w N (0
g. o c 4-
O L. T3 4- S /^
§O — — U O O
— o c o i- i- c
\ .c ID c — o — oo o
UJ o ~ >. -S j: — — 0) •- >.
aO*—'U4KU 0)UO 0)4-3
^14-— fflO 4-tOldC'DTJ'OO
Q * 8 — ' •- — '• Q. X X g "0 _= — t.
a CM"IOO o u. a: z z — _i !IJ z £
B-4
-------
,-.
+r
••
o
T
m
LU
CD
m
S
^
3
a
|
4-
C
._
o
v»-
8
C
O
3
.a
L.
4-
5
"O
9)
O
C
£
ii
-o|
<0
3-0
4- C
10 10
COCO
•o
"fe-o
L. C
3 <0
4- CO
10
CO
•^ o
O— J
u to
3 —
4-O
10
|±
to g
+- i
10 O
1 — 1
3
IS*
« —
C3CO
•o
0) E
4- 10
IH O
i__J
£*
U C
3 CO
•o
m
£
10
IB CO
1/1
C
0)
a
>•
i
j
4-
c
10
c
'i
10
4-
8
o
d
o
d
10
d
Wl
8
d
in
d
_
O
8
d
I!
fl
v.S-
Is
X.
.-* VI
o*—
!O> C
~v OJ
5^
^•S ">• >-
4^ V 4^ 4^
CD CM b ^
>, 4)
. . - '4-4- 4- O ......
< < o e — LTt O .- CA
o * *
— o o
X
X X
VO VO
CN f^ , • CM '• -CO ~ O
Z S Z -CM-rOZvOfS
0
c
— u
1 -g.. 1-
a. o io a>
EC C O
(D — l-
3 O ' 1- 1- <* O <0
c — jc O — S— c
a Q jz4- — >>3-ca}
•o — 4- tjc — .cc— o j:
jQ - jtf i. -H C C >-J= ffi i- (0
5 .iJ± g J* ^ o.-o t g
3EZZ O.O-CUCUJaCOH-^
B-5
-------
^
1
"7
CD
UJ
H
u
§
u
i
m
5
1
|
UJ
t
o
'o
H-
**-
5
c
0
Distribut
i
•o
v:
1. >
10 a
oot/J
XI
"fe-o
J0 o
1 1
3
^ GO
8
o
saturated
ndy Loam
^^
§
d
•o
Cl
•(-
o
1--D
•jj g
iQCn
5
1
1-
—
'5
.
0
o c
— o
CTO
°ih
if
^ u
^•s. en
fs
•t-
c
c
1
s
o
< .s < <
UJ — LLI UJ
CO C 10 <0
00 CO 1 ~ 00 CO COCO
• Ox >**~ • O\ • C^\
3 — —110 3— 3 —
o o o
O O O ON
0\
O O O 1^
O O O (M
VO
CM — —
CM CO —
docs
o> in
a\ CM vO O\
O*7OO^
1
O
X m
§•"• r**
0
CM o d —
CO 0 «•
• — 5 — z
0
O 10
§ -§.
— wo
1 ^ ^
s _ °.
O VI O
JZ 1- —
0 O >. 0
•— — c c
0
yS
u
1
10
1
^^
u
1
B-6
-------
from the literature for sandy and sandy-Toam .soils. No difference is
anticipated between unsaturated and saturated soils. The analyst should
select the soil condition most closely matched to soils found on the site
for selection of the Kd. .,...'
For organic contaminants, the Kd is a function of organic content in
soil. As a consequence, the analyst has two options:
1. If the organic content of the soil on the site is known, the
Koc value should be selected from Table B-l and the Kd
calculated from Equation B-2.
2. If the organic content of the soil on the site is not known,
the soil classification should be matched with those soil types
provided in Table 8-1 and the associated Kd value selected.
It is assumed that subsoils in the aquifer will not have organic matter and,
therefore, the Kd for organics in the saturated zone is equated to zero.
This is conservative in that research suggests that at low organic levels
(i.e., <0.1%), organics interact with clay minerals. However, these
interactions are not well understood and no means of prediction is currently
available. Therefore, retention in the saturated zone is not considered at
this time.
Whenever specific Kd values are available for the on-site soil, they
should be employed in place of the values provided in Table B-l. Use of
such data should be accompanied by detailed documentation on how they were
derived. ,; :
B.2. HENRY'S LAW CONSTANTS
The Henry's Law Constant allows one to calculate vapor concentrations
over a solution as a function of the contaminant's concentration in the
solution. If Henry's Law Constants (H) have not been derived
experimentally, they are estimated according to:
B-7
-------
H + Pvp/S (B-3)
where
P » vapor pressure of contaminant (atm)
S = solubility of contaminant in water (mol/m3)
Both S and P need to be measured at the same temperature. Hence, if
vapor pressures are given at a different temperature, they must be
adjusted. The Henry's Law Constant can also be determined from
thermodynamic data describing the free energy of solution if such data are
available. This approach considers the reaction for dissolution of a gas:
A(l) = A(g) (B-4)
The equilibrium constant for the above reaction is:
Ks ,=
where (A(g)} is the activity of constituent A in the atmosphere and
(A(l)} is the activity in solution. By definition, the activity in the
atmosphere is equal to the partial pressure and the activity in solution is
equal to the concentration in solution, multiplied times an activity
coefficient. Equation B-5 can then be rewritten as:
PA
Ks = _ ft
where
P.
(B-6)
, [A(1)JYA
= partial pressure of constituent A (atm)
[A(l)] = concentration of A in solution (M/S.)
Y. = activity coefficient of A (dimensionless)
Because PA/[A(1)] is the Henry's Law Constant, Equation B-6 can be
rewritten as:
H = Ks
(B-7)
B-8
-------
TABLE B-2
Activity Coefficients for Species of Various Charge
for Various Ionic Strengths
I (M/fc)
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0,08
0.09
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.24
0.26
0.28
0.30
0.32
Y0a
1.00
1.00
1.01
1.01
. 1.01
1.01
1 .02
1.02
1 .02
1 .02
1.03
1.03
1.04
1.04
1.05
1.05
1.06
1.06
1.07
1.07
1.08
Y±lb
0.901
0.867
0.844
0.825
0.810
0.797
0.786
0.776
0,767
0.776
0.764
0.755
0.747
0.740
0.734
0.728
0.724
0.720
0.716
0.713
0.710
Y±2C
0.658
0.565
0.507
0.464
0.431
0.404
0.382
0.362
0.346
0.363
0.325
0.324
0.311
0.299
0.290
0.281
0.274
0.268
0.263
0.258
0.254
B-9
-------
TABLE B-2 (cont.)
I (M/!l)
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
Yoa
1.08
1.09
1.09
1.10
1.10
1.11
1.11
1.12
1.12
Y±l
0.708
0.706
0.704
0.702
0.700
0.699
0.698
0.697
0.696
Y±2C
0.251
0.248
0.245
0.243
0.241
0.239
9.238
0.236
0.235
- Activity coefficient of uncharged species
« Activity coefficient of singly charged species
Cyj-2 - Activity coefficient of doubly charged species
B-10
-------
The activity coefficient depends on the ionic strength of the solution.
Representative values are given in Table B-2.
The equilibrium constant, Ks, can be calculated from the free energy of
reaction B-4:
-------
For charged species and 0.1 < I < 0.5:
-logy = 0.5 Z= / VI
I
- 0.21)
(Butler, 1964)
For unchanged species
log YQ = KI
where K is a constant. Unless otherwise given, K = 0.10 as suggested by
Butler (1964).
Values for H and H' were found in the literature or derived for the
contaminants of interest and are listed in Table B-3. The references
indicate where the values or the inputs for derivation of values were
obtained. The notes specify the method of derivation when published values
were not found.
B.3. POROUS MEDIA HYDROL06IC PROPERTIES
The methodology for evaluating disposal of municipal sewage sludges
requires the input of various site-specific values related to hydrologic
flow in soils and other geologic media. Some of these values must be
determined by direct measurement, while others can be selected from reported
values for given soil types or aquifer media. The following tables and
figures present typical values to assist the applicant and/or reviewer in
determining the reasonableness of values derived for specific applications:
o Table B-4 provides typical values for the slope df the moisture
retention curve for soils that may be found in the unsaturated
zone.
o Figure B-l provides ranges of values for saturated hydraulic
conductivity of different aquifer media.
o Table B-5 provides ranges of values for porosity of unsaturated
and saturated zone media.
B-l 2
-------
TABLE B-3
r™c+an
Constants
Con^tants <"> an<* Dimensionless Henry's Law
1) (Assumed Temperature: 20°C)a for Selected Contaminants
Contaminant
Aldrin
Arsenic
Benzene
Benzo(a)anthracene
Benzo(a)pyrene
Bis(2-ethylhexyl)
phthalate
Carbon tetrachloride
Cadmium
Chlordane
Chloroform
Chromium
Cobalt
Copper
Cyanide
DDT/ODE/ODD
2,4-Dichlorophenoxy-
acetic acid
Dieldrin
Dimethylnitrosamine
Fluoride
H
(atm-ma/mol)
1.4xlO~5
NV
5.5xlO-3
NV
NV
1.0
2.3xlO-2
NV
0.59
4.8X10-3
NV
NV
NV
1.9xlO~3
3.8x10-3
9xlO~s
2x1 0~7
4.9X10"4
(PH = 6):
1x10-'
(PH = 7):
IxlO-e
H1
(Dimensionless)
6.1x10-*
NV
2.4X1Q-1
NV
NV
40
9.7xlO-i
NV
24
2.0X10-1
NV
NV
NV
0.082
1.7xlO-3
3.7X10-3
8.9xlO-6
2.0xlO-2
(PH =6):
4.4x10-6
(PH = 7):
4.4x10-7
Reference
Lyman et al., 1982
Lyman et al., 1982
U.S. EPA, 1985afa
U.S. EPA, 1985ab
Lyman et al., 1982
U.S. EPA, 1985ab
Lyman et al., 1982
c/ for 10°C
Lyman et al., 1980
Oawson et at. , 1980b
Lyman et al., 1982
Dawson et al., 1980b
c/ for 10°C
B-13
-------
TABLE B-3 (cont.)
Contaminant
Heptachlor
Hexachlorobenzene
Hexachl orobutadi ene
Iron
Lead
Lindane
Malathion
Mercury
Methyl ene bis
(2-chloroaniline)
Methyl ene chloride
Methyl ethyl ketone
Molybdenum
Nickel
Nitrate
Pentachlorophenol
Phenanthrene
Phenol
Polychlorinated
biphenyls:
Aroclor 1242
Aroclor 1254
Aroclor 1248
Aroclor 1260
H
(atm-ma/mol)
7x10-=
3.7xlO-s
3.73
NV
NV
4.8x10-'
1.2X10-7
l.lxlO-2
5.1x10-7
3xlO-3
20.8
NV
NV
NV
3.4x10-*
3.9xlO-s
3xlO-7
5.6x10-*
2.7xlO-3
3.5xlO-3
7.1xlO-3
H1
(Dimensionless)
2.9X10-3
1.5X10-3
122
NV
NV
2.2xlO-5
5xlO-6
4.8X10"1
2.1X10-6
K3X10-1
900
NV
NV
NV
1.5x10-*
1.7xlO-3
1.2xlQ-s
2.4xlO-2
1.2X10-1
1.6X10-1
S.OxlQ-1
Reference
Dawson et al., 1980b
U.S. EPA, 1985ab
Verschueren, 1983b
Lyman et al., 1982
Dawson et al., 1980b
Lymam et al., 1982
U.S. EPA, 1985ab;
SRI, 1984&
Lymari et al . , 1982
U.S. EPA, 1985ab
Lyman et al., 1982
Lyman et al., 1982
U.S. EPA, 1985ab
Lyman et al., 1982
Lyman et al . , 1982
Lyman et al., 1982
Lyman et al . , 1982
Selenium
NV
NV
B-14
-------
TABLE B-3 (cont.)
Contaminant
Tetrachloroethylene
Toxaphene
Trichloroethylene
Tricresyl phosphate
Vinyl chloride
Zinc
H
(atm-m3/mol)
8.3xlO"3
5.4x10-2
1x10-2
1.5x10-2
2.4
NV
H1
(Dimensionless)
3.4xlO-i
2.2
4.2X10-1
0.61
99
NV
Reference
Lyman et al.,
Dawson et al . ,
Lyman et al.,
MSOSb
Lyman et al.,
1982
198Qb
1982
1982
Henry's Law Distant can be estimated by (Thibodeaux,
H, = 16 Pv (MW)
where
H'
PV
(MW)
SOL
T
(SOL) T
Henry's Law Constant (cm3/cm3)
saturation vapor pressure of the contaminant (mm Hg)
molecular weight of the compound (g/g mol)
contaminant's solubility in water (ppm)
ambient temperature (°K)
NV = A calculation of Henry's Law Constants for these materials is
not meaningful. No measurable vapor levels are anticipated.
B-15
-------
TABLE B-4
Typical Values for Slope of Soil Moisture Retention Curve*
Soil Texture
Clay
Silty clay
Silty clay loam
Clay loam
Sandy clay loam
Sandy silt loam
Silty loam
Sandy loam
Loamy sand
Sand
Value for Curve (b)
11.7
9.9
7.5
8.5
7.5
5.4
4.8
6.3
5.6
4.0
*Source: Hall et al., 1977
B-16
-------
cb
11
c
i5 <
^
I
0
Rocks
Unconsolidated /f /f K K K
Deposits ^ (darcy) (cm2) (cm/s) (m/s) (qai/dav/
ECQ 1
^•s 1
rJQ "D
0 c
0 C 0)
is CD CO-*
CO p- £5
y e « S
]
c
> o ^
- CD O
si I
73 ^"
£ E I
3 CO W ®
11 §1
"" en's
1 I0'
B
2
/>
— , co
1 CO
Conversion
1
Factors
c in
^2-» -!
2 o = J
«*^ <— O i
c: = CD C/
13 ^ C
CD O)
5~
I
1
c
CD
CO
*
<
3
5
5
CO
£
i .
C C
CO
0
f_
9
s
CO
?
CD
P
4.
J
J
I/
3
CO
LL
3
I 21$
^
4
(
~>
a
D
3
23
>
I
5
CD
«
CO
^
Permeability, k*
ft2
darcy
m/s
ft/s
gal/day/ft2
cm
r
9.29 X
1.02 X
3.11 X
5.42 X
in ft^ miiltin
110°-9 !'
10~3 1
08
06
m
ft
z
|-10S
-104
-103
-102
-10
•1
•10~1
-io-2
io-3
io-4
10~5
io-6
10~7
r-10-3
-io-4
-io-5
-io-6
-io-7
-io-8
-io-9
io-10
io-11
io-12
io-13
io-14
io-15
r-102
-10
-1
-io-1
•IO-2
-io-3
-io-4
-io-5
io-6
10~7
io-8
10~9
io-10
-1
r-106
-io-2
-io-3
-io-4
"10~5
-10~6
-io-7
io-8
io-9
io-10
io-11
10~12
0-8 g-16 1Q-11 Q-13
-10s
•104
-103
-102
10
I
_1
0
0"z
0~3
.
0
j-5
0~6
O"7
Hydraulic conductivity, K
darcy m/s ft/s gal/day/ft2
X 10~3 1.01 >
X 10~11 1
V
10-4 3.35 X
10~10 5.83 X
10"
6
io-7
io-13
^ .
A \.
C108 9.80 X102 3.22 X103 1.85 X 109
C 1010 9.11 X 105 2.99 X 10s 1.71 X 1012
5 y.oo X 10 3.17 X 10 1.82 X 10
3.15 X104 3.05 X10~1 '1 574 X105
5.49 X10~2 4.72 X 10~7 1.74 X10~6 1
FIGURE B-l
Representative Values for Saturated Hydraulic Conductivity
Source: Freeze and Cherry, 1979
B-l 7
-------
.TABLE, B-5
Porosity Values for Porous Media
A. Representati ve Values for Porosity
Material
Coarse gravel
Medium gravel
Fine gravel
Coarse sand
Medium sand
Fine sand
Silt
Clay
Porosity
28%
32%
34%
39%
39%
43%
46%
42%
B. Effective Porosities for General Hydrogeologic Classifications*
Generic Classification
Effective Porosity
(Dimensionless)
Fractured Crystalline Silicates
Fractured and Solutioned Carbonates
Porous Carbonates
Porous Silicates
Porous Unconsolidated Silicates
Fractured Shale
0.01
0.10
0.10
0.01
Average Value
0.16
0.01
*Source: Shafer et a!., 1984
B-18
-------
8.4. GEOCHEMICAL CONSIDERATIONS
The following series of figures are provided to convert unsaturated zone
contaminant concentratipns to resulting saturated zone concentrations based
on geochemical interactions. Each figure addresses a specific inorganic
contaminant (arsenic, 3.1; mercury, 3.2; lead, 3.3; copper, 3.4; and nickel,
3.5). Six curves are provided for each contaminant (a-f) depicting
relations for a different set of pH and Eh conditions. The pH values
included are 6.0 and 7.0. Eh values are -200 mv, +150 mv and +500 mv.
Directions for use of the curves can be found in Section 4.3.3.1.
B-19
-------
o
Is
|
o
O
•+•*
Q.
C
CO C
re
3
O
o
IT)
X
o
II
Q.
1/1
C
o
auoz
B-20
-------
o
'
c
o
It)
•f—
o
O)
o.
CO
CO i_
J3
w ^, ^ ~
< n S 5
-------
in
• °S
in
o
TJ-
in
.f»:
co
o
, p
co
o
Q)
o
c
o
O
*-*
Q.
in
.CM
c\i
O
11
O
O
c
a>
( //BLU 01-301.
UO!»BJiU80UOQ 0|U8SJV
8UOZ
B-22
-------
o
1
0)
u
o
o
+-•
Q.
53 o>
oo c
0>
LU U
Q£ C
=3 O
CD CJ
c
o>
(U
o
IV)
0>
(O
00
o
o
X
o
T3
II
3=
Q.
O
o
CJ
oiuaaty PSAIOSSIQ
auoz
B-23
-------
o
13
+-*
I
o
O
*^
Q.
u
'c
to
•»—
u
a.
oo
(D
0>
ITJ
C
o
OJUBSJV POAIOSSIQ
auoz
B-24
-------
g
OB
4^
c
co
o
o
O
Q.
C
g
0)
o
N
•a
2
co
1
CM
CD
(9
a
o
O)
a.
co
n.
03
C
O
re
a>
u
c
O
O
a>
v\
•o
0)
ra
oo
o
o
CVJ
I
X
o
TJ
• a>
:r
o.
C
O
o
o
{276LU U.-3QI. S9LU!i)
uo|}Bj;uaouoo omasjy PSAIOSSIQ |B;OI auoz
B-25
-------
o
j
1
co
CM
d
co
CM
d
•<*
CM
d
CM
CM
d
o
CM
d
co
d
CO
" d
" d
CM
B T~
, d
o
• T™
d
CO
-o
d
co
. o
d
•<*•
.0
CM
• O
d
0
d" — o « i ? : i o § $ ' " "a s § °
?i d d dodo do odd
C
g
"to
IE-
CD
o
o
0
"3
Q- _
C *v
— co
>>-^ i
3<: °°
« ? LU
CO O£i
*s "TZ ^3 '
•^ U o
CD t->
f* 1 1
O
N
•D
*2
to
1^
^
CO
V)
K.
O
-4^
A3
U
CU
CL
CO
4_
CU
-p
(O
C
o
•^~
•p
O
p-
II
3^
Q.
l/l
C
O
•IH)
•o
C
o
O
uo!iBj;ueouoo
euoz
B-26
-------
O
O
co
CM
I
O
o
•^>
CM
CM
CM
O
O
O
CM
CO
O
o
CO
I I
o o
o o
Tt CM
o
o
o
o
co
t
o
o
CO
I
o
o
co
CM
d
s- o
o> in
C
tJ n™*
c
O II
a:
>> o.
u
o>
O)
c
o
-o
O)
CO
OO
C
o
O
CM
_ O
o
q
d
PSAIOSSIQ
euoz
B-27
-------
I
c
c
J
]
co
CM
d
CO
CM
d
'd
CM
CM
1 d
o
. CM
0
co
" d
CO
• o
' d
CM
d
o
T-
d
CO
o
d
03
o
" d
-S
d
CM
- o
d
0
i O
1 1 1 T- 1— | i o
o o o o q q q q q
5O (^ CO Ifi •«* CO CM ••" ; O
g
OT
i_
"c
8
c
o
O
<4—
Q.
C
O O)
k. C
4)
CD
C
o
N
T3
13
3
13
CO
u
co
1
CQ
UJ
Q£
C9
h-4
U.
C
p
'£
n
u
>
• 3
U
CU
cu
c
o
-o
cu
2
2
(O
CO
i
o
0
C\J
1
II
X
o
T3
cu
OH
o
r-
II
^_
o.
..
c
o
4-*
T3
^Z
o
o
( //6ri 9-301. sauun)
uoijBJjueouoo Ajnojei/\| p9A|ossia
euoz
B-28
-------
O
O
O
•*-•
Q.
C
.
00)
u
U
C •!-
o -a
rvi c
o
-------
(O
CM
~T
CM
CM
CM
O
CM
T
oo
T 1 T 1 T
(O ^" CM O CO
T
co
T
CM
CO
oo e
1^ O
CD in
•H r-
rtJ
II
0)
CO
I
03
CM
O
O
T™*
d
oo
o
CO
q
o
••a-
o
CM
_ C3
CD
•S oe.
CD 1-4
•a
CD
to
CO
(O
c
Zl.
C X
o o
"'"* ^3
;4-> CU
ro oe.
s_
0>
u
CD
0)
t
O
cu
CO
ro
oo
-------
co
'Oil
c
O
•M
IQ
•r~
O
O)
Q.
OO
V.
OS
01
O
O
O
o o)
<5 5.
2D
Q)
I
1
in
CD
C
OJ
UJ O
a: c
=3 O „
CO CJ I'
vO
O
O)
-o
O)
-t->
(O
=3
+•>
fD
CO
a.
(/I
c
o
c
o
o
( J/6rl f-
uojiBJjueouoo Ajnoj9y\j
|e;oj. auoz
B-31
-------
O
O
O
q
cd
o
q
cd
O
o
o
o
+-f
Q.
C ,
na
O
i
ca
CO
" °.
CJ
(O
u
o>
Q.
co
O)
u-
(XJ
c
o
0>
u
c
o
o
d)
a>
C
O
-o
a>
(O
fO
to
>
o
o
o
p~
II
C
o
o
cj
o
to
CO
o
T
CO
CM
00
03
q
c\i
o
q
c>
( //BUI)
peAiossjQ
euoz
B-32
-------
_o
5
*^
0)
o
o
O
•*->
CL
C _
~~ «^j
TJ *•>
09 O>
o
N
ca
03
CO
C
ID
ca
o
C
o
u
§
T3
ITJ
O)
O)
C
o
-o
a>
oo
o
in
II
LU
*l
O
1/1
C
o
o
CJ
(7/Buj)
peel paA|ossia
euoz
B-33
-------
to
o
oo
°i
cd
I
•*
--
I
r-~
O
o
o
o
,0
o
.0
(£>
o
1
4-1
0)
c
o
o
CD o>
CD P
-23. os
o
• o
T3
I
_2
CO
D
o
— CD
Cvj
C
o
ro
•r~
U
o>
Q.
OO
s_
CD
(O
O
(d
cu
o
>-> o
U_ O
•o
03
O)
d)
C
o
l-sl
•o
o>
03
•M
to
O
o
CM
I
O
vO
II
c
0
O
CJ
o
• o
o
o
(g-Ot X J/6UI)
UOUBJ»U80UOQ PB8T p8A|OSS!Q
8UOZ
B-34
-------
o
1
•*-*
I
o
O
Q.
T3 -•
•s ^
T3
£
05
CO
O
03
u
o>
CL
oo
O)
c
o
0)
c:
O
TJ
(O
O)
c
o
"O
CD
HJ
o
o
in
O
10
II
I/I
c
o
TJ
C
o
o
peen peAiossjd iBtoi 9uoz
B-35
-------
o
1
•4-*
0)
u
o
O
•t^
a.
e
o
fO
Q.
00
o>
>
o
in
0)
I
CD
ID ^>
-------
o
•«r
co
52
a)
o
,q
o
o
.o
CO
o
,0
co
o
0
.0
"5
o
o
O
'3
Q.
C ,
s
CO
o
c\i
to
c
o
•Jj
«3
-M
c
CO
u
c
o
C_J
ro
o>
c
o
-o
O)
ro
oo
JZ.
UJ
•o
c
o
o
•* h-
j: o o
^ CM o
o o o
0
9_OI- X
PB9T P9AIOSSIQ |BJOi QUOZ
B-37
-------
'CM
CO
•5
CM
.CD
o
'2
c
o>
u
c
o
O
•^^
Q.
C.
a?
C
O
(O
u
CD
O
N
TO
0)
CD
(O
(U
u
c
o
Q.
Q.
O
C
O
r-j
03
II
3:
o.
c
o
o
o
CM
o
o
OJ
o
c?
CO
CM
O
CO
CM
O
CO
C4
-------
CM
O
r
in
co
0)
10
CO
d
CO
: T-~
d
o
d
CO
CO
c-
-CNJ
co
•5
C\l
CD
O
CO
0)
O
O
O
•+-*
Q.
Q.
O
O
0)
c
o
o
o>
Q.
i.
cu
c
o
•° •£
in 5°
i |
s 5
o
10
« s
oo
'd
in
N
o
co
CO
c
o>
o.
CL
o
o
II
a:
a.
c
o
cu •<-
O
CJ
•in
CM
a>
n:
10
O
o
,-f
LU
o
uoj;ej}usouoQ jaddoo
iejoi euoz
B-39
-------
o in co
t^ o TT
CO
•5
CXI
eb
CO
- 0
in
o
OB
CD
O
O
O
*-«
Q.
— ^ ^
o3 ^ co
Sg ui
o S a:
o
N
T3
co
co
o
•r—
•M
u
O)
Q.
§
c
a>
u
O)
Q.
OL
O
O)
C
O
h-J
T3
O)
(O
O
I—
II
in
c=
O
C
o
o
1 IO
CM
O
O
co cxi o> in 'r co T
T-; iq cq cxj 'H en co
^* co tn m ^^ co co
'111
o r*- in o co co o
cq ^ f^ en cxi cq o
d 2 cri 06 oci N^
X
uo!iBj;u80uoo jaddoo
auoz
8-40
-------
1 CM
cq
o
C\j
CD
O
2
0)
o
o
O
•*-
Q.
C .
oE
°
oq
ci
m
CD
o
M
co
CO
(O
Q.
00
L.
O)
c
o
2
4->
C"
0>
u
tg O
S o
^ s_
a>
Q.
Q.
O
o
I
OQ
O
O
LD
II
.c
II
CL
c
o
ro
O
O
in
CM
§
uu
OJ
rj;
CO
CO
oq
e\i
e\
Csj
e\i
05
in
CNJ
B-41
-------
•CM
£
d
T~
oo
5
eo
?
o
m
co
05
in
T
co
in
T
CO
T
o
m
m
cq
o
_g
os
*-•
CD
O
O
O
CM
CO
CO
O
If)
.
o
&
co
3
to
CO
o>
in
CO
Of
CD
H-)
c
o
10
•i—
o
cu
o.
CO
O)
(Q
c
o
(O
•»->
c
cu
<_>
c
o
o
o.
o.
o
o
cu
c
o
fsl
T3
cu
(O
(O
to
•m
CM
o
o
LU
CM o
-------
oq
-o
in
- CM
C£3
T"
O
o
"*3
£
I
o
o
3
Q.
C
g
o
N
T3
O
OB
CO
c
C
o
CO
u
o>
Q.
oo
o
o
CM
I
II
I/I
in
az u
3 I -
C " tt
o.
CL
3 5
l i
^ o
—. t-J
to
OJ T-
*. o
(8-301- X//6LU)
uojjejjuaouoo Jaddoo paAiossjQ
auoz
B-43
-------
o
1
4~i
CD
U
O
O
*^
Q.
•M
n)
•r-
U
U C
•t— O
Q> •!-
C T3
O C
M O
CJ
-o
O)
4->
2
3
-(->
(O
oo
tp in
p9A|ossiQ
auoz
B-44
-------
UOJIBJUJQOUOO |9>P!N
8-45
-------
o
.. °o
in
eo
T
o
in
o>
CM
T
o
cq
CO
CM
T
o
T
o
oo
OJ
LO
CO
co
a>
'"*.
O)
c
CD
O
O
O
*•*
Q.
u
vO
ca
co
o
•i. on
n =>
I
(X)
CO
c
CM
, h-
c
o
to
U
cu
s_
(D
c
o
c
O)
u
c
o
o
OJ
c
§
O)
3
OS
O
o
CM
I
II
a:
ex
i/l
C
o
-g
o
o
a>
LO
co
-co
OJ
o
o
•UJ
o
S o
|8>JO!N
IBIOJ.
B-46
-------
o
o £
I
CO
CO
c
c
o
IX}
•^
u
c
o
•f—
<& 2
en £
^ S
>
O
O
II
J=
UJ
»
O
r—
II
oc
Q.
•r- O
0)
C
o
•o
O)
-(->
ra
00
O
o
UOIJBJJU90UOO |9>|0!N p9A(OSS|a
9UOZ
B-47
-------
0>
o
'I
4->
0)
o
o
O
Q.
_ .—. t
« f ^
«H 1
C t-i
o u.
N
CO
in
c
(O
u
O)
o.
oo
0>
c
o
03
C
U
c
o
O)
o
in
o
•
r~
II
C
O
0> T-
O
r-j
0>
-!->
VO
(T3
OO
O
o
(//Buu)
|3>|3iN paA|ossia
3UOZ
B-48
-------
in
CO
in
o>
C\J
CD
CO
oo
I
o
iri
CM
in
CD
CD
o
CO
o
o
2
'c
CO
o
c
o
O
•4*1
Q.
(O
o
O)
Q.
oo
O)
o
o
CVJ
I
•JO
co
oo
o
C\J
TT
<0
o
N
•a
0)
CO
n
-------
B.4. .REFERENCES
Butler, 3.N. 1964. Ionic Equilibria^ Add1son-Wesley Pub!. Co., Menlo
Park, CA.
Chemical Rubber Co. Yearly. Handbook of Chem1s,try and Physics. Cleveland,
OH.
Dawson, G.W., C.J. English and S.E. Petty, ,1980. Physical Chemical
Properties of Hazardous Waste Constituents. Office of Solid Waste, U.S.
EPA, Washington, DC. . ,. , .... r
Freeze, R.A. and 3.A. Cherry. 1979. Groundwater. Prentice-Hall, Englewood
CUffs, NJ. .. •._••-.• ,.- -;•.' •• •-,. ••• <•• : • • •
Hall, D.G.H., A.J. Reeve, A.J. Thomasson and V.F. Wright. 1977. Water
Retention, Porosity, and Density of Field Soils. Soil Survey Tech. Monogr.
9. Rothamsted Experimental Station, Harpenden, England.
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt. 1982. Handbook of Chemical
Property Estimation Methods. HcGraw H111, San Francisco, CA.
HSDS (Material Safety Data Sheets). General Electric Company. 1977.
TMcresyl phosphate - No. 322. Schnectady, NY, and Monsanto Company.
1971. TMcresyl phosphate - No. OSHA-20 (44-14387). St. Louis, MO.
B-50
-------
O'Melia, C.R. and W. Stumm. 1967. Aggregation of silica dispersions by
iron (III). J. Colloid. Interface Sci. 23: 437-447. As referenced in
Battelle, Pacific Northwest Laboratory, 1984. Chemical Attenuation Rates.
Coefficients, and Constants in Leachate Migration. EPRI EA-5356. Volume 1.
Electric Power Research Institute, Palo Alto, CA.
Shafer, J.M., P.L. Oberlander and R.L. Skaggs. 1984. Mitigative techniques
and analysis of generic site conditions for groundwater contamination
associated with severe accidents. NUREG/CR-3681, PNL-5072. Nuclear
Regulatory Commission, Bethesda, MD.
Thibodeaux, L.J. 1979. Chemodynamics. John Wiley and Sons, New York, NY.
U.S. EPA. 1985a. Environmental Profiles and Hazard Indices for
Constituents of Municipal Sludges. Office of Water, Washington, DC.
U.S. EPA. 1985b. Sorption Protocol Evaluation for OSW Chemicals. Athens
Environmental Research Laboratory, Athens, GA.
*U.S. GOVERNMENT PRINTING OFFICE: 19 3 0 . 7 * 8 . 1 5 9/2 0 17 9
B-51
-------
-------
-------
00
TJ =•
m
O)
CD
CX>
2-9 5
=r o 2
-9 3
58"
Ml
0)1
------- |