EPA600/R-06/087
June 2006
Exposures and Internal Doses of
Trihalomethanes in Humans:
Multi-Route Contributions from
Drinking Water
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
NOTICE
The U.S. Environmental Protection Agency through its Office of Research and
Development funded and managed the research described here under Purchase Order
no. 3C-R3350WASX to Wilkes Technologies, Inc. It has been subjected to the
Agency's peer and administrative review and has been approved for publication as an
EPA document. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
ABSTRACT
The concentrations of the four commonly-identified trihalomethanes (THM;
chloroform, bromodichloromethane, chlorodibromomethane and bromoform) in U.S.
drinking water systems are regulated as a group. This report develops, applies and
communicates a method to estimate internal exposures to these simultaneously-
exposed chemicals. Because they are present in water used for drinking, bathing and
other household uses, and because they are highly volatile, this work evaluated the
development of internal doses via the oral, dermal and inhalation routes following
residential exposures. This was accomplished by integrating several data sets that
characterize human activity patterns, water use behavior, household volumes and
ventilation, and THM concentration in drinking water. Physiologically based
pharmacokinetic modeling was used to translate external exposures to internal doses
for the simulated adult male and female and the 6-year-old child. Results indicated that
inhalation exposures predominated and that children developed higher internal doses
(mg/kg body weight) than adults in the same household. This report demonstrates the
technical feasibility of combining stochastic and deterministic models and modeling
approaches with "real-world" concentrations of drinking water contaminants (here,
THMs) to estimate internal doses for risk evaluation and for the examination of
toxicokinetic interactions among mixtures of chemicals.
Preferred citation:
U.S. EPA. 2006. Exposures and Internal Doses of Trihalomethanes in Humans: Multi-
Route Contributions from Drinking Water. Office of Research and Development,
National Center for Environmental Assessment, Cincinnati, OH. EPA 600/R-06/087.
-------
TABLE OF CONTENTS
TABLE OF CONTENTS iii
LIST OF TABLES vi
LIST OF FIGURES xi
LIST OF ABBREVIATIONS xviii
PREFACE xix
AUTHORS, CONTRIBUTORS AND REVIEWERS xx
EXECUTIVE SUMMARY xxii
1. PROJECT OVERVIEW AND OBJECTIVES 1
1.1 OVERVIEW OF EXPOSURE AND PBPK MODEL APPROACH
AND LINKAGE 4
1.1.1. Modeling Theory and Numerical Methods 5
1.1.2 Simulating Water Uses 5
2. MODEL PARAMETERS 8
2.1. VOLATILIZATION MODEL PARAMETERS 9
2.1.1. Literature Review of Chemical Properties 11
2.1.2. Estimating Chemical Properties 11
2.2. BEHAVIORAL CHARACTERISTICS 16
2.2.1. Activity Patterns 17
2.2.2. Water-use Behaviors for Groups of Interest 21
2.3. INGESTION CHARACTERISTICS 27
2.3.1. Available Data Sources 28
2.3.2. Ingestion Behavior for the Three Populations: Results of
Analysis 29
-------
TABLE OF CONTENTS cont.
2.4. BUILDING CHARACTERISTICS 30
2.4.1. Representation of Household Volumes 30
2.4.2. Representation of Whole House Air Exchange Rates and
Interzonal Airflows 32
2.4.3. Model Representation of Building 35
2.5. CONCENTRATIONS IN WATER SUPPLY 35
2.5.1. Water Concentrations Selected as Model Inputs 38
2.5.2. Estimated Concentrations in Consumed Tap Water 39
2.6. UPTAKE AND SOLUBILITY PARAMETERS 42
2.6.1. Breathing Rates by Activity and Demographic Group 42
2.6.2. Skin Permeability Coefficients for Each Chemical 42
2.6.3. Partition Coefficients for Each Chemical 42
2.7. UPTAKE CALCULATIONS 42
2.7.1. Dermal Uptake Calculations 43
2.7.2. Inhalation Uptake Calculations 44
3. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK) MODEL 46
3.1. MODEL STRUCTURE 46
3.2. MASS BALANCE EQUATIONS 47
3.3. RELATIONSHIP BETWEEN THM METABOLISM AND TOXICITY 49
3.4. METABOLIC INTERACTIONS 50
4. TRANSFER FILE DEFINITIONS 54
4.1. BREATHING RATE FILES 54
4.2. DERMAL DATA FILES 55
4.3. INGESTION DATA FILES 55
4.4. INHALATION DATA FILES 56
IV
-------
TABLE OF CONTENTS cont.
5. RESULTS 58
5.1. EXPOSURE TO THE THMs THROUGH WATER USAGE 58
5.1.1.' Water Concentrations 58
5.2. INTERNAL DOSES OF THE THMs FROM WATER USAGE: AN
ILLUSTRATIVE CASE RESULTS 59
5.3 INTERNAL DOSES OF THE THMs FROM WATER USAGE:
POPULATION-BASED RESULTS 60
5.3.1. Population Results for Chloroform 60
5.3.2. Population Results for BDCM 61
5.3.3. Population Results for DBCM 61
5.3.4. Population Results for Bromoform 61
5.4. METABOLIC INTERACTIONS BETWEEN THE THMs 61
5.5. INFLUENCE OF WATER-USE PATTERNS ON INTERNAL
DOSIMETRY FOR THE THMs 62
5.6. LIMITED SENSITIVITY ANALYSIS OF THE PBPK MODEL 64
6. DISCUSSION 67
7. CONCLUSIONS 71
8. MODEL ASSUMPTIONS AND DATA QUALITY 73
8.1. DATA QUALITY 73
8.2. ACTIVITY PATTERN DATABASE OVERVIEW 73
8.3. OTHER ASSUMPTIONS 76
9. REFERENCES 78
APPENDIX: PBPK Model Code 86
-------
LIST OF TABLES
No. Title
1 Location Codes Recorded in the National Human Activity Pattern
Survey (NHAPS) T-1
2 Activity Codes Recorded in the National Human Activity Pattern
Survey (NHAPS) T-2
3 List of Chemicals for Exposure Assessment T-3
4 Physical Properties of Chemicals of Interest T-4
5 Estimated Values for Liquid Phase Diffusivity, Gas Phase Diffusivity, and
Dimensionless Henry's Law Constant T-5
6 Relevant Chemical Properties for the THMs T-6
7 Relevant Chemical Properties for the Predictor Chemicals T-7
8 Summary of Normalized Percent Difference (Equation 9) for the THMs
as a Function of Predictor Chemical T-8
9 Estimated Values for Overall Mass Transfer Coefficient (K0i_A) based
on Toluene T-9
10 Shower Frequency Values from NHAPS and REUWS Analyses T-10
11 Summary Statistics for Shower Duration, Volume and Flowrate from
REUWS Analyses T-10
12 Selected Model Parameters for Showers T-11
13 Bath Frequency and Duration Values from NHAPS Analyses T-12
14 Bath Volume and Flowrate Values from REUWS Analyses T-12
15 Selected Model Parameters for Bathing T-13
16 Frequency of Clothes Washer Use for 3-Person Households: REGS T-13
17 Typical Clothes Washer Parameters: Based on REUWS and
Experimental Data T-14
18 Selected Model Parameters for Clothes Washer Use T-15
19 Frequency of Dishwasher Use for 3-person Households: U.S. DOE, 1999....T-16
VI
-------
LIST OF TABLES cont.
No. Title
20 Manufacturer Supplied Dishwasher Information Summary T-17
21 Selected Model Parameters for Dishwasher Use T-19
22 Summary of Reported Toilet Use Characteristics from Literature T-20
23 Statistics for Toilet Flushes from REUWS T-21
24 Selected Parameters for Toilet Use T-21
25 Summary of Reported Faucet Frequency and Volume of Use
Characteristics in Literature T-22
26 Summary Statistics for Faucet Use from REUWS T-23
27 Selected Parameters for Faucet Use T-23
28 Tap Water Consumption Characteristics T-24
29 Parameters of Fitted Lognormal Distribution for Water Ingestion in
the United States T-26
30 Comparison of Consumption for Raw Data and Fitted Distributions
based on CSFII Data T-27
31 Selected Parameters for Tapwater Consumption Modeling Study T-28
32 Analysis of REGS for Total House Volume for 3-Person U.S.
Households (U.S. DOE, 1999) T-29
33 Estimated Range of Dimensions of Water-Use Zones Based on
Hoke, 1988, 1994 T-30
34 Summary Statistics for US Residential Air Exchange Rates T-31
35 Summary of Volume and Ventilation Parameters for Case 48 T-32
36 Pearson Correlation Coefficients between Chloroform and BDCM,
DBCM, and Bromoform based on all samples in the 85th to 95th
Percentile in the Cumulative Chloroform Concentrations T-32
VII
-------
LIST OF TABLES cont.
No. Title
37 Pearson Correlation Coefficients between BDCM and Chloroform,
DBCM, and Bromoform based on all samples in the 85th to 95th
Percentile in the Cumulative BDCM Concentrations T-33
38 Pearson Correlation Coefficients between DBCM and Chloroform,
BDCM, and Bromoform based on all samples in the 85th to 95th
Percentile in the Cumulative DBCM Concentrations T-33
39 Pearson Correlation Coefficients between Bromoform and Chloroform,
BDCM, and DBCM based on all samples in the 85th to 95th Percentile
in the Cumulative Bromoform Concentrations T-34
40 95th Percentile Chloroform (66 ppb) Values for ICR Surface Water
Treatment Plants T-35
41 95th Percentile BDCM (23.8 ppb) Values for ICR Surface Water
Treatment Plants T-36
42 95th Percentile DBCM (17 ppb) Values for ICR Surface Water
Treatment P lants T-37
43 95th Percentile Bromoform (5.6 ppb) Values for ICR Surface Water
Treatment Plants T-39
44 Summary of THM Concentrations Paired with the 95th Percentile for
Each THM for All ICR Samples T-40
45 Description of Variables Used in Analysis and Their Associated Attributes T-41
46 Summary of THM Concentrations Paired with the 95th Percentile for the
Analyzed THM Based on Analysis of the ICR Database T-42
47 Chemical Properties of Compounds (24° C) Studied by Howard and
Corsi(1996) T-45
48 Chemical properties of Compounds Being Modeled (24°C) T-45
49 Summary of Experimental Stripping Efficiencies for Cyclohexane,
Toluene, and Acetone T-46
VIM
-------
LIST OF TABLES cont.
No. Title
50 Estimated Rate Constants for Removal of THMs from a Storage
Container Based on Batterman et al T-47
51 Estimated Fractional Volatilization from a Storage Container as a
Function of Time for THMs for Cold, Room Temperature, and Hot Water T-48
52 THM Consumption Scenarios T-49
53 Recommended Consumption Model Inputs for the THMs T-51
54 Alveolar Ventilation Rates by Demographic Group and Activity T-51
55 Skin Permeability Coefficients T-52
56 Partition Coefficients Required for Fundamental Uptake Modeling in TEM T-53
57 Definition of Some Terms Commonly Used in PBPK Modeling T-54
58 Physiological Parameters Used in the PBPK Models T-55
59 Tissue partition coefficients for the THMs T-56
60 Metabolic Parameters for the THMs T-57
61 Description of Transfer File Naming Conventions T-58
62 Summary of THM Paired Concentrations for the Selected Factors T-59
63 Demographic Characteristics of Simulation Number 48 T-61
64 Water-Use Activity Pattern from the NHAPS Database for Simulation
Number 48 T-62
65 Predicted Chloroform Absorbed Dose Results T-66
66 BDCM Absorbed Dose Results T-68
67 DBCM Absorbed Dose Results T-70
68 Bromoform Absorbed Dose Results T-72
69 Water Concentrations Used to Investigate THM Metabolic Interactions T-74
70 Inhibition of DBCM Bioactivation by THMs T-75
IX
-------
LIST OF TABLES cont.
No. Title
71 Effect of Decreasing Enzyme Content on the Metabolic Interactions of
theTHMs T-76
72 Categories of Data Sources and Models T-78
73 Quality and Sources of Data Used in the Models T-79
74 Categories of Model Approaches and Algorithms T-82
75 Quality of Modeling Approaches and Algorithms T-83
-------
LIST OF FIGURES
No. Title
1 Population-Based Modeling Paradigm F-1
2 Schematic Representation of the Procedures Used for Simulating Water
Uses Based on a Sampled Activity Pattern F-2
3 Cumulative Distribution Function of Volume for 3-Person Households F-3
4 Comparison of REGS Data and the Fitted Probability Density Function of
Volume for 3-Person Households F-4
5 Schematic Representation of House Interzonal Air Flows F-5
6 Structure of the PBPK model used to analyze human exposures to THMs F-6
7 Predicted Chloroform Air Concentrations for the Example Case F-7
8 Predicted BDCM Air Concentrations for the Example Case F-7
9 Predicted DBCM Air Concentrations for the Example Case F-8
10 Predicted Bromoform Air Concentrations for the Example Case F-8
11 Predicted Personal Air Concentrations for the Adult Male for the
Exam pie Case F-9
12 Predicted Personal Air Concentrations for the Adult Female for the
Exam pie Case F-9
13 Predicted Personal Air Concentrations for the Child for the Example Case.... F-10
14 Predicted Concentrations of Metabolites Produced in the Liver over 24
hours (CM24) for the Adult Male in the Example Case F-10
15 Predicted Concentrations of Metabolites Produced in the Liver over 24
hours (CM24) for the Adult Female in the Example Case F-11
16 Predicted Concentrations of Metabolites Produced in the Liver over 24
hours (CM24)forthe Child in the Example Case F-11
17 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Kidney for the Adult Male in the Example Case F-12
XI
-------
LIST OF FIGURES cont.
No. Title
18 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Kidney for the Adult Female in the Example Case F-12
19 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Kidney for the Child in the Example Case F-13
20 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Genitals for the Adult Male in the Example Case F-13
21 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Genitals for the Adult Female in the Example Case F-14
22 Predicted Area under the Curve (AUC) for the parent THMs
Concentrations in the Genitals for the Child in the Example Case F-14
23 Histogram of Absorbed Chloroform Dermal Dose for Females, Males,
and Children F-15
24 Histogram of Absorbed Chloroform Inhalation Dose for Females,
Males, and Children F-15
25 Histogram of Absorbed Chloroform Ingestion Dose for Females,
Males, and Children F-16
26 Histogram of Total Absorbed Chloroform Dose for Females, Males,
and Children F-16
27 Histogram of Absorbed BDCM Dermal Dose for Females, Males,
and Children F-17
28 Histogram of Absorbed BDCM Inhalation Dose for Females, Males,
and Children F-17
29 Histogram of Absorbed BDCM Ingestion Dose for Females, Males,
and Children F-18
30 Histogram of Total Absorbed BDCM Dose for Females, Males, and
Children F-18
31 Histogram of Absorbed DBCM Dermal Dose for Females, Males,
and Children F-19
XII
-------
LIST OF FIGURES cont.
No. Title
32 Histogram of Absorbed DBCM Inhalation Dose for Females, Males,
and Children F-19
33 Histogram of Absorbed DBCM Inhalation Dose for Females, Males,
and Children F-20
34 Histogram of Total Absorbed DBCM Dose for Females, Males,
and Children F-20
35 Histogram of Absorbed Bromoform Dermal Dose for Females, Males,
and Children F-21
36 Histogram of Absorbed Bromoform Inhalation Dose for Females, Males,
and Children F-21
37 Histogram of Absorbed Bromoform Inhalation Dose for Females, Males,
and Children F-22
38 Histogram of Total Absorbed Bromoform Dose for Females, Males, and
Children F-22
39 Histogram of the Distribution of the AUC for Chloroform in the Kidneys of
Exposed Subjects from 1000 Different Water-use Patterns F-23
40 Histogram of the Distribution of the AUC for BDCM in the Kidneys of
Exposed Subjects from 1000 Different Water-use Patterns F-23
41 Histogram of the Distribution of the AUC for DBCM in the Kidneys of
Exposed Subjects from 1000 Different Water-use Patterns F-24
42 Histogram of the Distribution of the AUC for Bromoform in the Kidneys of
Exposed Subjects from 1000 Different Water-use Patterns F-24
43 Histogram of the Distribution of the AUC for Chloroform in the Genitals of
Exposed Subjects from 1000 Different Water-use Patterns F-25
44 Histogram of the Distribution of the AUC for BDCM in the Genitals of
Exposed Subjects from 1000 Different Water-use Patterns F-25
XIII
-------
LIST OF FIGURES cont.
No. Title
45 Histogram of the Distribution of the AUC for DBCM in the Genitals of
Exposed Subjects from 1000 Different Water-use Patterns F-26
46 Histogram of the Distribution of the AUC for Bromoform in the Genitals of
Exposed Subjects from 1000 Different Water-use Patterns F-26
47 Histogram of the Distribution of the Concentration of Chloroform
Metabolites (CM24) Formed in the Liver over 24 hr in Exposed Subjects
from 1000 Different Water-use Patterns F-27
48 Histogram of the Distribution of the Concentration of BDCM Metabolites
(CM24) Formed in the Liver Over 24 hr in Exposed Subjects from 1000
Different Water-use Patterns F-27
49 Histogram of the Distribution of the Concentration of DBCM Metabolites
(CM24) Formed in the Liver Over 24 hr in Exposed Subjects from 1000
Different Water-use Patterns F-28
50 Histogram of the Distribution of the Concentration of Bromoform
Metabolites (CM24) Formed in the Liver Over 24 hr in Exposed Subjects
from 1000 Different Water-use Patterns F-28
51 Effect of Varying the Maximal Rate of Metabolism (VmaxC) on the Liver
Concentration of Metabolites (CAM) for Chloroform F-29
52 Effect of Varying the Maximal Rate of Metabolism (VmaxC) on the Liver
Concentration of Metabolites (CAM) for Bromoform F-29
53 Effect of Varying Liver Blood Flow (QLC) on the Liver Concentration of
Metabolites (CAM) for Chloroform F-30
54 Effect of Varying Liver Blood Flow (QLC) on the Liver Concentration of
Metabolites (CAM) for Bromodichloromethane F-30
55 Effect of Varying Liver Blood Flow (QLC) on the Liver Concentration of
Metabolites (CAM) for Dibromochloromethane F-31
56 Effect of Varying Liver Blood Flow (QLC) on the Liver Concentration of
Metabolites (CAM) for Bromoform F-31
XIV
-------
LIST OF FIGURES cont.
No. Title
57 Effect of Varying the Maximal Rate of Metabolism (VmaxC) on the
Liver Area Under the Curve (AUCL) for Chloroform F-32
58 Effect of Varying the Maximal Rate of Metabolism (VmaxC) on the
Liver Area Under the Curve (AUCL) for Bromoform F-32
59 Effect of Varying KM (mg/L) on the Liver Concentration of Metabolites
(CAM) for Chloroform F-33
60 Effect of Varying KM (mg/L) on the Liver Area Under the Curve (AUCL)
for Chloroform F-33
61 Effect of Varying KM (mg/L) on the Liver Concentration of Metabolites
(CAM) for Bromoform F-34
62 Effect of Varying KM (mg/L) on the Liver Area Under the Curve (AUCL)
for Bromoform F-34
63 Effect of Varying Cardiac Output (QCC) on the Liver Concentration of
Metabolites (CAM) for Chloroform F-35
64 Effect of Varying Cardiac Output (QCC) on the Liver Concentration of
Metabolites (CAM) for Bromodichloromethane F-35
65 Effect of Varying Cardiac Output (QCC) on the Liver Concentration of
Metabolites (CAM) for Dibromochloromethane F-36
66 Effect of Varying Cardiac Output (QCC) on the Liver Concentration of
Metabolites (CAM) for Bromoform F-36
67 Effect of Varying Cardiac Output (QCC) on the Liver Area Under the
Curve (AUCL) for Chloroform F-37
68 Effect of Varying Cardiac Output (QCC) on the Liver Area Under the
Curve (AUCL) for Bromodichloromethane F-37
69 Effect of Varying Cardiac Output (QCC) on the Liver Area Under the
Curve (AUCL) for Dibromochloromethane F-38
xv
-------
LIST OF FIGURES cont.
No. Title
70 Effect of Varying Cardiac Output (QCC) on the Liver Area Under the
Curve (AUCL) for Bromoform F-38
71 Effect of Varying Liver Blood Flow (QLC) on the Liver Area Under the
Curve (AUCL) for Chloroform F-39
72 Effect of Varying Liver Blood Flow (QLC) on the Liver Area Under the
Curve (AUCL) for Bromodichloromethane F-39
73 Effect of Varying Liver Blood Flow (QLC) on the Liver Area Under the
Curve (AUCL) for Dibromochloromethane F-40
74 Effect of Varying Liver Blood Flow (QLC) on the Liver Area Under the
Curve (AUCL) for Bromoform F-40
75A Cumulative Total Absorbed Chloroform Dose F-41
75B Normalized Cumulative Total Absorbed Chloroform Dose F-41
76A Cumulative Total Absorbed Bromoform Dose F-42
76B Normalized Cumulative Total Absorbed Bromoform Dose F-42
77A Route-Specific Contributions to the Total Absorbed Chloroform Dose
for the Male Population Group F-43
77B Route-Specific Contributions to the Normalized Total Absorbed
Chloroform Dose for the Male Population Group F-43
78A Route-Specific Contributions to the Total Absorbed Chloroform Dose
for the Female Population Group F-44
78B Route-Specific Contributions to the Total Absorbed Chloroform Dose
for the Female Population Group F-44
79A Route-Specific Contributions to the Total Absorbed Chloroform Dose
for the Child Population Group F-45
79B Route-Specific Contributions to the Normalized Total Absorbed
Chloroform Dose for the Child Population Group F-45
XVI
-------
LIST OF FIGURES cont.
No. Title
80A Route-Specific Contributions to the Total Absorbed Bromoform Dose
for the Male Population Group F-46
806 Route-Specific Contributions to the Normalized Total Absorbed
Bromoform Dose for the Male Population Group F-46
81A Route-Specific Contributions to the Total Absorbed Bromoform
Dose for the Female Population Group F-47
81B Route-Specific Contributions to the Normalized Total Absorbed
Bromoform Dose for the Female Population Group F-47
82A Route-Specific Contributions to the Total Absorbed Bromoform Dose
for the Child Population Group F-48
82B Route-Specific Contributions to the Normalized Total Absorbed
Bromoform Dose for the Child Population Group F-48
83 Comparison of Chloroform and Bromoform Route-Specific Contributions
for the Female Population Group F-49
84 Effective Consumption Volume (Volume of Tap Water Consumed if
all of the Absorbed Dose originated from the Ingestion Route) F-49
XVII
-------
LIST OF ABBREVIATIONS
ACH Air exchange rate
AUC Area under the curve
BDCM Bromodichloromethane
CM24 Concentration of metabolites produced in the liver over 24 hours
CMFM Completely mixed flow model
CSFII Continuing Survey of Food Intake by Individuals
CYP2E1 Cytochrome P450 2E1
DBCM Dibromochloromethane
DBFs Disinfection byproducts
gpm Gallons per minute
ICR Information Collection Rule
KoiA Overall mass transfer coefficient
Kp Permeability coefficient of stratum corneum
MCL Maximum contaminant level
NHAPS National Human Activity Patters Survey
NIST National Institute of Standards and Technology
PBPK Physiologically based pharmacokinetic
REUWS Residential End Use Water Survey
REGS Residential Energy Consumption Survey
PFM Plug flow model
PFT Perfluorocarbon tracer
PID Potential inhalation dose
ROM Rest-of-House
TCE Trichloroethylene
TEM Total Exposure Model
THM Trihalomethane
TTHM Total trihalomethane
USDA U.S. Department of Agriculture
VOCs Volatile organic compounds
XVIII
-------
PREFACE
This report was prepared by NCEA's Cincinnati Division to support
considerations undertaken by the Office of Water, and other bodies who are concerned
with the transition of drinking water contaminants from finished drinking water to tissues
and organs within the body. It is a research project, aimed at demonstrating the
feasibility of linking several complex data sets through computer-based modeling to
estimate internal doses of four concurrently-exposed trihalomethanes compounds. The
outcome was successful, in that external concentrations were transformed to internal
target issue doses. Health Canada bases acceptable drinking water contaminant
levels, taking into account exposure by all routes based, where possible, on
extrapolation from internal dose metrics. Should the U.S. EPA consider such a concept,
the present report demonstrates its technical feasibility, and it identifies some research
needs that would increase the confidence in predicted internal exposures for these
compounds. The activities supporting this work were initiated in 2003. Internal review
of draft report was completed in 2005. A formal peer panel review was conducted in
June 2005. Subsequent to revisions reflecting internal and external peer review
comments, a final draft was developed in May 2006.
XIX
-------
AUTHORS, CONTRIBUTORS AND REVIEWERS
AUTHORS
John C. Lipscomb
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Gregory L. Kedderis
Independent Consultant
Chapel Hill, NC 27516
Andrea D. Mason
Wilkes Technologies, Inc.
Bethesda, MD
Laura L. Niang
Wilkes Technologies, Inc.
Bethesda, MD
Charles R. Wilkes
Wilkes Technologies, Inc.
Bethesda, MD
INTERNAL REVIEWERS
Rob Dewoskin
National Center for Environmental Assessment
Research Triangle Park, NC
Rick Hertzberg
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Cincinnati, OH
EXTERNAL REVIEWERS
Harvey Clewell
CUT Centers for Health Research
Research Triangle Park, NC
xx
-------
AUTHORS, CONTRIBUTORS AND REVIEWERS cont.
Gary L. Ginsberg
Connecticut Department of Public Health
Hartford, CT
Margaret MacDonell
Argonne National Laboratory
Argonne, IL
Moiz M. Mumtaz
Agency for Toxic Substances and Disease Registry (ATSDR)
Chamblee, GA
Clifford P. Weisel
Environmental & Occupational Health Sciences Institute (EOHSI)/UMDNJ
Piscataway, NJ
ACKNOWLEDGMENTS
Support for this work was provided by the NCEA Cincinnati Information
Management Team. Ruth Durham oversaw critical communications, Bette Zwayer
conducted and oversaw numerous editing and document preparation tasks, and
Barbara Cook fulfilled duties as Extramural Coordinator. The technical support of Lana
Wood, Dan Heing and Teresa Shannon, contractors with IntelliTech Systems, Inc., is
also gratefully acknowledged.
XXI
-------
EXECUTIVE SUMMARY
The concentrations of the four commonly-identified trihalomethanes (THM) in
U.S. drinking water systems are regulated as a group. These four compounds include
chloroform, bromodichloromethane, chlorodibromomethane and bromoform and are
common disinfection byproducts. Chloroform is the most well-studied of these
compounds; however, this report is concerned with developing, applying and
communicating a method to estimate internal exposures, rather than risk, from exposure
to these compounds. Each of these compounds is metabolized by the same enzyme,
and metabolism of chloroform by this route is responsible for the development of
toxicologically-active metabolites. Because of this situation, and because these
compounds are metabolized at different rates, it is possible that competition between
them for metabolism may increase or decrease risk. The present investigation was
undertaken to estimate human internal exposures to these compounds and to ascertain
whether the internal concentrations attained in humans exposed to them via drinking
water may be anticipated to result in metabolic interactions. Because these compounds
are present in water used for drinking, bathing, and other household uses, the present
investigation evaluated the development of internal doses via the oral, dermal and
inhalation routes. This was accomplished by integrating several data sets.
The goal of this project is to implement comprehensive exposure and PBPK
models to estimate population-based exposures and doses to the trihalomethane (THM)
species originating in the drinking water. The populations of concern in this project are
the following: (a) women of reproductive age (ages 15-45); (b) men of similar age (ages
15-45); and (c) children (age 6). This report presents and identifies the various model
parameters needed for running the exposure and PBPK models, specifically those
related to chemical volatilization, human activity patterns, ingestion, building
characteristics, chemical concentrations in the water supply, tissue partition coefficients,
and other physiological characteristics. In this study, information on the frequency and
duration of use for the six most common household water uses (showering, bathing,
using the clothes washer, dishwasher, toilet and faucet) were studied.
The initial phase of the investigation involved application of a model to estimate
"contact" dose. The Total Exposure Model (TEM) was employed to integrate data sets
describing (1) human water-use patterns, (2) human household activity patterns, and (3)
household ventilation patterns. Drinking water THM concentrations in finished and
distributed drinking water have been characterized through activities undertaken in
support of the U.S. EPA's Information Collection Rule (ICR); drinking water
concentrations of these disinfection byproducts used in this report were those
concentrations estimated at the 95th percentile for the distribution of their
concentrations in drinking water, representing a "high-end" concentration (and resulting
exposures). These concentration values were taken from an analysis of the information
presented in the ICR Database. For this analysis, a computer-based evaluation
developed thousands of "typical" water-use behaviors by sampling from publicly-
available databases. The water-use behavior parameters needed for TEM have been
XXII
-------
developed from the data presented in the National Human Activity Pattern Survey
(NHAPS), the Residential End Use Water Study (REUWS), Residential Energy
Consumption Survey (REGS), in appliance manufacturer data, and supplemented, as
necessary, by best judgment. These behavior parameters were based on the
characteristics defined (e.g., age, gender, etc); and water-use activities were
superimposed on the sampled activity pattern in accordance with specified
characteristics (e.g., frequency and duration of water uses, activities during which the
water use activity can occur, etc.).
While oral ingestion relied on concentrations of the THM compounds in drinking
water, estimates of contacted doses via the dermal and inhalation routes was not so
straightforward. Volatilization of these compounds into indoor air was generated from
water-use data employing physico-chemical characteristics obtained from the open
literature or predicted according to peer-reviewed methods. Airborne concentrations
were modified based on available data describing household volumes and airflow
patterns (ventilation). Specified concentrations in areas (rooms or groups of rooms)
were combined with data describing human household movements to develop a
description of concentrations of compounds in human breathing zones. For dermal
exposures, water-use behavior and exposed skin surfaces were combined via
physiologically based pharmacokinetic (PBPK) modeling which incorporated measures
of dermal penetrability to estimate internalized concentrations of THMs.
The exposure model (TEM) and the PBPK models were integrated such that the
relevant parameters were shared on an individual simulation basis. This integration
allowed the PBPK model to account for the variations in exposure as a result of
individuals' behavior and environment, including accounting for the effects of variations
in water-use behavior, concentrations in the water supply, physiological parameters
(e.g., breathing rate), location in the home, and building parameters (e.g., ventilation
rates).
To estimate internal exposures of these compounds in critical organs and tissues
(i.e., liver, kidney, genital tissues), PBPK modeling approaches previously applied to
chloroform were adapted. Specific human models were developed to simulate the body
characteristics of the adult male and female of child-bearing age as well as the 6 year
old child. Human water use and activity patterns for each of these individuals were
simulated. For all simulated humans, the estimated absorbed dose from the inhalation
route predominates the estimated total exposure, often surpassing the other route-
specific exposures (oral, dermal) by ten-fold or more. Results in Tables 65-68 indicate
that internal doses were bromoform < dibromochloromethane < bromodichloromethane
< chloroform. The present analysis indicated that children absorbed a lower total
amount (mg) of these compounds compared to adults, but their absorbed dose (mg/kg)
was consistently higher than the absorbed doses in adults. At the 50th percentile for the
exposure distributions for each simulated human (exposed to THMs present at their
respective 95th percentiles of their distributions) total internal doses resulting from the
combined oral, dermal and inhalation exposure routes were below the respective oral
Reference Dose (RfD) values for each THM.
XXIII
-------
With joint, concomitant exposure to these THM compounds, under near-extreme
drinking water concentrations, metabolic interaction was not apparent under typical
conditions of enzyme expression; when enzyme content was reduced by 10,000-fold,
metabolism of the individual THM chemicals was only impacted 20% or less, indicating
a low likelihood of metabolic interactions, given present knowledge about the enzyme
responsible and its quantitative distribution among humans. These results indicate that
the internal dosimetry of these compounds under the anticipated conditions of human
exposures is unlikely to be changed under the conditions of a simultaneous exposure to
the four-chemical mixture, versus single-chemical exposure.
This report is the second in a series that demonstrates the technical feasibility of
combining stochastic and deterministic models and modeling approaches with "real-
world" concentrations of drinking water contaminants (see U.S. EPA, 2003). In this
report, internal doses of THMs are estimated for risk evaluation and for the examination
of toxicokinetic interactions among mixtures of chemicals.
XXIV
-------
1. PROJECT OVERVIEW AND OBJECTIVES
Disinfection of drinking water is widely recognized for its significant role in
reducing illness due to waterborne pathogens that are responsible for numerous
diseases. Although disinfection is necessary for the elimination of these pathogenic
organisms, it can also lead to the generation of a variety of chemicals, known as
disinfection byproducts (DBFs), which are formed as a result of reactions of the
disinfectant with organic matter in the water. In the U.S., where the primary form of
disinfection is chlorination, public drinking water contains low levels of many DBFs and
is a potential source of exposure to these compounds. While the potential for exposure
is recognized as significant via the ingestion route, recent attention has also focused on
the inhalation and dermal routes of exposure. The importance of each route varies with
chemical characteristics, use patterns, physiological characteristics, and a variety of
other factors (Wilkes etal., 1996; Olin, 1998). For example, exposure to a volatile
drinking water contaminant occurs most significantly during large household water uses,
such as showering, bathing, and clothes washing activities. These activities release
volatile compounds that may persist in indoor air, prolonging exposures and resultant
doses via the inhalation route. Although all three primary exposure routes can be
significant, inhalation typically dominates the exposure for these volatile compounds
(U.S. EPA, 2003). For the less volatile compounds, ingestion and dermal contact play
more significant roles in exposure and uptake. The present work addresses exposure
to a mixture of four trihalomethane (THM) compounds via each of these three routes,
separately.
Exposure to DBFs originating in the drinking water is a very complex problem,
influenced by a multitude of factors, including chemical properties of the contaminant,
physical characteristics of the indoor environment, behavior of the individual relative to
the contaminant, and behavioral and physiological characteristics of the exposed
population. Previous modeling studies have demonstrated the considerable impact
human behavior has on an individual's exposure to waterborne contaminants (Wilkes et
al., 1996; Wilkes, 1999; U.S. EPA, 2003), demonstrating that differences in behavior
can produce exposures varying across more than an order of magnitude. Mathematical
-------
exposure and uptake models represent a realistic, cost-effective means for estimating
human exposure. Mathematical models within a probabilistic framework allow a close
examination of the factors that lead to exposures and provide a basis for addressing
higher risk populations. However, in the case of exposure to waterborne contaminants,
previous modeling studies (Wilkes et al., 1996; Wilkes, 1999) have shown that a strictly
probabilistic framework would fail to capture the effect of an individual's activities on his
or her exposure. The ideal model would therefore combine a probabilistic
representation of human behavior related to water use and exposure with a
deterministic calculation of the concentrations in the contact media leading to the
exposure (i.e., in the water and air). Such modeling frameworks also offer the ability to
evaluate the impacts of parameter uncertainty and variability, such that results may be
incorporated into meaningful and useful sets of outcomes.
The internal doses resulting from human exposures to chemicals can be
quantitated using physiologically based pharmacokinetic (PBPK) models. PBPK
models provide a biologically based framework for understanding the absorption,
distribution, metabolism and elimination of chemicals ingested by any route. PBPK
models are based on known anatomy, physiology and biochemistry, and thus
realistically reflect the flux of a chemical through an organism (Clewell and Andersen,
1994). The compartments in PBPK models represent organs or groups of organs of
known volume interconnected by blood flows at known rates. Physiological parameters
for human PBPK models can be obtained from the literature. Chemical-specific
parameters such as tissue partition coefficients and metabolic rates can be measured
experimentally or obtained from the literature. Experimental data can be obtained
through whole animal or human studies in vivo or from experiments with in vitro
systems. Extrapolation of laboratory animal data to humans is often problematic due to
species differences in biotransformation and other pharmacokinetic processes. In vitro
systems such as intact human cells or subcellular fractions can provide quantitative
data on human metabolic rates and interindividual differences in those rates (Kedderis
and Lipscomb, 2001). However, the in vitro data must be integrated into physiological
models to understand the true impact of differences in chemical bioactivation on the
target organ concentrations of toxicants.
-------
Exposure models provide estimates of the quantity of chemical that comes into
contact with an individual as a result of releases into the environment and individual
product uses. Often, these estimated exposures are used to infer a risk to an individual
or to a population group. However, many studies have demonstrated that external
exposure is not equivalent to internal dose (Kedderis, 1997). For example, exposure to
the same external concentration of furan vapors results in considerable differences
between the internal dose of furan in mice, rats and humans (Kedderis and Held, 1996).
The integration of exposure models with PBPK models provides a powerful method of
linking chemical use in the environment to internal tissue doses, both in individuals and
populations. Using the results of exposure modeling for a given environment as input to
an appropriate PBPK model can provide information on the target organ concentrations
of a toxicant and its metabolites in exposed humans. Variations in the exposure
scenarios will produce a series of time courses of target organ exposure to toxicants
that can be used to derive the relationship between external chemical exposures,
human activity patterns and target organ doses of toxicants. These exposure-dose
relationships can help assess the importance of various parameters in impacting the
target organ dose of toxicants and enable a more rigorous, scientifically based
approach to human health risk assessment.
The goal of this project is to implement comprehensive exposure and PBPK
models to estimate population-based exposures and doses to the trihalomethane (THM)
species originating in the drinking water. The populations of concern in this project are
the following: (a) women of reproductive age (ages 15-45); (b) men of similar age (ages
15-45); and (c) children (age 6). This report presents and identifies the various model
parameters needed for running the exposure and PBPK models, specifically those
related to chemical volatilization, human activity patterns, ingestion, building
characteristics, chemical concentrations in the water supply, tissue partition coefficients,
and other physiological characteristics. The exposure model being used in this study,
the Total Exposure Model (TEM), was used in the previous study (U.S. EPA, 2003) and
was developed specifically to accommodate studying population-based exposure to
water-borne contaminants. To assess the population doses associated with the
resultant exposures, the TEM and PBPK models will be integrated such that the
-------
relevant parameters, predicted time-varying concentrations and exposures are passed
between the models.
1.1. OVERVIEW OF EXPOSURE AND PBPK MODEL APPROACH AND LINKAGE
This modeling study integrates exposure and PBPK model algorithms to construct a
framework for predicting exposure, uptake and internal distribution of chemicals in
humans resulting from exposure to a contaminated water supply. An overview of this
approach is illustrated in Figure 1. As illustrated in the figure, the exposure model
assembles the critical factors influencing exposures, implements finite difference
algorithms to predict air and water concentrations, combines these predicted
concentrations with location and physiological characteristics, and estimates the uptake
as a result of contact with the contaminated air and water. The outputs from the
exposure model, including concentrations in the contact media and uptake as a function
of route of exposure, along with relevant physiological parameters are utilized by the
PBPK model to predict internal concentrations and doses.
To represent the occupant behavior, TEM samples activity patterns from an activity-
pattern database (e.g., National Human Activity Pattern Survey, NHAPS; Survey of
Activity Pattern of California Residences, GARB) based on the characteristics defined
by the user (e.g., age, gender, etc), and superimposes water-use activities on the
sampled activity pattern in accordance with user-entered characteristics (e.g., frequency
and duration of water uses, activities during which the water-use activity can occur,
etc.).
The physical properties, including the building description, air exchange and
ventilation rates, and appliance characteristics are all provided as the setting in which
the activities occur. Similarly, the chemical properties, source definition, and water
concentrations are defined for the study. Behavior-driven water uses initiate the
emissions, which are simulated using the mass-transfer models along with the defined
chemical properties.
For each simulation, once the input parameters are selected, a mass-balance model
is executed to predict air and water concentrations. Combining these predicted
concentrations with the location of the occupant yields an estimated exposure for one
representative member of the population group. These simulations are repeated to
-------
compile a distribution of expected exposures to numerous members of the population
group.
The outputs from the exposure model are a collection of files, referred to as
"transfer" files, which provide time-series of concentration, exposure, and physiological
information for each simulation as input to the PBPK model. Using this information
along with other physiological parameters, the PBPK model predicts tissue
concentrations as a function of time.
1.1.1. Modeling Theory and Numerical Methods. The modeling approach is based
on representing the building and physical environment as a collection of well-mixed
zones (or air parcels) interconnected by interzonal airflows, Numerous processes may
affect the contaminant concentration within these zones, including emissions, transport
in and out of the zone, and other removal mechanisms. This approach has been
developed and used by many others (Sindent, 1978, Sandberg, 1984, Axley 1989).
Setting up this modeling framework results in a set of simultaneous differential
equations. For a more complete description of the mathematical basis, see Wilkes,
1994.
TEM uses the fourth-order Runga-Kutta method (Mathews, 1992) for temporal
integration of the system of equations. This is an extremely accurate and stable
algorithm (errors on the order of At5). The PBPK model is programmed using
acslXtreme software (Aegis Technologies, Huntsville, AL).
1.1.2. Simulating Water Uses. TEM simulates water uses as a two-step process: (1)
all occurrences of a specified water-use type, and (2) the duration of each water-use
event. Using water-use frequency information provided by the modeler, water-use
occurrences are simulated as a Poisson process, consistent with the sampled activity
pattern as shown schematically in Figure 2. For example, when NHAPS is used to
sample activity patterns, the modeler defines specific NHAPS location and activity
codes, which are then used by the model to identify time periods during the day when
each water-use activity is eligible to occur. The location codes and activity codes
provided in NHAPS are given in Tables 1 and 2.
Activity and location code pairs are used to identify eligible water-use periods in
the sampled activity pattern. For example, the modeler may specify that periods with an
-------
activity and location code pair of "Personal needs and care - Bathing, etc."
(activity code 40) and "Home Bathroom (location code 104) are eligible for showering;
or, if the occupant is in the kitchen, the kitchen faucet water-use activity is eligible to
occur. After identifying all eligible periods for a given water use, the mean duration of
the water-use is subtracted from the end of each eligible time period to prevent the
water use from starting to close to the end of that time period. Then the modified
eligible time periods are mapped to a continuous 0 to 1 time scale, as shown in Figure
2.
Subsequently, a Poisson process is used to specify the time between starting
times of successive events, by sampling a random number from an exponential
distribution with the rate parameter, lambda (k), equal to the daily frequency. The
sampled number is used to place the starting time of the water-use event by adding it to
the start time of the previous event. This process is repeated until the next start time
falls beyond the end of the last eligible time period. This results in a simulated
frequency of water-use events, on average, equaling the specified frequency.
After the water use occurrences have been specified, the water-use durations
are simulated using the mean and standard deviations of event durations specified by
the modeler. These durations are typically based on information gathered in surveys of
water uses (e.g., EPA, 1997). The event duration is assumed to be lognormally
distributed, such that
where:
y = the event duration, a lognormally distributed random variable.
£,,§ = the parameters of the lognormal distribution.
The mean and standard deviation, characteristics of the water-use duration, are
converted to the parameters of a lognormal distribution by the following equations:
In
V
2
V J^
-------
,2
where:
(j, = the sample mean duration.
a = the sample standard deviation of the duration.
After a random number from the above distribution is simulated, the event duration
is assigned the value from Equation 1 A. If the water-use event continues beyond the end
of the occupant's current activity period, the water-use event will be truncated at the end
of this activity period. This will happen infrequently, since the eligible time period for
simulating the occurrences of the water-uses was shortened by the mean water-use
duration, as described above.
7
-------
2. MODEL PARAMETERS
The exposure and PBPK models rely on a variety of parameters that describe the
behavioral, physical, chemical, and physiological characteristics relevant to exposure to
the chemical, including rate constants, partition coefficients, volumes, and water-use
behavior that affect chemical concentrations, exposures, or physiological parameters.
These models (for the male, the female and the child; for chloroform,
bromodichloromethane, chlorodibromomethane and bromoform) address exposures via
the oral, dermal and inhalation pathways. For each model parameter, the values have
either been collected from published literature or estimated. An attempt has been made
to identify parameter values from multiple sources to assist in the execution of the
sensitivity and uncertainty analysis. The collected values are evaluated and a judgment
made, considering credibility of the source and consistency among multiple sources, to
select the most appropriate value(s) for use in the model execution.
The TEM input parameters include the following:
• Parameters needed for implementation of volatilization model
• Human behavior characteristics that drive the activity model, including location
and water-use behaviors
• Ingestion characteristics
• Building characteristics
• Chemical concentrations in water supply
The PBPK model input parameters include the following:
• Compartment volumes by demographic group
• Compartment blood flows by activity for each demographic group
• Alveolar ventilation rates for each demographic group
• Concentration of each chemical in inspired air
• Oral ingestion rates for each chemical
• Concentration of each chemical in drinking water
• Definition of the exposure scenarios for each exposure route for each chemical
• The compartment-to-blood partition coefficients for each chemical
• The skin permeation coefficients for each chemical
8
-------
• The rate constants for the gastrointestinal model to be used for each chemical
• The metabolism pathways for each parent chemical
• The metabolism rate constants Vmax, the maximal rate of metabolism, and KM,
the substrate concentration giving one-half Vmax, for each chemical to be
modeled, and the inhibitory effect of each chemical on the metabolism of the
others.
2.1. VOLATILIZATION MODEL PARAMETERS
While human exposures to drinking water contaminants immediately conjure
images of an oral ingestion scenario, drinking water supplies many water use
appliances in the home. The method of water use in these appliances (dishwashers,
clothes washers, showers, faucets, etc.) is a major contributor to airborne
concentrations of volatile contaminants, resulting in appreciable potential for an
inhalation exposure pathway. This has previously been demonstrated for volatile
drinking water contaminants, and previously employed methods and peer reviewed data
are applied to evaluate airborne concentrations and human exposures to the THMs
from drinking water, via the inhalation route.
Each of the water-using appliances or fixtures, when operated, represents an
opportunity for emission of waterborne chemicals. The emission behavior during a
given water use is a function of a variety of chemical and physical factors, including
water temperature, surface area, concentration, chemical diffusivities, and Henry's Law
constant.
To facilitate prediction of water and air concentrations, the emission behavior is
idealized using two types of models: the plug flow model (PFM) and the completely
mixed flow model (CMFM). The derivations of these models are presented elsewhere
(Olin, 1998).
The plug flow model is derived assuming a constant uniform flow and a volume
and surface area that remains essentially constant. The PFM is appropriate for use in
representing emissions during continuous flowing water uses such as faucets and
showers. Emissions for sources idealized as plug flow are represented by the following
equation:
-------
(2)
i ^ n jj
where:
Kv =QL(l-exp(Z)) (3)
^ n/ '« / A \
Z=_
OL
KOLA KLA HKGA
S = source emission rate (mass/time)
Kv = volatilization coefficient (volume/time)
Ci = contaminant concentration in the water supply prior to volatilization
(mass/volume)
Cg = concentration in the air surrounding the water stream (mass/volume)
H = dimensionless Henry's Law constant
QL = volumetric flow rate of the water (volume/time)
KOL = overall mass transfer coefficient (length/time)
A = interface area between water and air (length2)
KL = liquid phase mass transfer coefficient (length/time)
KG = gas phase mass transfer coefficient (length/time)
The rate of volatilization is maximized if Cg/H is negligible relative to C/.
Conversely, if Cg/H approaches C/, a state of chemical equilibrium may be achieved with
a corresponding suppression of volatilization. This equilibrium condition may occur for
sources that include a headspace with poor air exchange (e.g., dishwashers) or that
involve chemicals with low Henry's Law constants. The concentration of a contaminant
in the liquid phase may be effectively spatially uniform (e.g., in well-mixed systems such
as washing machines), or may vary with space (e.g., the flowing water film or droplets
associated with showers). The interfacial area, A, is typically difficult, if not impossible,
to determine for residential water uses. This is particularly true when significant
amounts of splashing occur (e.g., in kitchen wash basin), disintegrated films or droplets
occur (e.g., showers and dishwashers), and/or when entrained air bubbles are present
10
-------
(e.g., during the filling of bathtubs). Thus, interfacial area and overall mass transfer
coefficients are typically combined (K0i_A).
The completely mixed flow model assumes a well-mixed volume of water with a
constant surface area, and is appropriate for use in representing emissions from
standing water-type water uses. An example of a CMFM type source is a filled bathtub.
Emissions for sources idealized as CMFM are represented by the following equation:
(6)
The volatilization coefficient represents the rate of transfer across the liquid/gas
interface where the water is in contact with the air, while Henry's Law constant is used
to quantify the concentration gradient relative to equilibrium.
2.1.1. Literature Review of Chemical Properties. The chemicals of interest for this
study are the THMs as listed in Table 3. The properties of interest are Henry's Law
constant (dimensionless), liquid phase diffusivity (Iength2/time), gas phase diffusivity
(Iength2/time), octanol/water partition coefficient (dimensionless), and molecular weight.
Boiling point and volatility are additional properties of value for the study.
2.1.1.1. Literature Search — The literature was searched to identify
representative values for the desired chemical properties. Values were obtained from
chemical handbooks and dictionaries or online data banks. The results of the search
are summarized in Table 4. References to the relevant journal articles have been
provided where available.
2.1.2. Estimating Chemical Properties. Prediction methods are used to supplement
the literature review for chemical properties that were not found in the literature. Values
for the liquid and gas phase diffusivity, the dimensionless Henry's Law constant, and the
overall mass transfer coefficient are predicted and discussed in the following
subsections.
2.1.2.1. Estimating Liquid and Gas Phase Diffusivity and Henry's Law
Constant — The liquid phase diffusivity is predicted using the Hayduk and Laudie
method (Lyman et al., 1990). This method is reasonably accurate for a wide range of
compounds and has been validated using compiled measured data. The method uses
11
-------
the molal volume as predicted by the LaBas method and the viscosity of water to predict
the liquid phase diffusivity as a function of temperature. Similarly, the gas phase
diffusivity is predicted using the Wilke and Lee method (Lyman et al., 1990). According
to Lyman et al., this method was found to have an absolute average error of 4.3% when
compared to measured values for approximately 150 compounds. This method uses
the molecular weight, boiling point, the molal volume, and properties of air to predict the
chemical's diffusivity in air. The liquid- and gas-phase diffusivities are estimated as a
function of temperature to incorporate the effects of temperature into the estimate of the
overall mass-transfer coefficient. The estimated values for liquid and gas phase
diffusivities are given in Table 5.
Henry's Law constant can be found in current literature for most chemicals, but
often not at the temperature of interest. Therefore, a method to adjust H to the
designated temperature is necessary. The following equation is used to adjust Henry's
Law constant for temperature dependence:
where:
H = Henry's Law constant at desired temperature
He = Henry's Law constant at standard conditions
AH = enthalpy of solution
R = gas constant = 0.082057 liters atm K"1 mol"1
T = temperature (K)
6 = denotes standard condition (298.15 K)
The values for Henry's Law constant based on temperature are presented in Table 5.
2.1.2.2. Estimating Overall Mass Transfer Coefficients — Modeling
emissions of drinking water contaminants during water usage requires knowledge of the
overall mass transfer coefficient (K0i_A) as a function of the appliance, the water
temperature, the water flowrate, and the chemical. The mass-transfer rates for the
THMs have, in general, not been studied comprehensively. The possible exception,
chloroform, has been investigated by other researchers. The fractional volatilization
12
-------
(fraction of available chemical mass that is transferred to the air) during showering for
chloroform reported in the literature generally ranges from approximately 0.5 to 0.6. For
example, Giardino and Andelman (1996) reported a fractional volatilization for
chloroform ranging from 43-62% for a 10-minute shower with water temperature ranging
from 26-42°C and the water flowrate ranging from 5-10 L/min.
Because of the lack of measurement data for the mass-transfer parameters for
the THMs under the water temperature and flowrate conditions characteristic of the
general population, an estimation method is required. The volatilization coefficient, a
function of the overall mass-transfer coefficient (Koi_), is primarily a function of the water
temperature, surface area, and the chemical's diffusion coefficients in water and air.
Using a power relationship between liquid-phase and gas-phase diffusivities and the
liquid-phase and gas-phase mass transfer coefficients (KL xDLP and KG °cDGq), Little
(1992) derived the following equation for predicting the overall mass-transfer coefficient
for a desired chemical based on the measured coefficient for a reference chemical:
_ __ __
(KOLA), (KLA\ DU (KGA\ H, DGJ
where:
DL = liquid-phase diffusivity (L2/T)
DG = gas-phase diffusivity (L2/T)
/ = chemical for which the overall mass-transfer coefficient is being
estimated
r = reference chemical
p, q = power constants
Using Equation 7 and the observations of previous researchers that the ratio of
KG/KL is approximately constant for a given mass-transfer system (Little, 1992; U.S.
EPA, 2000a), U.S. EPA (2000a) presented the following equation:
(KoLA), (Du^fD^} (Hj_
(DLr) \DGr (Hr
DL\ +\DGG\ \KGl
(9)
13
-------
The above equation provides a means for estimation of KOLA for a chemical
based on measurements for another chemical based on the diffusivities, Henry's Law
constants, and the ratio of Kc/KL for the given system. U.S. EPA (2000a) conducted a
series of laboratory experiments to determine the values of K0i_A and Kc/KL. The
experiments were conducted for five reference chemicals (acetone, ethyl acetate,
toluene, ethylbenzene, and cyclohexane) and for five water-use types (sinks, showers,
bathtubs, wash machines, and dishwashers) covering a significant range of Henry's
Law constants and diffusivities. Using the measured values by U.S. EPA (2000a),
Equation 7 can be used to estimate the product of the overall mass transfer coefficient
and the interfacial surface area (K0i_A).
This estimation method requires identifying an appropriate predictor chemical
from the set of chemicals studied by U.S. EPA (2000a) based on physical and chemical
properties. U.S. EPA (2000a) conducted laboratory experiments and estimated the
overall mass transfer coefficients for common household water appliances for the
following five chemicals: acetone, ethylacetate, toluene, ethylbenzene, and
cyclohexane. A significant shortcoming for using any of these five predictor chemicals
is that they are very different in structure and are of lower molecular weight than the
THMs. In addition, important characteristics, such as Henry's Law constant, are
dissimilar. Since these factors greatly influence the mobility rates and equilibrium
conditions, the rate of mass-transfer between the water and air is likely to be influenced
leading to lower confidence in the predictions.
An evaluation of the normalized difference between the values for the liquid-
phase diffusivity, gas-phase diffusivity and Henry's Law constant for each of the five
predictor chemicals and each of the four THMs was conducted. The relevant chemical
properties for each chemical are given in Tables 6 and 7, with the results given in Table
8. A sample calculation for identifying the predictor chemical for chloroform is
presented, as follows:
SAMPLE CALCULATION: Normalized Difference
The normalized difference between the chemical properties for each predictor
chemical and chloroform is calculated as follows:
14
-------
[(Predictor Chemical Property,, -Chloroform Chemical Property,)
A/D.., = Absolute Value] ± ^ y '•' ^ y';
|_ Chloroform Chemical Property,
where:
A/D/j = normalized difference between the predictor chemical property / and the
chloroform property /
/ = chemical property
j = predictor chemical
EXAMPLE CALCULATION (for Toluene, Liquid Diffusivity at 20°C):
A/D, ,„,„,„,„„.„„,„ T^,,,_ = Absolute Value
Liquid Diffusivity Toluene - Liquid Diffusivity
Chloroform
*100 (11)
UDUquidDiffusMtyJoluene = Absolute Value
7.96E-06 - 9.21 E-06
*100 = 13.6% (12)
9.21 E-06
Because the overall mass transfer coefficients for THM compounds have not
been previously determined, the present analysis required a method to predict them.
Mass transfer coefficients for five chemicals (acetone, ethyl acetate, toluene,
ethylbenzene and cyclohexane) were previously determined (U.S. EPA, 2000a); the
mass transfer coefficients for these chemicals were evaluated for use as a normalizer
for the mass transfer coefficients predicted for each of the studied THM compounds.
Liquid diffusivity, gas diffusivity and Henry's Law constants (at 20°C and at 40°C) were
determined for each of the nine chemicals (four THM chemicals; five chemicals studied
by U.S. EPA, 2000a). Differences between values for the five studied chemicals and
the four unstudied (THM) chemicals were tabulated and toluene was determined to be
the optimal studied chemical for use in normalizing differences in predicted values for
overall mass transfer coefficient among the THM chemicals. Table 8 presents the
normalized percent differences between the predictor chemicals and the four THMs
based on 20°C. This evaluation considered Henry's Law constant as the most
significant indicator, since Henry's Law captures the solubility and vapor pressure
relationship of the compound and the diffusivities of the predictor and THM chemicals
were not dissimilar. The method for estimating the overall mass-transfer coefficient was
found to result in predictions that generally agree with the fractional mass of chemical
reported by Giardino and Andelman (1996) for chloroform. As a comparison, the
15
-------
calculation was repeated using ethylbenzene as the predictor chemical for BDCM. The
difference in predicted mass-transfer coefficients between the two predictor chemicals
for BDCM was less than 10% for all appliances. Therefore, since toluene as a predictor
chemical provided a reasonable prediction for the chloroform mass-transfer coefficient,
and because there was a relatively small difference in the predictions for BDCM, the
predicted mass transfer coefficients derived from using toluene as the predictor
chemical are used in this study, as presented in Table 9.
The estimated values for the overall mass transfer coefficient, presented in Table
9, are estimated based on Equation 7 using toluene as the predictor chemical and
assuming a water temperature and hydrodynamic conditions similar to those under
which the experiments were conducted (e.g., dropsize distribution, water flowrate, air
turbulence, etc.). The predicted overall mass-transfer coefficient was the average of
the predictions based on the experimental values measured for the following conditions:
water temperature = 35 C (approximate), shower flowrate = 9.1 liters per minute (2.4
gallons per minute) and 6.1 liters per minute (1.6 gallons per minute); and course and
fine droplet sizes, as reported by U.S. EPA (2000a). Temperature is a critical factor,
affecting mass transfer and uptake kinetics. There is a great deal of uncertainty in the
understanding of temperature and temperature effects, and this is an area where future
research is warranted.
2.2. BEHAVIORAL CHARACTERISTICS
Activity patterns and water-use behavior have been shown to have a significant
impact on predicted exposure (Wilkes et al., 1996). TEM represents the influence of
behavior by using activity pattern databases and analysis of other behaviors that
influence contaminant release and subsequent human exposure. The activity pattern
database was queried to obtain a subset of records having the desired demographic
characteristics. This subset is randomly sampled to obtain an activity pattern record,
and this record is used to specify locations within the household and opportunities for
conducting activities that may result in exposure. Using the activity code and location
code in the sampled activity pattern, a transition matrix is used to assign a location in
the modeled building as described in Section 2.2.1.2. The actual water uses are
simulated based on parameters defined from analysis of other water-use studies. This
16
-------
results in occupant-driven water uses, which ultimately lead to exposure to the
waterborne contaminants. See Section 2.2.1.2, below, for a description of the methods
for modeling activity patterns.
The chosen population for this exposure estimation modeling study is a three-
person family in which both parents are within their reproductive years. The family
consists of one male between the ages of 15 and 45, one female between the ages of
15 and 45, and one child approximately 6 years of age. Because there are few records
in the database reflecting 6-year-olds, the child is characterized by sampling the
database for children between the ages of one and nine. Although it is recognized that
there is significant difference in behavior between a toddler and a 9-year-old, it was
necessary to represent the child as a range of ages to allow a reasonable sample size
in the database. It is not entirely clear what the impact of this assumption is on the
ultimate exposure to drinking water contaminants. Younger children likely spend a
greater fraction of their day at home, and for higher volatility chemicals this may
increase their exposure. For less volatile chemicals, the impact of inhalation exposure
is minimal, and the resultant exposure is highly dependent upon the child's water-use
behavior.
2.2.1. Activity Patterns. In order to most accurately represent individuals' exposure to
waterborne contaminants, it is necessary to understand the frequency of each type of
water use (e.g., how often they shower), and the duration of the events (e.g., minutes
occupant spends in shower). In this study, the frequency and duration are described for
each of the six water-use activities most important to exposure: showering, bathing, and
using the clothes washer, dishwasher, toilet, and faucet. For some of these events, the
frequencies or durations are described as distributions from which individual usages will
be sampled, in other cases (e.g., dishwasher duration), the parameters are specified as
the best available estimate.
The water-use behavior parameters needed for TEM have been developed from
the data presented in the National Human Activity Pattern Survey (NHAPS), the
Residential End Use Water Study (REUWS), Residential Energy Consumption Survey
(REGS), in appliance manufacturer data, and supplemented, as necessary, by best
judgment, as described in Section 2.2.2. These databases are described below.
17
-------
2.2.1.1. Available Activity Pattern Databases —
2.2.1.1.1. NHAPS. The NHAPS database contains the results from a 2-year
nationwide activity pattern survey commissioned by the U.S. EPA National Exposure
Research Laboratory. During the period from October 1992 through September 1994,
9386 persons residing in the 48 contiguous United States were chosen using a
telephone random-digit dial method and interviewed over the phone (Tsang and
Klepeis, 1996). The interviewees were selected and their responses weighted
according to geographic, socioeconomic, time/season, and other demographic factors
to ensure that they were representative of the U.S. population (Tsang and Klepeis,
1996). The weighted sample is consistent with the U.S. population for gender, age,
region, and other factors.
First, respondents were asked to recall their activities and locations for the
previous 24 hours. The locations and activities were recorded as codes chosen from a
list of 83 possible locations and 91 possible activities. This diary section had minimal
information regarding water use. The only activity choice that specifically pertained to
water-use was "bathing." All of the other activities are more generally defined such as
"food clean-up", "plant care", "personal care". Location codes included 12 indoor home
locations such as "Home-bedroom," and "Home-bathroom," and a variety of other
outside the home locations such as "Office," "Grocery Store," "Work-Transit," and
"Outdoor-Park."
Then the respondents were asked a series of multiple-choice questions. Every
respondent was asked for specific demographic information, including date of birth,
gender, race, geographical region, level of education, etc., and they were asked a
multitude of questions, asking for demographic information as well as information about
various activities, most relating to possible exposure to contaminants in the air and
water, such as "How long did you spend in the shower?" or "Was a dishwasher used
yesterday when you were home?" Not everyone was asked the same questions as
there were two versions of the questionnaire. NHAPS did not acquire information on
toilet use, and acquired only limited information on faucet use.
2.2.1.1.2. REUWS. The REUWS database contains water-use data obtained
from 1,188 volunteer households throughout North America (Mayer et al., 1998). The
18
-------
REUWS study was funded by the American Water Works Association Research
Foundation. During the period from May 1996 through March 1998, approximately 100
single-family detached homes in each of 12 different municipalities (located in
California, Colorado, Oregon, Washington, Florida, Arizona, and Ontario) were outfitted
with a data-logging device (Meter Master 100 EL, manufactured by Brainard Co.,
Burlington, NJ) attached to their household water meter (on only magnetic driven water
meters). The data logger recorded the water flows at 10-second intervals for a total of
four weeks (two in warm weather and two in cool weather) at each household.
Following the study, the data were retrieved and analyzed by a flow trace analysis
software program, called Trace Wizard®, developed by Aquacraft, Inc., Boulder, CO,
which disaggregated the total flows into individual end uses (i.e., toilet, shower, faucet,
dishwasher, clothes washer, etc.) (Mayer et al., 1998). Trace Wizard® disaggregates
the household water-use signature by using a signal processing algorithm which
compares segments of the signature to general characteristics of each appliance. For
example, the volume of water used by a toilet falls into several small ranges, depending
upon the age of the toilet, and the fill rate is relatively consistent across all toilets. The
software attempts to identify all water-uses that fall into these pre-identified ranges and
label them as toilet uses. After identifying the type of water use (e.g., shower, faucet,
toilet), Trace Wizard® estimates the event durations, volumes, peakflows, and mode
measurements for each water-using event from the resultant disaggregated water-use
signature.
The REUWS database includes demographic information on each household
based on a mail-in survey. This information includes employment status (unemployed,
part-time, full-time), education level of the primary wage earner (less than high school,
high school graduate, some college, Bachelor's, Master's, Doctoral), and household
income. It also provides information on the number of adults (18 and over), children
(under 13) and teenagers (13-17), as well as a variety of information about the house
(type, number of floors, square footage, water-related amenities such as swimming
pools, etc.), household appliances and general water-use behavior. It does not give
information on age or gender.
19
-------
2.2.1.1.3. RECS. The Residential Energy Consumption Survey (REGS),
conducted nationwide in 1997, contains energy usage characteristics of 5900 residential
housing units. The information was acquired through on-site personal interviews with
residents; telephone interviews with rental agents of units where energy use is included
in the rent; and mail questionnaires to energy suppliers to the units. The database
contains information on physical characteristics of the housing units, demographic
information of the residents, heating and cooling appliances used, fuel types, and
energy consumption.
2.2.1.2. Modeling Activity Patterns — NHAPS represents the most
comprehensive survey of activities of U.S. residents available. However, water-use
behavior data associated with the survey data is sparse and incomplete. The 24-hour
record of locations and activities contains general locations (e.g., Home-kitchen, Home-
bathroom, etc.) and activities (e.g., personal care, cooking , cleaning, etc.). However,
the 24-hour activity record does not specify actual water-use events such as dishwasher
use, clothes washer use, and showering. To model the activity patterns, TEM samples
a 24-hour record from NHAPS for each occupant and, using a transition matrix, places
the occupant in the modeled house such that his/her location is consistent with the
recorded activity and location in the NHAPS database. For example, if the sampled
activity pattern identifies the location as "Home- Bathroom," and the activity as
"Personal needs and care - Bathing, etc.," the model places the adults in the "Master
Bathroom" or the child into the "Hall Bathroom" and assigns the specified breathing rate
for that activity. Information on water-use behavior gathered from other sources is then
used to simulate appropriate water-use activities for each occupant.
The water uses are incorporated into sampled activity patterns by simulating
activity-appropriate water uses with appropriate frequencies and durations of use. For
example, to simulate showering behavior, the portions of an activity pattern where a
showering activity is appropriate are identified. This is accomplished by identifying
eligible activity codes and locations codes (e.g., the combination of the NHAPS activity
code of "Personal-wash, etc.," and the NHAPS location code of "Home-Bathroom"
would be eligible for showering activities). Once eligible time periods are identified, the
time between the starting of successive water-use events is simulated by sampling a
20
-------
Poisson distribution with the rate parameter, lambda (A,), equal to the daily frequency.
The duration of the event is simulated based on a lognormal distribution with
parameters appropriate for the study population. The method for sampling activity
patterns and simulating appropriate water-uses is fully described in Wilkes (1999).
Water-use occurrences are simulated as a Poisson process using frequency data
obtained from analyses of NHAPS, REUWS, and REGS. The water-use activity
duration is also simulated based on, typically, a lognormal distribution, also resulting
from analyses of NHAPS, REUWS, and REGS. For more information on how the
activities are mapped to model locations and how the water-use simulation is
implemented, see Wilkes (1999).
2.2.2. Water-use Behaviors for Groups of Interest. Release of airborne
contaminants occurs as a result of typical household water uses. In addition, dermal
contact occurs during some household water uses like showers and baths. For this
reason, it is imperative to represent these water uses as accurately as is reasonable
within the daily activity patterns of the model occupants. From a population exposure
point of view, the water-use activities that have a significant impact are use of showers,
baths, clothes washers, dishwashers, toilets, and faucets. For each of these water
uses, the published literature and other data sources such as survey data have been
reviewed, analyzed, and summarized in the following sections.
By comparing the data in NHAPS to other, smaller population based studies of
water use, it was concluded that NHAPS provides reliable data on frequency of
occasional water-use events (e.g., showering and bathing), but is believed to provide
poor estimates of the event durations, because the values were based on recall (Wilkes
et al., 2004). The respondents tended to estimate event durations around 5-minute
intervals, and the values were not consistent with published literature (Wilkes et al.,
2004). In contrast, because REUWS is derived from direct water meter measurements,
REUWS provides reasonable data on the durations and volumes of some water-use
events, particularly showers, clothes washers, and toilets. However, since REUWS is
based on the entire household water use, personal frequencies of water-use events for
individual persons cannot be reliably discerned. In regard to clothes washer
frequencies, REGS provides the best data for our purposes.
21
-------
2.2.2.1. Showers — The model uses shower frequency, duration, water flowrate
and temperature to represent occupant showering behavior and subsequent
contaminant release and occupant exposure. A Poisson process is used to simulate
shower occurrence, and a lognormal distribution is sampled to simulate the duration.
Analysis has shown that showering characteristics vary among demographic groups. A
number of shower studies have been done throughout the United States to determine
typical shower frequency, durations, and volumes. These studies include a study of 162
U.S. households by the U.S. Department of Housing and Urban Development (HUD,
1984). A study was conducted of 25 homes in Tampa, Florida (Konen and Anderson,
1993), and a study of 25 homes in Oakland, California (Aheret al., 1991). In general,
these studies revealed an average frequency of around five showers per week and a
duration ranging from 6.0 to 10.4 minutes. The average flowrates measured in the
Tampa and Oakland studies ranged from 1.5 to 2.5 gpm.
In addition to the above studies, NHAPS and REUWS have been analyzed for
showering characteristics, as discussed above. The analysis conducted by Wilkes et al.
(2004) concluded that NHAPS provided the most reasonable basis for specifying
shower use frequency, and REUWS provided the most reasonable basis for specifying
shower duration characteristics. The results of the frequency analyses from both
NHAPS and REUWS are presented in Table 10. The results of the duration, volume
and flowrate analyses from REUWS are presented in Table 11. For a more detailed
discussion of these data sources and analyses, refer to Wilkes et al. (2004). The
selected parameter values for showering frequency, duration and flowrate used in this
modeling study are presented in Table 12. These values were selected based on the
data presented in Tables 10 and 11 assuming the 15-45 age group is similar to the
analyzed 18-48 age group and that a 6-year-old child is well represented by the analysis
of the 5-12 age group.
2.2.2.2. Baths — The model uses bath frequency, duration and water volume
and temperature to represent occupant bathing behavior and subsequent contaminant
release and occupant exposure. A Poisson process is used to simulate bath
occurrence, and a lognormal distribution is sampled to simulate the duration. Relatively
few studies have been conducted in the United States to determine typical bath
22
-------
frequencies, duration, and volumes. The HUD study in 1981-83 found that people who
only bathe (do not shower) take about 2.9 baths per week. The NHAPS database is
analyzed for bathing frequencies and duration. Although the bathing durations given in
NHAPS tended to cluster around 5-minute intervals, and are based on recall, it is the
only generalized population study of this behavior, and therefore is the best available
data. The REUWS database does not provide bathing durations, only the amount of
time it took to fill the tub. The results of the NHAPS bathing frequencies and durations
for the three subpopulations of interest are provided in Table 13. The results of the
REUWS analysis to determine bath flowrate is presented in Table 14. The bathtub
emission model uses a bathtub water volume, a fill duration, and a bath duration.
Although no studies have analyzed the volume of water used in bathing, HUD (1984)
estimated 50 gallons (189 L) based on the physical size of a typical bathtub. The fill
duration was set at 8 minutes, which is consistent with the amount of time required to fill
a 50-gallon bathtub, based on a mean flowrate of 25 L/minute (6.6 gal/minute). This
mean bath fill flowrate was derived by evaluating both field measurements and the
REUWS data. The flowrate in two independent field measurements in household
bathtubs were 8.9 and 9.3 gallons/minute (Appendix to Wilkes et al. 2004). The
REUWS analysis resulted in a mean bath fill flowrate of 4.9 gallons/minute, with a
standard deviation of 2.1 gallons/minute. The selected bath fill flowrate value of 6.6
gallons/minute is consistent with the REUWS study at approximately the 85th percentile.
The selected parameter values used in the modeling study are presented in Table 15.
These values were selected based on the data presented in Tables 13 and 14
assuming the 15-45 age group is similar to the analyzed 18-48 age group and that a
6-year-old child is well represented by the analysis of the 5-12 age group.
2.2.2.3. Clothes Washers — The model uses clothes washer frequency, the
number of cycles and information about each cycle, including fill duration, agitation
duration, water volume and water temperature to represent occupant use of clothes
washers and subsequent contaminant release and occupant exposure. A Poisson
process is used to simulate clothes washer use. Both the NHAPS and the REGS
surveys asked respondents questions about their clothes washer use. The two
questions asked in NHAPS were: "How often do you wash clothes in a machine?" and
23
-------
"How many separate loads of laundry were done when you were home?" The answers
for the first question were recorded as: Almost every day, 3-5 times a week, 1-2 times a
week, Less often, or Don't know. The answers for the second question were recorded
as actual number of loads under 10, or "over 10." The problem with the first question
was that the frequency range in the choices is too broad, and the question is unclear
whether it refers to how many actual loads or how many days per week they did laundry
regardless of how many sequential loads they did in one day. The major problem with
the second question is that it required the individual to be at home during the event. In
the REGS survey, the question relating to clothes washer use was more specific;
however, the answer choices likewise offered a range. The REGS question was: "In an
average week, how many loads of laundry are washed in your clothes washer?" The
answer choices were: 1 load or less each week, 2 to 4 loads, 5 to 9 loads, 10 to 15
loads, More than 15 loads, or Don't know.
REGS was analyzed for clothes washer frequency behavior (Wilkes et al., 2004)
because the questionnaire was less ambiguous than the one used for NHAPS. The
results for three-person families are presented in Table 16, showing the percentage of
the 3-person families in the REGS database that used the clothes washer the specified
number of times per week. The analysis of three-person families excluded families with
individuals over the age of 65 because we were attempting to represent families with
children. The REUWS and experimental data are analyzed for clothes washer volume
and durations of the various wash and rinse fills, and agitation cycles. The results of the
analysis are presented in Table 17. Table 18 presents selected parameters to be used
in modeling clothes washer use.
2.2.2.4. Dishwashers — The model uses dishwasher frequency, the number of
cycles and information about each cycle, including cycle duration, water volume and
water temperature to represent occupant use of dishwashers and subsequent
contaminant release and occupant exposure. A Poisson process is used to simulate
dishwasher use. There are very few studies on the water-use characteristics of
dishwasher use. In 1984, a HUD study reported that people generally used the
dishwasher 3.7 times per household per week, or 1.2 times per person per week. A
1983 Consumer Reports study (reported in HUD, 1984) found that dishwashers at the
24
-------
time were using from 8.5 to 12 gallons per load, and older dishwashers were using 14
gallons per load. Similar to the NHAPS clothes washer data, the NHAPS dishwasher
data is likewise unreliable as the questions pertaining to dishwashers were ambiguous.
The NHAPS questions relating to dishwashers were, "How often does (respondent) use
the dishwasher?" This does not indicate how often the family used the dishwasher.
However, the REGS respondents were asked, "Which category best describes how
often your household actually uses the automatic dishwasher in an average week?"
Their answer choices were as follows: less than 4 times a week, 4 to 6 times a week, or
at least once each day. The REGS data were analyzed for three person households,
excluding all families with a member over 65 years old in order to best represent
families with a child. Table 19 presents the percentage of 3-person families surveyed in
the REGS database that used the dishwasher either daily, 4-6 times per week, or less
than 4 times per week. Since dishwasher cycle volumes and durations have not been
well-characterized in any known surveys, data obtained from the manufacturers were
used for these parameters. These data are presented in Table 20. Table 21 presents
the values selected for use in the modeling study. The emissions during the drying
portion of the cycle have not been studied, and therefore drying is not considered.
2.2.2.5. Toilets — The model uses the frequency of flushing to incorporate toilet
use into the sampled activity pattern. Once a toilet flush has occurred the emission
models also require the volume of water for the flush. For modeling purposes, it is
assumed that a flush duration is instantaneous.
Several recent studies reported toilet flush frequency and volume. These studies
focused on the performance of ultra-low flow toilets, contrasting their performance after
retrofit with the performance of the low flow and older non-conserving toilets they
replaced. The Tampa, Florida study (Konen and Anderson, 1993) retrofitted the
showers and toilets in 25 single-family homes with ultra-low flow devices and monitored
their water usage for 30 days before and 30 days after retrofit. The Oakland, California
study (Aher et al., 1991) retrofitted 25 single-family homes with ultra-low flow toilets and
monitored their water usage for 21 days before and 21 days after retrofit. The HUD
(1984) study monitored 196 households with 545 persons found that people flushed
25
-------
toilets approximately 4 times per day. The results from these studies are presented in
Table 22.
REUWS also provides toilet use data. The data were derived from an analysis of
household water meter monitoring. Because the water meters record total water use for
the household, it is impossible to attribute each flush to any given individual. Therefore,
the average frequency of toilet use in REUWS was derived by analyzing the total
frequency of use for each family divided by number of persons in the household. The
data contained in REUWS has been analyzed for frequency of toilet use and water
volume characteristics. For a complete description of the analysis of REUWS refer to
Wilkes et al. (2004).
The frequency of toilet use was modeled as a Poisson process with a mean
frequency of 5.23 flushes per person per day. The volume per flush was found to be
best represented as a normal distribution with a mean of 3.5 gallons and a standard
deviation of 1.2 gallons. The results of the REUWS analysis are presented in Table 23.
The actual toilet use frequency and volume values used in the DBP modeling study are
presented in Table 24.
2.2.2.6. Faucets — Faucet use characteristics for bathrooms and kitchens were
researched in a study of 25 homes in the city of Tampa (Konen and Anderson, 1993).
The mean water flowrate was 2.4 gpm from the kitchen faucet and 3.4 gpm from the
bathroom faucet, each with the faucets fully open. HUD (1984) estimated that faucet
use in the homes they studied was 9.0 gallons/person/day. The frequency of faucet use
was not given. These data are presented in Table 25.
The faucet use characteristics reported in REUWS are analyzed and reported in
Table 26. The REUWS database should be used with caution in respect to faucet use,
since the techniques used to acquire the data in REUWS are unreliable, and it is
expected that many uses labeled as faucets are misclassified and that many of the uses
labeled as "leaks" and "unknown" could be faucets. For a complete discussion of the
analysis, refer to Wilkes et al. (2004). The actual faucet use parameter values selected
for use in the DBP modeling study are presented in Table 27. The frequency and
duration values were adjusted from those in the REUWS analysis because the room
locations and activity patterns sampled from NHAPS do not typically provide adequate
26
-------
opportunity for the frequency of faucet use reflected in the analysis of REUWS. Most
probably resulting from the fact that people don't often report being in the locations of
faucet use, they tend to under-report bathroom visits, and small water uses overall. In
addition, there is no reasonable information on which household faucet is being used
(e.g., bathroom, laundry, kitchen). Therefore, to compensate for the discrepancies (i.e.,
interface with activity patterns), the faucet frequencies were adjusted downward, while
the durations were increased.
2.3. INGESTION CHARACTERISTICS
The most obvious route of human exposure to waterborne contaminants is via
ingestion. Every day, people drink water directly and consume water indirectly in juices,
sodas, soups, foods, coffee, tea, etc. In order to assess a person's ingestion exposure
to chemicals found in the water system, it is important to appropriately represent and
estimate the amount of water the person consumes, and from what sources. In order to
understand the dynamics of exposure uptake and distribution in the body, we must first
consider the dynamics of direct and indirect consumption from an exposure perspective.
U.S. EPA (2000b) puts forth the following definitions:
Direct Water: defined by (U.S. EPA, 2000b) as plain water ingested directly as a
beverage.
Indirect Water: defined by (U.S. EPA, 2000b) as water added to foods and
beverages during final preparation at home, or by food service
establishments such as school cafeterias and restaurants.
In this study, direct consumption is defined as "direct water" consumed by the
occupant and indirect consumption is defined as "indirect water" consumed by the
occupant. For direct consumption, we must develop a methodology for representing the
number of drinks and volumes consumed, either assuming that the contaminant level
remains constant from tap to glass to body, or consider that some contaminant
volatilized during air contact. For indirect water consumption, such as via food or
reconstituted drinks, we also need to consider the quantity consumed, and also
evaluate whether the fraction of the contaminant remaining in the drink or food after
volatilization and preparation is still significant or should the drink or food be ignored in
the exposure calculation.
27
-------
2.3.1. Available Data Sources. Currently, the U.S. EPA typically assumes that adults
consume an upper-percentile quantity of 2 liters of tap water per day and infants (body
mass of 10 kg or less) consume 1 liter per day (U.S. EPA, 1997a). These rates include
the tap water consumed directly and the tap water consumed in other drinks like juices,
coffee, etc. Prior to 1995, the primary survey used to estimate tap water intake in the
U.S. was the U.S. Department of Agriculture's (USDA) 1977-1978 National Food
Consumption Survey (Ershow and Cantor, 1989 in Exposure Factors Handbook (U.S.
EPA, 1997a)). However, newer studies have been conducted that better reflect
consumption behavior for modern times, reflecting our changed habits such as drinking
more bottled or filtered water, and drinking more soda and other canned drinks.
Furthermore, water intake is assumed to vary with levels of physical activity and outdoor
temperatures and Americans are exercising more than ever.
There are two major recent surveys that prove useful when estimating the
amount of water people ingest per day. One is NHAPS and the other is the Combined
1994-1996 Continuing Survey of Food Intake by Individuals (CSFII) (U.S. EPA, 2000b)
conducted by the USDA. There are also a few other studies presented in the Exposure
Factors Handbook (U.S. EPA, 1997a).
2.3.1.1. Ingestion: Exposure Factors Handbook — The Exposure Factors
Handbook, Volume 1, Chapter 3 (U.S. EPA, 1997a) presents the key and relevant
drinking water intake studies prior to 1995. These surveys and studies include the
following: 1981 Tapwater Consumption in Canada study by the Canada Department of
Health and Welfare; 1977-78 Nationwide Food Consumption Survey by the U.S.
Department of Agriculture, analysis by Ershow and Cantor; 1978 Drinking Water
Consumption in Great Britain, analysis by Hopkins and Ellis; 1987 Bladder Cancer,
Drinking Water Source, and Tapwater Consumption study by the National Cancer
Institute, analysis by Cantor et al.; and the 1992-1994 National Human Activity Patterns
Survey (NHAPS) analysis by Tsang and Klepeis (1996). For a more complete
discussion of these studies, see Wilkes et al. (2004). The tapwater consumption data
from these studies are summarized in Table 28, specifically for the subpopulations that
most closely represent the three groups of interest identified previously.
28
-------
2.3.1.2. 1994-1996 USDA's Continuing Survey of Food Intake by Individuals
(CSFII) — The 1994-96 USDA's CSFII is the most recent and comprehensive
consumption database available. CSFII was conducted over the 3-year period between
January 1994 and January 1997. More than 15,000 persons in the United States were
interviewed on two non-consecutive days with questions about what drinks and foods
they consumed in the previous 24 hours. The U.S. EPA report, Estimated Per Capita
Water Ingestion in the United States (U.S. EPA, 2000b), presents estimates of per
capita water ingestion based on the CSFII data for direct and indirect water intake. The
study uses the following definitions:
• Direct water: plain water consumed directly as a beverage.
• Indirect Water: water used to prepare foods and beverages at home or in a
restaurant.
• Intrinsic Water: water contained in foods and beverages at the time of market
purchase before home or restaurant preparation. Intrinsic water includes both the
"biological water" of raw foods and any "commercial water" added during
manufacturing or processing.
In the survey, respondents were asked:
• What is the main source of water used for cooking? (Community water, private well,
spring, bottled, other?)
• What is the main source of water used for preparing beverages? (same)
• What is the main source of plain drinking water? (same)
• How many fluid ounces of plain drinking water did you drink yesterday?
• How much of this plain drinking water came from your home? (All, most, some,
none)
• What was the main source of plain drinking water that did not come from your
home? (Tap or drinking fountain, bottled, other, don't know)
• Recall everything they ate over the past 24 hours. Where was the food
obtained?
2.3.2. Ingestion Behavior for the Three Populations: Results of Analysis. Of the
available references providing water consumption data on the subpopulation groups of
interest for our study, the CSFII survey was chosen as the most useful because of its
current relevance and its comprehensive specification of water intake in its various
29
-------
forms. The intakes for the two days of the survey were averaged for each person,
providing the estimated mean two-day average. Table 29 lists the distribution
parameters (geometric mean and standard deviation) for direct and indirect tapwater
consumption in the U.S. for women and men over 20 and children between 1 and 10
from the CSFII study. Table 30 shows a comparison of the consumption percentiles for
the data set and the fitted lognormal distributions for each of the demographic groups.
The actual parameters selected for use in this DBP modeling study are presented in
Table 31.
2.3.2.1. Methodology for Distributing Water Consumption Throughout
Day — No studies were identified that quantify the manner in which water consumption
is distributed throughout the day. A reasonable, common sense approach is being
adopted for implementing this distribution. The water consumption will be distributed
into a specified number of consumption events represented by a Poisson process over
the daytime (nonsleep) period using parameters listed in Table 31. This results in an
exponential distribution with the mean frequency given in Table 31 across the modeling
study for each population group. The consumption volume is sampled from the
appropriate lognormal distribution as identified in Section 2.3.2 and Table 31, with the
total volume randomly placed among the consumption events.
2.4. BUILDING CHARACTERISTICS
Housing characteristics, including zonal volumes, interzonal airflows, and whole
house air exchange rates, also have a significant impact on the estimated exposures.
The important building parameters are volumes of the whole house, volumes of the
individual water-use zones, whole house air exchange rates, and interzonal airflows.
TEM will model each subject residence as a collection of individual water-use
zones in flow communication with a "Rest-of-House" (ROM) zone that aggregates the
zones that are free of water-use sources. In order to execute TEM for typical conditions
and building characteristics, information related to indoor volume and airflows is
needed.
2.4.1. Representation of Household Volumes. The Exposure Factors Handbook
(U.S. EPA, 1997b) recommends using 369 m3 as the central estimate of volume for
American residences, with a conservative value with respect to air concentrations of
30
-------
217 m3. These estimates are based on peer-reviewed data appraisals drawn from
statistically representative surveys of American households through the Residential
Energy Consumption Survey. The REGS survey was first conducted in 1978 and was
updated on a biennial basis until 1984, after which the survey was conducted
periodically, every 3 or 4 years. In addition to data related to energy consumption,
REGS solicits information on demographics, building characteristics, and other factors
that relate to the needs of TEM. The distribution of indoor residential volume contained
in the Exposure Factors Handbook was calculated based on the estimated floor area
(the estimated or measured square footage of conditioned floor space in each home)
assuming 8-foot (2.44 m) ceiling height.
Estimates for total house volume contained in the Exposure Factors Handbook
were derived primarily from REGS data collected in 1993 and published in 1995 (U.S.
DOE, 1995). Results of the 1997 survey (U.S. DOE, 1999) only became available after
the Exposure Factors Handbook was updated. Initial reviews of the 1997 REGS data
indicate that total house volume estimates derived from the 1997 REGS data would be
very similar to the earlier data. The REGS data was analyzed and the
representativeness of several distributions was evaluated. The REGS data was further
analyzed to characterize the volume of 3-person U.S. households. In Figure 3, these
house-volume data are fitted to a lognormal distribution, with a geometric mean of 317
m3 and a geometric standard deviation of 0.422. The probability density function for the
chosen lognormal distribution is compared to a histogram of housing volumes in Figure
4. Based on the fit, a lognormal distribution was chosen to represent the distribution of
volumes, as shown in Figure 5. Such housing corresponds to a modest (~1400 ft2)
residence occupied by 3 or 4 people. In addition to expected general appliances, all
such homes are equipped with a kitchen (which usually contains an automatic
dishwasher), and nearly all have two baths plus a laundry, as well as a basement. The
"average" house has a central forced-air system to support heating and cooling needs.
Selection of Total House Volume: Total house volume for 3-bedroom cases are
selected from the statistical distribution derived from the 1997 REGS data (Table
32). The distribution of total volume for 3-bedroom homes is lognormal (Figure
3), and is characterized by a geometric mean volume of 317 m3 (11,195ft3) and
geometric standard deviation of 0.4218.
31
-------
The REGS data does not identify volumes for individual water-use zones. Given
that indoor spaces are designed to meet specific patterns of use, the Architectural
Graphics Standards published through the American Institute of Architects (Hoke, 1988,
1994) provides a basis for assigning floor areas to specific zones. This resource
summarizes the range of basic dimensions for key zones for various sized households.
For example, the range of kitchen dimensions is keyed to the number of people in the
household. Table 33 summarizes this range for a household composed of 3-4 people
(the predominant household size for 3-bedroom U.S. homes). Bathroom dimensions,
on the other hand, are largely independent of the number of people. Floor areas have
been transformed to volume estimates assuming 8-foot (2.44 m) ceiling height.
This range of zonal volumes is largely unverified in the professional literature, but
the values in Table 33 have the intuitive appeal of being derived from an authoritative
source (Hoke, 1988, 1994) that guides residential design. Residential laundry facilities,
for the most part, are installed in a host space rather than taking up a separate room. In
homes featuring a heated basement, the laundry should be positioned in that zone. In
homes built to slab-on-grade and crawlspace designs, the laundry is usually assigned to
the kitchen, and the kitchen-laundry zone should be sized to accept both uses.
Selection of Indoor Volumes for Water-Use Zones'. The range for zonal sizes are
defined from the Architectural Graphics Standards. For each type of water-use
zone, each range listed in Table 33 will be used to define zone-specific uniform
distributions. Values assigned to individual model cases will be randomly
selected from these distributions within TEM.
2.4.2. Representation of Whole House Air Exchange Rates and Interzonal
Airflows. The Exposure Factors Handbook (U.S. EPA, 1997b) recommends using 0.45
as the "typical" value for air exchange rate (ACH) in American residences. The national
distribution of residential air exchange is described in the Exposure Factors Handbook
and summarized in Table 34. In the absence of comprehensive measurement surveys,
the distribution in Table 34 was derived from analysis of perfluorocarbon tracer (PFT)
data collected for a number of research programs since the early 1980s (Koontz and
Rector, 1995).
Selection of Air Exchange Rate'. The national distribution of residential air
exchange rates are defined from the Exposure Factors Handbook (see Table
32
-------
34). Values assigned to individual model cases will be randomly selected from
the distribution representing "All Regions" within TEM.
The water-use zones are set up in this model to have airflows with the "Rest of
the House (ROM)," but not with the outdoors. These zones exchange air with the ROM
at rates specified in Figure 3 of the report, rates which are a function of the whole house
air exchange rate (WHACH), but are somewhat lower, and subsequently, the decay rate
is slowed. These relationships were developed as a result of analysis of the PFT
database. Another important factor is the means of specifying the WHACH. The
WHACH is sampled from a representative distribution, which is lognormal with a
geometric mean of 0.46 hr-1 and a geometric standard deviation of 2.25. Sampling
from this distribution will result in a wide range of WHACHs, with some values lower
than 0.1 and other values greater than 10. The WHACH sampled for case 48 (the
illustrative case whose results are shown in the figures) was 0.1 1 hr"1. Table 35
summarizes the relevant parameters selected for case 48. All of the parameters are
consistent with the algorithms presented in this section and represented in Figure 3.
Given the simplified scenarios envisioned for the model runs, interzonal airflows
can be assigned through the air exchange rate. That is, interzonal airflows would be
sized by the air exchange terms. The next level of complexity utilizes the algorithms
developed by Koontz and Rector (1995) from their analysis of the PFT data cited above.
Under this scheme, the normalized interzonal airflow (QN, h"1) for any zonal pair is
defined as a function of the flow from zone 1 to zone 2 (Qi2), flow from zone 2 to zone 1
(Q2i), and total (V, m3) such that:
i
While the analysis showed differences in the correlation equations, the practical
differences are negligible in that both estimators produce a normalized interzonal airflow
term of 0.22 h"1 at an air exchange rate (I, h"1 ) of 0.45:
Bedroom: QN = 0.078+0.31 / (14)
Kitchen: QN = 0.046+0.39/ (15)
33
-------
It is expected that bathrooms are used with the door closed. Relatively little
direct data exists to define airflows. Experimental work by Giardino et al. (1992)
provides useful values published in a peer-reviewed journal. For a 13 m3 bath, these
determinations found exiting airflow from the bath to the adjacent hallway to be 4.2 m3
h"1 with the door closed and 15.1 m3 h"1 with the door open. Similarly, entering airflows
from the hallway to the bath were found to be 16.3 m3 h"1 with the door closed and 47.9
m3 h"1 with the door open. These flows were utilized in subsequent residential exposure
modeling of radon volatilized from various water-use scenarios (Rector et al., 1996). At
higher levels of complexity, dynamic and engineering estimators can be applied to
recognize the influences of weather and operation of the heating/cooling system.
A modeling study conducted by researchers at the National Institute of Standards
and Technology (NIST) developed simplified approaches to modeling interzonal
dispersal of indoor contaminants in homes served by central air-conditioning/heating
systems (Persily, 1998). Under the NIST study, patterns of fan operation were defined
by the following rules:
• Airflows were assumed to be 50 L s"1 (180 m3 h"1) at major supply registers and
25 L s"1 (90 m3 h"1) at minor supply registers when the central air handler was
running. (These values are consistent with standard guidance in ASHRAE
1992.)
• System on-time was assumed to be 60% (of the total timeframe) at design
conditions, (i.e., the highest outdoor temperature reached 98-99% of the time
during the cooling months, or the lowest outdoor temperature reached 98-99% of
the time during the heating months).
The NIST study also addressed local exhaust fans operating in the kitchen and
bathrooms under user control. Based on analysis of commercially-available equipment
and engineering judgment, kitchen exhaust flows were assigned to be 170 m3 h"1 (100
cfm), and bath exhaust flows in the NIST study were assigned to be 80 m3 h"1 (47 cfm).
Selection of Interzonal and Exhaust Airflows'. Interzonal airflows are scaled by
the air exchange rate using the algorithm developed by Koontz and Rector
(1995). Exhaust flows for the kitchen and bathrooms will be assigned in
conformance with the NIST study (170 m3 h"1 in the kitchen, 80 m3 h"1 in each
bath, under user control). These flows will be superimposed on the airflows that
prevail when the fans are not operating.
34
-------
2.4.3. Model Representation of Building. As described in Section 1.4.1, the house is
idealized as a collection of compartments where water-use zones are explicitly
represented and the remaining indoor zones are lumped into a common zone called
"Rest of House" (ROM). The volume parameters and the air exchange rate parameters
are specified in accordance with Sections 1.4.1 and 1.4.2. The idealized representation
of the house is presented in Figure 5.
2.5. CONCENTRATIONS IN WATER SUPPLY
The concentrations of DBFs in U.S. drinking water supplies vary significantly
across utilities largely influenced by the source water characteristics and the treatment
processes. The results of three national surveys and a limited 35-city survey of total
trihalomethane (TTHM) concentrations in finished U.S. drinking water both at the plant
and in the distribution systems found mean concentrations between 42 and 68 ug/L and
maximum concentrations ranging from 185 to 482 ug/L (McGuire et al., 2002). However,
these studies varied in methods of sample collection and laboratory analysis, and they
showed considerable variation in THM concentrations. Another recent case study in
two U.S. municipal water systems showed wide variation across individual distribution
systems (Lynberg et al., 2001).
Data on water concentrations of a variety of contaminants, including THMs, has
been collected as the result of requirements mandated by the Information Collection
Rule (ICR). These data encompass the period of July 1997 through December 1998 for
more than 330 water treatment facilities, which has been assembled as a database,
referred to herein as the ICR database (U.S. EPA, 2000c). A recent analysis of this
data by the U.S. EPA (McGuire et al., 2002) found significant variations as a function of
a variety of factors, including source of the water supply (e.g., surface water,
groundwater, etc.), season, EPA region, and plant disinfectant type (e.g., ozonation,
chlorine dioxide, chloramines, etc.).
A specific objective of this project is to examine the relationship between the
THM concentrations in the water supply and the blood and tissue concentrations in the
human body. The blood and tissue concentrations are affected by a variety of factors,
including exposures, uptake characteristics, metabolic and other removal processes.
The oxidative metabolism of the four THMs is catalyzed by the cytochrome P450 2E1
35
-------
(CYP2E1) isoform (Guengerich et al., 1991; Raucy et al., 1993). Metabolism of the
THMs by CYP2E1 forms phosgene or its brominated analogs, which can react with
tissue macromolecules to produce toxicity. Since the THMs are all substrates for
CYP2E1, mutual competitive inhibition of metabolism is expected to occur. During
mixed exposures to the THMs in water, each THM would inhibit the bioactivation of the
other THMs that are alternative substrates for CYP2E1. This inhibition would decrease
the formation of phosgene and its analogs from the THMs, leading to higher circulating
concentrations of the parent THMs in the body. The relative extent of these inhibitory
effects would depend upon the concentrations of the four THMs in the water supply and
their kinetic characteristics. The brominated THMs may also be metabolized by a
glutathione conjugation pathway (Ross and Pegram, 2003), but this study will focus on
the oxidative bioactivation pathways of the THMs.
Considering that the four THMs may not act independently with respect to
metabolic removal, it is important to appropriately represent the concentration of all
THMs when modeling metabolic processes. For this reason, we investigated the data in
the ICR database to evaluate the correlation among THMs at the upper ends of their
respective distributions (in the vicinity of the 90th percentile). In our analysis of the ICR
data, we addressed values reported as below the "Minimum Reporting Level (MRL)" by
assigning a concentration of half the MRL. The MRL reported by the ICR is 1 ug/L for
all four THMs, so the concentration 0.5 ug/L is used for reported values below the MRL.
To evaluate the correlation for a given compound, all samples in the 85thto 95th
percentile were selected, and then the Pearson correlation coefficient was calculated for
that compound with each of the other three THMs. This approach was applied
separately to chloroform, BDCM, DBCM, and bromoform. The results of these analyses
are presented in Tables 36-39.
The results, given in Tables 36-39, exhibit a significant correlation between each
individual THM compound and its closest neighbors in terms of the number of chlorine
or bromine atoms. For example as can be seen in Table 36, the chloroform
concentrations are significantly correlated with the BDCM concentrations, with minimal
correlation with the DBCM and bromoform concentrations. Similarly in Table 37, BDCM
36
-------
concentrations exhibit correlations with the chloroform and DBCM concentrations, but
minimal correlation with bromoform concentrations.
Based on the results of the correlations analysis, we considered using a specific
percentile (e.g., the 95th percentile) for the concentration of compound of interest, and
assigning the concentration of the remaining three THMs based on a regression
analysis, but that approach is somewhat artificial because of the inconsistent degree of
correlation in the concentrations of the individual chemicals. So we chose the
alternative of using concentrations from the actual sample corresponding to the desired
percentile. For example, for chloroform, we analyzed the data set to determine the
chloroform concentration at the desired percentile, identified the actual records (drinking
water samples) for which that chemical was present at the concentration representing
the 95th percentile for its overall distribution. The correspondingly measured
concentrations for all four THMs taken from that record are reported in Tables 40-43 to
demonstrate the fluctuation of the "other three" chemicals, versus the consistency of the
concentration of the "target" THM. For simplicity sake, only records from treatment
systems relying on surface water are reported in that series of tables. Table 44
demonstrates concentration results from all systems, those relying on surface water, as
well as those relying on ground water as source water. For each analysis, because of
the size of the dataset, a number of records were found to have the same 95th
percentile value (i.e., the same concentration which corresponded to the 95th
percentile). For example, Tables 40-43 present the ICR database records for the 95th
percentile concentration for chloroform, BDCM, DBCM, and bromoform, respectively. In
selecting an actual record to be used to represent the 95th percentile, we opted for the
record where the sum of the concentrations of the brominated compounds (i.e.,
bromoform DBCM, and BDCM) is maximized. We chose this condition to examine
exposure scenarios where the maximum inhibition of the oxidative metabolism of the
THMs (i.e., their bioactivation) is expected to occur. Using these criteria, the ICR
database was analyzed for the 95th percentile record for each THM as presented in
Table 44.
A number of water treatment factors were found to influence the THM
concentrations. We chose to evaluate the effect of the subset of these factors, listed in
37
-------
Table 45. The factors were chosen to investigate the effects of the raw water source,
the disinfection methods and seasonal factors. A number of other factors that are
expected to be important include retention time in the distribution system, the organic
content, and the effects of the hot water heater that were not considered in these
analyses because they were outside the scope of this project. The ICR database was
analyzed for each variable in Table 45 to identify the 95th percentile record for each
THM, using the maximized sum of the brominated compounds to identify the worst-case
record, as discussed above. The results of the analysis are presented in Table 46. The
concentrations presented in Tables 44 and 46 were used as inputs.
2.5.1. Water Concentrations Selected as Model Inputs. The concentrations of
THMs presented in Tables 44 and 46 represent the available water concentration inputs
for the exposure and dose analysis. From this dataset, the exposure and dose analysis
will be conducted using concentrations presented in Table 44 and 46 for the following
variable subgroups:
• The entire dataset,
• The "Surface Water Intake" treatment facilities,
• Treatment plants that include "all chlorine based disinfection,"
• And, "systems sampled between July and September (1997 and 1998).
Within each data set, the concentrations for each THM were selected so that
maximum exposure might occur. Under that condition, the "worst case" 95th percentile
concentration for each distribution was selected. Given their chemical nature, and
because of differences in treatment options, source water characteristics, etc., these
compounds are formed at different rates in different systems. Within each category of
treatment (i.e., All Systems Using Surface Water Intake), the 95th percentile values for
each THM were different, and were somewhat related. In Table 46, the first row
indicates that when chloroform is present at the 95th percentile of its distribution, the
BDCM is present at the 98th percentile of its distribution, DBCM is present at the 90th
percentile of its distribution, and bromoform is present at the 0th percentile of its
distribution - treatments that favor the formation of chloroform disfavor the formation of
bromoform. When bromoform is present at the 95th percentile for its distribution, then
38
-------
chloroform is present at the 34th percentile of its distribution, BDCM is present at the
96th percentile of its distribution and DBCM is present at the 98th percentile of its
distribution. To simulate a potential worst-case exposure, the exposure model
incorporated drinking water THM concentrations at the 95th percentile of their
distributions (e.g., chloroform @ 66 ppb, BDCM @ 23.8 ppb, DBCM @ 17.0 ppb and
bromoform @ 5.6 ppb).
Each paired set of data will be used and the resulting exposures and doses will
be estimated.
2.5.2. Estimated Concentrations in Consumed Tap Water. This section presents
the development of reasonable representations of the chemical concentrations in
consumed tap water for the four THMs. The volatilization of contaminant occurs during
the filling activity, from the water surface while sitting in a glass or storage and as a
result of any processing action. Each of these is analyzed below, and a combined
volatilization is calculated for a number of scenarios. The results of this calculation are
used to recommend estimated fractional volatilization and first order removal rate
constants for each chemical.
2.5.2.1. Volatilization During Filling — Volatilization during a filling activity
occurs in much the same way as during any other faucet use. There are differences in
the volatilization occurring in the pool of water in a partially filled glass of water and the
film of water in the bottom of a sink.
The experiments from Howard and Corsi (1996) as well as those performed by
Batterman et al. (2000) attempt to quantify this volatilization. Batterman et al.
implement an experiment meant to represent an "experimental procedure portray(ing)
the filling of a pitcher from the tap and then the filling of a glass from the pitcher." The
authors describe the procedure as follows:
The THM stock solution (2 mg/mL of each THM) was diluted in a filled 4 L
black bottle to obtain the test mixture containing 100 ug/L of each THM
compound and then transferred to a typical covered water pitcher
(Rubbermaid, capacity = 2.34 L, filled to 1.96 L, height = 21.7 cm, dia =
12.2 cm, material = resin) and used to fill glasses and mugs.
According to the authors, the "water transfers were done quickly (3-5 seconds) and at a
minimal (2 cm) pouring height."
39
-------
Unfortunately, neither the quick filling nor the filling height is typical of filling a
glass of water for consumption. Filling 1.96 L in 3-5 seconds yields a flowrate in the
range of 23.5-39.2 L/min. A typical faucet has a possible flowrate ranging from 0 (user
controlled) to approximately 11 L/min, with a typical faucet use being in the range of
about 2-8 L/min (Wilkes et al., 2004). The large flowrate used by Batterman et al. would
significantly lower the opportunity for volatilization. Although no behavioral studies have
been identified that quantify the distance the water must travel, it seems likely from
personal experience that 2 cm would represent a reasonable minimum, and a
reasonable maximum is probably on the order of 12-15 cm. The combination of the
large flowrate and low height of the filling in the Batterman et al. experiment has the
effect of significantly lowering volatilization, and therefore this research is not useful in
estimating the volatilization during filling.
Howard and Corsi (1996) conducted experiments measuring the volatilization
resulting from using the kitchen faucet. The most consequential differences between
the Howard and Corsi experiments and the filling of a glass or pitcher for consumption
are the larger height of the drop and the potential splashing that could occur when the
water lands in the sink. Both of these differences lead to a higher volatilization rate.
Howard and Corsi measured the stripping efficiency (percent of the chemical moving
from the water to the air during the process of filling the glass or pitcher with tap water)
for three compounds: cyclohexane, toluene, and acetone. The chemical properties
impacting the volatilization rate for the three compounds measured by Howard and
Corsi are given in Table 47. The chemical properties impacting volatilization for the
compounds being modeled are given in Table 48. Table 49 summarizes the stripping
efficiency measured by Howard and Corsi for the three compounds. These measured
values are used as a basis for estimating the stripping efficiency for each THM during
filling.
2.5.2.2. Volatilization During Storage — After preparation and prior to
consumption, the water may sit in a pitcher in the refrigerator or in a glass or cup on the
table. During this period, volatilization occurs at the liquid/air interface. Batterman et al.
studied the rate at which this occurred for the four trihalomethanes at a variety of
temperatures (4, 25, 30, and 100°C) and in two containers (tall glass, wide mouth glass)
40
-------
for a 2-hour period. Batterman et al. fit the resulting measurements to an exponential
decay model with good results (R2 values for chloroform ranged from 0.59 to 0.86).
Table 50 summarizes these results. The recommended fractions volatilized as a
function of time are summarized in Table 51 for three conditions (cold water, room
temperature water, and hot water). These data will be used in conjunction with the
estimated volatilization during filling to estimate the amount of each THM remaining in
the water at the time of consumption.
2.5.2.3. Volatilization During Processing — A wide variety of activities
influence the removal of compounds from tap water. These activities include primarily
heating and mixing activities that occur when using the water to make coffee, tea, other
water based beverages, and in the process of preparing food. Beverages made from
tap water fall into two primary categories: heated and non-heated beverages. The non-
heated beverages undoubtedly have some volatilization due to the process of mixing
the water with any additives, such as orange juice from concentrate. These losses have
not been quantified in the literature sources identified above. The heating of water
greatly reduces the concentration of volatile constituents. Batterman et al. report an
average chloroform loss of 81% resulting from bringing water to 100°C (presumably
from room temperature, although this is not stated) in a kettle. After pouring the water
into a mug, the measured fraction volatilized is an average of 85%.
2.5.2.4. Recommendations — Similar to volatilization from other water uses,
the volatilization during filling is correlated with the chemicals' Henry's Law constant, the
liquid phase diffusivity, and the gas phase diffusivity. Table 52 presents a variety of
consumption scenarios and estimated volatilization fraction as a result of each scenario
for each of the THMs. Table 53 presents recommended values for model inputs for the
THMs. The model uses an initial fraction volatilized and a rate constant to estimate the
amount of contaminant remaining at the time of consumption. The values presented in
Table 53 for the fraction of the compound remaining prior to consumption or storage
accounts for an estimate of the average amount volatilized as a result of filling a
container with tap water. The rate constant is used by the model to estimate the
volatilization during storage or while a glass of water is consumed over an extended
period (e.g., used to represent the volatilization from a glass of water over a period like
41
-------
30 minutes when someone slowly sips the water). The effects of expected heating,
stirring and other factors often encountered during food processing activities results in a
larger fraction of the original mass being released into the air a for indirect consumption,
as reflected in Table 53.
2.6. UPTAKE AND SOLUBILITY PARAMETERS
The PK model requires sets of input parameters by chemical, by exposure, by
compartment, by demographic group, and by activity.
2.6.1. Breathing Rates by Activity and Demographic Group. The breathing rates
(alveolar ventilation rates, QA) based on the Exposure Factors Handbook, Table 5.6
(U.S. EPA, 1997b) are presented in Table 54 for an adult male and female (15-45 years
old) and a child of approximately age 6 for two activity levels: resting and sedentary.
2.6.2. Skin Permeability Coefficients for Each Chemical. The skin permeation
coefficient, called the Permeability Coefficient of Stratum Corneum, Kp, is required for
each chemical to be modeled. For each of the four chemicals of interest, the Kp is
given in Table 55. Some of these values remain to be determined and are not available
at this time.
2.6.3. Partition Coefficients for Each Chemical. The partition coefficients between
the skin and blood and between the blood and air are required for the fundamental
uptake modeling in TEM. Partition coefficients for each physiological compartment are
given in Table 56 for the four DBPs of interest.
2.7. UPTAKE CALCULATIONS
The dermal uptake calculation implemented in TEM is based on membrane
equations developed by Cleek and Bunge (Olin, 1998). This representation uses two
simple functions, representing the non-steady-state and steady-state periods. The
dermal uptake does not account for issues such as skin hydration and skin temperature.
The ingestion uptake calculation implemented in TEM is based on the estimated
water concentrations at the time the water is consumed, and assumes that the entire
mass of the chemical in the consumed water is absorbed into the bloodstream.
The inhalation uptake calculation implemented in TEM is based on the predicted
air concentrations in the breathing zone. TEM implements an equilibrium calculation
42
-------
between the inhaled air and the bloodstream. This calculation is described below. The
inhalation uptake used for the PBPK model is described in Section 3.2.
2.7.1. Dermal Uptake Calculations. The dermal uptake is dependent upon a number
of factors, including the concentration in the water in contact with the skin, the duration
of the contact, the temperature of the water, and chemical properties. The water
concentration, which is dependent upon the amount of the chemical that has volatilized
prior to contact, is assumed to be the concentration in the incoming water for flowing
water type exposures (e.g., showers, facuets) and is calculated for the amount
volatilized for standing water type exposures (e.g., baths).
Dermal contact with the household water supply occurs primarily during two
types of water-using activities: (1) showering and bathing; and (2) faucet use. This
study assumes that for any bathing activity, 90 % of the skin is in contact with the
contaminated water, and for any faucet use, 5.2% of the skin (hands) is in contact. The
body surface areas corresponding to these assumptions are estimated to be 16,920
cm2 (adult female), 19,400 cm2 (adult male), and 7930 cm2 (child) for the entire body
(U.S. EPA, 1997a). The contact is assumed to occur for the time period that the water-
use is active at the initial water concentration. Water temperature is likely to impact the
rate of uptake (Gordon et al., 1997), but is not accounted for in the dermal uptake
calculation.
This case study implements the technique presented by Cleek and Bunge
(1993), as given by the following equations:
40 / p f
^ sc/w * scexp whenfexp>2.4f,ag (16)
and
M!n=AC°w(Psctexp +RSC/WLSC) when texp>2.4tlag (17)
where: Min = mass entering the skin
A = area of skin in contact with water
C° = concentration of solute in the aqueous solution
texp = duration of exposure
RSC/W= equilibrium partition coefficient stratum corneum (sc) and water (w)
43
-------
Lsc = diffusion path length through the stratum corneum (sc)
Psc = permeability coefficient of the stratum corneum (sc)
R L
t = sc/w sc = estimated time to reach steady state
/ag 6P
2.7.2. Inhalation Uptake Calculations. The inhalation uptake calculated by TEM is
based on the assumption that the lung-alveolar and lung blood achieve instantaneous
equilibrium at each breath, and that at equilibrium, the partition coefficient given in Table
56 describes the partitioning of the chemicals between the vapor and liquid (blood)
phases. The following steps are used to approximate chemical uptake into the blood:
The equilibrium concentrations in the alveolar blood and the alveolar air are as
follows:
Cbl,eq = Calv.eq * P (18)
where:
i, eq
= concentration in the lung blood after equilibrium is reached with
alveolar air
= concentration in the alveolar air after equilibrium is reached with lung
blood
= Blood/Air partition coefficient.
The equilibrium concentration in the lung blood and alveolar air is calculated by the
following equations:
Cbi.eq = (CW VW + Cor \/alv )/(V« + Va/v/P) (19)
and
' >
(20)
44
-------
where:
\laiv = alveolar volume for a time step = breathing rate multiplied by time step
Vw = lung blood volume for a time step = cardiac output multiplied by time step
Cbi = bulk blood concentration at the start of the time step
Cair = air concentration entering lungs.
The mass accumulated in the lung blood is assumed to accumulate in the body,
and the lung blood concentration is reset to zero between time steps. The cumulative
mass accumulated in the body is calculated as follows:
Mass absorbed = AC* V over all time steps (21)
The air concentration outside the body is assumed to be unaffected by the mass
transferred into the blood.
45
-------
3. PHYSIOLOGICALLY BASED PHARMACOKINETIC (PBPK) MODEL
PBPK modeling has proven useful in a number of different types of risk
assessment applications. Here, it is used to translate external concentrations and
exposure doses to internal doses, when the human exposure to four THM compounds
is simulated. The modeling approach makes use of the physicochemical properties of
the four THMs studies, physiological and anatomic characteristics of the demographic
groups evaluated, as well as the chemical-specific biochemical characteristics of
interest (e.g., metabolic rate constants). Some definitions of PBPK terms are provided
in Table 57.
3.1. MODEL STRUCTURE
The PBPK model for human exposure to THMs consists of seven compartments
representing organs or groups of organs: liver, kidney, rapidly perfused tissues, slowly
perfused tissues, genitalia, fat, and a gas-exchange lung (Figure 6). Skin is considered
to be a barrier in the PBPK model rather than a physiological compartment. The
compartments are interconnected by blood flows (Q). All biotransformation is assumed
to take place in the liver and follow Michaelis-Menten saturation kinetics. The PBPK
model is based on a 5-compartment PBPK model used to analyze the pharmacokinetics
of over 30 volatile organic compounds (VOCs) (Gargas et al., 1986, 1990), including the
THM chloroform. The PBPK model of Gargas et al. (1986, 1990) has been used as the
basis for more detailed PBPK models for chloroform (Corley et al., 1990) and
bromodichloromethane (Lilly et al., 1998). In the present study, separate PBPK models
for a human adult male, adult female, and 6-year-old male child will be used to analyze
tissue concentrations following exposure to THMs through water usage. The
physiological parameters are shown in Table 58. Because of concern for the potential
for developmental toxicity and reproductive toxicity of these compounds, the PBPK
model has been developed to include specific compartments for testes and ovary
tissues. In addition, it focuses on adults of reproductive age, broadly defined as ages
15-45. A single set pf parameter values for physiologic parameters (i.e., testes, as
percent of body weight) can be developed, inasmuch as these values do not change
appreciably with age - between ages 15 and 45. However, because of the rapid growth
46
-------
during childhood, with its accompanying changes in organ size and relative blood flow,
a single set of parameter values cannot be developed that would be descriptive of the
child at various ages. Thus, the present endeavor was limited to the hypothetical 6-year
old male child. The chemical-specific parameters for each of the four THMs are shown
in Table 59. Simulations of the PBPK model will be done using acslXtreme software
(Aegis Technologies, Huntsville, AL).
The most information about the solubility of the THMs in human tissues is
available for chloroform, where partition coefficients have been directly measured in
several human tissues (U.S. EPA, 2003). The human blood:air partition coefficients for
the other three THMs have been measured (Batterman et al., 2002), but solubility
measurements have not been made in other human tissues. Therefore, the human
blood:air partition coefficients for the three THMs have been divided by rat tissue:air
partition coefficients (da Silva et al., 1999) to obtain estimated human blood:tissue
partition coefficients (Table 59). Information regarding the solubility of the THMs in
human reproductive tissues was not available, so the solubilities of the THMs in these
tissues (testes, ovaries) were calculated based on tissue lipid and water content using
the algorithms of Krishnan (2002). These calculated tissue:air partition coefficients
were divided by the appropriate measured blood:air partition coefficients to yield
estimates of the testes:blood and ovaries:blood partition coefficients (Table 59).
3.2. MASS BALANCE EQUATIONS
For each of the 4 THMs, the rate of change of the concentration in arterial blood
(c/CA/cff; Equation 22) is described by accounting for the influx of the THM in arterial
blood from the lung (Cl), the venous concentration of the THM (CV) and dermal
exposure (DD):
dCNdt = (QC * CV + QP * CI)/(QC + (QP/PB)) + DD (22)
where QC is cardiac output, QP is the alveolar ventilation rate, PB is the blood:air
partition coefficient for each THM, and DD is the dermal dose. The DD was calculated
by TEM as described in Section 2.7 using the membrane equations developed by Cleek
and Bunge and the skin permeability coefficients given in Table 55. When dermal
exposures occurred in the activity scenarios, the dermal dose was calculated for each
47
-------
5-minute exposure interval. The results of the TEM dermal exposure calculations were
then inputted to the PBPK model at each 5-minute interval for the duration of the dermal
exposure episode.
The rate of change of the amount of each THM in non-metabolizing tissues
(c/AT/cff; Equation 23) (kidney, fat, genitalia, rapidly perfused, and slowly perfused) is
described by:
cfAT/df = QT(CA - CVT) (23)
where QT is the blood flow to tissue T and CVT is the venous blood concentration of the
THM leaving tissue T. The concentration of each THM in the venous blood leaving
tissue T (CVT; Equation 24) is given by:
CVT = AT/(VT*PT) (24)
where AT is the amount of each THM in tissue T, VT is the volume of tissue T, and PT
is the partition coefficient of each THM in tissue T.
The rate of change of the amount of each THM in the metabolizing tissue liver
(dALIdt, Equation 25) is described by:
c/AL/df = QL(CA - CVL) - RAM + RAO (25)
where QL is blood flow to the liver, CVL is the venous blood concentration of the THM
leaving the liver (described by Equation 24), RAM is the rate of metabolism (discussed
in Section 3.3) and RAO is the rate of oral absorption. RAO is a first-order process
described by Equation 26:
RAO = KA*DOSE*e(-KA*t) (26)
where KA is the oral absorption rate constant for each THM, DOSE is the amount of
THM administered orally, and t is time.
The concentration of each THM in venous blood (CV; Equation 27) is described
by:
CV = (QF * CVF + Q L * CVL + QK * CVK + QG * CVG + QR * CVR + QS * CVS)/QC (27)
48
-------
where Q is blood flow and CV is the venous blood concentration leaving each tissue F
(fat), L (liver), K (kidneys), G (genitalia), R (rapidly perfused) and S (slowly perfused).
The overall mass balance for each THM in the PBPK model is given by the sum
of the amount of the THM in each tissue or tissue group.
3.3. RELATIONSHIP BETWEEN THM METABOLISM AND TOXICITY
The target organ toxicity produced by chloroform requires metabolic activation by
cytochromes P450, specifically CYP2E1 (Constan et al., 1999). The oxidative
bioactivation of chloroform proceeds via CYP2E1 oxidation to trichloromethanol, which
eliminates HCI to form phosgene (Pohl et al., 1980). Phosgene is likely to be the
reactive metabolite that acylates proteins (Potts et al., 1949) to produce hepatic
centrilobular necrosis and renal proximal tubular necrosis (llett et al., 1973). An
analogous oxidative bioactivation occurs with the brominated THMs involving
elimination of HBr and formation of brominated analogues of phosgene or phosgene
itself in the case of bromodichloromethane (Lilly et al., 1997). Oxidation of the lower
energy C-Br bond would occur more readily than oxidation of the C-CI bond (March,
1968). The most appropriate dosimeter for the metabolite-mediated hepatic toxicity of
the THMs is CM24, the concentration of metabolites produced in the liver over 24 hours.
CM24 represents the integrated exposure of the liver to the reactive metabolites of THM
oxidation by CYP2E1 over 24 hours. The reactive metabolites formed from THM
oxidation (phosgene and its brominated analogs) are transient and do not accumulate in
the liver. Thus, CM24 represents exposure of the liver to the metabolites, not the
concentration of metabolites in the liver at any given time.
Local metabolic activation of the THMs in the extrahepatic target organs kidney
and genitals may occur. Renal metabolism has been observed with chloroform (Corley
et al., 1990; Constan et al., 1999) and BDCM (Lily et al., 1997, 1998). Extrahepatic
metabolism was not described in the present PBPK model for the THMs because the
enzymes involved and the kinetics of these potential metabolic pathways have not been
characterized. Therefore, the area under the curve (AUC) for the parent THMs provides
the most appropriate dosimeter for the exposure of the extrahepatic target tissues
kidneys and genitals to each THM. The AUC represents the integrated exposure of the
organ to the parent THM over 24 hours.
49
-------
3.4. METABOLIC INTERACTIONS
The cytochrome P450 2E1 isoform (CYP2E1) is the principal catalyst of the
oxidative metabolism of the THMs (Guengerich et al., 1991; Raucy et al., 1993). This
has been unequivocally demonstrated for chloroform (Constan et al., 1999) and
bromodichloromethane (Allis and Zhao, 2002; Zhao and Allis, 2002) and inferred for
dibromochloromethane and bromoform (da Silva et al., 1999). The oxidative
bioactivation of chloroform proceeds via CYP2E1 oxidation to trichloromethanol, which
eliminates HCI to form phosgene (Pohl et al., 1980). Phosgene is likely to be the
reactive metabolite that acylates proteins (Potts et al., 1949) to produce hepatic
centrilobular necrosis and renal proximal tubular necrosis (llett et al., 1973). An
analogous oxidative bioactivation occurs with the brominated THMs involving
elimination of HBr and formation of brominated analogues of phosgene or phosgene
itself in the case of bromodichloromethane (Lilly et al., 1997). Oxidation of the lower
energy C-Br bond would occur more readily than oxidation of the C-CI bond (March,
1968). While the lower energy of the C-Br bond would also be expected to allow
nucleophilic displacement of Br by glutathione S-transferases (Ross and Pegram,
2003), this study will focus on the oxidative bioactivation pathways of the THMs. In
addition, because the GST-mediated pathway is not active for each of these
compounds, because it is expected that the oxidative pathway accounts for a
substantially higher fraction of a metabolized dose of compounds that are also
metabolized by this pathway, and because this study focuses on estimating internal
dose without estimating toxicity or risk, this approach seems valid.
Corley et al. (1990) postulated that high concentrations of chloroform produced
inactivation of cytochrome P450 via metabolic activation. They inferred this suicide
inactivation pathway from gas uptake studies with mice where metabolic uptake was
observed to decrease after several hours exposure to 10,000 ppm chloroform. The
suicide inactivation pathway was not invoked for rats or humans (Corley et al., 1990).
However, there is no direct evidence for inactivation of cytochrome P450 by chloroform.
Experiments at CUT with isolated mouse hepatocytes in vitro were not consistent with
cytochrome P450 inactivation by chloroform (Kedderis and Held, unpublished
observations; Kedderis et al., 1993; Held et al., 1994). The freshly isolated cells were
50
-------
incubated with high but sublethal (Ammann et al., 1998) concentrations of chloroform
(~1 mM) for 2 hours, washed in fresh medium by centrifugation, resuspended in fresh
medium, and assessed for viability by microscopy and metabolic capability toward
chloroform by gas chromatography. Treatment with chloroform under these conditions
did not affect cell viability or the metabolic capacity of the cells toward chloroform
(Kedderis and Held, unpublished observations). These results suggest that the suicide
inactivation pathway for cytochrome P450 postulated by Corley et al. (1990) is not
operative in mice. Necropsy of mice following gas uptake studies at 10,000 ppm
chloroform revealed macroscopic and microscopic evidence of liver toxicity (Kedderis
and Held, unpublished observations), suggesting that acute liver injury rather than
inactivation of cytochrome P450 was responsible for the decreased metabolism of
chloroform observed by Corley et al. (1990) in their gas uptake studies. Therefore, the
postulated suicide inactivation of cytochrome P450 by chloroform was not included in
the PBPK model used in this study to describe THM metabolism.
Since the THMs are all substrates for CYP2E1, mutual competitive inhibition of
metabolism is expected to occur. During mixed exposures to the THMs, each THM
would inhibit the bioactivation of the other THMs that are alternative substrates for
CYP2E1. For competitive alternative substrate inhibition, the inhibition constant for
each substance would be the same as the KM, the substrate concentration giving one-
half the maximal velocity (Vmax) (Segel, 1975). For each THM, the general rate equation
describing metabolism in the presence of 3 competitive inhibitors (Segel, 1975) is given
in Equation 28:
RAM1 = (Vmaxl * CVL1 )/(KM1 * (1 + CVL2/KM2 + CVL3/KM3 + CVL4/KM4) + CVL1) (28)
where RAM1 is the rate of metabolism of THM1, Vmaxl is the maximal rate of
metabolism of THM1, KM/ is the Michaelis constant for each THM/ (/ = 1-4), and CVL/ is
the venous blood concentration of THM/ leaving the liver (/ = 1 -4). CVL represents the
concentration of the substrates presented to the liver for metabolism via hepatic blood
flow, and thus is equivalent to the substrate concentration in the Michaelis-Menten
equation. The rate of metabolism of each THM would be given by a separate equation
51
-------
analogous to Equation 28 for RAM2, RAMS, and RAM4. The values of Vmax and KM
used in the PBPK model are given in Table 60.
Michaelis-Menten kinetic parameters (Vmax, KM) are obtained from experiments
where the concentration of the substrate is varied under conditions where the initial rate
of the enzyme-catalyzed reaction is linear with time and added source of enzyme
(Kedderis, 1997). This experimental approach can be used in vitro with a variety of
systems including isolated cells, tissue homogenates or subcellular fractions, or purified
enzymes. The in vitro data can be extrapolated to the intact organism based on
hepatocellularity or enzyme content, since the initial rates of enzyme-catalyzed
reactions are directly proportional to the enzyme content (Kedderis, 1997). Michaelis-
Menten kinetic parameters can also be estimated from in vivo pharmacokinetic studies
but in general it is difficult to obtain accurate estimates of the kinetic parameters from in
vivo data, particularly for rapidly metabolized compounds like the THMs. The difficulties
arise from the inhomogeneous distribution of substrates in tissues, incomplete
absorption of substrates from the gastrointestinal tract and other barriers, excretion
pathways that compete with metabolism such as exhalation of the substrate, and the
limitation of metabolism by blood flow delivery to the liver. The hepatic blood flow
limitation of metabolism essentially prevents accurate measurement of the initial rate of
metabolism of rapidly metabolized compounds such as the THMs (Kedderis, 1997). In
vivo pharmacokinetic studies with rapidly metabolized chemicals can yield reasonable
estimates of Vmax but generally the estimates of KM values are upper limits. This is
because other processes such as blood flow limit the overall rate of metabolism of the
chemical and essentially mask the more rapid initial rate of metabolism (Kedderis,
1997). The chloroform kinetic parameters in Table 60 were obtained from in vitro
experiments with human liver microsomes from adults and children (U.S. EPA, 2006).
Other PBPK models for chloroform report much higher values for KM. This is explained
by PBPK model optimization routines which communicate only the highest value that
provides an adequate fit to the data. The present work incorporated biochemically-
derived values for KM, extrapolated from in vitro studies conducted with samples of rat
and human liver microsomal preparations (Lipscomb et al., 2005; U.S. EPA, 2006).
Both the previously available (optimized) values and the presently-developed lower
52
-------
(biochemically-derived) values provide adequate fit to the observed data (not shown).
The parameters for the other THMs were estimated from in vivo gas uptake studies in
rats (da Silva et al., 1999) and assumed to be the same for humans. The KM values for
the brominated THMs in Table 60 are likely to be upper limits. Lower values of KM (i.e.,
more rapid initial rates) are likely to fit the gas uptake data just as well as the upper
limits. Human metabolic data for the brominated THMs were not available.
53
-------
4. TRANSFER FILE DEFINITIONS
The exposure model (TEM) and the PBPK model communicate by passing
parameters, media concentrations, physiological, and other necessary data in "transfer
files." These transfer files are text files (ASCII) formatted such that the data are passed
in fixed column format. Each model run, which encompasses the model predictions for
a single household simulation, in this case a family of three, generates a group of files
with related filenames. A description of the file naming convention is given in Table 61.
Each of the files listed in Table 61 contains data in a fixed column format. For each of
the file types listed in Table 61, a description of the file format is presented below:
4.1. BREATHING RATE FILES
The breathing rate files contain a definition of the time-varying breathing rate of
the subject. The file format is a comma separated ASCII file with the following
characteristics:
• Each line contains a record of the subject's breathing rate for a time interval. The
first number is the time that the breathing rate interval started in hours, the is the
breathing rate in L/hour. These numbers are separated by commas.
• A breathing rate defined by a given line is in effect until the start time of the
subsequent line in the file.
• Any line that contains a semi-colon (;) in the first column is ignored (indicates a
comment).
Example: The data for a scenario where the breathing rate is modeled as
540 L/hour from midnight to 7:05 am, 600 L/hour from 7:05 am to 10:40
pm, and 540 L/hour from 10:40 pm to midnight. The resulting breathing
rate data file contained the following data:
; THIS IS THE Breathing Rate FILE IN THE FOLLOWING FORMAT
; TIME(hours),Breathing Rate (L/hour)
0.000000,540.000000
7.08333,600.000000
22.6667,540.000000
54
-------
4.2. DERMAL DATA FILES
The dermal data files contain a definition of the time-varying doses due to water
contact with portions of the subject's skin along with the time and duration of the event.
The file format is a comma separated ASCII file with the following characteristics:
• Each line contains a record of an exposure event with the first number being the
time that the dermal exposure event started in hours, the second number is the
dose in ug, and the third number is the duration of the event in hours. These
numbers are separated by commas.
• Any line that contains a semi-colon (;) in the first column is ignored (indicates a
comment).
Example: The data for a scenario where three dermal exposure events
occur: (1) from 6:00 to 6:04:21 am, the dose is 0.0000125 ug
corresponding to a handwashing activity; (2) from 6:09:00 to 6:16:30 am,
the dose is 0.000085 ug, corresponding to a showering activity; and (3)
7:22:30 to 7:23:41 am, the dose is 0.00000164 ug, corresponding to a
handwashing activity. The resulting breathing rate data file contained the
following data:
; THIS IS THE Dermal Contact FILE IN THE FOLLOWING FORMAT
; TIME(hours),Dose(ug),Duration(hours)
6. 000000,1.24726e-005,0.0725
6.15,8.5e-005,0.125
7.375, 1.64623e-006,0.019722
4.3. INGESTION DATA FILES
The ingestion data files contain a definition of the ingestion doses due to water
consumption. The file format is a comma separated ASCII file with the following
characteristics:
• Each line contains a record of an consumption event with the first number being
the time that the event started in hours, followed by an "I" or a "D" indicating
either direct or indirect consumption, followed by the dose due to the
consumption event in ug, and then followed by the duration of the event in hours.
These numbers are separated by commas.
• Any line that contains a semi-colon (;) in the first column is ignored (indicates a
comment).
55
-------
Example: The data for a scenario where three ingestion exposure events
occur: (1) from 5:39:50 to 5:40:20 am, ingestion route is direct, and the
dose is 0.00009663 ug; (2) from 6:47:50 to 6:50:10 am, ingestion route is
indirect, and the dose is 0.000007059 ug; and (3) 7:41:50 to 7:43:23 am,
ingestion route is indirect, and the dose is 0.0000585795 ug. The
ingestion data file contained the following data:
1 THIS IS THE Consumption Rate FILE IN THE FOLLOWING FORMAT
; TIME(hrs),! or D for Indirect or Direct, Consumption Mass (ug), Duration (hrs)
5. 6639,D, 9.6663e-005,0.00333
6.7972,I,7.05986e-006,0.03888
7.6972,I,5.85795e-005,0.02583
4.4. INHALATION DATA FILES
The inhalation data files contain a definition of the time-varying inhalation
concentrations. The file format is a comma separated ASCII file with the following
characteristics:
• Each line contains a time-varying record of an inhalation concentrations with the
first number being the time that the event started in hours, the second number is
the current inhaled concentration in ug/m3. These numbers are separated by
commas.
• Any line that contains a semi-colon (;) in the first column is ignored (indicates a
comment).
Example: The data for an example 24-hour period are presented as follows:
56
-------
; THIS IS THE Chloroform Concentration FILE IN THE FOLLWING FORMAT
; TIME(hours),Chloroform Concentration(ug/m3)
0.0000,1.8506283E-04
3.2500,1.6588291E-04
4.3375,1.9244952E-04
4.8375,2.1207979E-04
5.8375,2.3687192E-04
7.8375,2.5063853E-04
8.0833,3.3065418E-04
8.1250,3.9391501E-04
8.1708,7.4332117E-04
8.2167,9.3008879E-04
8.4667,2.5130055E-04
9.1708,2.9071942E-04
9.9208,3.2748350E-04
12.9208,3.4774606E-04
13.1708,0
15.8417,8.9983922E-03
15.9208,9.4782892E-03
15.9500,1.1975224E-02
15.9708,1.3346261E-02
16.1333,1.5013896E-02
16.2250,1.6454347E-02
16.3167,8.0999167E-04
16.4167,9.6454433E-04
16.5125,1.0124308E-03
16.5875,1.2882611E-03
16.7500,1.7599108E-03
17.0000,3.0958622E-04
21.9792,2.7701297E-04
24.0000,2.5127959E-04
Note: The concentration reported at 13.1708 as 0 reflects the subject
location of outdoors. The outdoor concentration is assumed to be zero.
57
-------
5. RESULTS
5.1. EXPOSURE TO THE THMs THROUGH WATER USAGE
5.1.1. Water Concentrations. The analyses presented in Section 2.5, Table 46,
provided the basis for selecting representative water concentrations for the THMs.
Considering a variety of factors, including the effect of chlorination, the impact of
organics found at higher levels in surface water systems, and the impact of the warmer
summer period on the formation of DBFs in the drinking water distribution system, we
chose to limit our set of data to investigate these factors. The pertinent concentrations
from our analyses are presented in Table 62.
To maximize the likelihood of identifying chemical interactions, we biased our
investigation to rely on the highest from among several possible measures of an upper
bound for THM concentration data. The ICR database was analyzed to determine
which factors resulted in higher values for the 95th percentile for the distribution of
resulting individual THM compounds. The ICR database evaluated drinking water
treatment and can be characterized in several ways. Table 62 presents some of the
more conventional categorization of the systems. Systems were initially divided into
either systems relying on groundwater or surface water as source water. Systems
relying on groundwater demonstrated lower values for concentrations at the 95th
percentile of the distribution (data not shown). Reliance on chlorine, rather than ozone
as primary disinfectant resulted in appreciably higher levels of THM compounds formed
(ozonation results not shown). With respect to timing of the sampling period, samples
taken between July and September (summer months) demonstrated higher
concentrations of THM compounds than samples taken at other times during the year
(data from other seasons not shown). Further subdivision of systems into categories
(i.e., systems employing ozonation and relying on groundwater as source water) was
not undertaken, in part due to a reluctance to further reduce the number of samples
available for distributional analysis. Finally, the complete database was analyzed.
After a closer inspection of the concentration data, it was clear that there is very
little difference between three of the four subgroups (Surface Water Intake, Systems
using Chlorine, and All Samples). Furthermore, the concentrations reported in the
58
-------
"Systems Sampled between July and September" exhibit the largest concentrations at
the 95th percentile of these subgroups. Since we are examining the effect of these
contaminants, we are limiting our analysis to the July to September subgroup.
5.2. INTERNAL DOSES OF THE THMS FROM WATER USAGE: AN
ILLUSTRATIVE CASE RESULTS
As a method of demonstrating the predicted water-use behavior, the result from
one of the cases will be presented. We chose a case that demonstrated a variety of
water uses to show how the water-use behavior and occupant activities leads to air
concentrations and exposure by each of the three routes (inhalation, ingestion, and
dermal). For this purpose, simulation case 48 was chosen. The demographic
characteristics of the sampled and modeled individuals are presented in Table 63. The
water uses modeled for this simulation are given in Table 64.
Table 64 is broken down by location. The first line indicates that the kitchen
dishwasher was started by the female. The dishwasher was turned on at approximately
10:17 a.m. (10.29 hours past midnight), and ended 74.9 minutes later at approximately
11:31 a.m. (11.527 hours past midnight). This dishwasher event is indicated in the
upper panel of Figures 7-13 which demonstrate room air concentrations and personal
air concentrations of the four THMs. Personal air concentration profiles differ from room
air profiles due to the influence of the human activity pattern - humans move from room
to room during the course of the day. A series of three master bathroom events for the
female occurred in series beginning at approximately 9:30 a.m. These events may
have been "brushing teeth", "washing hands" and "drying hands" for example.
The water uses and the resultant air concentrations predicted in each of the
modeled compartments are displayed in Figures 7 through 10 for chloroform, BDCM,
DBCM and bromoform, respectively. In addition, the personal concentration in the
breathing zone of each occupant, determined by the model by considering the predicted
air concentrations along with the location of the occupant throughout the day, are given
in Figures 11 through 13 for the adult male, the adult female and the child, respectively.
The personal air concentrations provided in Figures 11, 12, and 13 are a union between
the predicted air concentrations given in Figures 7, 8, 9 and 10 and the location of each
occupant in the home. In addition, the personal air concentration is assumed to be zero
if the occupant is outside of the home. The movement from one location in the house
59
-------
with a lower air concentration to another location with a higher air concentration leads to
the sharp rise in personal air concentration shown in Figures 11, 12, and 13.
The resultant concentrations internal doses for this case are shown to
demonstrate the predictions. As discussed below, the concentration of metabolites
produced in the liver over 24 hours (CM24) was used as the internal dosimeter for THM
bioactivation. The CM24 is given in Figures 14-16 for the male, female, and child,
respectively. Also, as discussed below, the AUC for the parent THMs provides the most
appropriate dosimeter for the exposure of the non-metabolizing target tissues kidneys
and genitals to each THM. Figures 17-19 show the predicted AUC for the kidneys for
each of the three subjects, and Figures 20-22 show the predicted AUC for the genitals for
each of the three subjects.
5.3. INTERNAL DOSES OF THE THMS FROM WATER USAGE: POPULATION-
BASED RESULTS
The simulation predictions (results) for absorbed dose are analyzed for each
chemical as a function of route (dermal, ingestion, and inhalation) and presented in the
following sections. For each chemical, a table containing the absorbed dose is
presented as a function of route, population group, and percentile of the population. In
addition, the cumulative distribution function is plotted along with histograms.
Tables 65-68 present the predicted distribution of total absorbed dose (mg) for
the four THM compounds, as well as the predicted dose via route. In order to compare
doses per kg body mass, the total absorbed dose at the 50th percentile for each THM
was divided by body weight used for PBPK modeling (given in Table 58). For
chloroform, the doses at the 50th percentile were 0.0044, 0.0052 and 0.0078 mg/kg for
the adult female, adult male and child, respectively. For BDCM, the doses at the 50th
percentile were 0.0014, 0.0017 and 0.0026 mg/kg, respectively. For DBCM, the doses
at the 50th percentile were 0.00094, 0.0012 and 0.0016 mg/kg, respectively. For
bromoform, the doses at the 50th percentile were 0.00029, 0.00035 and 0.00049 mg/kg,
respectively.
5.3.1. Population Results for Chloroform. Table 65 presents the predicted absorbed
dose of chloroform from the analysis of the dermal, ingestion and inhalation exposure
routes for each of the population groups: female age 15-45, male age 15-45, and child
age 6. Figures 23, 24 and 25 present the histograms for absorbed dermal dose,
60
-------
inhalation dose, and ingestion dose, respectively, for the female, male and child
populations. Figure 26 presents the total absorbed chloroform dose.
5.3.2. Population Results for BDCM. Table 66 presents the predicted absorbed dose
of BDCM from the analysis of the dermal, ingestion and inhalation exposure routes for
each of the population groups: female age 15-45, male age 15-45, and child age 6.
Figures 27, 28 and 29 present the histograms for absorbed dermal dose, inhalation
dose, and ingestion dose, respectively, for the female, male and child populations.
Figure 30 presents the total absorbed BDCM dose.
5.3.3. Population Results for DBCM. Table 67 presents the predicted absorbed dose
of DBCM from the analysis of the dermal, ingestion and inhalation exposure routes for
each of the population groups: female age 15-45, male age 15-45, and child age 6.
Figures 31, 32 and 33 present the histograms for absorbed dermal dose and inhalation
dose, respectively, for the female, male and child populations. Figure 34 presents the
total absorbed DBCM dose.
5.3.4. Population Results for Bromoform. Table 68 presents the predicted absorbed
dose of bromoform from the analysis of the dermal, ingestion and inhalation exposure
routes for each of the population groups: female age 15-45, male age 15-45, and child
age 6. Figures 35, 36 and 37 present the histograms for absorbed dermal dose,
inhalation dose, and ingestion dose, respectively, for the female, male and child
populations. Figure 38 presents the total absorbed bromoform dose.
5.4. METABOLIC INTERACTIONS BETWEEN THE THMs
Since the THMs are all substrates for the same isoform of cytochrome P450,
CYP2E1 (Guengerich et al., 1991; Raucy et al., 1993), the THMs are expected to be
alternative substrate competitive inhibitors upon coexposure. The inhibition constants
for each THM are the same as their KM values (Segel, 1975). The extent of inhibition
observed during an exposure event will depend upon the exposure concentrations of
the THMs and the capacity of the metabolizing enzyme CYP2E1.
The metabolic interactions between the THMs were investigated using NHAPS
water-use activity pattern 728 for the male subject, which was the 86th percentile for
both chloroform and DBCM exposures. Data sets of the four THM concentrations
corresponding to the 95th percentile for each THM were determined from the complete
61
-------
data set for water systems sampled from July to September 1997-1998 and used to
investigate the interactions between the THMs. Use of near-maximally expected
concentrations was accomplished to maximize the likelihood of detecting metabolic
interactions. Interactions observed under these circumstances would indicate a need to
closely examine the possibility of interactions at lower concentrations/doses. Lack of
interaction at these (high) exposures would indicate that interactions at lower
concentrations/doses would not be anticipated. The concentration data sets are shown
in Table 69. Concentrations corresponding to the 95th percentile for each THM were
chosen to represent those concentrations having the maximum potential for interactions
of all the THM concentration data.
The present report surpasses the level of detail previously developed (U.S. EPA,
2003), in that metabolic interactions have been examined. Metabolic interactions
between the THMs were investigated by simulating the exposure to each THM
individually and comparing the results to simulations of exposure to all the THMs
together (Table 69) for each water concentration scenario sampled July-September
1997 and 1998 (see Table 46). Since metabolism represents bioactivation for each of
the THMs, the concentration of metabolites produced in the liver over 24 hours (CM24)
was used as the internal dosimeter for THM bioactivation. Under these exposure
conditions, inhibition of the metabolic activation of the THMs was not observed with the
95th percentile water concentration scenarios for chloroform, BDCM, or bromoform. A
very slight inhibition of DBCM bioactivation (0.0001%) was observed with the DBCM
95th percentile water concentration scenario (Table 70). The inhibition of DBCM
bioactivation was due to the combination of chloroform and BDCM (Table 70), as each
THM alone did not produce any inhibition of metabolism under these exposure
conditions.
One reason that metabolic interactions were not evident between the four THMs
may be that the metabolic capacity of CYP2E1 was large enough to metabolize all of
the THMs the subject was exposed to from the water-use scenarios. This hypothesis
was tested by lowering the capacity of CYP2E1 in the PBPK model by decreasing
Vmaxc, since the maximal velocity of an enzyme-catalyzed reaction is directly
proportional to the amount of enzyme present and is an indication of the amount of
62
-------
active enzyme (Segel, 1975). Table 71 shows that as CYP2E1 capacity (Vmaxc)
decreased, more inhibition of THM metabolism became evident. These results are
consistent with the interpretation that the lack of metabolic interactions between the
THMs following exposure from water usage was due to the large capacity of the
metabolizing enzyme, CYP2E1. However, the extent of inhibition of THM metabolism
was only approximately 0.2% at most after lowering Vmaxc by a factor of one million
(Table 71). At such a low enzyme capacity, the exposure of the liver to THM
metabolites (indicated by CM24) became vanishingly small. Taken together, these
results indicate that inhibitory interactions between the four CYP2E1 substrate THMs
would not be expected to be significant under the low level, intermittent exposures
encountered through water use in the home setting.
5.5. INFLUENCE OF WATER-USE PATTERNS ON INTERNAL DOSIMETRY FOR
THE THMs
Just as external exposure to the THMs was dependent upon water use patterns,
the internal dosimetry of the THMs was also dependent upon water use patterns. In the
present PBPK model for the THMs, metabolism was assumed to only take place in the
liver and not in the extrahepatic target organs. While the vast majority of THM
metabolism does take place in the liver such that the overall pharmacokinetics of the
THMs can be accurately described assuming that metabolism only takes place in the
liver (da Silva et al., 1999), local metabolism of the THMs in the extrahepatic target
organs kidney and genitals may occur. Renal metabolism has been observed with
chloroform (Corley et al., 1990; Constan et al., 1999) and BDCM (Lily et al., 1997,
1998). Extrahepatic metabolism was not described in the present PBPK model for the
THMs because the enzymes involved and the kinetics of these potential metabolic
pathways in these organs have not been quantified. Because of this limitation, the AUC
for the parent THMs provides the next most reliable dosimeter for the exposure of the
extrahepatic target tissues kidneys and genitals to each THM. The AUC (mg-hr)
represents the integrated exposure of the organ to the parent THM over 24 hours. The
distributions of the AUCs for each of the four THMs in the male, female and child
subjects are shown in Figures 39-42 for the kidney and in Figures 43-46 for the genitals.
The most appropriate dosimeter for exposure of the metabolizing target organ
liver is CM24, the concentration of metabolites produced in the liver over 24 hours. CM24
63
-------
represents the integrated exposure of the liver to the reactive metabolites of THM
oxidation by CYP2E1 over 24 hours. The reactive metabolites formed from THM
oxidation (phosgene and its brominated analogs) are transient and do not accumulate in
the liver. Thus, CM24 represents exposure of the liver to the metabolites, not the
concentration of metabolites in the liver at any given time. The distributions of the CM24
values for each of the four THMs in the livers of the male, female and child subjects are
shown in Figures 47-50. The distributions of CM24 in the liver for the THMs were similar
to the distributions of the AUCs for the parent THMs in extrahepatic tissues.
5.6. LIMITED SENSITIVITY ANALYSIS OF THE PBPK MODEL
The sensitivity of the PBPK model to changes in selected parameter values was
determined from simulations of exposure of an adult male to the THMs from water
usage described by activity scenario 2, which involves household activities leading to
inhalation, dermal and oral ingestion exposures to the THMs. Sensitivity was estimated
by varying the values of QCC (cardiac output, L/hr/kg) and QLC (liver blood flow,
fraction of cardiac output) on the concentration of metabolites in the liver (CAM, mg/L)
and the area under the curve for the liver concentration of the parent THM (AUCL,
mg-hr). Since TEM specifies the values of QPC (alveolar ventilation, L/hr/kg) used in
the PBPK model, the effects of systematically varying QPC could not be readily
determined at this time, but those effects are anticipated to be similar to the effects of
varying QCC since these two physiological parameters are linked. Additionally,
sensitivity to changes in the metabolic parameters VmaxC (mg/hr/kg) and KM (mg/L) for
chloroform and bromoform were determined.
The CAM for both chloroform and bromoform was not sensitive to changes in
VmaxC in the range of 2 to 20 mg/hr/kg (Figures 51 and 52). Similar results are
expected for bromodichloromethane and dibromochloromethane. These results are
consistent with the interpretation that the rate of THM metabolism in the liver is limited
by the rate of hepatic blood flow (Figures 53-56). Figures 51 and 52 clearly
demonstrate that VmaxC values greater than 2 mg/hr/kg do not significantly increase
CAM. Thus, interindividual variability in VmaxC (a reflection of variability in enzyme
content) at values greater than approximately 2 mg/hr/kg would not translate to
interindividual variability in THM metabolism or CAM (a risk-related endpoint), as has
64
-------
been previously been demonstrated in PBPK simulations for chloroform (U.S. EPA,
2004). In contrast, the AUCL for parent chloroform (Figure 57) and bromoform (Figure
58) decreased with increasing VmaxC. Similar results are expected for
bromodichloromethane and dibromochloromethane. As expected, increasing
metabolism strongly affected the concentration of the parent THM in the liver.
Increasing the value of KM, the Michaelis constant for metabolism that
represents the THM concentration that yields one-half the Vmax, decreased the values
of CAM and increased the values of AUCL for chloroform and bromoform (Figures
59-62). Similar results are expected for bromodichloromethane and
dibromochloromethane. The effect of increasing KM is to essentially decrease the initial
rate of metabolism V/K. The sensitivity of CAM and AUCL to KM also reflects the
relatively low levels of THM exposure in the water usage exposure scenario, such that
the THM tissue concentrations are well below metabolic saturation.
Figures 63-66 show the effects of varying QCC on the CAM for each of the four
THMs. Figures 67-70 show the effects of varying QCC on the AUCL for each of the four
THMs. In general, as QCC increased, the values of CAM and AUCL for each of the four
THMs increased. This is because increased blood flow increased the amount of each
THM in the liver. This is further illustrated by the effects of varying liver blood flow, QLC
(Figures 53-56, 71-74). As liver blood flow increases, more of each THM is brought to
the liver, increasing both CAM and AUCL.
The CAM for both chloroform and bromoform was not sensitive to changes in
VmaxC in the range of 2 to 20 mg/hr/kg (Figures 51 and 52). Similar results are
expected for bromodichloromethane and dibromochloromethane. These results are
consistent with the interpretation that the rate of THM metabolism in the liver is limited
by the rate of hepatic blood flow (Figures 53-56). Figures 51 and 52 clearly
demonstrate that VmaxC values greater than 2 mg/hr/kg do not significantly increase
CAM. Thus, interindividual variability in VmaxC (a reflection of variability in enzyme
content) at values greater than approximately 2 mg/hr/kg would not translate to
interindividual variability in THM metabolism or CAM (a risk-related endpoint), as has
been previously been demonstrated in PBPK simulations for chloroform (EPA, 2004).
In contrast, the AUCL for parent chloroform (Figure 57) and bromoform (Figure 58)
65
-------
decreased with increasing VmaxC. Similar results are expected for
bromodichloromethane and dibromochloromethane. As expected, increasing
metabolism strongly affected the concentration of the parent THM in the liver.
Increasing the value of KM, the Michaelis constant for metabolism that
represents the THM concentration that yields one-half the Vmax, decreased the values
of CAM and increased the values of AUCL for chloroform and bromoform (Figures
59-62). Similar results are expected for bromodichloromethane and
dibromochloromethane. The effect of increasing KM is to essentially decrease the initial
rate of metabolism V/K. The sensitivity of CAM and AUCL to KM also reflects the
relatively low levels of THM exposure in the water usage exposure scenario, such that
the THM tissue concentrations are well below metabolic saturation.
66
-------
6. DISCUSSION
This modeling study allows examination of several factors central to the manner
in which people are exposed to the four THMs and how these chemicals are regulated.
In this study, specific population groups were modeled to compare the effects of
chemical-specific processes (including mass-transfer effects and their effects on route-
specific exposures and uptake); the effects of physiological processes; and the effects
of activity patterns. Because the entire population was subjected to the same water
concentrations, the effects of differing activity patterns across the same population
group provides insight into the role activities play in the eventual dose.
The four THMs represent a moderately broad range of volatilities, as evident in
the range of values for the dimensionless Henry's Law constant (0.15 to 0.022) and the
vapor pressures (160 mm Hg to 5.6 mm Hg) between chloroform and bromoform (see
Table 4). These chemicals are found in the drinking water supply at considerably
different concentrations. Generally, chloroform is found at the highest concentrations,
while the other THMs are generally found at progressively lower concentrations as the
number of chlorine atoms decreases and the number of bromine atoms increases. In
addition to volatility, other chemical properties, such as partition and permeability
coefficients, also impact the magnitude of route-specific exposure and uptake, as well
as the removal rates of each of the THMs.
The variability in the predicted absorbed dose, given in Figures 23-34 and Tables
65-68 is largely due to differences in activity patterns across a given population group.
Other effects, including variations in house size and interzonal airflows, have lesser
effects on the variation in absorbed dose across the population. It is evident from the
results in the figures that the variation across the population is largely attributable to the
inhalation route, which has a far larger range of exposures and doses for all four THMs.
Wilkes et al. (1996) closely examined the effect of activity patterns on the inhalation
route in a modeling study. In that study, Wilkes et al. defined the potential inhalation
dose (PID) as the amount of contaminant entering the lungs and available for uptake, by
modeling a two-person household with a water supply contaminated with
trichloroethylene (TCE). The parameters other than activity patterns were held constant
67
-------
across each population group, including the configuration of the household, the size of
rooms, the interzonal airflows, the whole-house air exchange rate, the appliance
characteristics, and the breathing rate. Wilkes et al. found very strong correlations
between a number of activities and the resultant PID. The strongest correlations were
found with the following factors (in order of importance):
• Shower duration
• Time spent in the bathroom
• The fraction of time spent in the home multiplied by the total volume of water use
in the home
• Bath duration
These correlations were found to be similar for households with one adult and
households with two adults. Furthermore, the Wilkes et al. study found that the effect of
two people sharing a household resulted in approximately a 20% increase in PID over a
single person in the same household.
The Wilkes et al. study did not examine the effect of building-related parameters,
such as the size of the house and the air-exchange rate, on the inhalation dose.
However, some of these factors were also examined in the Wilkes et al. (2002) study
(U.S. EPA, 2003). In that study, a sensitivity analysis was conducted to examine the
sensitivity of the model to a variety of parameters, including behavior and activity factors
(such as shower and bath durations) as well as a number of building related factors
(such as air exchange rate and room volumes). For chloroform, the model was found to
be relatively sensitive to these factors, in addition to confirming the earlier study's
finding that the inhalation dose was very sensitive to several behavioral factors.
The exposure as a function of its route (i.e., ingestion, dermal, and inhalation)
can be evaluated for each THM based on the predicted uptake. Since the four THMs
are all present in the water supply, a comparison of route-specific contributions to the
total absorbed dose will provide insight into the importance of volatility for each route.
Since much of the Environmental Protection Agency's regulatory approach for setting
maximum contaminant levels (MCLs) relies on using the ingestion route as an indicator
68
-------
of exposure and dose, the relationship between the dose due to the ingestion route and
overall dose warrants examination.
The total absorbed doses for chloroform and bromoform are shown in Figures 75
and 76 as functions of the percentile of the population. From these figures, it is clear
that the highly exposed portions of the population (e.g., the population exposed at
greater than the 90th percentile) are expected to be exposed to considerably more than
the mean absorbed dose. For these simulation results given in Tables 65-68 and
Figures 75 and 76, the absorbed dose for the 90th percentile case for all four THMs was
typically 4 to 8 times the absorbed dose for the 50th percentile case for the adult male
and female population groups. The absorbed dose for the 99th percentile case for the
four THMs was typically 30 to 40 times the absorbed dose for the 50th percentile case.
Similarly for the child population group, the absorbed dose for the 90th percentile case
for all four THMs was typically approximately 4 times the absorbed dose for the 50th
percentile case, while the absorbed dose for the 99th percentile case for the four THMs
was typically approximately 10 times the absorbed dose for the 50th percentile case.
Examining the same data set for route-specific contributions yields Figures
77-82. In these figures, the contribution of each route is displayed in a cumulative
fashion, such that the top line is the total absorbed dose, and each shaded area
represents the respective contribution of the route it represents. This analysis indicates
that the inhalation route contributes an increasing percentage as the total dose the
increases. For chloroform in all population groups above the 20th percentile of total
absorbed dose, inhalation is the dominant route of exposure. Above the 50th percentile
in total dose, the inhalation route contributes more than 70%, and above the 90th
percentile, the inhalation route contributes more than 90%. It is clear looking at Figures
77-79 that the contribution of the oral route has very little or no systematic increase with
the increasing total dose, while nearly all the increase is due to the inhalation route.
Figure 83 compares the chloroform and bromoform contributions by the oral and
inhalation routes for the female population group. From this comparison, it is apparent
that the bromoform inhalation route fractional contribution is slightly smaller, but
generally similar to the inhalation route contribution by chloroform. Bromoform is a
borderline semi-volatile, based on the generally accepted definition for VOCs (boiling
69
-------
point less than 150°C and a vapor pressure of less than 0.1 mm Hg at standard
temperature and pressure, refer to Table 4 for the properties of bromoform). The lower
volatility results in a somewhat lower fraction by the inhalation route, however, the
inhalation route still dominates exposure.
Exposure by the ingestion route is typically used as a basis for regulating
drinking water contaminants. Figure 84 shows the amount of water that would need to
be consumed if all of the exposure was from the ingestion route. This analysis assumes
that the consumed water is at the same concentration as the tap water, and further
assumes 100% uptake in the digestive system. This analysis suggests that using
consumption as a predictor for overall dose is problematic and not acceptable for
volatile chemicals. For volatile chemicals, consumption is not a significant portion of the
overall dose nor can consumption can be used as a predictor of the overall dose. If the
MCL is set based on the amount of tap water consumed, the vast majority of the dose
will not be considered, leaving much of the population exposed to higher than expected
doses.
70
-------
7. CONCLUSIONS
The four THMs have a reasonably wide range of volatility, with vapor pressures
ranging from 5.6 mm Hg for bromoform to 160 mm Hg for chloroform. As volatility
increases, the relative importance of the inhalation route increases. This analysis
indicates that exposure to waterborne THMs varies widely across a population,
influenced by a variety of factors. These factors include THM concentrations in the
water supply, building characteristics that impact ventilation and therefore airborne
concentrations, water-use activities that lead to release of THMs into the air, activities
that bring the subject into the vicinity of high airborne concentrations, dermal contact,
consumption, and physiological factors that affect uptake such as breathing rate.
Excluding the water concentration, the factors that have the largest impact across the
population are the activity patterns that impact exposure to the occupants, including
activities that use water and activities that bring the subject into the vicinity of the water
uses. Furthermore, as discussed above, pairing two people in the same household
results in approximately a 20% increase in inhalation dose over a single person
occupying the same house, due to the water-use activities of other occupants. For all
four THMs, the inhalation route plays an extremely large role in the total dose,
especially for the highly exposed portions of the population. Since water-use behavior
and other activity pattern exposure factors can vary substantially across the population,
the dose by the inhalation route reflects this variability for volatile chemicals, and
subsequently, the inhalation route is largely responsible for higher exposures in the
population.
It is evident, based on the route-specific analysis and on the analysis of effective
consumption presented in Figure 84, that the exposure and dose by the ingestion route
is a poor proxy for the total exposure, and that the estimated dose by the ingestion route
cannot be used in an effective manner for estimating total dose. The effective
consumption for chloroform, a fairly volatile chemical, is very different than the effective
consumption for bromoform, a borderline volatile, semi-volatile chemical. In the case of
bromoform, 2 liters per day may properly represent the exposure and dose to over 75%
of the population, whereas for chloroform, an effective consumption of more than 15
71
-------
liters per day is likely warranted. It is evident from this analysis, that the route specific
dose is very dependent of chemical properties and human activity patterns. Therefore,
all routes of exposure must be considered when assessing exposure and dose to water-
borne contaminants.
Just as external exposure to the THMs was dependent upon water use patterns,
the internal dosimetry of the THMs was also dependent upon water use patterns. Since
the THMs are all substrates for the same isoform of cytochrome P450, CYP2E1
(Guengerich et al., 1991; Raucy et al., 1993), the THMs are expected to be alternative
substrate competitive inhibitors upon coexposure. Under the exposure conditions to the
THMs from water use patterns (Table 69), significant inhibition of the metabolic
activation of the THMs was not observed with the 95th percentile water concentration
scenarios for chloroform, BDCM, DBCM or bromoform. The reason that metabolic
interactions were not evident between the four THMs was that the metabolic capacity of
CYP2E1 was large enough to metabolize all of the THMs the subject was exposed to
from the water-use scenarios. This conclusion was verified by lowering the amount of
CYP2E1 in the simulations (Table 71). Decreasing Vmaxc one million times showed
inhibition of THM metabolism by approximately 0.2% (Table 71). These results indicate
that inhibitory interactions between the four CYP2E1 substrate THMs would not be
expected to be significant under the low level, intermittent exposures encountered
through water use in the home setting.
72
-------
8. MODEL ASSUMPTIONS AND DATA QUALITY
This study is the implementation of a variety of stochastic and deterministic
modeling techniques. The data used in calculations, methods and models used to
derive quantitative measures, including those of internal exposure, tissue dosimetry,
and risk were taken from publications and other sources subjected to peer review where
possible. These publications include peer reviewed journals and other open literature.
The sources of all data contained within this report have been documented by reference
or footnote describing the source of the data. In addition, a discussion of shortcomings
of data used in this study is included in the text of this report in the section where the
data are introduced.
8.1. DATA QUALITY
Many diverse types of data are used in this study, including behavioral data,
physical data, chemical data, and physiological data. These data are taken from a
variety of sources including databases, peer-reviewed publications, and estimation
techniques. In addition, numerous models are used to develop the exposure, dose and
tissue concentrations, including fate and transport models, mass-transfer models,
models to represent behavior, uptake and pharmacokinetic models. A general
summary of the models and data utilized in this study are presented in Tables 72-75.
The data fall into seven general categories, as described in Table 72. The sources of
the major data utilized in this study are categorized and described in Table 73. The
models and model algorithms utilized in this study are categorized and described in
Tables 74 and 75.
8.2. ACTIVITY PATTERN DATABASE OVERVIEW
This report uses data that was analyzed by Wilkes et al. (2004) for water-use
behavioral characteristics. Wilkes et al. analyzed four primary data sources: (1)
NHAPS, (2) REUWS, (3) REGS, and (4) CSFII. The survey conducted to compile
NHAPS (Tsang and Klepeis, 1996) was designed to gather exposure-related
information, and as such, quantifying duration and frequency of appliance use was a
goal of the survey. REUWS (Mayer et al., 1998) and REGS (U.S. DOE, 1995) were
gathered for other purposes, but also contain useful information. REUWS was
73
-------
conducted to better understand how much water is used by the various household
appliances and issues related to water conservation. REGS was conducted with a
primary focus on energy consumption. CSFII (U.S. EPA, 2000b) is a study of food
intake, which is analyzed for tap-water consumption.
The National Human Activity Pattern Survey (NHAPS) database contains the
results from a two-year, nationwide, activity pattern survey. The NHAPS study was
commissioned by the EPA National Exposure Research Laboratory. During the period
from October 1992 through September 1994, 9386 persons residing in the 48
contiguous United States were interviewed over the phone. The households were
chosen using a telephone random-digit dial method such that the database would
statistically represent the U.S. population. The interview was composed of two parts,
which will hereafter be referred to as the "Diary" and the "Main Questionnaire." NHAPS
data was analyzed by Wilkes et al. (2004) for a variety of household water uses. In
addition, the database was sampled in this study for activity pattern (location and
activity), as described in Section 2.2.
The Residential End Use Water Study (REUWS) database contains water-use
data obtained from 1188 volunteer households throughout North America. The REUWS
study was funded by the American Water Works Association Research Foundation.
During the period from May 1996 through March 1998, approximately 100 single-family
detached homes in each of 12 different municipalities (located in California, Colorado,
Oregon, Washington, Florida, Arizona, and Ontario) were outfitted with a data-logging
device (Meter Master 100 EL, manufactured by F.S. Brainard and Co.1) attached to their
household water meter (on only magnetic-driven water meters). The data logger
recorded the water quantities at 10-second intervals for a total of 4 weeks (2 in warm
weather and 2 in cool weather) at each household. Following the study, the data were
retrieved and analyzed by a flow-trace analysis software program, called Trace
Wizard®, developed by Aquacraft Engineering, Inc.,2 (DeOreo et al., 1996), which
disaggregated the total water volumes into individual end uses (i.e., toilet, shower,
faucet, dishwasher, clothes washer, etc.) (Mayer et al., 1998). In addition to identifying
1 F.S. Brainard and Company, P.O. Box 366, Burlington, NH 08016.
2 Aquacraft Engineering, Inc., 2709 Pine Street, Boulder, CO 80304.
74
-------
the type of water use (e.g., shower, faucet, toilet), Trace Wizard® identified the event
durations, volumes, peak flows, and mode measurements for each water-using event.
REUWS data was analyzed by Wilkes et al. (2004) for a variety of household water
uses.
The Residential Energy Consumption Survey is a nationwide survey conducted
in 1997 to obtain household energy-use information. The resultant REGS database
contains energy-usage characteristics of 5900 residential housing units. The
information was acquired through on-site personal interviews with residents; telephone
interviews with rental agents of units where energy use was included in the rent; and
mail questionnaires to energy suppliers to the units. The database contains information
on physical characteristics of the housing units, demographic information of the
residents, heating and cooling appliances used, clothes washer and dishwasher-use
frequency information, fuel types, and energy consumption. The REGS database was
analyzed by Wilkes et al. (2004) to quantify estimates on household clothes-washer and
dishwasher usage.
The 1994-96 USDA's Continuing Survey of Food Intake by Individuals (CSFII) is
the most recent and comprehensive consumption database available. CSFII was
conducted over the 3-year period between January 1994 and January 1997. A
nationally representative total of 15,303 persons in the United States were interviewed
on two non-consecutive days with questions about what drinks and foods they
consumed in the previous 24 hours. The dietary recall information was collected by an
interviewer who came to the participants' homes and provided instructions and standard
measuring cups and spoons to assist in recalling consumption quantities. The U.S.
EPA (2000b) report, "Estimated Per Capita Water Ingestion in the United States,"
explains the details of the study and presents the results. The CSFII data were
analyzed by Wilkes et al. (2004) for purposes of quantifying estimates of per capita
water ingestion for both direct water (plain water consumed as a beverage) and indirect
water (water used to prepare foods and beverages).
These data sources had a number of shortcomings. For NHAPS, the frequency
was calculated in one of two ways, depending upon how the data were gathered. Some
of the frequency data was reported in the form of a range of values, while others gave a
75
-------
specific number of events over a given time period, and in some cases, the frequency
range is truncated. For example, the clothes-washer frequency data was provided as
daily, 3-5 times per week, 1-2 times per week, or less than once per week, and
showers, where the frequencies of 10 and greater reported as "greater than 10." In the
Wilkes et al. (2004) analysis for binned data, the midpoint of the range was assumed in
the calculation. For truncated data, the calculation for overall frequency assumed the
first number in the truncated range (i.e., 11 was assumed for the truncated range
"greater than 10").
Though REUWS offers a tremendous amount of useful information, the database
is not a statistically representative sample of our nation's population (as is NHAPS).
The sampled households were located within only six U.S. states (five of which are in
the western U.S.) and one Canadian province, and the participants were all volunteers
who may not be representative of the entire population. The REUWS database
presents a potentially significant data source toward the understanding of household
water-use behavior. However, the quality of the data relies heavily on the
disaggregation algorithms employed by the Trace Wizard® software. In a recent small,
evaluation study of Trace Wizard® (see Wilkes et al., 2004, Appendix A), flaws in Trace
Wizard's® analysis techniques were uncovered. Though fairly acceptable in classifying
single, non-overlapping water-uses, the software quite often misclassified water-uses
when two or more water uses overlapped. In the evaluation study, over 83% of single
water uses were classified correctly, and less than 25% of multiple, overlapping water-
uses were classified correctly.
8.3. OTHER ASSUMPTIONS
There are a number of other issues that are likely to be important, but they are
poorly understood. Below is a partial list of issues that were not addressed in this study:
1. Water Heater. The water heater, as a storage device that maintains a relatively
high temperature, represents an opportunity for further reactions. It is possible
that the THM content of the water leaving the water-heater is substantially higher
than the water entering the water heater
2. Dishwasher. The dishwasher, like the water heater, contains water that is heated
and provided an opportunity for further reactions, thereby generating additional
THMs.
76
-------
3. Furthermore, chlorinated detergent is often added and is in the presences of
waste food matter, providing additional reactants.
4. Activity Pattern Databases. The NHAPS activity pattern data base does not
include specific demarcation in its time-activity records. As discussed earlier in
this report, this shortcoming is addressed by appropriately simulating water-uses
consistent with the population groups' characteristics in appropriate places in the
time-activity pattern (see Section 2.2). Although there is every expectation that
this generates realistic exposure scenarios, it is not possible to be sure. In
addition, the activity patterns are independent, and therefore when a family is
simulated, the correlation in activities that likely exists in actual families, is most
likely not captured. Wilkes et al. (1996) discusses this issue.
5. The house volumes, room volumes, air exchange rates, interzonal airflows, and
other building-related parameters are developed from a variety of sources, as
discussed in Section 2.4. These represent general population characteristics;
however individual nuances and peculiarities, such as whole-house fans,
opening and closing of windows and doors, etc., are unlikely to have been
captured.
6. Dermal Contact. Dermal contact is assumed to be a constant fraction of the skin
for each activity. The following factors are not considered:
• variations in amount of skin in contact with the water throughout a given
activity,
• the impact of water and air temperature on dermal uptake,
• the dermal uptake rate is averaged over all areas of the body and there is no
difference in the location of dermal contact.
7. Ingestion. Ingestion is randomly distributed throughout the day, as described in
Section 2.3. The manner in which actual consumption behavior is distributed
throughout the day is not well quantified.
77
-------
9. REFERENCES
Aher, A., A. Chouthai, L. Chandrasekar, W. Corpening, L. Russ and B. Vijapur. 1991.
East Bay Municipal Utility District Water Conservation Study. Prepared for East Bay
Municipal Utility District, Oakland, CA: October 1991. Stevens Institute of Technology.
Report No. R219.
Allis, J.W. and G. Zhao. 2002. Quantitative evaluation of bromodichloromethane
metabolism by recombinant rat and human cytochrome P450s. Chem. Biol. Interact.
140(2): 137-153.
Ammann, P., C.L. Laethem and G.L. Kedderis. 1998. Chloroform-induced cytolethality
in freshly isolated male B6C3F1 mouse and F-344 rat hepatocytes. Toxicol. Appl.
Pharmacol. 149(2):217-225.
ASHRAE (American Society of Heating, Refrigerating, and Air-Conditioning Engineers).
1992. Systems and Equipment Handbook. American Society of Heating, Refrigerating,
and Air-Conditioning Engineers, Atlanta, GA.
Axley, J.W. 1989. Multi-zone dispersal analysis by element assembly. Building
Environ. 24(2): 113-130.
Batterman, S., A. Huang, S. Wang and L. Zhang. 2000. Reduction of ingestion
exposure to trihalomethanes due to volatilization. Environ. Sci. Technol. 34:4418-4424.
Batterman, S., L. Zhang, S. Wang and A. Franzblau. 2002. Partition coefficients for the
trihalomethanes among blood, urine, water, milk and air. Sci. Total Environ.
284(1-3):237-247.
Canadian Ministry of National Health and Welfare. 1981. Tapwater Consumption in
Canada. Public Affairs Directorate, Department of National Health and Welfare,
Ottawa, Canada. Document No. 82-EHD-80.
Cantor, K.P., R. Hoover, P. Hartge et al. 1987. Bladder cancer, drinking water source,
and tapwater consumption: A case-control study. J . Natl. Cancer Inst.
79(6): 1269-1279.
Cleek, R.L. and A.L. Bunge. 1993. A new method for estimating dermal absorption
from chemical exposure. 1. General approach. Pharm. Res. 10:497-506.
Clewell, H.J. and M.E. Andersen. 1994. Physiologically based pharmacokinetic
modeling and bioactivation of xenobiotics. Toxicol. Ind. Health. 10(1-2): 1-24.
78
-------
Constan, A.A., C.S. Sprankle, J.M. Peters et al. 1999. Metabolism of chloroform by
cytochrome P450 2E1 is required for induction of toxicity in the liver, kidney, and nose
of male mice. Toxicol. Appl. Pharmacol. 160(2): 120-126.
Corley, R.A., A.L. Mendrala, F.A. Smith et al. 1990. Development of a physiologically
based pharmacokinetic model for chloroform. Toxicol Appl Pharmacol.
103(3):512-527.
da Silva, M.L., G. Charest-Tardif, K. Krishnan and R. Tardif. 1999. Influence of oral
administration of a quaternary mixture of trihalomethanes on their blood kinetics in the
rat. Toxicol. Lett. 106:49-57.
DeOreo, W.B., J.P. Heaney and P.W. Mayer. 1996. Flow trace analysis to assess
water use. JAWWA. 88(1):79-90.
Ershow, A.G. and K.P. Cantor. 1989. Total water and tapwater intake in the United
States: Population-based estimates of quantities and sources. Life Sciences Research
Office, Federation of American Societies for Experimental Biology.
Ershow, A.G., L.M. Brown and K.P. Cantor. 1989. Intake of tapwater and total water by
pregnant and lactating women. Am. J. Pub. Health. 81:328-334.
Gangolli, S., Ed. 1999. The Dictionary of Substances and Their Effects, 2nd ed.
Volumes 1-7. Royal Society of Chemistry, Cambridge, UK.
Gargas, M.L., M.E. Andersen and H.J. Clewell III. 1986. A physiologically based
simulation approach for determining metabolic constants from gas uptake data. Toxicol.
Appl. Pharmacol. 86(3):341-352.
Gargas, M.L., H.J. Clewell III and M.E. Andersen. 1990. Gas uptake inhalation
techniques and the rates of metabolism of chloromethanes, chloroethanes, and
chloroethylenes in the rat. Inhal. Toxicol. 2(3):295-319.
Giardino, N.J. and J.B. Andelman. 1996. Characterization of the emissions of
trichloroethylene, chloroform, and 1,2-dibromo-3-chloropropane in a full-size,
experimental shower. J. Expo. Anal. Environ. Epidemiol. 6(4):413-423.
Giardino, N.J., E. Gumerman, N.A. Esmen et al. 1992. Shower volatilization exposures
in homes using tap water contaminated with trichloroethylene. J. Expo. Anal. Environ.
Epidemiol. Suppl. 1:147-158.
Gordon, S., P. Callahan, M. Brinkman, K. Kenny and L. Wallace. 1997. Effect of water
temperature on dermal exposure to chloroform. Presented at the 7th Annual Meeting of
the International Society of Exposure Analysis.
79
-------
Guengerich, P.P., D.H. Kim and M. Iwasaki. 1991. Role of human cytochrome P450
IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res.
Toxicol. 4(2): 168-179.
Held, S.D., M.L. Gargas and G.L. Kedderis. 1994. Kinetics of chloroform
biotransformation determined in freshly isolated male and female rodent hepatocytes.
Int. Soc. Study Xenobiotics Proc. 6:168. [abstract]
Hoke, J.R., Ed. 1988. Architectural Graphic Standards, 8th ed. John Wiley and Sons,
New York, NY.
Hoke, J.R., Ed. 1994. Architectural Graphic Standards, 9th ed. John Wiley and Sons,
New York, NY.
Hopkins, S.M. and J.C. Ellis. 1980. Drinking Water Consumption in Great Britain: A
Survey of Drinking Habits with Special Reference to Tap-water-Based Beverages.
Technical Report 137, Water Research Centre, Wiltshire, Great Britain.
Howard, C. and R.L. Corsi. 1996. Volatilization of chemicals from drinking water to
indoor air: Role of the kitchen sink. J. Air Waste Manage. Assoc. 46:830-837.
HSDB (Hazardous Substances Data Bank). 2001. U.S. National Library of Medicine.
Accessed January 18, 2001. Available at http://www.toxnet.nlm.nih.gov/.
HUD (U.S. Department of Housing and Urban Development). 1984. Residential Water
Conservation Projects, Summary Report. Prepared by Brown and Caldwell Consulting
Engineers for the Office of Policy Development and Research, Washington, DC. Report
No. HUD-PDR-903.
ICRP (International Commission on Radiological Protection). 1975. Report of the Task
Group on Reference Man. Pergamon, Oxford.
llett, K.F., W.D. Reid, I.G. Sipes and G. Krishna. 1973. Chloroform toxicity in mice:
Correlation of renal and hepatic necrosis with covalent binding of metabolites to tissue
macromolecules. Exp. Mol. Pathol. 19(2):215-229.
Kedderis, G.L. 1997. Extrapolation of in vitro enzyme induction data to humans in vivo.
Chem. Biol. Interact. 107:109-121.
Kedderis, G.L. and S.D. Held. 1996. Prediction of furan pharmacokinetics from
hepatocyte studies: Comparison of bioactivation and hepatic dosimetry in rats, mice,
and humans. Toxicol. Appl. Pharmacol. 140:124-130.
Kedderis, G.L. and J.C. Lipscomb. 2001. Application of in vitro biotransformation data
and pharmacokinetic modeling to risk assessment. Toxicol. Ind. Health.
17(5-10):315-321.
80
-------
Kedderis, G.L., S.D. Held, A.C. Pearson and M.A. Carfagna. 1993. Isolated mouse
hepatocytes as an in vitro model for the in vivo metabolism and cytolethality of
chloroform (CF). Toxicologist. 13:198. [abstract]
Konen, T.P. and D.L. Anderson. 1993. The Impact of Water Conserving Plumbing
Fixtures on Residential Water Use Characteristics: A Case Study in Tampa, FL.
Stevens Institute of Technology, Hoboken, NJ and Ayres Associates, Tampa, FL.
Koontz, M.D. and H.E. Rector. 1995. Estimation of Distributions for Residential Air
Exchange Rates. Technical report, prepared for the U.S. Environmental Protection
Agency, Office of Pollution Prevention and Toxics, Washington, DC under Contract No.
68-D9-0166, Work Assignment No. 3-19.
Krishnan, K. 2001. University of Montreal, Canada. Personal Communication to J.
Lipscomb, U.S. EPA, Cincinnati, OH.
Krishnan, K. 2002. University of Montreal, Canada. Personal communication to J.
Lipscomb, U.S. EPA, Cincinnati, OH.
Lide, D.R., Ed. 2000. CRC Handbook of Chemistry and Physics. CRC Press, Boca
Raton, FL. Accessed January 19, 2001. Available at http://208.254.79.26/.
Lilly, P.O., M.E. Andersen, T.M. Ross and R.A. Pegram. 1997. Physiologically based
estimation of in vivo rates of bromodichloromethane metabolism. Toxicology.
124:141-152.
Lilly, P.O., M.E. Andersen, T.M. Ross and R.A. Pegram. 1998. A physiologically based
pharmacokinetic description of the oral uptake, tissue dosimetry, and rates of
metabolism of bromodichloromethane in the male rat. Toxicol. Appl. Pharmacol.
150(2):205-217.
Lipscomb, J. 2001. U.S. EPA, Cincinnati, OH. Personal communication to K.
Krishnan, University of Montreal, Canada.
Little, J.C. 1992. Applying the two-resistance theory to contaminant volatilization in
showers. Environ. Sci. Technol. 26(7): 1341-1349.
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt, Ed. 1990. Handbook of Chemical
Property Estimation Methods. American Chemical Society, Washington DC.
Lynberg, M.L., J.R. Nuckols, P. Langlois et al. 2001. Assessing exposure to
disinfection by-products in women of reproductive age living in Corpus Christi, Texas,
and Cobb County, Georgia: Descriptive results and methods. Environ. Health
Perspect. 109(6): 597-604.
81
-------
Mackay, D. and W.Y. Shiu. 1981. A critical review of Henry's Law constants for
chemicals of environmental interest. J. Phys. Chem. Ref. Data. 10(4): 1175-1199.
March, J. 1968. Advanced Organic Chemistry: Reactions, Mechanisms, and Structure.
McGraw-Hill, New York. pp. 24-27.
Mathews, J.H. 1992. Numerical Methods for Mathematics, Science, and Engineering.
2nd Ed. Prentice Hall, Englewood Cliffs, NJ.
Mayer, P.W., W.B. DeOreo, E.M. Opitz et al. 1998 . Residential End Uses of Water.
American Water Works Association Research Foundation, Denver, CO. Project No.
241A. Available at
http://www.awwarf.org/research/topicsandproiects/execSum/241.aspx.
McGuire, M.J., J.L. McLain and A. Obolensky, Ed. 2002. Information Collection Rule
Data Analysis. American Water Works Association Research Foundation, Denver, CO.
Miles, A.M., P.C. Singer, D.L. Ashley et al. 2002. Comparison of trihalomethanes in tap
water and blood. Environ. Sci. Technol. 36(8): 1692-1698.
Moore, R.M., C.E. Geen and V.K. Tait. 1995. Determination of Henry's Law constants
for a suite of naturally occurring halogenated methanes in seawater. Chemosphere.
30:1183-1191.
Nicholson, B.C., B.P. Maguire and D.B. Bursill. 1984. Henry's Law constants for the
trihalomethanes: Effects of water com position and temperature. Environ. Sci. Technol.
18:518-521.
Olin, S.S., Ed. 1998. Exposure to Contaminants in Drinking Water. Estimating Uptake
Through the Skin and by Inhalation. International Life Sciences Institute, Washington,
DC. CRC Press, Boca Raton, FL.
Persily, A.K. 1998. A Modeling Study of Ventilation, IAQ and Energy Impacts of
Residential Mechanical Ventilation. National Institute of Standards and Technology,
Gaithersburg, MD. Report No. NISTIR6162. Available at
http://fire.nist.gov/bfrlpubs/build98/PDF/b98008.pdf.
Pohl, L.R., J.L. Martin and J.W. George. 1980. Mechanism of metabolic activation of
chloroform by rat liver microsomes. Biochem. Pharmacol. 29(24):3271-3276.
Potts, A.M., F.P. Simon and R.W. Gerard. 1949. The mechanism of action of
phosgene and diphosgene. Arch. Biochem. 24:329-337.
Price, K., S. Haddad and K. Krishnan. 2003. Physiological modeling of age-specific
changes in the pharmacokinetics of organic chemicals in children. J. Toxicol. Environ.
Health A. 66(5):417-433.
82
-------
Raucy, J.L., J.C. Kranerand J.M. Lasker. 1993. Bioactivation of halogenated
hydrocarbons by cytochrome P4502E1. Grit. Rev. Toxicol. 23:1-20.
Rector, H.E., C.R. Wilkes and N.J. Giardino. 1996. Effects of human behavior on
inhalation exposure to radon volatilized from domestic water uses. In: Proc. 1996
International Radon Symposium, McClean, VA. American Association of Radon
Scientists and Technologists, p 1-8.1 - I-8.8.
Rector, H.E., C.R. Wilkes and A.D. Mason. 2001. Techniques for Modeling Building
Systems in TEM. Draft report. Prepared for U.S. Environmental Protection Agency,
Office of Research and Development.
Risk Assessment Information System. 2001. Chemical-specific factors. Oak Ridge
National Laboratory. Accessed January 22, 2001. Available at
http://risk.lsd.ornl.gov/cgi-bin/tox/TOX select?select=csf.
Ross, M.K. and R.A. Pegram. 2003. Glutathione transferase theta 1-1-dependent
metabolism of the water disinfection byproduct bromodichloromethane. Chem. Res.
Toxicol. 16(2):216-226.
Sandberg, M. 1984. The multi-chamber theory reconsidered from the viewpoint of air
quality studies. Building Environ. 19(4):221-233.
Sander, R. 2001. Compilation of Henry's Law constants for inorganic and organic
species of potential importance in environmental chemistry. Accessed January 24,
2001. Available at http://dionvsos.mpch-mainz.mpg.de/~sander/res/henrv.html.
Sangster, J. 1989. Octanol-water partition coefficients of simple organic compounds.
J. Phys. Chem. Ref. Data. 18(3):1111-1229.
Segel, I.H. 1975. Enzyme Kinetics. John Wiley & Sons, New York, NY. p. 100-118.
Sinden, F.W. 1978. Multi-chamber theory of air infiltration. Building Environ. 13:21-28.
SRC (Syracuse Research Corporation). 2001. Interactive PhysProp Database Demo.
Accessed January 29, 2001. Available at http://esc.syrres.com/interkow/physdemo.htm.
Staudinger, J. and P.V. Roberts. 1996. A critical review of Henry's Law constants for
environmental applications. Grit. Rev. Environ. Sci. Technol. 26:205-297.
Tsang, A.M. and N.E. Klepeis. 1996. Descriptive Statistics Tables from a Detailed
Analysis of the National Human Activity Pattern Survey (NHAPS) Data 1996. U.S.
Environmental Protection Agency, Washington, DC. EPA/600/R-96/074.
83
-------
U.S. DOE. 1995. Residential Energy Consumption Survey (REGS): Housing
Characteristics 1993. U.S. Department of Energy, Energy Information Administration,
Washington, DC. Report No. DOE/EIA-0314(93).
U.S. DOE. 1999. A Look at Residential Energy Consumption in 1997. U.S.
Department of Energy. Energy Information Administration. Washington, DC. Report
No. DOE/EIA-0632(97).
U.S. EPA. 1997a. Exposure Factors Handbook. Volume I. General Factors. U.S.
Environmental Protection Agency, Office of Research and Development, National
Center for Environmental Assessment, Washington, DC. EPA/600/P-95/002Fa.
Available at http://www.epa.gov/clariton/clhtml/pubindex.html.
U.S. EPA. 1997b. Exposure Factors Handbook. Volume III. Activity Factors. U.S.
Environmental Protection Agency, Office of Research and Development, National
Center for Environmental Assessment, Washington, DC. EPA/600/P-95/002Fc.
Available at http://www.epa.gov/clariton/clhtml/pubindex.html.
U.S. EPA. 2000a. Volatilization Rates from Water to Indoor Air, Phase II. U.S.
Environmental Protection Agency, Office of Research and Development, Washington,
DC. EPA 600/R-00/096.
U.S. EPA. 2000b. Estimated Per Capita Water Ingestion in the United States, Based
on Data Collected by the USDA's 1994-96 Continuing Survey of Food Intakes by
Individuals. U.S. Environmental Protection Agency, Office of Water, Washington, DC.
EPA/822/R-00/008.
U.S. EPA. 2000c. ICR Auxiliary 1 Database. Version 4.0 Query Tool-Version 1.0
(CDROM). U.S. Environmental Protection Agency, Washington, DC.
EPA/815/C-00/001.
U.S. EPA. 2000d. Estimated Per Capita Water Ingestion in the United States: Based
on Data Collected by the United States Department of Agriculture's 1994-96 Continuing
Survey of Food Intakes by Individuals. U.S. Environmental Protection Agency, Office of
Water, Washington, DC. EPA/822/R-00/008.
U.S. EPA. 2003. The Feasibility of Performing Cumulative Risk Assessments for
Mixtures of Disinfection By-Products in Drinking Water. U.S. Environmental Protection
Agency, Office of Research and Development, National Center for Environmental
Assessment, Cincinnati, OH. EPA/600/R-03/051. Available at
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=56834.
84
-------
U.S. EPA. 2004. Use of Physiologically Based Pharmacokinetic Models to Quantify the
Impact of Human Age and Interindividual Differences in Physiology and Biochemistry
Pertinent to Risk: Final Report for Cooperative Agreement CR828047010. U.S.
Environmental Protection Agency, Office of Research and Development, National
Center for Environmental Assessment, Cincinnati, Ohio. NCEA-C-1467.
U.S. EPA. 2006. Use of Physiologically Based Pharmacokinetic Models to Quantify the
Impact of Human Age and Interindividual Differences in Physiology and Biochemistry
Pertinent to Risk (Final Report). U.S. Environmental Protection Agency, Washington,
D.C. EPA/600/R-06/014A. March 2006.
Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=151384
Verschueren, K. 1983. Handbook of Environmental Data on Organic Chemicals, 2nd
ed. Van Nostrand Reinhold, New York, NY.
Wilkes, C.R. 1994. Modeling exposure to VOCs from residential water. Ph.D.
Dissertation, Carnegie Mellon University, Department of Civil Engineering, Pittsburgh,
PA.
Wilkes, C.R. 1999. Exposure to Contaminants in Drinking Water: Estimating Uptake
Through the Skin and by Inhalation. S.S. Olin, Ed. International Life Sciences Institute
(ILSI). CRC Press, Boca Raton, FL.
Wilkes, C.R., M.J. Small, C.I. Davidson and J.B. Andelman. 1996. Modeling the effects
of water usage and co-behavior on inhalation exposures to contaminants volatilized
from household water. J. Exp. Anal. Environ. Epidemiol. 6(4):393-412.
Wilkes, C.R., A.D. Mason, L.L. Niang et al. 2002. Developing Individual Human
Exposure Estimates for Individual DBPs. Prepared for U.S. Environmental Protection
Agency, Office of Research and Development, National Center for Environmental
Assessment, Cincinnati under GSA Contract No. GS-10F-0154K.
Wilkes, C.R., A.D. Mason, L.L. Niang and K.L. Jensen. 2004. Quantification of
Exposure Related Water Uses for Various U.S. Subpopulations. Report prepared for
U.S. Environmental Protection Agency, Las Vegas, NV. EPA/600/R-04/066.
Zhao, G. and J.W. Allis. 2002. Kinetics of bromodichloromethane metabolism by
cytochrome P450 isoenzymes in human liver microsomes. Chem. Biol. Interact.
140(2): 155-168.
85
-------
APPENDIX
PBPK Model Code
$TARGET_FUNCTION
#include "ia.hpp"
SEND
PROGRAM THM
! PBPK model for THMs in water, interactions between 4 THMs
! THMs: 1=CF, 2=BDCM, 3=DBCM, 4=BF
! 2/6-23/04 GLK based on volatiles
INITIAL
CONSTANT SUBNUM = 1 ISUBJECTNUMBER 1=MALE,2=FEMALE,3=CHILD
CONSTANT QPC = 15. lalveolar ventilation rate (L/hr/kg)
CONSTANT QCC = 15. ICARDIAC OUTPUT (I/HR/KG)
CONSTANT QLC = 0.26 FRACTIONAL BLOOD FLOW TO LIVER
CONSTANT QFC = 0.05 FRACTIONAL BLOOD FLOW TO FAT
CONSTANT QKC = 0.034 FRACTIONAL BLOOD FLOW TO KIDNEY
CONSTANT QGC = 0.013 FRACTIONAL BLOOD FLOW TO GENITALS
CONSTANT BW = 70 IBODY WEIGHT (KG)
CONSTANT VLC = 0.026 FRACTION LIVER TISSUE
CONSTANT VFC = 0.19 FRACTION FAT TISSUE
CONSTANT VKC = 0.004 FRACTION KIDNEY TISSUE
CONSTANT VGC = 0.0004 FRACTION GENITAL TISSUE
CONSTANT BVC = 0.06 FRACTION BLOOD VOL
CONSTANT VABC = 0.35 FRACTION ARTERIAL BLOOD VOL
CONSTANT WBC = 0.65 FRACTION VENOUS BLOOD VOL
CONSTANT PL1 = 1.6 ILIVER/BLOOD PARTITION COEFF CF
CONSTANT PL2 = 1.15 ILIVER/BLOOD PARTITION COEFF BDCM
CONSTANT PL3 = 2.56 ILIVER/BLOOD PARTITION COEFF DBCM
CONSTANT PL4 = 2.06 ILIVER/BLOOD PARTITION COEFF BF
CONSTANT PF1 = 31.0 FAT/BLOOD PARTITION COEFF CF
CONSTANT PF2 = 19.8 FAT/BLOOD PARTITION COEFF BDCM
CONSTANT PF3 = 39.0 FAT/BLOOD PARTITION COEFF DBCM
CONSTANT PF4 = 40.4 FAT/BLOOD PARTITION COEFF BF
CONSTANT PK1 = 1.3 IKIDNEY/BLOOD PARTITION COEFF CF
CONSTANT PK2 = 1.24 IKIDNEY/BLOOD PARTITION COEFF BDCM
CONSTANT PK3 = 2.56 IKIDNEY/BLOOD PARTITION COEFF DBCM
CONSTANT PK4 = 1.69 IKIDNEY/BLOOD PARTITION COEFF BF
CONSTANT PG1 = 1.1 IGENITAL/BLOOD PARTITION COEFF CF
CONSTANT PG2 = 0.69 IGENITAL/BLOOD PARTITION COEFF BDCM
CONSTANT PG3 = 1.5 IGENITAL/BLOOD PARTITION COEFF DBCM
CONSTANT PG4 = 1.18 IGENITAL/BLOOD PARTITION COEFF BF
CONSTANT PS1 = 1.5 ISLOWLY PERF/BLOOD PART COEFF CF
CONSTANT PS2 = 0.47 ISLOWLY PERF/BLOOD PART COEFF BDCM
CONSTANT PS3 = 1.13 ISLOWLY PERF/BLOOD PART COEFF DBCM
86
-------
CONSTANT PS4 = 1.12 ISLOWLY PERF/BLOOD PART COEFF BF
CONSTANT PR1 = 1.6 IRICHLY PERF/BLOOD PART COEFF CF
CONSTANT PR2 = 1.15 IRICHLY PERF/BLOOD PART COEFF BDCM
CONSTANT PR3 = 2.56 IRICHLY PERF/BLOOD PART COEFF DBCM
CONSTANT PR4 = 2.06 IRICHLY PERF/BLOOD PART COEFF BF
CONSTANT PB1 = 11.34 IBLOOD/AIR PARTITION COEFF CF
CONSTANT PB2 = 26.6 IBLOOD/AIR PARTITION COEFF BDCM
CONSTANT PBS = 49.2 IBLOOD/AIR PARTITION COEFF DBCM
CONSTANT PB4 = 102.3 IBLOOD/AIR PARTITION COEFF BF
CONSTANT MW1 = 119.4 ICF MOLECULAR WEIGHT (G/MOL)
CONSTANT MW2 = 163.83 IBDCM MOLECULAR WEIGHT (G/MOL)
CONSTANT MW3 = 208.29 IDBCM MOLECULAR WEIGHT (G/MOL)
CONSTANT MW4 = 252.75 IBF MOLECULAR WEIGHT (G/MOL)
CONSTANT VMAXC1 = 8.956 ICF MAXIMAL VELOCITY (MG/HR/KG)
CONSTANT VMAXC2 = 8.01 IBDCM MAXIMAL VELOCITY (MG/HR/KG)
CONSTANT VMAXC3 = 13.7 IDBCM MAXIMAL VELOCITY (MG/HR/KG)
CONSTANT VMAXC4 = 10.4 IBF MAXIMAL VELOCITY (MG/HR/KG)
CONSTANT KM1 = 0.012 ICF MICHAELIS-MENTEN CONSTANT (MG/L)
CONSTANT KM2 = 0.302 IBDCM MICHAELIS-MENTEN CONSTANT (MG/L)
CONSTANT KM3 = 0.72 IDBCM MICHAELIS-MENTEN CONSTANT (MG/L)
CONSTANT KM4 = 0.42 IBF MICHAELIS-MENTEN CONSTANT (MG/L)
CONSTANT ODOSE1 = 0. ICF ORAL DOSE (MG/KG)
CONSTANT ODOSE2 = 0. IBDCM ORAL DOSE (MG/KG)
CONSTANT ODOSE3 = 0. IDBCM ORAL DOSE (MG/KG)
CONSTANT ODOSE4 = 0. IBF ORAL DOSE (MG/KG)
CONSTANT KA = 2.0 IORAL UPTAKE RATE (/HR)
CONSTANT DDOSE1 = 0. ICF DERMAL DOSE (MG TOTAL)
CONSTANT DDOSE2 = 0. IBDCM DERMAL DOSE (MG TOTAL)
CONSTANT DDOSE3 = 0. IDBCM DERMAL DOSE (MG TOTAL)
CONSTANT DDOSE4 = 0. IBF DERMAL DOSE (MG TOTAL)
ICONSTANT IVDOSE = 0. IIV DOSE (MG/KG)
CONSTANT CONC1 = 0. ICF INHALED CONG (PPM)
CONSTANT CONC2 = 0. IBDCM INHALED CONG (PPM)
CONSTANT CONC3 = 0. IDBCM INHALED CONG (PPM)
CONSTANT CONC4 = 0. IBF INHALED CONG (PPM)
I TIMING COMMANDS
CONSTANT TSTOP = 24. ILENGTH OF EXPT (HR)
CONSTANT TCHNG = 24. ILENGTH OF EXPOSURE (HR)
ICONSTANTTINF = 0.002 ILENGTH OF IV INFUSION (HR)
CONSTANT TDER = 0.167 ILENGTH OF DERMAL EXPOSURE (HR)
CONSTANT POINTS = 17280 INUMBER OF POINTS IN PLOT
CINT = TSTOP/POINTS ICOMMUNICATION INTERVAL
ISCALED PARAMETERS
QC = QCC*BW**0.74
QP = QPC*BW**0.74
QL = QLC*QC
87
-------
QF = QFC*QC
QK = QKC*QC
QG = QGC*QC
QS = 0.24*QC - QF
QR = 0.76*QC-QL-QK-QG
VL = VLC*BW
VF = VFC*BW
VK = VKC*BW
VG = VGC*BW
VS = 0.82*BW-VF
VR = 0.09*BW - VL - VK - VG
BV = BVC*BW
VAB = BV*VABC
WB = BV*VVBC
VMAX1 =VMAXC1*BW**0.7
VMAX2 = VMAXC2*BW**0.7
VMAX3 = VMAXC3*BW**0.7
VMAX4 = VMAXC4*BW**0.7
DOSE1 =ODOSE1*BW
DOSE2 = ODOSE2*BW
DOSES = ODOSE3*BW
DOSE4 = ODOSE4*BW
!IVR = IVDOSE*BW/TINF
END !OF INITIAL
DYNAMIC
IALG = 2 IGEAR METHOD FOR STIFF SYSTEMS
DERIVATIVE
!CI = CONG IN INHALED AIR (MG/L)
CIZONE = RSW(T.GT.TCHNG,0.,1.)
CM =CIZONE*CONC1*MW1/24450
CI2 = CIZONE*CONC2*MW2/24450
CIS = CIZONE*CONC3*MW3/24450
CI4 = CIZONE*CONC4*MW4/24450
!AI = AMOUNT INHALED (MG)
RAM = QP*CI1
RAI2 = QP*CI2
RAI3 = QP*CI3
RAW = QP*CI4
AM =INTEG(RAI1,0.)
AI2 = INTEG(RAI2,0.)
AI3 = INTEG(RAI3,0.)
AI4 = INTEG(RAI4,0.)
!MR = AMOUNT REMAINING IN STOMACH (MG)
RMR1 = -KA*MR1
RMR2 = -KA*MR2
RMR3 = -KA*MR3
88
-------
RMR4 = -KA*MR4
MR1 = DOSE1*EXP(-KA*T)
MR2 = DOSE2*EXP(-KA*T)
MRS = DOSE3*EXP(-KA*T)
MR4 = DOSE4*EXP(-KA*T)
!CA = CONG IN SYSTEMIC ARTERIAL BLOOD (MG/L)
CA1 = (QC*CV1 + QP*CI1)/(QC + (QP/PB1)) + DE1A/AB
CA2 = (QC*CV2 + QP*CI2)/(QC + (QP/PB2)) + DE2A/AB
CAS = (QC*CV3 + QP*CI3)/(QC + (QP/PB3)) + DE3A/AB
CA4 = (QC*CV4 + QP*CI4)/(QC + (QP/PB4)) + DE4A/AB
AUCB1 =INTEG(CA1,0.)
AUCB2 = INTEG(CA2,0.)
AUCB3 = INTEG(CA3,0.)
AUCB4 = INTEG(CA4,0.)
!AX = AMOUNT EXHALED (MG)
CX1 = CA1/PB1
CX2 = CA2/PB2
CX3 = CA3/PB3
CX4 = CA4/PB4
CXPPM1 =(0.7*CX1 +0.3*CI1)*24450./MW1
CXPPM2 = (0.7*CX2 + 0.3*CI2)*24450./MW2
CXPPM3 = (0.7*CX3 + 0.3*CI3)*24450./MW3
CXPPM4 = (0.7*CX4 + 0.3*CI4)*24450./MW4
RAX1 = QP*CX1
RAX2 = QP*CX2
RAX3 = QP*CX3
RAX4 = QP*CX4
AX1 = INTEG(RAX1,0.)
AX2 = INTEG(RAX2,0.)
AX3 = INTEG(RAX3,0.)
AX4 = INTEG(RAX4,0.)
!AS = AMOUNT IN SLOWLY PERFUSED TISSUES (MG)
RAS1 =QS*(CA1 -CVS1)
RAS2 = QS*(CA2 - CVS2)
RAS3 = QS*(CA3 - CVS3)
RAS4 = QS*(CA4 - CVS4)
AS1 = INTEG(RAS1,0.)
AS2 = INTEG(RAS2,0.)
ASS = INTEG(RAS3,0.)
AS4 = INTEG(RAS4,0.)
CVS1 =AS1/(VS*PS1)
CVS2 = AS2/(VS*PS2)
CVS3 = AS3/(VS*PS3)
CVS4 = AS4/(VS*PS4)
CS1 =AS1A/S
CS2 = AS2A/S
89
-------
CSS = AS3A/S
CS4 = AS4A/S
IAMOUNT IN RAPIDLY PERFUSED TISSUES (MG)
RAR1 =QR*(CA1 -CVR1)
RAR2 = QR*(CA2 - CVR2)
RAR3 = QR*(CA3 - CVR3)
RAR4 = QR*(CA4 - CVR4)
AR1 =INTEG(RAR1,0.)
AR2 = INTEG(RAR2,0.)
AR3 = INTEG(RAR3,0.)
AR4 = INTEG(RAR4,0.)
CVR1 =AR1/(VR*PR1)
CVR2 = AR2/(VR*PR2)
CVR3 = AR3/(VR*PR3)
CVR4 = AR4/(VR*PR4)
CR1 =AR1A/R
CR2 = AR2A/R
CR3 = AR3A/R
CR4 = AR4A/R
!AF = AMOUNT IN FAT (MG)
RAF1 =QF*(CA1 -CVF1)
RAF2 = QF*(CA2 - CVF2)
RAF3 = QF*(CA3 - CVF3)
RAF4 = QF*(CA4 - CVF4)
AF1 = INTEG(RAF1,0.)
AF2 = INTEG(RAF2,0.)
AF3 = INTEG(RAF3,0.)
AF4 = INTEG(RAF4,0.)
CVF1 =AF1/(VF*PF1)
CVF2 = AF2/(VF*PF2)
CVF3 = AF3/(VF*PF3)
CVF4 = AF4/(VF*PF4)
CF1 =AF1A/F
CF2 = AF2A/F
CF3 = AF3A/F
CF4 = AF4A/F
!AK = AMOUNT IN KIDNEY (MG)
RAK1 =QK*(CA1 -CVK1)
RAK2 = QK*(CA2 - CVK2)
RAK3 = QK*(CA3 - CVK3)
RAK4 = QK*(CA4 - CVK4)
AK1 =INTEG(RAK1,0.)
AK2 = INTEG(RAK2,0.)
AK3 = INTEG(RAK3,0.)
AK4 = INTEG(RAK4,0.)
CVK1 =AK1/(VK*PK1)
90
-------
CVK2 = AK2/(VK*PK2)
CVK3 = AK3/(VK*PK3)
CVK4 = AK4/(VK*PK4)
CK1 =AK1A/K
CK2 = AK2A/K
CK3 = AK3A/K
CK4 = AK4A/K
!AG = AMOUNT IN GENITALS [TESTES, OVARIES] (MG)
RAG1 =QG*(CA1 -CVG1)
RAG2 = QG*(CA2 - CVG2)
RAGS = QG*(CA3 - CVG3)
RAG4 = QG*(CA4 - CVG4)
AG1 = INTEG(RAG1,0.)
AG2 = INTEG(RAG2,0.)
AG3 = INTEG(RAG3,0.)
AG4 = INTEG(RAG4,0.)
CVG1 =AG1/(VG*PG1)
CVG2 = AG2/(VG*PG2)
CVG3 = AG3/(VG*PG3)
CVG4 = AG4/(VG*PG4)
CG1 =AG1A/G
CG2 = AG2A/G
CG3 = AG3A/G
CG4 = AG4A/G
!AL = AMOUNT IN LIVER (MG)
RAL1 = QL*(CA1 - CVL1) - RAM1 + RA01
RAL2 = QL*(CA2 - CVL2) - RAM2 + RA02
RAL3 = QL*(CA3 - CVL3) - RAMS + RA03
RAL4 = QL*(CA4 - CVL4) - RAM4 + RA04
AL1 =INTEG(RAL1,0.)
AL2 = INTEG(RAL2,0.)
AL3 = INTEG(RAL3,0.)
AL4 = INTEG(RAL4,0.)
CVL1 =AL1/(VL*PL1)
CVL2 = AL2/(VL*PL2)
CVL3 = AL3/(VL*PL3)
CVL4 = AL4/(VL*PL4)
CL1 =AL1A/L
CL2 = AL2A/L
CL3 = AL3A/L
CL4 = AL4A/L
AUCL1 =INTEG(CL1,0.)
AUCL2 = INTEG(CL2,0.)
AUCL3 = INTEG(CL3,0.)
AUCL4 = INTEG(CL4,0.)
!AM = AMOUNT METABOLIZED, P450 SATURABLE PATHWAY (MG)
91
-------
RAM1 = (VMAX1*CVL1)/(KM1*(1+CVL2/KM2+CVL3/KM3+CVL4/KM4)+CVL1)
RAM2 = (VMAX2*CVL2)/(KM2*(1+CVL1/KM1+CVL3/KM3+CVL4/KM4)+CVL2)
RAMS = (VMAX3*CVL3)/(KM3*(1+CVL1/KM1+CVL2/KM2+CVL4/KM4)+CVL3)
RAM4 = (VMAX4*CVL4)/(KM4*(1+CVL1/KM1+CVL2/KM2+CVL3/KM3)+CVL4)
RAMM1 = RAM1*1000./MW1
RAMM2 = RAM2*1000./MW2
RAMM3 = RAM3*1000./MW3
RAMM4 = RAM4*1000./MW4
AM1 = INTEG(RAM1,0.)
AM2 = INTEG(RAM2,0.)
AM3 = INTEG(RAM3,0.)
AM4 = INTEG(RAM4,0.)
CAM1 =AM1A/L
CAM2 = AM2A/L
CAMS = AM3A/L
CAM4 = AM4A/L
DM1 =CAM1*1000./MW1
DM2 = CAM2*1000./MW2
DM3 = CAM3*1000./MW3
DM4 = CAM4*1000./MW4
!AO = TOTAL MASS INPUT FROM STOMACH (MG)
RA01 = KA*MR1
RA02 = KA*MR2
RA03 = KA*MR3
RA04 = KA*MR4
A01 =DOSE1 -MR1
A02 = DOSE2 - MR2
A03 = DOSES - MRS
A04 = DOSE4 - MR4
!IV = IV INFUSION DOSE (MG)
!IV=IVR*(1.-STEP(TINF))
IDE = DERMAL DOSE (MG)
DDR1 = DDOSE1/TDER
DDR2 = DDOSE2/TDER
DDR3 = DDOSE3/TDER
DDR4 = DDOSE4/TDER
DE1 = DDR1*(1. - STEP(TDER))
DE2 = DDR2*(1. - STEP(TDER))
DE3 = DDR3*(1. - STEP(TDER))
DE4 = DDR4*(1. - STEP(TDER))
!CV = MIXED VENOUS BLOOD CONG (MG/L)
CV1 = (QF*CVF1+QL*CVL1+QS*CVS1+QR*CVR1+QK*CVK1+QG*CVG1)/QC
CV2 = (QF*CVF2+QL*CVL2+QS*CVS2+QR*CVR2+QK*CVK2+QG*CVG2)/QC
CVS = (QF*CVF3+QL*CVL3+QS*CVS3+QR*CVR3+QK*CVK3+QG*CVG3)/QC
CV4 = (QF*CVF4+QL*CVL4+QS*CVS4+QR*CVR4+QK*CVK4+QG*CVG4)/QC
ITMASS = MASS BALANCE (MG)
92
-------
TMASS1 = AF1+AL1+AS1+AR1+AK1+AG1+AM1+AX1+MR1
TMASS2 = AF2+AL2+AS2+AR2+AK2+AG2+AM2+AX2+MR2
TMASS3 = AF3+AL3+AS3+AR3+AK3+AG3+AM3+AX3+MR3
TMASS4 = AF4+AL4+AS4+AR4+AK4+AG4+AM4+AX4+MR4
TMASS = TMASS1 + TMASS2 + TMASS3 + TMASS4
TERMT(T.GE.TSTOP)
END !OF DERIVATIVE
discrete readdata
interval readinterval = 0.01667
QP = get_breathingrate(simulationnumber, SUBNUM, t)
CONC1 = get_airconcentration(simulationnumber, SUBNUM, 1, t)
CONC2 = get_airconcentration(simulationnumber, SUBNUM, 2, t)
CONC3 = get_airconcentration(simulationnumber, SUBNUM, 3, t)
CONC4 = get_airconcentration(simulationnumber, SUBNUM, 4, t)
DDOSE1 = get_dermaldose(simulationnumber,SUBNUM,1,t,-1)
DDOSE2 = get_dermaldose(simulationnumber,SUBNUM,2,t,-1)
DDOSE3 = get_dermaldose(simulationnumber,SUBNUM,3,t,-1)
DDOSE4 = get_dermaldose(simulationnumber,SUBNUM,4,t,-1)
TDER = get_dermaldose(simulationnumber,SUBNUM,1,t,1)
ODOSE1 = get_oraldose(simulationnumber,SUBNUM,1,t,BW)
ODOSE2 = get_oraldose(simulationnumber,SUBNUM,2,t,BW)
ODOSE3 = get_oraldose(simulationnumber,SUBNUM,3,t,BW)
ODOSE4 = get_oraldose(simulationnumber,SUBNUM,4,t,BW)
!callput_dataout(t,QP,CONC1,CONC2,CONC3,CONC4,simulationnumber,SUBNUM)
Icall
put_dataout2(DDOSE1,DDOSE2,DDOSE3,DDOSE4,TDER,ODOSE1,ODOSE2,ODOS
E3,ODOSE4)
end ! of discrete readdata
discrete writedata
interval writeinterval = 0.083
call resultsoutputl (SUBNUM,simulationnumber,t,QP,QC)
call resultsoutput2(AM1 ,AM2,AM3,AM4)
call resultsoutput3(CAM1 ,CAM2,CAM3,CAM4)
call resultsoutput4(AUCL1 ,AUCL2,AUCL3,AUCL4)
call resultsoutput5(CL1,CL2,CL3,CL4)
call resultsoutput6(CF1 ,CF2,CF3,CF4)
call resultsoutput7(CS1 ,CS2,CS3,CS4)
call resultsoutput8(CR1 ,CR2,CR3,CR4)
call resultsoutput9(CK1 ,CK2,CK3,CK4)
call resultsoutputl0(CG1 ,CG2,CG3,CG4)
call resultsoutputl 1 (CA1 ,CA2,CA3,CA4)
call resultsoutputl 2(CV1,CV2,CVS,CV4)
call resultsoutputl 3(DM1,DM2,DM3,DM4)
end ! of discrete writedata
END !OF DYNAMIC
93
-------
TERMINAL
call resultsoutputl (-1 ,-1 ,-1 ,-1 ,-1)
call put_dataout(-1 ,-1 ,-1 ,-1 ,-1 ,-1 ,-1 ,-1)
END !OF TERMINAL
END !OF PROGRAM
94
-------
TABLE 1
Location Codes Recorded in the National Human Activity Pattern Survey (NHAPS)
Location
Code
100
101
102
103
104
105
106
107
108
110
111
112
113
114
120
199
200
201
202
203
204
205
206
207
208
210
211
212
213
214
220
299
300
301
302
303
304
305
306
307
308
310
311
312
313
314
320
Name
Home - Other
Home - Kitchen
Home - Living / Family / Den
Home - Dining
Home - Bathroom
Home - Bedroom
Home - Study / Office
Home - Garage
Home - Basement
Home - Utility / Laundry
Home - Pool, spa (outdoors)
Home - Yard, other outdoors
Home - Room to room
Home - In /out of house
Home - Other (verified)
Home - Ref
Other's House - Other
Other's House - Kitchen
Other's House - Living / Family / Den
Other's House - Dining
Other's House - Bathroom
Other's House - Bedroom
Other's House - Study / Office
Other's House - Garage
Other's House - Basement
Other's House - Utility / Laundry
Other's House - Pool, spa (outdoors)
Other's House - Yard, other outdoors
Other's House - Room to room
Other's House - In / out of house
Other's House - Other (verified)
Other's House - Ref
Transit - Other
Transit - Car
Transit - Truck (Pick-up / Van)
Transit - Truck (Others)
Transit - Motorcycle / Moped / Scooter
Transit - Bus
Transit - Walking
Transit - Bicycle / Skateboard / RSkates
Transit - Stroller/ Carried by adult
Transit - Train / Subway / Rapid transit
Transit - Airplane
Transit - Boat
Transit - Waiting for Bus, train, ride (stop)
Transit - Waiting for travel, indoors
Transit - Other (verified)
Location
Code
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
499
500
501
502
503
504
505
506
507
508
510
511
520
599
Name
Transit - Ref
Indoor- Other
Indoor - Office Bldg / Bank / Post Office
Indoor - Industrial plant / Factory / Warehouse
Indoor - Grocery store / Convenience store
Indoor - Shopping mall / Non-grocery store
Indoor - Bar / Night club / Bowling alley
Indoor - Auto repair shop / Gas station
Indoor - Gym / Sports or health club
Indoor - Public Bldg / Library / Museum /
Theater
Indoor- Laundromat
Indoor - Hospital / Health care facility / Dr's
Office
Indoor - Beauty parlor / Barber shop / Hair
dresser
Indoor - Work (no specific main location)
Indoor - School
Indoor - Restaurant
Indoor - Church
Indoor - Hotel / Motel
Indoor- Dry cleaner
Indoor - Other repair shop
Indoor - Indoor parking garage
Indoor - Other (verified)
Indoor - Ref
Outdoor - Other
Outdoor - Sidewalk / Street / Neighborhood
Outdoor - Parking lot
Outdoor - Service station / Gas station
Outdoor - Construction site
Outdoor - School grounds / Playground
Outdoor - Sports stadium
Outdoor - Park / Golf course
Outdoor - Pool, river, lake
Outdoor - Restaurant / picnic (outdoors)
Outdoor - Farm
Outdoor - Other (verified)
Outdoor - Ref
T-1
-------
TABLE 2
Activity Codes Recorded in the National Human Activity Pattern Survey (NHAPS)
Activity
Code
1
2
3
5
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
47
48
49
Description
Work - Main job
Work - Unemployment
Work - Travel
Work - Second job
Work - Break
Work - Travel to/from
Household work - Food preparation
Household work - Food cleanup
Household work - Cleaning house
Household work - Outdoor cleaning
Household work - Clothes care
Household work - Car repair/maintenance (by
respondent)
Household work - Other repairs (by respondent)
Household work - Plant care
Household work - Animal care
Household work - Other
Child care - Baby care
Child care - Child care
Child care - Helping/Teaching
Child care - Talking/Reading
Child care - Indoor playing
Child care - Outdoor playing
Child care - Medical care
Child care - Other
Child care - Cleaning
Child care - Travel
Obtaining goods and services - Food shopping
Obtaining goods and services - Clothes/Household
shopping
Obtaining goods and services - Personal services
Obtaining goods and services - Medical appointments
Obtaining goods and services -
Government/Financial service
Obtaining goods and services - Car repair services
Obtaining goods and services - Other repair services
Obtaining goods and services - Other services
Obtaining goods and services - Errands
Obtaining goods and services - Travel
Personal needs and care - Bathing, etc.
Personal needs and care - Medical care
Personal needs and care - Help and care
Personal needs and care - Eating
Personal needs and care - Personal hygiene
Personal needs and care - Sleeping / Napping
Personal needs and care - Dressing, etc.
Personal needs and care - NA activity
Personal needs and care - Travel
Activity
Code
50
51
54
55
56
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
Description
Education and Training - Student classes
Education and Training - Other classes
Education and Training - Homework
Education and Training - Library
Education and Training - Other
Education and Training - Travel
Organizational activities - Professional/Union
Organizational activities - Special interest
Organizational activities - Political/Civic
Organizational activities - Volunteer/Helping
Organizational activities - Religious groups
Organizational activities - Religious practice
Organizational activities - Fraternal
Organizational activities - Child/Youth/Family
Organizational activities - Other
Organizational activities - Travel
Social - Sports events
Social - Entertainment events
Social - Movies/Videos
Social - Theatre
Social - Museums
Social - Visiting
Social - Parties
Social - Bars/Lounges
Social - Other
Social - Travel
Recreation - Active sports
Recreation - Outdoor
Recreation - Exercise
Recreation - Hobbies
Recreation - Domestic Crafts
Recreation - Art
Recreation - Music/Drama/Dance
Recreation - Games
Recreation - Computer use
Recreation - Travel
Communication - Radio
Communication - Television
Communication - Records/Tape
Communication - Read books
Communication - Read magazines, etc.
Communication - Read newspaper
Communication - Conversations
Communication - Letters/Write/Paperwork
Communication - Think/Relax
Communication - Travel, Passive Leisure
T-2
-------
TABLE 3
List of Chemicals for Exposure Assessment
DBP Subclass
Chemical Name
CAS Number
Trihalomethanes
Chloroform
67-66-3
(THMs)
Bromodichloromethane (BDCM)
75-27-4
Dibromochloromethane (DBCM)
124-48-1
Bromoform
75-25-2
T-3
-------
TABLE 4
Physical Properties of Chemicals of Interest
Chemical
Henry's Law Constant
Dimensionless
H
O
d.
E
AH
RT
( K)
o
Q)
Q)
C£
Diffusivity in Water
(cm2/s)
O
d.
F
o
Q)
Q)
C£
Diffusivity in Air
(cm2/s)
O
d.
F
o
(D
C£
Octanol/H2O
Partition Coef.
Log Kow
o
(D
C£
-C
D)
JS
O
Boiling
Point
Tb
O
(C)
o
(D
C£
Vapor Pressure
r vp
(mmHg)
o
d.
F
o
(D
C£
Trihalomethanes (THMs)
Chloroform
(CAS: 67-66-3)
CHCI3
Bromodichloro-
methane
(CAS: 75-27-4)
CHBrCb
Dibromochloro-
methane
(Chlorodibromo-
methane)
(CAS: 124-48-1)
CHBr2CI
Bromoform
(CAS: 75-25-2)
CHBr3
0.150
0.150 (a)
0.151 (a)
0.163 (a)
0.0656
0.0866 (a)
0.065
0.095
0.085
0.102
0.037
0.034
0.056
0.0219
0.0219 (a)
0.0255 (a)
0.0240 (a)
25
24
25
25
25
25
25
25
25
25
25
25
25
25
25
25
4500
4700
4700
5200
1
2
5a
5b
1
2
5c
5c
5c
5d
5c
5c
5d
1
2
5a
5b
1.0x10'5
1.06x10'5
lOSxIO'5
1.03x10'5
b
b
b
b
1
1
1
1
0.1040
0.0298
0.0196
0.0149
25
b
b
b
1
1
1
1
1.96
1.97
1.88
2.00
2.09
2.16
2.24
2.30
2.40
1
2
1
2
1
2
5
1
2
119.38
163.83
208.28
252.73
61.17
90
90
120
119-120
149.1
150-151
3
3
3
4
3
4
160
(vapor
density
4.12)
50
76
5.6
(vapor
density
8.7)
20
20
20
25
4
6
4
4
a. Henry's law constant is reported in the literature with concentration and partial pressure units. The value reported in the table was converted to dimensionless H.
b. Temperature at which diffusivity is measured is not reported.
REFERENCES: Risk Assessment Information System, 2001; HSDB, 2001; Lide, 2000; Gangolli, 1989; Sander, 2001; 6a Mackay, 1981; 6b Staudinger, 1996; 6c
Nicholson, 1984; 6d Moore, 1995; 7 SRC, 2001.
T-4
-------
TABLE 5
Estimated Values for Liquid Phase Diffusivity, Gas Phase Diffusivity, and Dimensionless Henry's Law Constant
Temp
°C
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Chloroform
DL/(1 E-6)
(L2/T)
8.200
8.443
8.699
8.951
9.206
9.465
9.726
9.992
10.260
10.532
10.807
11.085
11.368
11.653
1 1 .942
12.233
12.527
12.825
13.126
13.431
13.739
14.050
14.362
14.680
15.001
15.324
15.649
15.976
16.310
16.644
16.981
17.322
17.668
18.012
18.362
DG
(L2/T)
0.0894
0.09
0.0906
0.0912
0.0917
0.0923
0.0929
0.0935
0.0941
0.0947
0.0953
0.0959
0.0965
0.0971
0.0977
0.0983
0.0989
0.0995
0.1002
0.1008
0.1014
0.102
0.1026
0.1032
0.1039
0.1045
0.1051
0.1057
0.1063
0.107
0.1076
0.1082
0.1089
0.1095
0.1101
H*
0.1086
0.1105
0.1123
0.1142
0.1161
0.1236
0.131
0.1384
0.1459
0.1533
0.1617
0.1701
0.1785
0.1869
0.1953
0.2037
0.2122
0.2207
0.2291
0.2376
0.2475
0.2575
0.2674
0.2773
0.2872
0.2981
0.3093
0.3209
0.3328
0.3451
0.3577
0.3707
0.3841
0.3979
0.4121
BDCM
DL/(1 E-6)
(L2/T)
8.077
8.316
8.5685
8.8162
9.0674
9.3226
9.5802
9.8418
10.106
10.374
10.645
10.919
11.197
11.478
11.763
12.05
12.339
12.633
12.929
13.229
13.533
13.839
14.147
14.46
14.776
15.094
15.415
15.736
16.065
16.394
16.727
17.062
17.403
17.742
18.086
DG
(L2/T)
0.0849
0.0854
0.086
0.0866
0.0871
0.0877
0.0882
0.0888
0.0894
0.09
0.0905
0.0911
0.0917
0.0922
0.0928
0.0934
0.094
0.0946
0.0951
0.0957
0.0963
0.0969
0.0975
0.0981
0.0987
0.0993
0.0999
0.1005
0.1011
0.1017
0.1023
0.1029
0.1035
0.1041
0.1047
H*
0.0586
0.0618
0.0651
0.0686
0.0722
0.076
0.0799
0.0841
0.0884
0.0929
0.0976
0.1025
0.1076
0.1129
0.1185
0.1243
0.1303
0.1366
0.1431
0.1499
0.157
0.1643
0.172
0.1799
0.1882
0.1968
0.2057
0.2149
0.2245
0.2345
0.2449
0.2556
0.2667
0.2782
0.2902
DBCM
DL/(1 E-6)
(L2/T)
7.9588
8.1943
8.4431
8.6872
8.9347
9.1862
9.44
9.6978
9.9579
10.222
10.489
10.759
11.034
11.31
11.59
11.873
12.159
12.448
12.74
13.035
13.335
13.637
13.94
14.248
14.559
14.873
15.189
15.506
15.83
16.154
16.482
16.812
17.148
17.482
17.822
DG
(L2/T)
0.0814
0.082
0.0825
0.0831
0.0836
0.0841
0.0847
0.0852
0.0858
0.0863
0.0869
0.0874
0.088
0.0885
0.0891
0.0897
0.0902
0.0908
0.0913
0.0919
0.0925
0.093
0.0936
0.0942
0.0947
0.0953
0.0959
0.0965
0.097
0.0976
0.0982
0.0988
0.0994
0.0999
0.1005
H*
0.0245
0.0259
0.0274
0.0289
0.0305
0.0322
0.0339
0.0358
0.0377
0.0397
0.0418
0.044
0.0463
0.0487
0.0512
0.0538
0.0565
0.0594
0.0624
0.0654
0.0687
0.072
0.0756
0.0792
0.083
0.087
0.0911
0.0954
0.0999
0.1045
0.1093
0.1143
0.1195
0.1249
0.1306
Bromoform
DL/(1 E-6)
(L2/T)
7.8452
8.0773
8.3226
8.5632
8.8072
9.055
9.3052
9.5593
9.8157
10.076
10.339
10.606
10.876
11.149
1 1 .425
11.704
11.985
12.27
12.558
12.849
13.145
13.442
13.741
14.045
14.352
14.661
14.972
15.285
15.604
15.924
16.247
16.572
16.903
17.233
17.567
DG
(L2/T)
0.0787
0.0792
0.0797
0.0802
0.0808
0.0813
0.0818
0.0823
0.0829
0.0834
0.0839
0.0845
0.085
0.0856
0.0861
0.0866
0.0872
0.0877
0.0883
0.0888
0.0894
0.0899
0.0905
0.091
0.0916
0.0921
0.0927
0.0932
0.0938
0.0944
0.0949
0.0955
0.0961
0.0966
0.0972
H*
0.0134
0.0142
0.0151
0.016
0.017
0.018
0.0191
0.0203
0.0214
0.0227
0.024
0.0254
0.0269
0.0284
0.03
0.0317
0.0335
0.0353
0.0373
0.0393
0.0415
0.0437
0.0461
0.0485
0.0511
0.0538
0.0566
0.0595
0.0625
0.0657
0.0691
0.0726
0.0762
0.08
0.084
* Henry's law constants in this table are based on a combination of literature reported values and
estimates derived from procedures presented in Section 2.1.2.1.
T-5
-------
TABLE 6
Relevant Chemical Properties for the THMs
Property
Molecular Weight
Liquid Diffusivity
(cm2/sec)
Gas Diffusivity
(cm2/sec)
Vapor Pressure
(mm Hg)
Solubility (mg/L)
Henry's Law
Constant
Chloroform
119.38
9.21E-06(20°C)
1.50E-05(40°C)
0.091 75 (20°C);
0.1 0386 (40°C)
160 (B)
8000 (B)
0.116(20°C)
0.287 (40°C)
BDCM
163.8
9.07E-06 (20°C)
1.48E-05(40°C)
0.0871 (20°C)
0.104(40°C)
57.4 (A)
3030 (A)
0.0722 (20°C)
0.188(40°C)
DBCM
208.03
8.94E-06 (20°C)
1.46E-05(40°C)
0.0836 (20°C)
0.0947 (40°C)
15.6 (A)
2700 (B)
0.0305 (20°C)
0.0830 (40°C)
Bromoform
252.77
8.81E-06(20°C)
1 .44E-05 (40°C)
0.0808 (20°C)
0.091 6 (40°C)
5.4 (A)
3 100 (A)
0.0170(20°C)
0.0511 (40°C)
Notes: Liquid-phase diffusivity, and gas-phase diffusivity are estimated based on the Hayduk
and Laudie method (Lyman et al., 1990, pg. 17-200) and the Wilke and Lee Method (Lyman et
al., 1990, pg 17-13), respectively. The Henry's law constant is estimated based on the method
described in section 2.1.3.1, above.
Vapor pressure and solubility data are from the following sources:
a Risk Assessment Information System, 2001.
b Verschueren, 1983. Value reported at 20°C.
T-6
-------
TABLE 7
Relevant Chemical Properties for the Predictor Chemicals
Property
Molecular Weight
Liquid-phase diffusivity
@ 20°C (cm2/sec)
Liquid-phase diffusivity
@ 40°C (cm2/sec)
Gas-phase diffusivity
@ 20°C (cm2/sec)
Gas-phase diffusivity
@ 40°C (cm2/sec)
Vapor pressure
(mm Hg)
Solubility (mg/L)
Henry's law const @
20°C
Henry's law const @
40°C
Acetone
58.08
1.05E-05
1.71E-05
0.110
0.124
231 (A)
1 E 06 (A)
0.0011
0.00298
Ethylacetate
88.1
8.36E-06
1.36E-05
0.0880
0.0997
72.8 (B)
8.6E04 (B)
0.00445
0.0132
Toluene
92.1
7.96E-06
1.30E-05
0.0831
0.0942
22 (B)
515 (B)
0.215
0.456
Ethylbenzene
106.17
7.19E-06
1.17E-05
0.0753
0.0853
7 (B)
152 (B)
0.252
0.642
Cyclohexane
84.16
7.96E-06
1.30E-05
0.0853
0.0966
77 (B)
55 (B)
6.18
11.62
Notes: Liquid-phase diffusivity, and gas-phase diffusivity are estimated based on the
Hayduk and Laudie method (Lyman et al., 1990, pg 17-200 and the Wilke and Lee
Method (Lyman et al., 1990, pg 17-13), respectively. The Henry's law constant is
estimated based on the method described in section 2.1.3.1, above.
Vapor pressure and solubility data are from the following sources:
a Risk Assessment Information System, 2001.
b Verschueren, 1983. Value reported at 20°C.
T-7
-------
TABLE 8
Summary of Normalized Percent Difference (Equation 9) for the THMs as a Function of
Predictor Chemical
Property
Acetone
Ethylacetate
Toluene
Ethylbenzene
Cyclohexane
Chloroform
Liquid Diffusivity
Gas Diffusivity
Henry's Law
Constant
14.0%
19.9%
99.1%
9.2%
4.1%
96.2%
13.6%
9.4%
85.3%
21.9%
17.9%
117.2%
13.6%
7.0%
5227.6%
BDCM
Liquid Diffusivity
Gas Diffusivity
Henry's Law
Constant
15.8%
26.3%
98.5%
7.8%
1 .0%
93.8%
12.2%
4.6%
197.8%
20.7%
13.5%
249.0%
12.2%
2.1%
8459.6%
DBCM
Liquid Diffusivity
Gas Diffusivity
Henry's Law
Constant
17.4%
31.6%
96.4%
6.5%
5.3%
85.4%
11.0%
0.6%
604.9%
19.6%
9.9%
726.2%
11.0%
2.0%
20162.3%
Bromoform
Liquid Diffusivity
Gas Diffusivity
Henry's Law
Constant
19.2%
36.1%
93.5%
5.1%
8.9%
73.8%
9.6%
2.8%
1164.7%
18.4%
6.8%
1382.4%
9.6%
5.6%
36252.9%
Note: Values evaluated at 20°C
T-8
-------
TABLE 9
Estimated Values for Overall Mass Transfer Coefficient (K0i_A) based on Toluene
Appliance
Shower
Temp
°C
40
Estimated K0i_A (m3/hr)
Chloroform
0.432
BDCM
0.428
DBCM
0.415
Bromoform
0.402
Bath:
Fill
Bathing
35
35
0.245
0.0780
0.228
0.0735
0.186
0.0625
0.153
0.0531
Clothes Washer
Fill
Wash
Rinse
Toilets
35
35
35
25
0.317
0.113
0.403
0.00468
0.265
0.0637
0.265
0.00368
0.174
0.0293
0.122
0.00312
0.124
0.0177
0.0735
0.00265
Faucets
Kitchen
Bathroom
Laundry Room
35
35
30
0.128
0.128
0.117
0.116
0.116
0.104
0.0913
0.0913
0.0792
0.0731
0.0731
0.0613
T-9
-------
TABLE 10
Shower Frequency Values from NHAPS and REUWS Analyses
Statistic
Shower
Frequency per
person-day
Population
Children
5-1 2 years
(NHAPS)
0.6
Men
18-48
years
(NHAPS)
1.2
Women
18-48
years
(NHAPS)
1.1
All
Households
(NHAPS)
1.0
All Households
(REUWS)
0.8
TABLE 1 1
Summary Statistics for Shower Duration, Volume and Flowrate from REUWS Analyses
Statistic for All
Households
(REUWS)
Shower Duration
Shower Volume
(adults only)
Shower Flowrate
Geometric Mean
6.8 minutes
15.8 gallons/shower
2.0 gallons/minute
Geometric
Standard
Deviation
1.64 minutes
1.75 gallons
1.58 gallons/min
Arithmetic Mean
7.7 minutes
18.6 gallons/shower
2.4 gallons/minute
T-10
-------
TABLE 12
Selected Model Parameters for Showers
Statistic
Shower Frequency per person per day
Children 6 years
Men 1 5-45 years
Women 15-45 years
Shower Duration (Geometric Mean)
Shower Duration (Geometric Standard Deviation)
Shower Flowrate
Value
0.6
1.2
1.1
6.8 minutes
1.64 minutes
2.4 gallons/minute
T-11
-------
TABLE 13
Bath Frequency and Duration Values from NHAPS Analyses
Statistic (NHAPS)
Bath frequency (events per
person per day)
Bath Duration
Geometric Mean
(minutes)
Geometric Standard
Deviation (minutes)
Arithmetic Mean
(minutes)
Population
Men
1 8-48 years
0.2
17.2
1.99
20.8
Women
1 8-48 years
0.4
17.8
2.05
21.5
Children
5-1 2 years
0.5
18.6
1.66
20.8
All
Households
0.3
17.6
1.88
20.9
TABLE 14
Bath Volume and Flowrate Values from REUWS Analyses
Statistic for All
Households
(REUWS)
Geometric Mean
Geometric Standard
Deviation
Arithmetic Mean
Bath Flowrate
4.4 gallons/minute
1.71 gallons/minute
4.9 gallons/minute
T-12
-------
TABLE 15
Selected Model Parameters for Bathing
Statistic
Bathing Frequency
per person per day
Bathing Duration
Bath Volume
Bath Fill Duration
Men
1 5-45 years
0.2
20.8 minutes
50 gallons
8 minutes
Women
1 5-45 years
0.4
21.5 minutes
50 gallons
8 minutes
Children
6 years
0.5
20.8 minutes
50 gallons
8 minutes
TABLE 16
Frequency of Clothes Washer Use for 3-Person Households: REGS
Frequency
1 5+ loads/wk
10-15 loads/wk
5-9 loads/wk
2-4 loads/wk
1 load or less/wk
Total
Estimated Mean Frequency
3-Person Family
%
3.4
14.8
50.2
28.8
2.9
100.0
6.7 loads per week
T-13
-------
TABLE 17
Typical Clothes Washer Parameters: Based on REUWS and Experimental Data
Parameter
Cycle 1
Volume
Time to Fill
Time to Agitate
Time to Drain/Spin
Cycles 2, 3 and 4
Volume
Time to Fill
Time to Agitate
Time to Drain/Spin/Spray
Cycle 2 is 100% likely to occur
Cycle 3 is 18.7% likely to occur
Cycle 4 is 0.8% likely to occur
Average Total Time for Washing
Event (for this configuration)
Typical Top-
Loaded
Clothes
Washer
Wash
16.6 gallons
3.8 minutes
12.0 minutes
4.0 minutes
Rinse
15.3 gallons
7.5 minutes
4.0 minutes
8.0 minutes
43.1 minutes
Comments
Mean volume for first fills (REUWS)
Based on experimental data3 on time to fill
for a typical wash cycleb
Based on experimental data on time to
agitate for a typical wash cycleb
Based on experimental data on time to
drain and spin for a typical wash cycleb
Mean volume for second fills (REUWS)
Based on experimental data on time to fill
for a typical rinse cycleb
Based on experimental data on time to
agitate for a typical rinse cycleb
Based on experimental data on time to
drain, spin and spray for a typical rinse
cycle1
Based on REUWS data
Time for 1st cycle (19.8 minutes) plus (1.0 +
0.187 + 0.008) multiplied by time for rinse
cycle (19.5 minutes)
1 Average calculated using only settings to high-water level.
2 For experimental data see Wilkes et al., 2004
T-14
-------
TABLE 18
Selected Model Parameters for Clothes Washer Use
Parameter
Temperature
Value Used in Modeling
35°C
Wash
Fill Duration
Agitation Duration
Volume
3.3 minutes
7.4 minutes
16.6 gallons
Rinse
Fill Duration
Agitation Duration
Volume
Frequency
4.2 minutes
9.8 minutes (5 min. added for spin rinse)
21.0 gallons
1 .0 events per day for 3 person household
Note: The model is currently set up to handle 2 complete cycles. The first event is the
wash cycle, consisting of the wash fill and the wash agitation and drain, the second
event is a combination of all the rinse activities, which are represented as 1.2 rinse
cycles.
T-15
-------
TABLE 19
Frequency of Dishwasher Use for 3-person Households: U
Frequency
Daily
4-6 times/week
Less than 4 times/week
Total
Estimated Mean Frequency
3-Person
.S. DOE (1999)
Family (%)
17.7
29.9
52.4
100.0
3.8
times/week
T-16
-------
TABLE 20
Manufacturer Supplied Dishwasher Information Summary
Condition
Total Volume,
gal
Number of Fills
Average
Volume per Fill,
gal
Dishwasher Model: Whirlpool GU980SCG3
Rinse Only - Heavy Soil
Rinse Only - Light Soil
Quick Wash - Heavy Soil
Quick Wash - Light Soil
China- Heavy Soil
China - Light Soil
Low Energy - Heavy Soil
Low Energy - Light Soil
Normal - Heavy Soil
Normal - Medium Soilb
Normal - Light Soil
Heavy - Heavy Soil
Heavy - Medium Soil
Heavy - Light Soil
4.3
2.2
6.9
4.8
8.6
6.5
8.6
6.5
10.8
8.6
6.9
10.8
10.8
8.6
2
2
2
2
3
3
3
3
3 or 4
3 or 4
3 or 4
5
5
5
2.15
1.1
3.45
2.4
2.87
2.17
2.87
2.17
3.60-2.7
2.87-2.15
2.30-1.725
2.16
2.16
1.72
Dishwasher Model: Whirlpool DU920PFGa
Rinse Only
Low Energy/China
Normalb
Heavy
Pots-N-Pans
2.2
6.5
6.9
8.6
8.6
2
3
3
5
5
1.1
2.17
2.3
1.72
1.72
T-17
-------
TABLE 20 cont.
Condition
Total Volume,
gal
Number of Fills
Average
Volume per Fill,
gal
Dishwasher Model: Whirlpool DU850DWG3
Rinse Only
Light Wash
Normal"
Pots-N-Pans
2.9
5.8
7.2
8.6
2
4
5
6
1.45
1.45
1.44
1.43
Dishwasher Model: GE Potscrubberc
Rinse and Hold
Short Wash
Water Saver Cycle
China/Crystal Cycle
Light Wash Cycle
Normal Wash Cycleb
Potscrubber Cycle
3
7
6.1
7.3
7
8.5
10.1
2
5
4
5
5
6
7
1.5
1.4
1.53
1.46
1.4
1.42
1.44
awhirlpool(S)in-response.com 9/2000
bNormal cycles used for calculations in following table of selected model parameters.
canswerctr(S)exchange.appl.ge.com 2001
T-18
-------
TABLE 21
Selected Model Parameters for Dishwasher Use
Characteristic
Volume of Water
Number of Cycles (without drying)
Volume of Water per Cycle
Duration per Cycle
Frequency
Average*
8.5 gallons
2 Cycles
4.3 gallons
30 minutes
0.5 events per day for 3 person
households
*Based on the average of the "normal" cycles of selected dishwashers
T-19
-------
TABLE 22
Summary of Reported Toilet Use Characteristics from Literature
Toilet Type
Low-Flow
(Avg. 3.6 gpf)
Ultra-low Flow
(rated 1.6 gpf)
Low-Flow
(Avg. 4.0 gpf)
Ultra-low Flow
(rated 1.6 gpf)
Variety of
toilets (33%
low volume
models or
devices)
Reported
Frequency
(fpcd)a
Mean = 3.8
Min=1.8
Max = 8.4
Mean = 4.5
Min = 1.7
Max =12.8
Mean = 3.2
or 12.8
fphdb
Mean = 3.7
or 15.9fphd
Mean = 4.0
Volume
(gal/flush)
Mean =3.6
Min = 1.7
Max = 5.6
Mean = 1.6
Min = 1.1
Max =3.0
Mean = 4.0
Mean = 1.8
Min = 1.3
Max = 2.4
Population/
Sample Size
Tampa,
Florida,
25 single family
homes
Tampa, Florida
25 single family
homes
Oakland,
California,
25 single family
homes
Oakland,
California,
25 single family
homes
CA, CO, D.C.,
VA, WA, 196
households,
545 persons,
356 toilets
Reference
Konen
and
Anderson,
1993
Konen
and
Anderson,
1993
Aher et
al.,1991
Aher et
al.,0ct.
1991
HUD,
1984
Special
Study
Conditions
Study
comparison
of low flow
to ultra-low
flow retrofit
(average
2.9
persons/
home)
Study
comparison
of low flow
to ultra-low
flow retrofit
(average
4.4
persons/
home)
Study
subjects
recorded
toilet flush
counts
fpcd: Flushes per capita day
bfphd: Flushes per home per day
T-20
-------
TABLE 23
Statistics for Toilet Flushes from REUWS
Minimum
Maximum
Mean
Standard
Deviation
Number of
Records or
Households3
All Flushes
Frequency
(flushes/
person/day)
0.0
42.7
5.2
3.15
2,145b
Family
Size
0.0
9.0
2.8
1.37
2,158
Sampling
Days
1.0
16.0
10.7
1.63
2,158
Single Flushes Only
Duration
of Tank
Fill
(seconds)
10.0
2,720.0
71.4
29.77
245,328
Volume
(gallons)
0.3
9.8
3.5
1.18
245,331
Mode Flow
(gallons per
minute)
0.0
14.1
3.9
1.31
245,331
Number of households reflects the combined total of homes participating in the first
sampling period (1,173) and second sampling period (985).
b13 surveys indicated "0" for Q.31 or Q.30 regarding the number of people in selected
age groups (households aggregated from 295,660 records).
TABLE 24
Selected Parameters for Toilet Use
Statistics
Frequency
Volume of water used per flush
Value
6 flushes/person/day
3.5 gallons/flush
Note: model assumes instant filling
T-21
-------
TABLE 25
Summary of Reported Faucet Frequency and Volume of Use Characteristics in
Literature
Type of
Appliance
Conventional
Conventional
Conventional
Location
Kitchen
Bathroom
Not given
Frequency
Not given
Not given
Not given
Volume
(gpm)
Maximum flow3
Mean = 2.4
Min = 1.5
Max = 3.8
Maximum flow3
Mean = 3.4
Min = 0.9
Max = 7.9
9.0 gal/person/
dayb
Population/
Sample Size
Tampa, Florida,
25 single family
homes (avg.
2.9
persons/home)
Tampa, Florida,
25 single family
homes (avg.
2.9
persons/home)
Nationwide
Reference
Konen and
Anderson,
March
1993
Konen and
Anderson,
March
1993
HUD, 1984
3Measured flowrates with faucets in fully open position
Estimated value
T-22
-------
TABLE 26
Summary Statistics for Faucet Use from REUWS
Minimum
Maximum
Mean
Standard Deviation
Number of Records
Duration
(minutes)
0.0
90.0
0.6
0.76
973,717
Volume
(gallons)
0.01
37.6
0.7
0.98
973,717
Mode
Flow
(gpm)
0.0
10.7
1.2
0.68
973,717
Frequency of Use per
day per person
2.3
143.0
17.4
11.6
965 (households)
TABLE 27
Selected Parameters for Faucet Use
Statistic
Faucet Use Duration
Flowrate
Frequency of Faucet Use
Value
Range from 1.1 to 1.7 minutes
1 .20 gallons per minute
15.5 events per person per day
T-23
-------
TABLE 28
Tap Water Consumption Characteristics
Population
Average Consumption (units)
Canadian Department of Health3: 970 individuals, 295 households
Children, 3-5 Years
Children, 6-17 Years
Females, 18-34 Years
Females, 35-54 years
Males, 1 8-54 Years
Average Daily Consumption, (All)
90th Percentile
48 ml/kg
26 ml/kg
23 ml/kg
25 ml/kg
19 ml/kg
1.34L/day
2.36 L/day
1978 Drinking Water Consumption in Great Britain13: N = 3,564 People
Females, 5-1 1 Years
Females, 18-30 Years
Females, 31-54 Years
Males, 5-1 1 Years
Males, 1 8-30 Years
Males, 31-54 Years
0.533 L/day
0.991 L/day
1.091 L/day
0.550 L/day
1.006 L/day
1.201 L/day
1987 National Cancer Institute Study0: N = 8,000 White Adults
Females, 21-84 Years
Males, 21-84 Years
Females and Males, 18-44 Years
1.35 L/day
1 .4 L/day
1.3 L/day
1977 - 78 USDA Nationwide Food Consumption Survey (NFCS)d: N = 26,000
Adults, 20 to 75 or older Years
90th Percentile
1 .2 L/day
2.1 L/day
T-24
-------
TABLE 28 cont.
Population
Adults, 15-1 9 Years6
Adults, 20-44 Years6
Children, 4-6 Years6
Pregnant Womenf
Lactating Womenf
Non-Pregnant, Non-Lactating Women,
15-49Yearsf
Average Consumption (units)
999 ml/day (N = 2998)
1, 255 ml/day (N = 71 71)
37.9 ml_/kg-day(N = 1702)
2,076 ml/day (N = 188)
2,242 ml/day (N = 77)
1, 940 ml/day (N = 6201)
All references discussed and cited in Exposure Factors Handbook, U.S. EPA (1997a)
aCanadian Ministry of National Health and Welfare (1981)
bHopkins and Ellis (1980)
cCantoretal. (1987)
dErshow and Cantor (1989)
6Ershow and Cantor (1989)
trshowetal. (1989)
T-25
-------
TABLE 29
Parameters of Fitted Lognormal Distribution for Water Ingestion in the United States
Population
Women, direct (20+ years)
Women, indirect (20+ years)
Men, direct (20+ years)
Men, indirect (20+ years)
Children, direct (1-10 years)
Children, indirect (1-10 years)
All ages, direct
All ages, indirect
Geometric Mean
ml/day
394
384
389
418
188
97
321
290
Geometric Standard
Deviation
2.52
2.20
2.69
2.33
2.50
2.51
2.79
2.53
Source: Fitted to data from Table A1 in U.S. EPA (2000d)
T-26
-------
TABLE 30
Comparison of Consumption for Raw Data and Fitted Distributions based on CSFII Data
Percentile
1
5
10
50
90
95
99
Men, 20+ years
Direct
Consumption
(ml/d)
CO
"CD
Q
—
—
—
352
1,450
1,891
3,773
Fitted
Lognormal
Distribution
39
77
110
390
1,380
1,980
3,897
Indirect
Consumption
(ml/d)
CD
"CD
Q
—
—
—
412
1,210
1,597
3,094
Fitted
Lognormal
Distribution
58
104
142
419
1,235
1,682
3,000
Women, 20+ years
Direct
Consumption
(ml/d)
CD
"CD
Q
—
—
—
349
1,395
1,865
3,062
Fitted
Lognormal
Distribution
46
86
121
394
1,285
1,799
3,386
Indirect
Consumption
(ml/d)
CD
"CD
Q
—
—
—
365
1,080
1,394
2,367
Fitted
Lognormal
Distribution
61
105
140
385
1,057
1,410
2,421
Children, 1-10 years
Direct
Consumption
(ml/d)
CD
"CD
Q
—
—
—
174
696
919
1,415
Fitted
Lognormal
Distribution
22
42
58
189
611
854
1,601
Indirect
Consumption
(ml/d)
CD
"CD
Q
—
—
—
84
352
457
734
Fitted
Lognormal
Distribution
11
21
30
97
316
441
828
Total Population
Direct
Consumption
(ml/d)
CD
"CD
Q
—
—
—
290
1,270
1,769
3,240
Fitted
Lognormal
Distribution
30
60
87
322
1,193
1,734
3,499
Indirect
Consumption
(ml/d)
CD
"CD
Q
—
—
—
262
1,008
1,334
2,373
Fitted
Lognormal
Distribution
33
63
88
290
952
1,336
2,523
T-27
-------
TABLE 31
Selected Parameters for Tapwater Consumption Modeling Study
Statistic
Man (age 1 5-45 years)
Direct
Consumption
Indirect
Consumption
Woman (age 15-45 years)
Direct
Consumption
Indirect
Consumption
Child (age 6 years)
Direct
Consumption
Indirect
Consumption
Volume (Liters/day)
Geometric Mean
Geometric Standard
Deviation
0.3895
0.988
0.419
0.8449
0.3943
0.9228
0.3848
0.4894
0.1889
0.9173
0.0974
0.9187
Duration (minutes to consume water)
Geometric Mean
Geometric Standard
Deviation
Arithmetic Mean
Arithmetic Standard
Deviation
Mean Frequency
Time of Day
2.236
1.269
5
10
8
5 am-10 pm
3.162
1.517
10
30
8
5 am-10 pm
2.236
1.269
5
10
8
5 am-10 pm
3.162
1.517
10
30
8
5 am-10 pm
2.236
1.269
5
10
8
5 am-10 pm
3.162
1.517
10
30
8
5 am-10 pm
T-28
-------
TABLE 32
Analysis of REGS for Total House Volume for 3-Person U.S. Households
(U.S. DOE, 1999)
Percentile
4.1
22.3
60.4
79.7
90.5
96.6
Area, ft2
0-600
601-999
1000-1599
1600-1999
2000-2399
2400-2999
Area, m2
55.7
92.8
148.6
185.7
222.9
278.6
Volume, ft3
4800
7992
12792
15992
19192
23992
Volume, m3*
135.9
226.3
362.3
452.9
543.5
679.5
*Volumes were calculated by assuming an 8 ft ceiling height
T-29
-------
TABLE 33
Estimated Range of Dimensions of Water-Use Zones Based on Hoke (1988, 1994)
Zone
Hall Bath
Master Bath
Kitchen
Laundry
Shower
Dimension
Area (m2)
Volume (m3)
Area (m2)
Volume (m3)
Area (m2)
Volume (m3)
Area (m2)
Volume (m3)
Area (m2)
Volume (m3)
Small
3.2
7.9
2.0
4.9
6.3
15.4
5.5
13.5
1.2
2.9
Large
6.1
14.9
3.5
8.5
7.4
18.1
10.4
25.4
1.8
4.5
Source: Hoke (1988, 1994)
T-30
-------
TABLE 34
Summary Statistics for U.S. Residential Air Exchange Rates
Arithmetic Mean (h~1)
Arithmetic Standard
Deviation (h~1)
Geometric Mean (h~1)
Geometric Standard
Deviation
10thPercentile(IT1)
50thPercentile(IT1)
90thPercentile(IT1)
Maximum (h~1)
West
Region
0.66
0.87
0.47
2.11
0.20
0.43
1.25
23.32
North
Central
Region
0.57
0.63
0.39
2.36
0.16
0.35
1.49
4.52
Northeast
Region
0.71
0.60
0.54
2.14
0.23
0.49
1.33
5.49
South
Region
0.61
0.51
0.46
2.28
0.16
0.49
1.21
3.44
All
Regions
0.63
0.65
0.46
2.25
0.18
0.45
1.26
23.32
Source: U.S. EPA(1997b)
T-31
-------
TABLE 35
Summary of Volume and Ventilation Parameters for Case 48
Parameter
WHACH
Whole House Volume
Laundry Volume
Kitchen Volume
Hall Bath Volume
Master Bath Volume
Shower Volume
ROH Volume
ROH to Outdoor Flowrate
Laundry to ROH Flowrate
Kitchen to ROH Flowrate
Hall Bath to ROH Flowrate
Master Bath to ROH Flowrate
Shower to Master Bath Flowrate
Value
0.11
311.1
18.3
17.3
12.6
8.4
4.4
250.0
35.4
2.08
1.96
1.43
0.95
50
Units
hr1
m3
m3
m3
m3
m3
m3
m3
m3/hr
m3/hr
m3/hr
m3/hr
m3/hr
m3/hr
TABLE 36
Pearson Correlation Coefficients between Chloroform and BDCM, DBCM, and
Bromoform based on all samples in the 85th to 95th Percentile in the Cumulative
Chloroform Concentrations
BDCM
DBCM
Bromoform
CHLOROFORM
Pearson Correlation
0.528
0.001
Sig. (2-tailed)
N
0.000
15974
0.857
15950
-0.183
0.000
15926
T-32
-------
TABLE 37
Pearson Correlation Coefficients between BDCM and Chloroform, DBCM, and
Bromoform based on all samples in the 85th to 95th Percentile in the Cumulative BDCM
Concentrations
BDCM
Pearson Correlation
Sig. (2-tailed)
N
Chloroform
0.528
0.000
15974
DBCM
0.685
0.000
16089
Bromoform
0.149
0.000
16057
TABLE 38
Pearson Correlation Coefficients between DBCM and Chloroform, BDCM, and
Bromoform based on all samples in the 85th to 95th Percentile in the Cumulative
DBCM Concentrations
DBCM
Pearson Correlation
Sig. (2-tailed)
N
Chloroform
0.001
0.857
15950
BDCM
0.685
0.000
16089
Bromoform
0.640
0.000
16022
T-33
-------
TABLE 39
Pearson Correlation Coefficients between Bromoform and Chloroform, BDCM, and
DBCM based on all samples in the 85th to 95th Percentile in the Cumulative Bromoform
Concentrations
Bromoform
Pearson Correlation
Sig. (2-tailed)
N
Chloroform
-0.183
0.000
15926
BDCM
0.149
0.000
16057
DBCM
0.640
0.000
16022
T-34
-------
TABLE 40
95th Percentile Chloroform (66 ppb) Values for ICR Surface Water Treatment Plants
EVENT ID
1406409
1439209
6755111
2041103
3265607
4030312
676561 1
1417215
3277801
5855003
5589303
5400913
4075701
5401316
CHCI3
66
66
66
66
66
66
66
66
66
66
66
66
66
66
BDCM
2.6
2.8
10
11
12
9.3
11
14
15
19
19.3
22
29
24
DBCM
0.5
0.5
1
1.2
1.3
1.4
1.6
3.1
3.6
4.1
5
6.5
12
11
CHBr3
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1.1
Disinfectant
CL2
CL2_CLM
CL2_CLM
CL2
CLX
CL2
CL2
03
CLX
CL2
CL2_CLM
CL2_CLM
CL2
CL2_CLM
Utility Name
East Bay Municipal Utility
District
East Bay Municipal Utility
District
Newport News Waterworks
City of Sacramento
Cobb County-Marietta Water
Auth
City of New Bedford Water
Depart
City of Norfolk, Dept. of
Utilities
East Bay Municipal Utility
District
Cobb County-Marietta Water
Auth
PRASA - Aguadilla Urbano
Philadelphia Water
Department
Lawton Water Treatment
Plant
Washington Suburban
Sanitary Com.
Lawton Water Treatment
Plant
Note: the shaded row indicates the row selected based on the stated criteria.
T-35
-------
TABLE 41
95th Percentile BDCM (23.8 ppb) Values for ICR Surface Water Treatment Plants
Event ID
5971217
5971117
6783101
CHCI3
26.1
26.5
59.5
BDCM
23.8
23.8
23.8
DBCM
17.7
17.4
5.1
CHBr3
2.6
2.6
0.5
Disinfectant
CLM
CLM
CL2
Utility Name
Charleston
Charleston
City of Portsmouth, DPU
Note: the shaded row indicates the row selected based on the stated criteria.
T-36
-------
TABLE 42
95th Percentile DBCM (17 ppb) Values for ICR Surface Water Treatment Plants
Event ID
6380918
6381018
6385218
1461008
6385109
6385312
6381009
6385303
6385203
6261102
6411009
6571607
6571707
6571310
6540810
6540710
5640602
2085603
2165004
2161013
6392606
6392206
2165013
6393006
1715402
CHCI3
1.2
1.4
1.7
1.3
3.7
2.3
4.4
3.1
3
4.6
6
8.7
6
7.4
8.1
8.6
10.2
11
12
12
13
13
14
14
17
BDCM
5.2
5.3
5.7
6
7.8
7.9
8.9
9.1
9.2
9.3
9.7
10
11
13
13
13
14.7
16
18
20
20
20
20
20
20
DBCM
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
CHBr3
11
11
11
17
7.9
9.9
7.6
8.3
8.1
8.6
24
7.8
8.2
6.4
6.8
6.1
5
4.3
4.1
3.4
3.3
3.6
3.3
3.3
3
Disinfectant
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CLX
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2_CLM
CL2
CL2
CL2_CLM
CL2_CLM
CLM
CLM
CL2_CLM
CLM
CL2_CLM
Utility Name
City of Lubbock Water Utilities
City of Lubbock Water Utilities
City of Lubbock Water Utilities
Contra Costa Water District
City of Lubbock Water Utilities
City of Lubbock Water Utilities
City of Lubbock Water Utilities
City of Lubbock Water Utilities
City of Lubbock Water Utilities
El Paso Water Utilities
City of Corpus Christ!
City of Laredo, Texas
City of Laredo, Texas
City of Laredo, Texas
City of Austin
City of Austin
Pittsburgh Water and Sewer Authority
Cucamonga County Water District
Helix Water District
Helix Water District
City of Waco Utility Department
City of Waco Utility Department
Helix Water District
City of Waco Utility Department
Metro Water Dist of So Calif
T-37
-------
TABLE 42 cont.
Event ID
6545404
6391406
1295114
1720805
1295211
2155102
6401215
6401615
6401715
5971214
2155202
1271116
1271216
2190703
2155105
6392412
1271513
2155005
6393012
1242211
1271610
3790804
6402418
1261813
CHCI3
20
13
14
15
24
20
22
22
22
27.9
21
29
31
23
23
26
29
26
28
45
39
58.7
27
140
BDCM
20
21
21
21
21
23
23
23
23
23.9
24
25
25
26
26
26
26
27
27
27
28
31.7
34
44
DBCM
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
CHBr3
4.4
3.5
4.1
2.4
3.8
3.2
3.9
4.1
4.1
2.4
3
2
1.9
1.7
2.7
1.8
2.3
2.8
1.6
4.2
2
1.5
1.8
3.7
Disinfectant
CL2_CLM
CLM
CL2
CL2_CLM
CL2
CL2
CLM
CLM
CLM
CLX
CL2
CL2
CL2
CL2_CLM
CL2
CLM
CL2
CL2
CLM
CL2
CL2
CL2
CLM
CL2
Utility Name
City of Austin
City of Waco Utility Department
City of Scottsdale
Metro Water Dist of So Calif
City of Scottsdale
City of Escondido
City of Waco Utility Department
City of Waco Utility Department
City of Waco Utility Department
Charleston
City of Escondido
City of Glendale
City of Glendale
City of San Diego Water Utilities
City of Escondido
City of Waco Utility Department
City of Glendale
City of Escondido
City of Waco Utility Department
Chandler Municipal Water Department
City of Glendale
Evansville Water & Sewer Utility
City of Waco Utility Department
City of Glendale
Note: the shaded row indicates the row selected based on the stated criteria.
T-38
-------
TABLE 43
95th Percentile Bromoform (5.6 ppb) Values for ICR Surface Water Treatment Plants
Event ID
6530613
6145313
6571716
1201012
1285101
5609803
1321404
1321604
CHCI3
5.9
1.4
7.1
35.9
10.1
23.4
13
14
BDCM
12.4
4
15
24.4
15.3
30.4
25
25
DBCM
5.4
6.2
13
17.2
20.9
25.4
26
26
CHBr3
5.6
5.6
5.6
5.6
5.6
5.6
5.6
5.6
Disinfectant
CLX
CLX
CL2_CLM
CL2
CLX
CL2_CLM
CL2
CL2
Utility Name
City of Abilene Water Utilities
Brownsville Public Utilities
Board
City of Laredo, Texas
Phoenix Municipal Water
System
City of Mesa, Utility
Philadelphia Water
Department
City of Tempe
City of Tempe
Note: the shaded row indicates the row selected based on the stated criteria.
T-39
-------
TABLE 44
Summary of THM Concentrations Paired with the 95th Percentile for Each THM for All
ICR Samples
Variable
Subgroup
All
Reported
Samples
Description
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
MEAN
Number of Reported Samples
Concentration, ppb (percentile)
Chloroform
65.9 (95)
10.0 (37)
120.0 (1)
36.0 (79)
21.9
15987
BDCM
18.1 (90)
23.0 (95)
58.0 (1)
46.0 (1)
8.2
16144
DBCM
3.4 (65)
42.0 (1)
16.0 (95)
37.0 (1)
4.1
16109
Bromoform
0.5 (0)
24.0 (99)
0.5 (0)
6.1 (95)
1.6
16103
Note: The ICR database contained 18,214 records of analyzed THM samples. Some
records were incomplete for a variety of quality control reasons. No records containing
reported values were excluded from this analysis.
T-40
-------
TABLE 45
Description of Variables Used in Analysis and Their Associated Attributes
ICR
Database
Variable
Name
MSRC_CAT
WTP_DIS
SAMP_QTR
Variable Definition
Characterization of the
plant water resource type
Categorical description of
disinfection practices in
the treatment plant
Definition of the sampling
quarter
Variable Attribute to be Analyzed
o Surface Water Intake
o Ground Water Intake
o Ozonation
o All chlorine-based disinfection
systems
o Chlorine Only
o Chloramine
o Chlorine dioxide
o Chlorine followed by chloramine
o July-September of 1997 & 1998
o October-December of 1997 &
1998
o January-March 1998
o April-June 1998
T-41
-------
TABLE 46
Summary of THM Concentrations Paired with the 95th Percentile for the Analyzed THM Based
on Analysis of the ICR Database.
Variable
Subgroup
All Systems
Using Surface
Water Intake
(N = 12,440)
All Systems
Using Ground
Water Intake
(N = 4,318)
All Systems
Using
Ozonation as
Primary
Disinfectant
(N = 645)
Systems Using
any type of
Chlorine
Disinfectant
Process (N -
14,015)
THM Analysis
Description
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Concentration, ppb (Percentile)
Chloroform
66.0 (95)
26.1 (62)
140.0(100)
14.0 (34)
44.8 (95)
42.6 (94)
0.5 (0)
3.5 (64)
65.2 (95)
71.0 (95)
40.0 (87)
2.0 (27)
66.0 (95)
9.1 (26)
140.0 (100)
5.5 (18)
BDCM
29.0 (98)
23.8 (95)
44.0 (100)
25.0 (96)
7.2 (83)
15.3 (95)
3.6 (67)
9.0 (87)
16.0 (91)
20.0 (95)
18.0 (93)
5.3 (57)
29.0 (98)
24.0 (95)
44.0 (100)
12.3 (76)
DBCM
12.0(90)
17.7 (95)
17.0 (95)
26.0 (98)
0.5 (0)
5.1 (84)
12.0 (95)
17.0 (98)
2.5 (49)
4.9 (76)
9.7 (95)
7.8 (90)
12.0 (90)
46.0(100)
17.0 (95)
6.0 (78)
Bromoform
0.5 (0)
2.6 (89)
3.7 (92)
5.6 (95)
0.5 (0)
0.5 (0)
23.0 (99)
8.3 (95)
0.5 (0)
0.5 (0)
0.5 (0)
5.8 (95)
0.5 (0)
34.0 (100)
3.7 (91)
6.0 (95)
T-42
-------
TABLE 46 cont.
Variable
Subgroup
Systems Using
Chlorine as
Primary
Disinfectant (N
,137)
Systems Using
Chlorine
followed by
Chloramine (N
o O/l O\
- 6,242)
Systems Using
Chloramine as
Primary
Disinfectant
Systems Using
Chlorine
Dioxide as
Primary
Disinfectant
THM Analysis
Description
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Concentration, ppb (Percentile)
Chloroform
69.1 (95)
9.1 (27)
14.0 (38)
21.0 (52)
59.7 (95)
13.0 (28)
35.0 (77)
22.0 (54)
47.2 (95)
30.0 (87)
27.0 (83)
0.5 (0)
70.0 (95)
16.0 (53)
5.8 (31)
11.0 (42)
BDCM
24.9 (96)
24.0 (95)
6.6 (48)
29.0 (98)
9.4 (53)
24.0 (95)
40.0 (100)
37.0 (99)
13.3 (77)
27.6 (95)
34.0 (98)
1.5 (15)
8.7 (56)
22.0 (95)
10.5 (67)
24.0 (97)
DBCM
5.8 (82)
46.0(100)
15.0 (95)
20.0 (97)
0.5 (0)
33.0 (99)
20.0 (95)
31.0 (99)
1.8 (46)
21.0 (97)
17.2 (95)
5.1 (59)
0.5 (0)
23.0 (97)
18.4 (95)
35.0 (99)
Bromoform
0.5 (0)
34.0 (100)
45.0 (100)
4.2 (95)
0.5 (0)
11.0 (97)
3.3 (89)
6.6 (95)
0.5 (0)
2.3 (74)
1.8 (69)
12.0 (95)
0.5 (0)
9.9 (90)
5.3 (79)
13.9 (95)
T-43
-------
TABLE 46 cont.
Variable
Subgroup
Systems
Sampled
between July
and September
(1997 and
1998)
Systems
Sampled
between
October and
December
(1997 and
1998)
Systems
Sampled
between
January and
March (1998)
Systems
Sampled
between April
and July (1998)
THM Analysis
Description
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th Percentile
DBCM 95th Percentile
Bromoform 95th
Percentile
Concentration, ppb (Percentile)
Chloroform
74.8 (95)
14.0 (40)
170.0 (100)
30.0 (64)
55.0 (95)
22.1 (67)
38.8 (87)
20.0 (62)
49.0 (95)
11.0 (46)
1.5 (16)
14.1 (57)
70.3 (95)
24.0 (59)
60.0 (92)
75.0 (96)
BDCM
17.6 (84)
27.0 (95)
70.0 (100)
37.0 (99)
29.0 (99)
21.7 (95)
30.8 (99)
36.0 (99)
32.0 (99)
19.0 (95)
5.2 (50)
23.0 (98)
6.9 (55)
21.0 (95)
23.0 (96)
37.0 (100)
DBCM
4.0 (65)
43.0(100)
20.0 (95)
33.0 (99)
17.0 (97)
14.7 (95)
14.6 (95)
30.0 (99)
5.3 (80)
20.0 (98)
14.1 (95)
16.9 (97)
0.5 (0)
17.0 (97)
14.0 (95)
32.0(100)
Bromoform
0.5 (0)
23.0 (99)
1.6 (77)
6.5 (95)
2.3 (86)
2.6 (87)
2.3 (86)
6.4 (95)
0.5 (0)
5.5 (95)
40.0 (100)
5.3 (95)
0.5 (0)
3.8 (92)
5.8 (95)
5.0 (95)
T-44
-------
TABLE 47
Chemical Properties of Compounds (24°C) Studied by Howard and Corsi (1996)
Chemical
Cyclohexane
Toluene
Acetone
H (unitless)
7.1
0.27
0.0012
DL (cm2/sec)a
9.0 E -6
9.1 E-6
1.1 E-5
DG (cm2/sec)b
0.088
0.085
0.11
aDL is estimated using the Hayduk and Laudie method (Lyman et al., 1990, pp 17-20).
bDG is estimated using the Wilke and Lee method (Lyman et al., 1990, pp 17-13).
TABLE 48
Chemical properties of Compounds Being Modeled (24° C)
Chemical
Chloroform
BDCM
DBCM
Bromoform
H (unitless)
0.15
0.088
0.038
0.021
DL (cm2/sec)a
1.03 E-5
1.01 E-5
9.96 E -6
9.82 E -6
DG (cm2/sec)b
0.094
0.089
0.086
0.083
DL is estimated using the Hayduk and Laudie method (Lyman et al., 1990, pp 17-20)
DDG is estimated using the Wilke and Lee method (Lyman et al., 1990, pp 17-13).
T-45
-------
TABLE 49
Summary of Experimental Stripping Efficiencies for Cyclohexane, Toluene,
and Acetone
Flowrate
4.8
7.9
4.8
7.9
4.8
6.3
7.9
Aerator
None
None
Screen
Screen
Bubble Aerator
Bubble Aerator
Bubble Aerator
Stripping Efficiency (%)*
Cyclohexane
24
19
19
18
33
35
44
Toluene
21
17
13
14
23
22
23
Acetone
4.9
2.2
1.7
1.1
1.4
1.5
1.6
*Measured by Howard and Corsi (1996) for Kitchen Sink Experiments; water
temperature approximately 23°C.
T-46
-------
TABLE 50
Estimated Rate Constants for Removal of THMs from a Storage Container Based on Batterman et al.
Condition
Tall glass, full, water at 4°C
Tall glass, full, water at 25°C
Tall glass, half full, water at 25°C
Wide mouth glass, full, water at 25°C
Tall glass, full, water at 30°C
Tall glass, half full, water at 30°C
Wide mouth glass, full, water at 30°C
Coffee mug, full, water at 100°C
Chloroform
k(h-1)
0.088
0.055
0.070
0.180
0.183
0.248
0.411
1.50
R2
0.77
0.63
0.77
0.59
0.69
0.83
0.62
0.86
BDCM
k(h-1)
0.076
0.046
0.064
0.110
0.135
0.205
0.427
1.52
R2
0.78
0.53
0.64
0.30
0.65
0.90
0.80
0.82
DBCM
k(h-1)
0.080
0.047
0.063
0.108
0.142
0.177
0.392
1.41
R2
0.75
0.47
0.76
0.61
0.74
0.90
0.82
0.80
Bromoform
k(h-1)
0.080
0.044
0.062
0.140
0.158
0.193
0.332
1.40
R2
0.84
0.33
0.56
0.71
0.85
0.89
0.76
0.85
T-47
-------
TABLE 51
Estimated Fractional Volatilization from a Storage Container as a Function of Time forTHMs for Cold, Room Temperature, and Hot Water
Condition
Cold Water
(4'C)
Room Temp
(25°C)
Hot Water
(100°C)
Chemical
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Rate
Const,
k (h-1)
0.09
0.076
0.080
0.080
0.18
0.11
0.108
0.14
1.50
1.52
1.41
1.40
Fraction Volatilized
Time, minutes
0
0
0
0
0
0
0
0
0
0
0
0
0
5
0.007
0.006
0.07
0.07
0.015
0.009
0.009
0.012
0.12
0.12
0.11
0.11
10
0.015
0.013
0.013
0.013
0.030
0.018
0.018
0.023
0.22
0.22
0.21
0.21
15
0.022
0.019
0.020
0.020
0.044
0.027
0.027
0.034
0.31
0.32
0.30
0.30
30
0.044
0.037
0.039
0.039
0.086
0.054
0.053
0.068
0.53
0.53
0.51
0.50
60
0.086
0.073
0.077
0.077
0.16
0.104
0.102
0.13
0.78
0.78
0.76
0.75
75
0.11
0.091
0.095
0.095
0.20
0.13
0.13
0.16
0.85
0.85
0.83
0.83
90
0.13
0.11
0.11
0.11
0.24
0.15
0.15
0.19
0.89
0.90
0.88
0.88
105
0.15
0.13
0.13
0.13
0.27
0.18
0.17
0.22
0.93
0.93
0.92
0.91
120
0.16
0.14
0.15
0.15
0.30
0.20
0.19
0.24
0.95
0.95
0.94
0.94
180
0.24
0.20
0.21
0.21
0.42
0.28
0.28
0.34
0.99
0.99
0.99
0.99
240
0.30
0.26
0.27
0.27
0.51
0.36
0.35
0.43
1.0
1.0
1.0
1.0
360
0.42
0.37
0.38
0.38
0.66
0.48
0.48
0.57
1.0
1.0
1.0
1.0
420
0.47
0.41
0.43
0.43
0.72
0.54
0.53
0.62
1.0
1.0
1.0
1.0
480
0.51
0.46
0.47
0.47
0.76
0.59
0.58
0.67
1.0
1.0
1.0
1.0
T-48
-------
TABLE 52
THM Consumption Scenarios
Scenario
Glass of water, room
temperature, immediate
consumption (over 5-10
minutes)
Glass of water, room
temperature,
consumption over 1 hour
Glass of ice water,
immediate consumption
(over 5-10 minutes)
Glass of ice water,
consumption over 1 hour
Hot beverage (e.g.,
coffee or tea), consumed
immediately (over 5-10
minutes)
Chemical
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Fraction Volatilized
Filling
0.12
0.075
0.044
0.035
0.12
0.075
0.044
0.035
0.12
0.075
0.044
0.035
0.12
0.075
0.044
0.035
0.12
0.075
0.044
0.035
Storage3
0.013
0.008
0.008
0.010
0.084
0.053
0.052
0.067
0.007
0.006
0.006
0.006
0.044
0.037
0.039
0.039
0.11
0.11
0.11
0.11
Processing
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.85C
0.80C
0.72C
0.63C
Totalb
0.13
0.08
0.05
0.04
0.19
0.12
0.09
0.10
0.13
0.08
0.05
0.04
0.16
0.11
0.08
0.07
0.88
0.84
0.76
0.68
T-49
-------
TABLE 52 cont.
Scenario
Hot beverage (e.g.,
coffee or tea), consumed
immediately (over 20
minutes)
Prepared and stored
beverages (e.g., pitcher
of orange juice),
prepared, stored cold
(assume average = 4
hours), poured,
consumed over 5-10
minutes
Prepared and stored
beverages (e.g., pitcher
of orange juice),
prepared, stored cold
(assume average = 4
hours), poured,
consumed over 30
minutes
Chemical
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Chloroform
BDCM
DBCM
Bromoform
Fraction Volatilized
Filling
0.12
0.075
0.044
0.035
0.1 2d
0.1 2e
0.075d
0.0756
0.044d
0.0446
0.035d
0.0356
0.1 2e
0.1 2e
0.075d
0.0756
0.044d
0.0446
0.035d
0.0356
Storage3
0.23
0.23
0.22
0.22
0.29f
0.0079
0.25f
0.0069
0.26f
0.0069
0.26f
0.0069
0.29f
0.029
0.25f
0.029
0.25f
0.029
0.26f
0.029
Processing
0.85 g
0.80 g
0.72 g
0.63 g
0
0
0
0
0
0
0
0
Totalb
0.90
0.86
0.79
0.72
0.38
0.36
0.33
0.32
0.39
0.37
0.33
0.32
Calculated using weighted averages for the appropriate time categories, with fractional
volatilization as given in Table 48.
bTotal is calculated in a consecutive manner by multiplying fraction remaining after each
activity (i.e., for coffee, hot, consumed immediately; the initial concentration is reduced
for filling by 18% to yield 82%, then the 82% is reduced by 85% because of heating to
yield 12.3%, and finally the 12.3% is reduced by 23% to account for storage losses to
yield 9%, or a fractional volatilization of 0.91).
cTaken from Batterman et al. (2002).
Volatilization attributed to preparation.
^Volatilization attributed to pouring from the pitcher into the glass.
Volatilization attributed to storage in the pitcher.
9Volatilization while in the glass.
T-50
-------
TABLE 53
Recommended Consumption Model Inputs for the THMs
Chemical
Chloroform
BDCM
DBCM
Bromoform
Average Fraction
Remaining Prior to
Storage or Consumption
Direct
0.80
0.90
0.95
0.95
Indirect
0.15
0.2
0.25
0.3
Volatilization Rate Constant
(h-1)
Direct
0.07
0.06
0.06
0.06
Indirect
0.4
0.4
0.4
0.4
TABLE 54
Alveolar Ventilation Rates by Demographic Group and Activity
Activity Level
Rest
Sedentary
Alveolar Ventilation Rate (Liters/Hour)*
Male
(Age 15-45)
540
600
Female
(Age 15-45)
430
480
Child
(Age 6)
410
435
Trom Exposure Factors Handbook, Table 5-6, U.S. EPA (1997b)
T-51
-------
TABLE 55
Skin Permeability Coefficients
Chemical
Name
Chloroform
BDCM
DBCM
Bromoform
Kp
(cm/hr)
(measured)
0.13
—
—
—
Kp
(cm/hr)
(Krishnan, 2001)
0.0156-0.0393
0.0184-0.0478
0.0215-0.0577
0.0247-0.0681
Kpa
(cm/hr)
(est. possible
range)
0.015-0.15
0.018-0.18
0.021-0.22
0.24-0.25
Kpb
Value Used
as Model
Input
(cm/hr)
0.13
0.0331
0.0396
0.0464
Range of possible Kp values estimated based on predictions and on
measured/predicted values for other compounds in the same class. For classes other
than the THMs, no measurements have been identified, so the range itself is somewhat
uncertain.
bThe midpoint of the estimate range by Krishnan was used unless alternative
information was available.
T-52
-------
TABLE 56
Partition Coefficients Required for Fundamental Uptake Modeling in TEM
Chemical Name
Chloroform
BDCM
DBCM
Bromoform
Skin/Blood
1.62a
2.0C
3.82C
5.51C
Blood/Air
11.34(adult)b
12.41 (child)b
26.6d
49.2d
102.3d
Estimates for CHCI3 from Corley et al. (1990)
bData from U.S. EPA (2003)
Estimates from Krishnan (2001) and Lipscomb (2001)
dData from Batterman et al. (2002)
T-53
-------
TABLE 57
Definition of Some Terms Commonly Used in PBPK Modeling
Qaiv - alveolar ventilation
dnh - concentration of agent in inhaled air
Caiv - concentration of agent in exhaled air
Qc - cardiac output
CART - concentration of the agent in arterial blood; CA
QF - fraction of the cardiac output directed to the fat compartment
Qs - fraction of the cardiac output directed to the slowly perfused tissue compartment
QR - fraction of the cardiac output directed to the rapidly perfused tissue compartment
QT - fraction of the cardiac output directed to the testes
Qo - fraction of the cardiac output directed to the ovaries
QK - fraction of the cardiac output directed to the kidneys
QL - fraction of the cardiac output directed to the liver
QP - fraction of the cardiac output directed to the pulmonary (alveolar) region; Qc
CVF - concentration of the agent in venous blood leaving the fat compartment
Cvs - concentration of the agent in venous blood leaving the slowly perfused tissue
compartment
CVR - concentration of the agent in venous blood leaving the rapidly perfused tissue
compartment
CVT - concentration of the agent in venous blood leaving the testes
Cvo - concentration of the agent in venous blood leaving the ovaries
CVK - concentration of the agent in venous blood leaving the kidneys
CVL - concentration of the agent in venous blood leaving the liver
CVEN - concentration of the agent in pooled venous blood
PB - blood:air partition coefficient
PL - liver:blood partition coefficient
PK - kidney:blood partition coefficient
PT - testes:blood partition coefficient
PO - ovary: blood partition coefficient
PF -fat:blood partition coefficient
PR - rapidly perfused tissue:blood partition coefficient
PS - slowly perfused tissue: blod partition coefficient
Vmax - theoretical maximal initial rate of the metabolic reaction
Vmaxc - Vmax scaled as a function of body weight, i.e., Vmax*BW°7
KM - concentration of the agent producing a metabolic rate one-half that of Vmax,
concentration expressed in the CVL compartment
ROD - a first-order rate equation describing dermal absorption
RAM - rate of metabolism
RAO - rate of oral absorption
T-54
-------
TABLE 58
Physiological Parameters Used in the PBPK Models3
Parameter
Value3
Male
Female
Childb
Weights
Body (kg)
Liver (% Body Weight)
Kidney (% Body Weight)
Fat (% Body Weight)
Testes (% Body Weight)
Ovaries (% Body Weight)
Rapidly Perfused (% Body Weight)
Slowly Perfused (% Body Weight)
70
2.6
0.4
19.0
0.04
-
5.96
63.0
60
2.6
0.4
21.0
-
0.014
5.986
61.0
21.7
2.9
0.6
17.0
0.0083
-
5.492
65.0
Flows (l/hr/kg)c
QCc (cardiac output)
QPC (alveolar ventilation)
Liver (QL, % Cardiac Output)
Kidney (QK, % Cardiac Output)
Fat (QF, % Cardiac Output)
Testes (QT, % Cardiac Output)
Ovaries (Q0, % Cardiac Output)
Rapidly Perfused (QR, % Cardiac Output)d
Slowly Perfused (Qs, % Cardiac Output)6
15.0
15.0
26.0
3.4
5.0
1.3
-
45.3
19.0
15.0
15.0
26.0
3.4
5.0
-
0.12
46.48
19.0
15.0
15.0
7.95
5.1
5.3
0.07
-
62.88
18.7
3Values from ICRP (1975) unless otherwise indicated.
bFrom Price et al. (2003)
CQC = QCC * BW0.74; Qalv = QPC * BW0.74
dQR = 76-QL-QK-QT-QO
3QS = 24-QF
T-55
-------
TABLE 59
Tissue Partition Coefficients for the THMs
Value
Blood:Air (PB)
LiverBlood (PL)
Kidney: Blood (PK)
Testes: Blood (PT)d
Ovaries: Blood (P0)d
Fat: Blood (PF)
Rapidly Perfused: Blood (PR)
Slowly Perfused: Blood (Ps)
THM
Chloroform3
11. 34 (adult)
12.41 (child)
1.6 (adult)
1.4 (child)
1.3 (adult)
1.0 (child)
1.1 (adult)
0.99 (child)
0.78
31.0 (adult)
19.6 (child)
1.6 (adult)
1.4 (child)
1.5 (adult)
2.8 (child)
BDCM
26.6b
1.15°
1.24C
0.69
0.49
19.8C
1.15C
0.47C
CDBM
49.2b
2.56C
2.56C
1.5
1.03
39. Oc
2.56C
1.13C
Bromoform
102.3b
2.06C
1.69C
1.18
0.79
40.4C
2.06C
1.12°
aDatafrom U.S. EPA (2003)
bData from Batterman et al. (2002)
Calculated from rat tissue:air data (da Silva et al., 1999) and human blood:air data
(Batterman etal., 2002).
dCalculated based on tissue lipid and water content using the algorithms of Krishnan
(2002) and human blood:air data (Batterman et al., 2002).
T-56
-------
TABLE 60
Metabolic Parameters for the THMsa
THM
Chloroform c
BDCM
CDBM
Bromoform
Vmaxc (mg/hr/kg)b
8.96 (adult)
7.6 (child)
8.01
13.7
10.4
KM(mg/L)
0.012
0.302
0.72
0.42
Data from da Silva et al. (1999) unless otherwise noted.
bVmax = Vmaxc * BW ° 7°
cData from U.S. EPA (2003)
T-57
-------
TABLE 61
Description of Transfer File Naming Conventions.
File Type
Naming
Convention
Description
Breathing Rate
Files
BXYYYY.pk
B indicates the file contains breathing rate data
X identifies the subject; where A is subject 1, B is
subject 2, etc.
YYYY is the simulation number, for example 0005
is the 5th simulation
Dermal Data File
DXYZZZZ.pk
D indicates the file contains dermal data
X identifies the subject; where A is subject 1, B is
subject 2, etc.
Y is the chemical identifier (e.g., A is for
Chloroform, B for BDCM. Etc.)
ZZZZ is the simulation number, for example 0005
is the 5th simulation
Ingestion Data
File
GXYZZZZ.pk
G indicates the file contains ingestion data
X identifies the subject; where A is subject 1, B is
subject 2, etc.
Y is the chemical identifier (e.g., A is for
Chloroform, B for BDCM. Etc.)
ZZZZ is the simulation number, for example 0005
is the 5th simulation
Inhalation Data
File
IXYZZZZ.pk
I indicates the file contains inhalation data
X identifies the subject; where A is subject 1, B is
subject 2, etc.
Y is the chemical identifier (e.g., A is for
Chloroform, B for BDCM. Etc.)
ZZZZ is the simulation number, for example 0005
is the 5th simulation
T-58
-------
TABLE 62
Summary of THM Paired Concentrations for the Selected Factors
(Based on Analysis of the ICR Database)
Variable
Subgroup
All Systems
Using Surface
Water Intake
(N = 12,440)
Systems
Using any type
of Chlorine
Disinfectant
Process (N =
14,015)
Systems
Sampled
between July
and
September
(1997 and
1998)
THM Analysis
Description
Chloroform 95th
Percentile
BDCM 95th
Percentile
DBCM 95th
Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th
Percentile
DBCM 95th
Percentile
Bromoform 95th
Percentile
Chloroform 95th
Percentile
BDCM 95th
Percentile
DBCM 95th
Percentile
Bromoform 95th
Percentile
Concentration, ppb (Percentile)
Chloroform
66.0 (95)
26.1 (62)
140.0( 100)
14.0 (34)
66.0 (95)
9.1 (26)
140.0 (100)
5.5 (18)
74.8 (95)
14.0 (40)
170.0 (100)
30.0 (64)
BDCM
29.0 (98)
23.8 (95)
44.0(100)
25.0 (96)
29.0 (98)
24.0 (95)
44.0 (100)
12.3 (76)
17.6 (84)
27.0 (95)
70.0 (100)
37.0 (99)
DBCM
12.0(90)
17.7 (95)
17.0 (95)
26.0 (98)
12.0 (90)
46.0(100)
17.0 (95)
6.0 (78)
4.0 (65)
43.0(100)
20.0 (95)
33.0 (99)
Bromoform
0.5 (0)
2.6 (89)
3.7 (92)
5.6 (95)
0.5 (0)
34.0 (100)
3.7 (91)
6.0 (95)
0.5 (0)
23.0 (99)
1.6 (77)
6.5 (95)
T-59
-------
TABLE 62 cont.
Variable
Subgroup
All Samples
(18,214
records)
THM Analysis
Description
Chloroform 95th
Percentile
BDCM 95th
Percentile
DBCM 95th
Percentile
Bromoform 95th
Percentile
Concentration, ppb (Percentile)
Chloroform
65.9 (95)
10.0 (37)
120.0 (1)
36.0 (79)
BDCM
18.1 (90)
23.0 (95)
58.0 (1)
46.0 (1)
DBCM
3.4 (65)
42.0 (1)
16.0 (95)
37.0 (1)
Bromoform
0.5 (0)
24.0 (99)
0.5 (0)
6.1 (95)
T-60
-------
TABLE 63
Demographic Characteristics of Simulation Number 48.
NHAPS Record Number
2758
2677
4142
Population Group
Male, Ages 15-45
Female, Ages 15-45
Child, Ages 1-9
Sampled Occupant Age
18
39
8
T-61
-------
TABLE 64
Water-Use Activity Pattern from the NHAPS Database for Simulation Number 48
Source Name
Dishwasher
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Bathroom
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Model Location
Kitchen
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Occupant
Female
Male
Female
Female
Female
Female
Female
Female
Female
Female
Female
Female
Female
Child
Female
Female
Female
Female
Female
Female
Time On,
hours
10.279
5.508
9.512
9.563
9.592
17.133
17.157
17.164
17.177
17.278
17.367
17.396
17.404
8.365
10.271
10.292
10.387
10.411
10.412
10.432
Time Off,
hours
11.527
5.532
9.550
9.581
9.629
17.147
17.164
17.177
17.213
17.316
17.386
17.404
17.412
8.372
10.279
10.300
10.411
10.412
10.428
10.471
Duration,
min
74.9
1.4
2.3
1.1
2.2
0.9
0.4
0.8
2.2
2.3
1.1
0.5
0.5
0.4
0.5
0.5
1.5
0.1
1.0
2.4
T-62
-------
TABLE 64 cont.
Source Name
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Kitchen
Faucet - Laundry
Model Location
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Kitchen
Laundry
Occupant
Male
Male
Child
Child
Child
Child
Child
Child
Child
Child
Child
Male
Male
Male
Male
Male
Child
Male
Child
Male
Child
Time On,
hours
12.157
12.203
16.007
16.123
16.331
16.481
17.626
17.729
17.733
17.956
17.977
18.020
18.063
18.064
18.122
18.235
18.248
18.271
18.320
18.499
19.449
Time Off,
hours
12.167
12.210
16.016
16.131
16.336
16.500
17.640
17.733
17.809
17.967
18.013
18.063
18.064
18.077
18.140
18.248
18.271
18.288
18.329
18.518
19.461
Duration,
min
0.6
0.4
0.5
0.5
0.3
1.2
0.9
0.2
4.6
0.7
2.2
2.6
0.1
0.8
1.1
0.8
1.4
1.0
0.6
1.1
0.7
T-63
-------
TABLE 64 cont.
Source Name
Faucet - Laundry
Faucet - Laundry
Faucet - Laundry
Faucet - Laundry
Hall Bath
Hall Toilet
Hall Toilet
Hall Toilet
Hall Toilet
Hall Toilet
Hall Toilet
Hall Toilet
Hall Toilet
Shower
Shower
Shower
Toilet
Toilet
Toilet
Toilet
Toilet
Model Location
Laundry
Laundry
Laundry
Laundry
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Hall Bath
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Occupant
Child
Child
Child
Child
Child
Child
Child
Child
Child
Child
Child
Child
Child
Male
Male
Female
Male
Male
Female
Female
Female
Time On,
hours
19.461
19.491
19.664
20.25
19.25
19.265
19.319
19.347
19.380
19.390
19.579
19.663
20.290
5.532
5.620
17.004
5.511
5.716
9.594
9.655
9.694
Time Off,
hours
19.491
19.504
19.670
20.380
19.703
19.270
19.322
19.362
19.388
19.396
19.612
19.680
20.296
5.620
5.691
17.122
5.537
5.781
9.630
9.692
9.721
Duration,
min
1.8
0.8
0.3
7.8
27.2
0.3
0.2
0.9
0.5
0.3
2.0
1.1
0.4
5.3
4.2
7.1
1.6
4.0
2.1
2.2
1.6
T-64
-------
TABLE 64 cont.
Source Name
Toilet
Toilet
Toilet
Toilet
Toilet
Model Location
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Master Bathroom
Occupant
Female
Female
Female
Female
Female
Time On,
hours
17.220
17.308
17.364
17.428
17.465
Time Off,
hours
17.263
17.325
17.420
17.459
17.566
Duration,
min
2.6
1.0
3.4
1.8
6.0
T-65
-------
TABLE 65
Predicted Chloroform Absorbed Dose Results
Percentile
Chloroform Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Female, Age 15-45
1
5
10
25
50
75
90
95
99
0.021646
0.045655
0.073058
0.149855
0.310254
0.630598
1.145596
1.649123
8.393424
0*
0*
0.001117
0.002829
0.026129
0.045304
0.089087
0.121611
0.197200
0.006553
0.009483
0.011450
0.016664
0.027141
0.047809
0.084438
0.106439
0.248118
0.000838
0.010973
0.033892
0.091219
0.226843
0.540248
1.031325
1.551581
8.352551
Male, Age 15-45
1
5
10
25
50
75
90
95
99
0.019457
0.041836
0.067577
0.138960
0.311573
0.687820
1 .259366
2.044970
10.03811
0*
0*
0*
0.002399
0.024272
0.046043
0.082338
0.121381
0.237628
0.005736
0.009278
0.011848
0.017798
0.031017
0.052064
0.090232
0.134513
0.209462
0.000529
0.011465
0.024670
0.073703
0.237161
0.590267
1.131236
1.975072
9.968789
T-66
-------
TABLE 65 cont.
Percentile
Chloroform Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Child, Age 6
1
5
10
25
50
75
90
95
99
0.009934
0.020267
0.034366
0.076542
0.171478
0.377213
0.699450
0.941091
1 .444239
Oa
Oa
Oa
0.000739
0.006855
0.027164
0.051160
0.065190
0.088984
0.002120
0.003645
0.004697
0.007633
0.013569
0.024119
0.039602
0.051362
0.096345
0.000289
0.004391
0.013448
0.043171
0.137133
0.337850
0.634908
0.876666
1 .424603
*The zeroes entered in the dermal category represent the portion of the population that
has no dermal contact with the water supply during the simulated day. For the female
(age 15-45) population group, 6.9% had no dermal contact. For the male (age 15-45)
population group, 6.9% had no dermal contact. For the child (age 6) population group,
11.1% had no dermal contact.
T-67
-------
TABLE 66
Predicted BDCM Absorbed Dose Results
Percentile
BDCM Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Female, Age 15-45
1
5
10
25
50
75
90
95
99
0.008631
0.016902
0.025717
0.048507
0.099072
0.203035
0.370423
0.523761
2.814060
0*
0*
0.000189
0.000466
0.003149
0.005263
0.009307
0.012348
0.019722
0.002925
0.004124
0.004858
0.007103
0.011414
0.019801
0.034635
0.043487
0.101010
0.000267
0.003761
0.011244
0.031282
0.075642
0.178904
0.345126
0.504863
2.806662
Male, Age 15-45
1
5
10
25
50
75
90
95
99
0.008356
0.015445
0.023455
0.044175
0.101447
0.222273
0.401879
0.661688
3.432694
0*
0*
0*
0.000403
0.002924
0.005421
0.008875
0.012353
0.023330
0.002559
0.004011
0.005106
0.007653
0.013053
0.021847
0.037423
0.054970
0.085477
0.000174
0.003761
0.008234
0.024811
0.079238
0.195328
0.366215
0.649494
3.419677
T-68
-------
TABLE 66 cont.
Percentile
BDCM Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Child, Age 6
1
5
10
25
50
75
90
95
99
0.004433
0.008829
0.012558
0.025508
0.056992
0.125885
0.221136
0.300591
0.456096
0*
0*
0*
0.000123
0.000876
0.002900
0.005229
0.006434
0.008918
0.001296
0.002031
0.002667
0.004103
0.007079
0.011615
0.017947
0.022104
0.041569
9.2E-05
0.001432
0.004451
0.014526
0.045818
0.110828
0.209114
0.290700
0.449681
*The zeroes entered in the dermal category represent the portion of the population that
has no dermal contact with the water supply during the simulated day. For the female
(age 15-45) population group, 6.9% had no dermal contact. For the male (age 15-45)
population group, 6.9% had no dermal contact. For the child (age 6) population group,
11.1% had no dermal contact.
T-69
-------
TABLE 67
Predicted DBCM Absorbed Dose Results
Percentile
DBCM Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Female, Age 15-45
1
5
10
25
50
75
90
95
99
0.006838
0.012561
0.018498
0.034433
0.066284
0.133292
0.232392
0.333569
1.794922
0*
0*
0.000210
0.000521
0.003221
0.005348
0.008948
0.011876
0.018525
0.002529
0.003472
0.004094
0.005932
0.009375
0.015826
0.027289
0.034284
0.079193
0.000170
0.002202
0.007019
0.019891
0.048070
0.110357
0.211710
0.310299
1.787844
Male, Age 15-45
1
5
10
25
50
75
90
95
99
0.006539
0.011250
0.016953
0.032167
0.069418
0.143467
0.253583
0.413262
2.175275
0*
0*
0*
0.000447
0.002959
0.005548
0.008660
0.012051
0.021618
0.002196
0.003388
0.004283
0.006285
0.010803
0.017817
0.030253
0.043168
0.067179
0.000111
0.002338
0.005262
0.016069
0.049303
0.120803
0.223912
0.402837
2.124613
T-70
-------
TABLE 67 cont.
Percentile
DBCM Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Child, Age 6
1
5
10
25
50
75
90
95
99
0.002988
0.005530
0.008306
0.016842
0.036099
0.078574
0.139613
0.189552
0.282424
0*
0*
0*
0.000136
0.000926
0.002844
0.005002
0.006104
0.008271
0.000765
0.001253
0.001663
0.002652
0.004543
0.007909
0.012820
0.016558
0.031030
5.93E-05
0.000865
0.002877
0.009110
0.028304
0.069483
0.128593
0.179028
0.276845
*The zeroes entered in the dermal category represent the portion of the population that
has no dermal contact with the water supply during the simulated day. For the female
(age 15-45) population group, 6.9% had no dermal contact. For the male (age 15-45)
population group, 6.9% had no dermal contact. For the child (age 6) population group,
11.1% had no dermal contact.
T-71
-------
TABLE 68
Predicted Bromoform Absorbed Dose Results
Percentile
Bromoform Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Female, Age 15-45
1
5
10
25
50
75
90
95
99
0.002289
0.003976
0.005844
0.010873
0.019824
0.038290
0.065453
0.092500
0.505940
0*
0*
8.86E-05
0.000219
0.001330
0.002190
0.003593
0.004765
0.007186
0.000902
0.001217
0.001430
0.002087
0.003246
0.005282
0.008956
0.011241
0.025824
4.69E-05
0.000591
0.001928
0.005638
0.013413
0.030515
0.057593
0.083673
0.503191
Male, Age 15-45
1
5
10
25
50
75
90
95
99
0.002211
0.003744
0.005498
0.010434
0.020922
0.041260
0.070014
0.113307
0.599301
0*
0*
0*
0.000188
0.001215
0.002279
0.003517
0.004822
0.008450
0.000760
0.001189
0.001484
0.002187
0.003725
0.006019
0.010251
0.014303
0.021969
3.22E-05
0.000654
0.001464
0.004575
0.013943
0.033213
0.061013
0.109606
0.571568
T-72
-------
TABLE 68 cont.
Percentile
Bromoform Absorbed Dose, mg
Total
Dermal
Ingestion
Inhalation
Child, Age 6
1
5
10
25
50
75
90
95
99
0.000937
0.001829
0.002613
0.005179
0.010647
0.021956
0.039021
0.053294
0.077309
0*
0*
0*
5.75E-05
0.000383
0.001144
0.002008
0.002387
0.003211
0.000265
0.000425
0.000578
0.000892
0.001539
0.002640
0.004231
0.005402
0.010172
1 .66E-05
0.000249
0.000792
0.002564
0.007748
0.018976
0.034882
0.049472
0.075052
*The zeroes entered in the dermal category represent the portion of the population that
has no dermal contact with the water supply during the simulated day. For the female
(age 15-45) population group, 6.9% had no dermal contact. For the male (age 15-45)
population group, 6.9% had no dermal contact. For the child (age 6) population group,
11.1% had no dermal contact.
T-73
-------
TABLE 69
Water Concentrations Used to Investigate THM Metabolic Interactions
95th Percentile
Chloroform
BDCM
DBCM
Bromoform
THM (|jg/L)
Chloroform
74.8
14.0
170.0
30.0
BDCM
17.6
27.0
70.0
37.0
DBCM
4.0
43.0
20.0
33.0
Bromoform
0.5
23.0
1.6
6.5
T-74
-------
TABLE 70
Inhibition of DBCM Bioactivation by THMs
Exposure
Chloroform alone
Chloroform all
BDCM alone
BDCM all
Bromoform alone
Bromoform all
DBCM alone
DBCM all
DBCM + Chloroform
DBCM + BDCM
DBCM + Bromoform
DBCM + Chloroform + BDCM
DBCM + Chloroform + Bromoform
DBCM + BDCM + Bromoform
CM24 (mg/L liver)
2.97977 e-3
2.97977 e-3
1.81045e-3
1.81045e-3
4.7322 e-5
4.7322 e-5
6.95279 e-4
6.95278 e-4
6.95279 e-4
6.95279 e-4
6.95279 e-4
6.95278 e-4
6.95279 e-4
6.95279 e-4
% Inhibition
0
0
0
0
0
0
0
0.0001
0
0
0
0.0001
0
0
T-75
-------
TABLE 71
Effect of Decreasing Enzyme Content on the Metabolic Interactions of the THMs
Vmaxc
(% control)
100
10
1
0.1
0.01
THM
DBCM
Chloroform
BDCM
Bromoform
DBCM
Chloroform
BDCM
Bromoform
DBCM
Chloroform
BDCM
Bromoform
DBCM
Chloroform
BDCM
Bromoform
DBCM
Chloroform
BDCM
Bromoform
CM24 Alone3
6.95279 e-4
2.97977 e-3
1.81045e-3
4.7322 e-5
4.7301 e-4
2.91157e-3
1.30214e-3
3.63856 e-5
1.11495e-4
2.3686 e-3
3.39948 e-4
1.080966-5
1.28725e-5
8.24953 e-4
4.04642 e-5
1.339226-6
1.30737e-6
1.096476-4
4.12475e-6
1.371736-7
CM24 Allb
6.95278 e-4
2.97977 e-3
1.81045e-3
4.7322 e-5
4.72992 e-4
2.911566-3
1.3021 e-3
3.63847 e-5
1.11452e-4
2.36855 e-3
3.39799 e-4
1.080626-5
1 .28565 e-5
8.24895 e-4
4.0431 e-5
1.337796-6
1 .3053 e-6
1 .09637 e-4
4.11679e-6
1.369856-7
% Inhibition
0.0001
0
0
0
0.004
0.0003
0.003
0.002
0.04
0.002
0.04
0.03
0.12
0.007
0.08
0.11
0.16
0.009
0.19
0.14
T-76
-------
TABLE 71 cont.
Vmaxc
(% control)
0.001
0.0001
THM
DBCM
Chloroform
BDCM
Bromoform
DBCM
Chloroform
BDCM
Bromoform
CM24 Alone3
1.30942e-7
1.13375e-5
4.132750-7
1.375070-8
1.309630-8
1.13761 0-6
4.133550-8
1.37540-9
CM24 Allb
1.307280-7
1.133640-5
4.124540-7
1.373130-8
1.307480-8
1.13751 0-6
4.125320-8
1.373460-9
% Inhibition
0.16
0.01
0.2
0.14
0.16
0.009
0.2
0.14
aCM24 for ©ach THM alon©.
bCM24 for ©ach THM in a mixtur© of all 4 THMs.
T-77
-------
TABLE 72
Categories of Data Sources and Models
Category
1
II
III
IV
V
VI
VII
Description
Taken from peer reviewed literature, used for the purpose
the measurement
Taken from peer reviewed literature, used for the purpose
intended by the measurement.
intended by
other than
Taken from peer-reviewed database compiled for the purposes in which
it is being used.
Taken from non peer-reviewed database compiled for the
other than those for which it is being used.
purposes
Taken from other non peer-reviewed source.
Estimated based on peer-reviewed method or data.
Estimated based on non peer-reviewed method.
T-78
-------
TABLE 73
Quality and Sources of Data Used in the Models
Variables
Mass-Transfer
Coefficient
Gas- and liquid-
phase diffusivities
Henry's Law
Constant
Exposure-
Related Behavior
Water Use
Behavior
Ingestion
Behavior
House Volume
Water-Use Zones
Whole House Air
Exchange Rate
Category
VI
1, VI
1, II, VI
III
III, IV,
V
III
1, IV
VII
1
Description
Predicted based on peer reviewed algorithms
Diffusivities are used in the prediction algorithm
for the mass transfer coefficient, as described in
Section 3.1 . The sources of the diffusivities vary.
Several were obtained from the Department of
Energy, Risk Assessment Information System
(RAIS) database. The values for many of the
diffusivities were estimated using peer reviewed
prediction algorithms as described in Section
3.1.3.
Reported in literature or in databases at specific
temperatures. A temperature adjustment was
applied based on a peer-reviewed method as
described in Section 3.1 .3.
Activity patterns are sampled from the NHAPS
database
Compiled from a variety of databases including
REUWS, REGS, and NHAPS. NHAPS was
compiled for this purpose; REUWS and REGS
were compiled for other purposes.
Taken from the CSFII database
Household volumes are based on an analysis of
REGS data from 1993 and 1997. The 1993 data
are analyzed and presented in the Exposure
Factors Handbook.
Volumes are estimated based on architectural
design standards.
Sampled from the national distribution
recommended by the Exposure Factors
Handbook.
Citation
Corsi and Howard,
2000
Risk Assessment
Information System,
Oak Ridge National
Laboratory
Lyman etal., 1990
Various, see Table 2
and Section 2.1 fora
listing of data
sources and
temperature
adjustment algorithm
Described in Section
2.2.1
Described in Section
2.2.2
U.S. EPA, 2000d
U.S. DOE, 1995
U.S. DOE, 1997a
U.S. EPA, 1997b
Hoke, 1988
Hoke, 1994
U.S. EPA, 1997b
T-79
-------
TABLE 73 cont.
Variables
Interzonal
Airflows
Water
Concentrations
Ingestion
Concentrations
Breathing Rates
Body Weight
Body
Compartmental
Blood Flow Rates
Body
Compartmental
Volumes
Skin Permeability
Coefficients
Skin Partition
Coefficients
Gastro-lntestinal
Absorption Rate
Category
1
1
1, VII
1
1
1
1
VI
VI
1
1
VI
Description
Interzonal airflows are based on several sources.
The interzonal airflows between the non-water
using zones and the kitchen and laundry room
are based on a correlation from Koontz and
Rector, 1995. The flows between the non-water
using zones and the bathrooms are based on
Giardino etal., 1992.
The water concentrations were characterized
based on the published Information Collection
Rule (ICR) measurement data.
The ingestion concentrations were estimated for
a plausible set of activities based on published
results lab measurements.
Alveolar ventilation rates were assigned based
on two assumed activity levels: resting and
sedentary.
Calculated from the Exposure Factors
Handbook, Tables 7.2 and 7.3, adjusted for
clothes
Taken from ICRP (1975). Child values from
Price et al. (2003)
Taken from ICRP (1975). Child values from
Price et al. (2003)
Taken from methods used by Krishnan -
Krishnan, Personal Communication
Chloroform -
BDCM, DBCM, Bromoform
Estimated based on Corley et al (1990)
Citation
Koontz and Rector,
1995
Giardino etal., 1992
McGuire et al., 2002
Howard and Corsi,
1996
Batterman et al.,
2000
U.S. EPA, 1997b
U.S. EPA, 1997a
ICRP, 1975; Price et
al., 2003
ICRP, 1975; Price et
al., 2003
Krishnan, 2001, 2002
Krishnan, 2001,2002
Corley etal., 1990
Batterman et al.,
2002
Corley etal., 1990
T-80
-------
TABLE 73 cont.
Variables
Tissue Partition
Coefficients
Metabolism Rate
Constants
Category
1
1
VI
VII
1
II
Description
Chloroform partition coefficients from U.S. EPA
(2003)
BDCM, DBCM, Bromoform blood:airfrom
Batterman et al.(2002)
BDCM, DBCM, Bromoform calculated from rat
tissue:airdata (da Silva et al., 1999) and human
blood:airdata (Batterman et al., 2002).
Ovaries and Testes calculated based on tissue
lipid and water content using the algorithms of
Krishnan (2002).
Chloroform metabolic parameters are from EPA
(2003)
BDCM, DBCM and Bromoform metabolic
parameters are from rat studies by da Silva et al.
(1999)
Citation
U.S. EPA, 2003
Batterman et al.,
2002
da Silva etal., 1999;
Batterman et al.,
2002
Krishnan, 2002
U.S. EPA, 2003
da Silva etal., 1999
T-81
-------
TABLE 74
Categories of Model Approaches and Algorithms
Category
A
B
C
Description
Widely accepted modeling approach
Approach similar to commonly used and accepted approaches,
adapted to satisfy project specific requirements
Novel approach addressing specific requirements of estimating
and uptake of water borne contaminants
but
exposure
T-82
-------
TABLE 75
Quality of Modeling Approaches and Algorithms
Model
Category
Description
Representation
of the building
B
Building is represented as a collection of water using
zones and a lumped non-water using zones. Similar
approaches are widely used in the literature.
Fate and
transport
modeling
A
Commonly accepted approach based on mass balance.
Method assumes well mixed zones, each zone
constrained by mass and volumetric balance.
Fate and
Transport Model
Integration
Method
th
Model solves set of differential equations using the 4
order Runge-Kutta method (Mathews, 1992). This
method is widely cited, is very stable, self starting, and
accurate.
Behavior Models
The behavior is sampled from the NHAPS database, but
is modified to address known deficiencies in the dataset
and to accommodate water-use related behavior not
included in NHAPS.
Water Use
Models
C,A
Approach to simulating water uses incorporate
techniques for simulating water use occurrences as well
as the duration of water uses. The occurrences of water
uses are simulated based on survey data from NHAPS
and REUWS using a Poisson process. The durations of
the water uses are simulated by sampling from
representative lognormal distributions. These techniques
are used for similar purposes in peer-reviewed literature,
but the implementation in this modeling effort is unique
to exposure to water borne contaminants. This work has
been published in several peer reviewed publications
(Wilkes, 1999, Wilkes et al., 1996)
Exposure
Models
The exposure model used in this study, TEM, has been
published in several journal articles. The basic model
algorithms have been validated (Wilkes, 1994).
T-83
-------
TABLE 75 cont.
Model
Category
Description
Inhalation
Uptake Model
A, C
Dermal Uptake
Model
A, C
The exposure model uptake algorithms are described in
Section 2.7. These algorithms are taken from peer-
reviewed literature (Olin, 1998), but there integration into
an exposure model framework is unique to this exposure
model.
Physiologically
Based
Pharmacokinetic
Model
A, B
The physiologically based pharmacokinetic model is
based on a model used for over 30 volatile organic
chemicals (Gargas et al., 1986, 1990) including
chloroform (Corley et al., 1990) and BDCM (Lilly et al.,
1998). The model accounts for the potential metabolic
interactions between the four THMs as competitive
inhibitors of each others metabolism.
T-84
-------
Model Inputs
Demographic/Behavioral
Variables
> Age, gender, etc.
> Water-use behavior
> Location and activities
Physical Properties
> Building representation, etc.
> Water-use appliances
descriptions; flowrate, water
temperature, etc.
Chemical Properties
> Mass-transfer rates
> Henry's law
> Water concentrations
Other Physiological
Parameters
> Partition coefficients
> Rate constants
> Tissue volumes
Model Algorithms
Predictions/Outputs
Total Exposure Model
> Activity pattern sampling
algorithms
> Probabilistic representation
of water-use behavior
> Mass-transfer models
> Solution of set of differential
equations to predict air and
water concentrations
Media Concentrations
Function(time)
> Water
> Air
Exposure Function(time,
route, media)
> Inhalation, dermal,
ingestion
Physiological Variables
> Breathing rate, gender,
activity, etc.
Intermediate (Transfer)
Files
T
PBPK Model
> Solution of set of differential
equations to predict tissue
concentrations and doses
^
Tissue Concentrations
and Doses
function(time)
Figure 1 Population-Based Modeling Paradigm
F-1
-------
Eligible Activity/Location Codes
Identify eligible
activities and
•emove mean
;vent duration
Map to continuous timeline
and simulate water uses as a
Poisson process
Map back to original timeline and
simulate water-use duration as a,--'
ognormal distribution ,x''
uses
Figure 2 Schematic Representation of the Procedures Used for Simulating Water Uses
Based on a Sampled Activity Pattern
F-2
-------
_aj
'c
100
500
1000
1500
Floor Area, ft
2000 2500
3000
3500
4000
Properties of the Fitted Lognormal Distribution
Geometric Mean = 5.758 (= 317 m3)
Geometric STDEV = 0.4218
Volumes are calculated from the fitted distribution.
0 100 200 300 400 500 600 700 800 900 1,000
House Volume, m
Figure 3. Cumulative Distribution Function of Volume for 3-Person Households.
Source: Analysis of REGS 1997 data
F-3
-------
2.0E+06
1.8E+06
1.6E+06
1.4E+06
1.2E+06
o
c
01
g- 1.0E+06
01
8.0E+05
6.0E+05
4.0E+05
2.0E+05
O.OE+00
"Probability Density Function (PDF)
Geometric Mean = 5.758 (= 317 m3)
"Geometric STDEV = 0.4218
NOTES:
1. The area of the housing were surveyed and reported in RECS in ranges of
square feet. The values were converted to volumes by assuming an 8 ft
ceiling height (2.44 m). The values are reported in the following ranges, along
with the corresponding assumed volumes:
FLOOR AREA CALCULATED HOUSE VOLUME
0 - 600 ft2 (0 - 55.7 m2)
600-1000 ft2 (55.7-92.9m2)
1000 - 1600 ft2 (92.9 - 148.6 m2)
1600 - 2000 ft2 (148.6 - 185.8 m2)
2000 - 2400 ft2 (185.8 - 223.0 m2)
2400 - 3000 ft2 (223.0 - 278.9 m2)
0-135.9m
135.9-226.5m3
226.5 - 362.5 m3
362.5 - 453.1 m3
453.1 - 543.7 m3
543.7 - 679.6 m3
3000 ff (> 278.9 rrO > 679.6m3
2. The values are averaged in each category for display purposes
Values reported as > 680 m3(> 3000 ft2)
Assumed distribution for the purposes of
plotting.
200
400
600
Volume, m
800
1000
1200
Figure 4. Comparison of RECS Data and the Fitted Probability Density Function of Volume for
3-Person Households.
Source: Analysis of RECS 1997 data
F-4
-------
Rest of House (ROH)
V=LN (316.7m3, 0.4218)-
Z Water Using Zones
Water-Using Zones
Q = (0.078 + 0.31 * WHACH) * 24
Laundry
V = U (13.5, 25.4)
Q = (0.078 + 0.31 * WHACH) * 24
Kitchen
V = U(15.4, 18.1)
Q = (0.078 + 0.31 * WHACH) * 24
Hall Bath
V = U(7.9, 14.9)
Q = (0.078 + 0.31 * WHACH) * 24
Master Bath
V = U(4.9,8.5)
o
o
CN
Shower
V = U(2.9,4.5)
(N
#
O
o
a
WHACH = LN (0.46 , 2.25)
V = Zone Volume
Q = Air Flowrate (m3/day)
WHACH = Whole House Air Exchange Rate (h'1)
WHVOL = Whole House Volume (m3)
Notation:
LN (a , b) indicates that this parameter is sampled from a Log Normal
distribution
with geometric mean, a, and standard deviation, b.
U (a . b} indicates that this parameter is sampled from a Uniform distribution
Figure 5. Schematic Representation of House Interzonal Air Flows
F-5
-------
Qalv
Qc
Cinh
Cven
CVF
Cvs
CVR
CVT or Cvo
CVK
CVL
ALVEOLAR
LUNG BLOOD
FAT
Qalv
Calv
SLOWLY
PERFUSED
TISSUES
RAPIDLY
PERFUSED
TISSUES
TESTES OR
OVARIES
KIDNEYS
LIVER
Vma
KM
Qc
'ART
QF
'ART
Qs
CART
QR
'ART
QT or Qo
'ART
QK
'ART
QL
'ART
METABOLISM
Figure 6. Structure of the PBPK model used to analyze human exposures to THMs.
F-6
-------
1.6 -
1.4 -
WATERUSES: Indicates when
;ach aDDhance is in use)
Dishwasher
Faucet-M Bathroom
Faucet-Kitchen
Faucet-Laundry
Bath-Hall Bathroom
Toilet-Hall Bathroom
Shower-M Bathroom
Toilet-M Bathroom
Shower
Mster Bath
Kitchen
Laundry
Note: Water Concentration = 74.8 ug/L
Time, hours
Figure 7. Predicted Chloroform Air Concentrations for the Example Case
0.6
WATERUSES: (Indicates when each appliance is in use)
0.5 -
Dishwasher
Faucet-M Bathroom
Faucet-Kitchen
Faucet-Laundry
Bath-Hall Bathroom
Toilet-Hall Bathroom
Shower-M Bathroom
Toilet-M Bathroom
§0.3
c
o
o
—A— Master Bathroom
-*-Kitchen
O Laundry
0.1
0
0246
Note: Water Concentration = 27.0 ug/L
10 12 14
Time, hours
16
18
20
22
24
Figure 8. Predicted BDCM Air Concentrations for the Example Case
F-7
-------
WATERUSES: Indicates when each appliance Is In use
Dishwasher
Faucet-M Bathroom
Faucet-Kitchen
Faucet-Laundry
Bath-Hall Bathroom
Toilet-Hall Bathroom
Shower-M Bathroom
Toilet-M Bathroom
Shower
Master Bathroom
Kitchen
Laundry
0246
Note: Water Concentration = 20.0 ug/L
10 12 14
Time, hours
Figure 9. Predicted DBCM Air Concentrations for the Example Case
0.15
0.12
WATERUSES: (Indicates when each appli
Dishwasher
Faucet-M Bathroom
Faucet-Kitchen
Faucet-Laundry
Bath-Hall Bathroom
Toilet-Hall Bathroom
0 2 4 6 8 10 12 14
Note: Water Concentration = 6.5 ug/L Time, hours
Figure 10. Predicted Bromoform Air Concentrations for the Example Case
F-8
-------
o
o
1.2 -
WATERUSES: (Indicates when each appliance is in use)
Dishwash
Faucet-IV
Faucet-K
Faucet-Ls
Bath-Hall
Toilet-Ha
Shower-l\,
Toilet-M E
er
Bathroom •
tchen
Bathroom
Bathroom
Bathroom
iathroom
-
I
— -fr'xw— v
•
-
•
Male, C
Male, E
— 'Male, C
Male, E
a
-
-
wm»
—
•
hloroform Personal Concentration
DCM, Personal Concentration
BCM, Personal Concentration
romoform, Personal Concentration
f\
1
0 2 4 6 8 10 12 14 16 18 20 22 24
Time, hours
Figure 11. Predicted Personal Air Concentrations for the Adult Male for the Example Case
1.2 -
n.
centratior
D C
J> C
Air Cor
D C
^ C
WATERUSES: (Indicates when
Dishwash
Faucet-IV
Faucet-K
Faucet-L;
Bath -Hall
Toilet-Ha
Shower-l\
Toilet-M E
er
Bathroom
:chen
imHry
Bathroom
Bathroom
Bathroom
lath room
I
•
-
^^^"Female, Chloroform,
Concentration
— '
—
-emale, BDCM, Pers
Concentration
=emale, DBCM, Pers
Concentration
=emale, Bromoform,
Concentration
aach applii
Personal
onal
onal
Personal
nee is in u
-
/I
se)
-
I
.
*•• ••*
-
]
I
0
tm»
~
'
10 12 14 16 18 20 22 24
Time, hours
Figure 12. Predicted Personal Air Concentrations for the Adult Female for the Example
Case
F-9
-------
1.4
1.2 -
c 0.8
o
c
HI
£0.6
o
o
0.4
0.2
WATERUSES: (Indicates when each appliance is in use)
Dishwash
Faucet-IV
Faucet-K
Faucet-Ls
Bath-Hall
Toilet-Ha
Shovver-IV
Toilet-M I
er
Bathroom •
tchen
Bathroom
Bathroom
Bathroom
iathroom
-
^^^— Child, Chloroform, Personal
Concentration
_
Child, BDCM, Perso
Concentration
Child, DBCM, Perso
Concentration
Child, Bromoform, P
Concentration
nal
nal
ersonal
•
_
•
-
-
wm»
—
•
/ —
- .
6 8 10 12 14 16 18 20 22 24
Time, hours
Figure 13. Predicted Personal Air Concentrations for the Child for the Example Case
0)
0)
0.0014
0.0012
0.001
O
•Chloroform
-BDCM
DBCM
Bromoform
8 12
Time, hours
16
20
24
Figure 14. Predicted Concentrations of Metabolites Produced in the Liver over 24 hours
(CM24) for the Adult Male in the Example Case
F-10
-------
0)
•4-1
"o
£1
re
•55
O)
E
O
• DBCM
Bromoform
8 12
Time, hours
16
20
24
Figure 15. Predicted Concentrations of Metabolites Produced in the Liver over 24 hours
(CM24) for the Adult Female in the Example Case
0.0012
0.001
oj 0.0008
"5
si
o> 0.0006
E
O)
E
,3 0.0004
O
0.0002
Chloroform
-BDCM
— DBCM
Bromoform
8 12
Time, hours
16
20
24
Figure 16. Predicted Concentrations of Metabolites Produced in the Liver over 24 hours
(CM24) for the Child in the Example Case
F-11
-------
0.00004
0.000035
= 0.00003
o
|* 0.000025
£ 0.00002
o
0.000015
•Chloroform
-BDCM
•DBCM
- Bromoform
12 16
Time, hours
20
24
Figure 17. Predicted Area under the Curve (AUC) for the parent THMs Concentrations in
the Kidney for the Adult Male in the Example Case
0.00004
0.000035
•DBCM
Bromoform
12 16
Time, hours
20
24
Figure 18. Predicted Area under the Curve (AUC) for the parent THMs Concentrations in
the Kidney for the Adult Female in the Example Case
F-12
-------
0.00012
0.0001
3
O
-------
0.000025
,_ 0.00002
D
O
O)
E
j/f
re
'E
0)
O
D
0.000015
0.00001
0.000005
Chloroform
BDCM
-DBCM
Bromoform
12 16
Time, hours
20
24
Figure 21. Predicted Area under the Curve (AUC) for the parent THMs Concentrations in
the Genitals for the Adult Female in the Example Case
3
O
.E
*
O)
E
/T
c
0)
O
O
D
0.00006
0.00005
0.00004
0.00003
0.00002
0.00001
Chloroform
BDCM
— -DBCM
Bromoform
12 16
Time, hours
20
24
Figure 22. Predicted Area under the Curve (AUC) for the parent THMs Concentrations in
the Genitals for the Child in the Example Case
F-14
-------
c
o
a.
o
Q.
c
o
*->
o
ra
\J.I o
07
n RE;
OR
Occ
Oc
OAC.
OA
OT;
n •}
OQC
00
0-IC
O-i
One;
n
r
[
N
V
s
,
y
^
V
^
V
V
s
1-1
^
V
V
V
^
V
• Female, Age 15-45
C3 Male, Age 15-45
D Child, Age 6
i-H-|
J I 5-,nsL KI
Figure 23.
ID
O
ID
O
(N
O
ID
CS|
O
ro
CD
ID
ro
ID
1-^
ro
CD
A
Absorbed Dermal Chloroform Dose, mg
Histogram of Absorbed Chloroform Dermal Dose for Females, Males, and
Children
U.I -J
07
0 65
Oc
c 0 55
o
~s n z,
ra u'b
3 0 45
Q.
o n A
S. u-4
u_ n T^
o
i- n ^
C U.o
o
ra n 9
n 1 ^
O-i
o
V
V
s
V
V
V
s
V
V
s
V
• Female, Age 15-45
E3 Male, Age 15-45
D Child, Age 6
fen
^
K l~S~r^3
I R I S I H~lri3-i*3-.— - rfi
LOCOIOOOIOI-^IOIO
^CD^^fNcsi-:^
o T- CM ro °?
ro
A
Absorbed Inhalation Chloroform Dose, mg
Figure 24. Histogram of Absorbed Chloroform Inhalation Dose for Females, Males, and
Children
F-15
-------
\J.I o
07
Oc
o
•— n z,
JO
Q.
On A
CL
"o
Cn ^
0
S? no
it °'2
0-1C
n 1
n
s
S
S
^
•
L
$
J
>i
Ji
-,
• Female, Age 15-45
Q Male, Age 15-45
D Child, Age 6
^
k
B r£ „
B KilELn^fcurtSL-^ „ _ __
Figure 25.
OOOOOOO A
Absorbed Ingestion Chloroform Dose, mg
Histogram of Absorbed Chloroform Ingestion Dose for Females, Males, and
Children
0.75
0.7
0.65
0.6
0.55
I 0.5
3 0 45
Q.
o 0.4
D_
O
c
o
0.3
D.25
0.2
0.15
0.1
0.05
0
-K
v
y
S
\
S
V
s
V
1-1
• Female, Age 15-45
H Male, Age 15-45
D Child, Age 6
j]
s~|
V
^ k r^
5 R ISll5hrf3-,«7B_ rfsi
iq
d
(N
iq
cxi
ro
ID
ro
A
Absorbed Total Chloroform Dose, mg
Figure 26. Histogram of Total Absorbed Chloroform Dose for Females, Males, and Children
F-16
-------
\J.I o
07
n R1)
Oc
c n £,£,
o U-S:D
^ n ^
re u.o
3D A1!
Q.
o n A
w U.4
.. n T;
o ""^
_ n "}
•2 n o>^
nj no
"-01=1
n 1
0.1
n nc;
n
J
V
V
V
I ^ I ^
o
• Female, Age 15-45
Q Male, Age 15-45
D Child, Age 6
_
I y
V
S r§
V b~l
S R i-i.
I IMnKL^ « ^
lOt-iocxiioroioio
oot-ocxioroi-^
PdPdPcDPg
o o o o H
o
A
Absorbed Dermal BDCM Dose, mg
Figure 27. Histogram of Absorbed BDCM Dermal Dose for Females, Males, and Children
U./ vJ
07
OfiC
Oc
— n ^
o
•— n <^
re
~^ n A^
Q.
On A
u- n T^
Cn "}
o
^ n 9^
•g u.^a
ro n 9
LL
01 ^
0-1
n n1)
o
v
V
V
V
V
V
V
s
V
y
V
V
i-i
• Female, Age 15-45
E3 Male, Age 15-45
D Child, Age 6
I ^n
V
^ |~3-|
3_lj_SJ SnrlS-,r-m_ ^3
(N'
-------
U. / J
07
n PC
Oc
c n cc
o U-5:D
•^ n c
re u.t>
3n /it;
Q.
o n A
o u.4
.. n ic.
O
_- n •?
C U.o
•— n oc.
o
nj no
£ u.^
"- 0.15
O-i
Ont;
I
IS
• Female, Age 15-45
NMale, Age 15-45
DChild, Age 6
I
p
I— h
h
h
h
h
h
h
h
5
y
y
y
i
y
y
y
rH
OLOCOLOCDLOCOLOLO
t-o^i-or^oocN
°. d °. d °. d ^. 11
o o o o ^
o
A
Absorbed Ingestion BDCM Dose, mg
Figure 29. Histogram of Absorbed BDCM Ingestion Dose for Females, Males, and Children
Oyc
07
Occ
Oc
i- n E.E.
O
•^ n c
13 °'5
^ n A^
Q.
On A
Q.
.. n T;
o u"3b
Cn "^
0
•^ n 9^
"o U-^!D
2n 9
LL
0-IC
O-i
Of)C
n
i-.
V
V
s
V
V
V
V
V
V
s
V
• Female, Age 15-45
H Male, Age 15-45
D Child, Age 6
1 -S
V
V
V
^
1
1\
— — — — ^S
CM
(N
CD
CD
CO
oq
CD
CN
CO
LO
LO
CD
Absorbed Total BDCM Dose, mg
Figure 30. Histogram of Total Absorbed BDCM Dose for Females, Males, and Children
F-18
-------
0.75
0.7
0.65
0.6
c
o
• Female, Age 15-45
Q Male, Age 15-45
D Child, Age 6
Absorbed Dermal DBCM Dose, mg
Figure 31. Histogram of Absorbed DBCM Dermal Dose for Females, Males, and Children
U./ vJ
n 7
Oc
Cn £,£,
"re
•s n A1^
a.
On A
CL
o
Cn ^
O
2 02
LL
01 ^
O-i
O
V
V
s
V
V
s
s
V
s
V
l-l
• Female, Age 15-45
H Male, Age 15-45
D Child, Age 6
I Vn
V
^ Jli-JS
5 R nNiris-n-ra^ ^
•t-CNro-^-iocDi-^ir)
CDCDCDCDCDCDCD1^
Absorbed Inhalation DBCM Dose, mg
Figure 32. Histogram of Absorbed DBCM Inhalation Dose for Females, Males, and Children
F-19
-------
\J.I o
07
n RE;
OR
Cn E;E;
O
•r; n E;
re °'b
3 n 4.E;
Q.
o n 4
0 U.4
"o
_ n •}
° 025
o
re n 9
n -IE;
01
OnE;
o
V
S
^
-
L
R
ft
6
k
R
1 1?
o
o
• Female, Age 15-45
C3 Male, Age 15-45
D Child, Age 6
1
R
&
R ~^ HS
1 ^ ^lMsi-,nSLrta_r-CM ^N_ r, -^
O O O O O O 1-^
CD CD CD CD CD CD P
O
A
Absorbed Ingestion DBCM Dose, mg
Figure 33. Histogram of Absorbed DBCM Inhalation Dose for Females, Males, and Children
07
OR
_ n E;E;
4-1 U.b
re
Q.
On A
Q.
O
Cn ^
O
re 02
LL
01 E;
01
O
V
V
s
s,
V
V
s
s
V
V
s
s
s
l-l
• Female, Age 15-45
E3 Male, Age 15-45
D Child, Age 6
1 \
V
S $-,
^ ^Irs
Tt CO CM CD 1-^ •* CO 0)
t-(N'
-------
\J.I o
07
n R1)
Oc
c n £,£,
o U-S:D
^ n ^
re u.o
3D A1!
Q.
o n A
w U.4
.. n T;
o ""^
_ n •}
•2 n o>^
nj no
"-01=1
n 1
0.1
n IT;
|
[^
\
s.
V
a
I~F
^
1
1
i
• Female, Age 15-45
Q Male, Age 15-45
D Child, Age 6
JS 1
n K iii~§
Ik IS 1 R 1 §ll Inf^BsLK^-^ „ a _™
t-(NrO-
-------
\J.I o
07
0 65
Oc
o
^ n c;
If U-D
Q.
o n A
" U.4
*•— 0 35
•- 025
*j U.ZO
8 02
TO U.zi
"-01=1
0-1
n
,
V
V
y
V
V
1 ^
o
n
|
v
s
V
\
\
Is
IT
o
• Female, Age 15-45
Q Male, Age 15-45
D Child, Age 6
r§
Is jg
I Rll M-,rfLrfO_^pa _^^ ^n
t- IT) CM IT) CO IT) IT)
CD CD CD P
Figure 37.
o
A
Absorbed Ingestion DBCM Dose, mg
Histogram of Absorbed Bromoform Inhalation Dose for Females, Males, and
Children
U./ O
07
Occ
Oc
n i^
o
•— n ^
13 °'5
^ n A^
D U.4O
On A
D_
O
CD "}
O
ti °-25
2 no
LJ_
01 ^
01
n
V
s
s,
V
V
s
V
V
s
V
s
l-l
r—
• Female,
SMale, Ac
D Child, Ac
i
\
s-
V
5 1
^ Slf@n»L_ .
Age 15-45
e 15-45
je6
rf3
O O O O O O A
Absorbed Total Bromoform Dose, mg
Figure 38. Histogram of Total Absorbed Bromoform Dose for Females, Males, and Children
F-22
-------
800
700
100
1
11
l~hn ru, m
0 0.00002 0.00004 0.00006 0.00008 0.0001 0.00012 0.00014 0.00016 0.00018 0.0002 More
CF AUC Kidney
Figure 39. Histogram of the distribution of the AUC for chloroform in the kidneys of exposed
subjects from 1000 different water-use patterns.
1000
900
100
D Child
• Female
DMale
FL n^
0 0.000025 0.00005 0.000075 0.0001 0.000125 0.00015 0.000175 0.0002 0.000225 0.00025 More
BDCM AUC Kidney
Figure 40. Histogram of the distribution of the AUC for BDCM in the kidneys of exposed
subjects from 1000 different water-use patterns.
F-23
-------
800 -
700 -
requenc
^ c
II ~tv\J
200 -
n -
-
D Child
• Female
DMale
fki
II fin PL FL_ „ „ „
0 0.00003 0.00006 0.00009 0.00012 0.00015 0.00018 0.00021 0.00024 0.00027 0.0003 More
DBCM AUC Kidney
Figure 41. Histogram of the distribution of the AUC for DBCM in the kidneys of exposed
subjects from 1000 different water-use patterns.
900
800
700
600
500
.*• 400
300
200
100
D Child
• Female
DMale
0 0.0000070.0000140.0000210.0000280.0000350.0000420.0000490.0000560.000063 0.00007 More
BF AUC Kidney
Figure 42. Histogram of the distribution of the AUC for bromoform in the kidneys of exposed
subjects from 1000 different water-use patterns.
F-24
-------
800
700
600
500
u
c
d)
3 400
300
200
100
DChild
Female
DMale
HUkn
0 0.000015 0.00003 0.000045 0.00006 0.000075 0.00009 0.000105 0.00012 0.000135 0.00015 More
CF AUC Genitals
Figure 43. Histogram of the distribution of the AUC for chloroform in the genitals of exposed
subjects from 1000 different water-use patterns.
800
700
600
500
c
3 400
300
200
100
D Child
Female
DMale
n _ ^ _ n-n
0 0.0000060.0000120.0000180.000024 0.00003 0.0000360.0000420.0000480.000054 0.00006 More
BDCM AUC Genitals
Figure 44. Histogram of the distribution of the AUC for BDCM in the genitals of exposed
subjects from 1000 different water-use patterns.
F-25
-------
800
700
600
500
u
c
d)
3 400
300
200
100
D Child
Female
DMale
0 0.00001 0.00002 0.00003 0.00004 0.00005 0.00006 0.00007 0.00008 0.00009 0.0001 More
DBCM AUC Genitals
Figure 45. Histogram of the distribution of the AUC for DBCM in the genitals of exposed
subjects from 1000 different water-use patterns.
800
700
BF AUC Genitals
Figure 46. Histogram of the distribution of the AUC for bromoform in the genitals of exposed
subjects from 1000 different water-use patterns.
F-26
-------
600
500
100
fill
rfTlrfTI
HT1
_i-n
0.00025 0.0005 0.00075 0.001 0.00125 0.0015 0.00175 0.002 0.00225 0.0025
Chloroform CM24
More
Figure 47.
Histogram of the distribution of the concentration of chloroform metabolites
(CM24) formed in the liver over 24 hr in exposed subjects from 1000 different
water-use patterns.
700
600
D Child
• Female
DMale
0.00025 0.0005 0.00075 0.001 0.00125 0.0015 0.00175 0.002 0.00225 0.0025
BDCM CM24
More
Figure 48.
Histogram of the distribution of the concentration of BDCM metabolites (CM24)
formed in the liver over 24 hr in exposed subjects from 1000 different water-use
patterns.
F-27
-------
>
o
c
at
3 300 -
£
LL
n-i-i
D Child
DMale
•
[HI ril fbn ^ ^ rifl
0 0.00015 0.0003 0.00045 0.0006 0.00075 0.0009 0.00105 0.0012 0.00135 0.0015 More
Figure 49.
DBCM CM24
Histogram of the distribution of the concentration of DBCM metabolites (CM24)
formed in the liver over 24 hr in exposed subjects from 1000 different water-use
patterns.
600
500
400
u
c
3 300
200
100
n-
0.00007 0.00014 0.00021 0.00028 0.00035 0.00042 0.00049 0.00056 0.00063 0.0007
Bromoform CM24
More
Figure 50. Histogram of the distribution of the concentration of bromoform
metabolites (CM24) formed in the liver over 24 hr in exposed subjects from 1000
different water-use patterns
F-28
-------
1.2 n
1 -
0.8 -
0)
X
0.6 -
<
O
0.4 -
0.2 -
10
15
20
25
Vmaxd
Figure 51. Effect of varying the maximal rate of metabolism (VmaxC) on the liver
concentration of metabolites (CAM) for chloroform.
2 n
1.5 -
up
Q)
O
0.5 -
10
15
VmaxC
20
25
Figure 52. Effect of varying the maximal rate of metabolism (VmaxC) on the liver
concentration of metabolites (CAM) for bromoform.
F-29
-------
1.4 -,
1.2 -
1 -
0.8 -
X
<
o
0.6 -
0.4-
0.2 -
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 53. Effect of varying liver blood flow (QLC) on the liver concentration of
metabolites (CAM) for chloroform.
7 n
6 -
5 -
0) 4
<
O
3
2 -
1 -
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 54. Effect of varying liver blood flow (QLC) on the liver concentration of
metabolites (CAM) for bromodichloromethane.
F-30
-------
5 -1
4.5 -
4
3.5 H
up" 3
Q)
X
— 2.5
<
O
2 -
1.5 -
1
0.5 -
0
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 55. Effect of varying liver blood flow (QLC) on the liver concentration of
metabolites (CAM) for dibromochloromethane.
2 n
1.1
1.6 -
1.4 -
< 0.8
O
0.6 -
0.4 -
0.2 -
0
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 56. Effect of varying liver blood flow (QLC) on the liver concentration of
metabolites (CAM) for bromoform.
F-31
-------
10 15
Vmaxd
20
25
Figure 57. Effect of varying the maximal rate of metabolism (VmaxC) on the liver area
under the curve (AUCL) for chloroform.
7 n
6 -
5 -
2 -
10 15
VmaxC
20
25
Figure 58. Effect of varying the maximal rate of metabolism (VmaxC) on the liver area
under the curve (AUCL) for bromoform.
F-32
-------
KM (mg/L)
Figure 59. Effect of varying KM (mg/L) on the liver concentration of metabolites (CAM)
for chloroform.
70 -,
KM (mg/L)
Figure 60. Effect of varying KM (mg/L) on the liver area under the curve (AUCL) for
chloroform.
F-33
-------
up
d)
X
<
o
15
20 25
KM (mg/L)
30
35
40
45
Figure 61. Effect of varying KM (mg/L) on the liver concentration of metabolites (CAM)
for bromoform.
35 n
15
20 25 30
KM (mg/L)
35
40
45
Figure 62. Effect of varying KM (mg/L) on the liver area under the curve (AUCL) for
bromoform.
F-34
-------
1.4 -,
1.2 -
X
0.8 -
<
o
0.6 -
0.4 -
0.2 -
10
15
QCC
20
25
30
Figure 63. Effect of varying cardiac output (QCC) on the liver concentration of
metabolites (CAM) for chloroform.
7 -,
6 -
5 -
0) 4
X
3
O
2 -
1 -
10
15
QCC
20
25
30
Figure 64. Effect of varying cardiac output (QCC) on the liver concentration of
metabolites (CAM) for bromodichloromethane.
F-35
-------
0)
X
O
5 -1
4.5 -
4
3.5 -
3
2.5 -
2 -
1.5 -
1
0.5 H
0
10
15
QCC
20
25
30
Figure 65. Effect of varying cardiac output (QCC) on the liver concentration of
metabolites (CAM) for dibromochloromethane.
0)
X
O
1.8 -1
1.6 -
1.4 -
1.2 -
1 -
0.8 -
0.6 -
0.4 -
0.2 -
0 -
10
15
QCC
20
25
30
Figure 66. Effect of varying cardiac output (QCC) on the liver concentration of
metabolites (CAM) for bromoform.
F-36
-------
3 i
2.5 -
_ 2-
oo
Q)
X
^ 1-5 ^
O
0.5 -
10
15
QCC
20
25
30
Figure 67. Effect of varying cardiac output (QCC) on the liver area under the curve
(AUCL) for chloroform.
3 n
2.5 -
2 -
Q)
1.5 -
O
0.5 -
10
15
QCC
20
25
30
Figure 68. Effect of varying cardiac output (QCC) on the liver area under the curve
(AUCL) for bromodichloromethane.
F-37
-------
6 i
5 -
_ 4
Q)
X
3
O
<
1 -
10
15
QCC
20
25
30
Figure 69. Effect of varying cardiac output (QCC) on the liver area under the curve
(AUCL) for dibromochloromethane.
1.4 -,
1.2 -
O 0.6
<
0.4-
0.2 -
10
15
QCC
20
25
30
Figure 70. Effect of varying cardiac output (QCC) on the liver area under the curve
(AUCL) for bromoform.
F-38
-------
3 i
2.5 -
_ 2-
oo
Q)
X
^ 1-5 ^
O
0.5 -
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 71. Effect of varying liver blood flow (QLC) on the liver area under the curve
(AUCL) for chloroform.
3 n
2.5 -
2 -
Q)
1.5 -
O
0.5 -
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 72. Effect of varying liver blood flow (QLC) on the liver area under the curve
(AUCL) for bromodichloromethane.
F-39
-------
7 n
6 -
5 -
PO
2 -
1 -
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 73. Effect of varying liver blood flow (QLC) on the liver area under the curve
(AUCL) for dibromochloromethane.
1.6 -,
1.4 -
1.2 -
0)
>£
-^ 0.8 H
_l
o
< 0.6 H
0.4-
0.2 -
0
0
0.1
0.2
0.3
QLC
0.4
0.5
0.6
Figure 74. Effect of varying liver blood flow (QLC) on the liver area under the curve
(AUCL) for bromoform.
F-40
-------
•c
€
o
1/1
.a
Is
o
•c
01
8 «
OQ g
3
Q.
o
a.
•5
Ol
c
Ol
Ol
a.
Male, Total Dose
Female, Total Dose
0
2 3
Total Absorbed Chloroform Dose, mg
Figure 75A. Cumulative Total Absorbed Chloroform Dose.
100
.a
i_
o
«
.a
1
•a
ai
8 «
OQ g
.1°
4-1
ro
Q.
O
a.
•5
Ol
c
ai
ai
a.
Male, Normalized Total Dose
^^—Female, Normalized Total Dose
Child, Normalized Total Dose
0
0.02 0.04 0.06 0.08 0.1 0.12 0.14
Normalized Total Absorbed Chloroform Dose, mg/kg of Body Weight
Figure 75B. Normalized Cumulative Total Absorbed Chloroform Dose.
0.16
F-41
-------
100
O)
oT
in
o
Q
O
m
•a
01
&
o
VI
.a
<
1
Male, Total Dose
Female, Total Dose
Child, Total Dose
10
0
0
0.1 0.2 0.3
Percentile of Population Based on Total Absorbed Dose
Figure 76A. Cumulative Total Absorbed Bromoform Dose.
100
O)
oT
tn
o
Q
o
m
•a
01
&
o
VI
.a
<
1
Male, Normalized Total Dose
Female, Normalized Total Dose
Child, Normailized Total Dose
10
0
0.4
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
Percentile of Population Based on Total Absorbed Dose
Figure 76B. Normalized Cumulative Total Absorbed Bromoform Dose.
F-42
-------
1.4
1.2
O)
aT 1 n
in ' -w
8
E
£ 0.8
E
o
o
•c
€
o
1/1
.a
13 Male, Inhalation Dose
DMale, Dermal Dose
• Male, Oral Dose
0
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 77A. Route-Specific Contributions to the Total Absorbed Chloroform Dose for the
Male Population Group.
0.040
NOTE: Assumed Body Weight for Male = 70 kg
El Male, Inhalation Dose
DMale, Dermal Dose
• Male, Oral Dose
0.000
0
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 77B. Route-Specific Contributions to the Normalized Total Absorbed Chloroform
Dose for the Male Population Group.
F-43
-------
1.4
1.2 --
O)
8 1-0
o
O
•c
01
O
1/1
H Female, Inhalation Dose
D Female, Dermal Dose
• Female, Oral Dose
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 78A. Route-Specific Contributions to the Total Absorbed Chloroform Dose for the
Female Population Group.
0.040
o
O)
0.030
0.035 1
E
oT
I
E
£ £ 0.025
E
o
0.000
Q Female, Inhalation Dose
D Female, Dermal Dose
• Female, Oral Dose
NOTE: Assumed Body Weight for Female = 60 kg
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 78B. Route-Specific Contributions to the Total Absorbed Chloroform Dose for the
Female Population Group.
F-44
-------
O)
of
8
Q
E
I
.
o
o
m
Q Child, Inhalation Dose
D Child, Dermal Dose
• Child, Oral Dose
0.0 -M
0
10
90
20 30 40 50 60 70 80
Percent!le of Population Based on Total Absorbed Dose
Figure 79A. Route-Specific Contributions to the Total Absorbed Chloroform Dose for the
Child Population Group.
o
O)
0.040
0.035
0.030
£ £ 0.025
VI
o
O
E
o
il
O >, 0.020
11
o
-------
O)
aT
in
o
Q
E
00
O
1/1
.Q
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
HMale, Inhalation Dose
DMale, Dermal Dose
• Male, Oral Dose
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 80A. Route-Specific Contributions to the Total Absorbed Bromoform Dose for the
Male Population Group.
NOTE: Assumed Body Weight for Male = 70 kg
ED Male, Inhalation Dose
DMale, Dermal Dose
• Male, Oral Dose
0.0000
0
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 806. Route-Specific Contributions to the Normalized Total Absorbed Bromoform
Dose for the Male Population Group.
F-46
-------
1.4
1.2 --
O)
8 1-0
o
O
•c
01
O
1/1
H Female, Inhalation Dose
D Female, Dermal Dose
• Female, Oral Dose
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 81 A. Route-Specific Contributions to the Total Absorbed Bromoform Dose for the
Female Population Group.
NOTE: Assumed Body Weight for Female = 60 kg
Q Female, Inhalation Dose
D Female, Dermal Dose
D Female, Oral Dose
0.0000
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 81B. Route-Specific Contributions to the Normalized Total Absorbed Bromoform
Dose for the Female Population Group.
F-47
-------
O)
aT
in
o
Q
s
00
•a
€
o
«
.a
0.05
0.04
0.03
0.02
0.01
0.00
E3 Child, Inhalation Dose
D Child, Dermal Dose
• Child, Oral Dose
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 82A. Route-Specific Contributions to the Total Absorbed Bromoform Dose for the
Child Population Group.
0.0020
•5
o) 0.0018
E
VI
o
0.0016
0.0014
o 0)0.0012
C fl)
m >. 0.0010
0.0008
-c 0.0006
01
N
| 0.0004
0.0002
0.0000
H Child, Inhalation Dose
D Child, Dermal Dose
• Child, Oral Dose
NOTE: Assumed Body Weight for Child = 21.7 kg
0
10
90
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Figure 82B. Route-Specific Contributions to the Normalized Total Absorbed Bromoform
Dose for the Child Population Group.
F-48
-------
100%
90%
c
o
.a
£
c
o
o
o
Q.
«9
3
o
a:
•Female, Inhalation Chloroform Dose
-Female, Inhalation Bromoform Dose
— -Female, Oral Chloroform Dose
Female, Oral Bromoform Dose
,'v '\
•l,X'
' \i V i
A
M
0%
10
90
100
Figure 83.
20 30 40 50 60 70 80
Percentile of Population Based on Total Absorbed Dose
Comparison of Chloroform and Bromoform Route-Specific Contributions for
the Female Population Group.
100
90
I
o
8
OQ
a.
8.
IB
C
01
Ol
Q_
Female, Effective Consumption, Chloroform Dose
Female, Effective Consumption, Bromoform Dose
Male, Effective Consumption, Chloroform Dose
Male, Effective Consumption, Bromoform Dose
—~ — Child, Effective Consumption, Bromoform Dose
Child, Effective Consumption, Chloroform Dose
0
18
20
22
24
Figure
2 4 6 8 10 12 14 16
Effective Consumption, L
84. Effective Consumption Volume (Volume of Tap Water Consumed if all of the
Absorbed Dose originated from the Ingestion Route).
F-49
------- |