DRAFT EPA/600/R-06/044A
DO NOT CITE OR QUOTE September 2007
External Review Draft
An Exploratory Study: Assessment of Modeled Dioxin
Exposure in Ceramic Art Studios
NOTICE
THIS DOCUMENT IS A PRELIMINARY DRAFT. It has not been formally
released by the U.S. Environmental Protection Agency and should not at this
stage be construed to represent Agency policy. It is being circulated for comment
on its technical accuracy and policy implications.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document is a draft for review purposes only and does not constitute U.S.
Environmental Protection Agency policy. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
ABSTRACT
The purpose of this report is to describe an exploratory investigation of potential dioxin
exposures to artists/hobbyists who use ball clay to make pottery and related products. Dermal,
inhalation and ingestion exposures to clay were measured at the ceramics art department of Ohio
State University in Columbus, OH. The measurements were made in two separate studies, one in
April 2003 and one in July 2004. This assessment combines the results of these two studies.
Estimates of exposure were made based on measured levels of clay in the studio air, deposited on
media representing food and on the skin of artists. Dioxin levels in the clay were based on levels
reported in the literature for commercial ball clays commonly used by ceramic artists.
Hypothetical dioxin dose estimates were calculated for each subject assuming that all
used a 20% ball clay blend with 162 pg TEQ/g. The single-day total doses across the 10 subjects
were estimated to range from 0.49 to 20.81 pg TEQ/day, with an average of 3.45 pg TEQ/day.
The dermal dose was the major contributor to total dose, exceeding 78% for all subjects. A
Monte Carlo simulation suggested that ball clay exposures in a broad population of artists could
extend to levels lower or higher than the levels estimated for the 10 subjects. Comparing US
average background intakes (adjusted to an absorbed basis) to the 10 subject average dose from
ball clay use, indicates that the average ball clay dose is 10% of the background CDD/CDF dose
(34.4 pg TEQ/day).
Preferred Citation:
U.S. Environmental Protection Agency (EPA). (2007) An exploratory study: Assessment of
modeled dioxin exposure in ceramic art studios [draft]. National Center for Environmental
Assessment, Washington, DC; EPA/600/R-06/044A. Available from the National Technical
Information Service, Springfield, VA, and online at http://www.epa.gov/ncea.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 ii DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS
LIST OF TABLES v
LIST OF FIGURES vi
LIST OF ABBREVIATIONS AND ACRONYMS vii
PREFACE ix
AUTHORS, CONTRIBUTORS, AND REVIEWERS x
1. INTRODUCTION AND BACKGROUND 1
2. APPROACH OVERVIEW 3
2.1. GENERAL STRATEGY 3
2.2. CHARACTERIZATION PROCEDURES 4
2.2.1. Dermal Contact 4
2.2.2. Inhalation 4
2.2.3. Ingestion 4
3. SAMPLING METHODS 6
3.1. SAMPLE COLLECTION 6
3.1.1. Personal Air Sampling 6
3.1.2. Area Air Sampling 7
3.1.3. Skin Sampling 8
3.1.4. Surface Wipe Sampling 9
3.1.5. Surrogate Food and Beverage 10
3.2. SAMPLE PREPARATION AND ANALYSIS 10
3.2.1. Filtration and Drying 10
3.2.2. Gravimetric Analysis 11
3.2.3. Quality Control Samples 11
4. DIOXIN CONTENT OF CLAY AND STUDIO RESIDUES 13
5. DOSE ESTIMATION PROCEDURES 19
5.1. DERMAL CONTACT 19
5.1.1. Estimating Particle Loading on Skin 19
5.1.2. Estimating Monolayer Load 19
5.1.3. Estimating Fraction Absorbed 21
5.1.4. Calculating Dermal Dose 23
5.2. INHALATION 24
5.3. INGESTION 24
5.4. TOTAL DOSE 25
6. QUESTIONNAIRE RESULTS 26
7. COMPARING EXPOSURES ACROSS SUBJECTS 28
7.1. DERMAL CONTACT 29
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 iii DRAFT—DO NOT CITE OR QUOTE
-------
CONTENTS (continued)
7.1.1. Clay Loads on Surfaces 33
7.1.2. Dermatologist Report 35
7.2. INHALATION 35
7.2.1. Particle Levels in Air 35
7.2.2. Inhalation Dose 39
7.2.3. Classroom Exposure 40
7.3. INGESTION 40
7.4. TOTAL DOSE 40
8. MONTE CARLO SIMULATION OF THE EXPOSURE DATA 45
9. UNCERTAINTY 51
9.1. GENERAL UNCERTAINTY IS SUES 51
9.2. DERMAL EXPO SURE UNCERTAINTIES 51
9.3. INHALATION UNCERTAINTIES 54
9.4. INGESTION UNCERTAINTIES 54
10. CONCLUSIONS 56
REFERENCES 58
APPENDIX A: SUBJECT QUESTIONNAIRE RESULTS A-l
APPENDIX B: PICTURES OF ARTISANS PRIOR TO SKIN RINSE
PROCEDURE B-l
APPENDIX C: REAL-TIME PARTICLE CONCENTRATION DATA C-l
APPENDIX D: RESPICON, CASCADE IMP ACTOR, PDR-1000, CLIMET
DATA FOR EACH INDIVIDUAL SUBJECT D-l
APPENDIX E: SEM AND EDS DATA BY SUBJECT E-l
APPENDIXF: MONTE CARLO SIMULATION RESULT GRAPHICS F-l
APPENDIX G: EVALUATION OF CLAY DUST MODELING G-l
APPENDIX H: SKIN RINSING DATA H-l
APPENDIX I: ALTERNATIVE METHOD FOR ESTIMATING DERMAL
ABSORPTION 1-1
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 iv DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF TABLES
1. Raw ball clay dioxin concentrations 14
2. Processed ball clay dioxin concentrations (pg/g) 15
3. Percentage ball clay in the clay mixtures used during this study 16
4. Particle size distribution of Tennessee ball clay 20
5. Adjustments to Roy et al. (1990) dermal absorption data 22
6. Questionnaire questions on duration and frequency of subject clay work 26
7. Questionnaire questions about clay work 27
8. Artisan activities of each subject 30
9. Hypothetical estimates of dermal dose 31
10. Percent contribution to dermal dose by body part 33
11. Comparing clay loads on surfaces to clay loads on hands 34
12. Particle concentrations in air and mass median aerodynamic diameter
(MMAD) based on cascade impactor 36
13. Hypothetical estimates of inhalation dose 39
14. Clay deposition and hypothetical estimates of ingestion dose 41
15. Hypothetical estimates of total dioxin dose (pg TEQ/day) 42
16. Percent contribution to total dioxin dose 43
17. Dose estimates by activity 44
18. Monte Carlo simulation input parameters and sampling distributions 46
19. Clothing scenarios based on questionnaire responses 47
20. Descriptive statistics of dioxin doses from ball clay use, based on a Monte
Carlo simulation 49
21. Physical properties of dioxin congeners and concentration in processed clay 53
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 v DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF FIGURES
1. Scanning electron microscopy (SEM) and energy dispersive spectroscopy
(EDS) data 18
2. Scatter plot of adjusted absorption data versus time with linear trend line 23
3. Real-time particle concentration for Subject 3 using the CI-500 particle counter 37
4. Sculpture session 1 with dog present 38
5. Sculpture session 2 with dog present 38
6. Frequency distribution of total dose (pg TEQ/d) based on Monte Carlo
simulation 49
7. Sensitivity analysis based on percent contribution to variance 50
8. Sensitivity analysis based on rank correlation 50
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 vi DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS
°C degrees Centigrade
CDD Chlorinated dibenzo-p-dioxin
CDD/F Chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans
CDF Chlorinated dibenzofuran
cm centimeter
d day
DI Deionized
EDS Energy dispersive spectroscopy
EPA U.S. Environmental Protection Agency
ET Extrathoracic
FDA U.S. Food and Drug Administration
g gram
OFF Glass fiber filters
HpCDD Heptachlorodibenzo-p-dioxin
hr hour
FIRMS High-resolution mass spectrometry
HxCDD Hexachlorodibenzo-p-dioxin
IRB Institutional Review Board
kg kilogram
Kow Octanol-water partition coefficient
L liter
L/min liters per minute
LRB Laboratory record book
m meter
mg milligram
min minute
mL milliliter
mm millimeter
MMAD Mass median aerodynamic diameter
ND Nondetect
ng nanogram
NR Not reported
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 vii DRAFT—DO NOT CITE OR QUOTE
-------
LIST OF ABBREVIATIONS AND ACRONYMS (continued)
OCDD Octachlorodibenzo-p-dioxin
OSHA Occupational Safety and Health Administration
OSU Ohio State University
oz ounces
PCDD Polychlorinated dibenzo-p-dioxin
PCDF Polychlorinated dibenzofuran
PCD/F Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans
PeCDD Pentachlorodibenzo-p-dioxin
pg picogram
PU Pulmonary
QA Quality assurance
QC Quality control
r2 Regression coefficient squared
SD Standard deviation
SEM Scanning electron microscopy
TB Tracheobronchial
TCDD Tetrachlorodibenzo-p-dioxin
TEF Toxic equivalency factor
TEQ Toxic equivalent
TOC Total organic carbon
TWA Time-weighted average
USGS U.S. Geological Survey
WHO World Health Organization
wt Weight
jig microgram
jiL microliter
|im micrometer
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 viii DRAFT—DO NOT CITE OR QUOTE
-------
PREFACE
Dioxins were discovered in ball clay in 1996 as a result of an investigation to
determine the sources of elevated dioxin levels in two chicken samples from a national survey of
poultry. The investigation indicated that the contamination source was ball clay added to
chicken meal as an anti-caking agent. The purpose of this study is to evaluate another potential
exposure scenario associated with ball clay, namely its use in ceramic art studios. This
exploratory investigation makes preliminary exposure estimates that can be used to evaluate
whether more detailed follow-up analyses are needed. Hypothetical dioxin exposure estimates
were calculated using an assumption of dioxin levels in the ball clay based on measurements
from other studies. The study was conducted during 2003 and 2004 by the National Center for
Environmental Assessment with contract support provided by Battelle in Columbus, Ohio.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 ix DRAFT—DO NOT CITE OR QUOTE
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Principal Author
John Schaum, National Center for Environmental Assessment, U.S. Environmental Protection
Agency, Washington, DC (EPA Project Manager)
Authors
Ryan James, Battelle (Battelle Project Manager)
James Brown, National Exposure Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC
Dwain Winters, Office of Pollution Prevention and Toxic Substances, U.S. Environmental
Protection Agency, Washington, DC
Contributors
Ian MacGregor and Christine Lukuch of Battelle served as the technicians for the project.
Reviewers
Mark F. Boeniger, National Institute for Occupational Safety and Health, Cincinnati, OH
David Crawford, Office of Solid Waste and Emergency Response, U.S. Environmental
Protection Agency, Washington, DC
Mike Dellarco, Office of Research and Development, U.S. Environmental Protection Agency,
Washington, DC
Kim Hoang, Region 9, U.S. Environmental Protection Agency, San Francisco, CA
Chong Kim, Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC
Sid Soderholm, National Institute for Occupational Safety and Health, Washington, DC
Dan Stralka, Region 9, U.S. Environmental Protection Agency, San Francisco, CA
Acknowledgments
The authors would like to thank the Ohio State University (OSU) Ceramics Area and the
OSU Division of Dermatology for their cooperation during this study. In addition, the authors
thank Joe Ferrario at the U.S. EPA/Environmental Chemistry Laboratory (Stennis Space Center,
Mississippi) for his assistance in evaluating dioxin levels in ball clay.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 x DRAFT—DO NOT CITE OR QUOTE
-------
1 1. INTRODUCTION AND BACKGROUND
2
3 Ball clay is a natural clay mined commercially in the United States, primarily in
4 Kentucky, Tennessee, and Mississippi. A total of 1.21 million metric tons was mined in the
5 United States in 2005. Its plasticity makes ball clay an important commercial resource for a
6 variety of commercial uses. In 2005, it was used as follows: floor and wall tile - 40%, sanitary
7 ware (sinks, toilets, etc.) - 25%, exports - 17%, ceramics - 11%, fillers, extenders and binders -
8 4%, pottery - 1.5%, and miscellaneous purposes - 1.9% (USGS, 2007).
9 Dioxins were discovered in ball clay in 1996 as a result of an investigation to determine
10 the sources of elevated dioxin levels in two chicken samples from a national survey of poultry
11 (Ferrario et al., 1997). The investigation indicated that soybean meal added to chicken feed was
12 the source of the dioxin contamination. Further investigation showed that the dioxin
13 contamination occurred when ball clay was mixed with the soybean meal as an anti-caking agent
14 (Ferrario et al., 2000b; U.S. FDA, 2000). In 1997, the Food and Drug Administration (FDA)
15 asked producers or users of clay products in animal feeds to cease using ball clay in all animal
16 feeds and feed ingredients (U.S. FDA, 1997).
17 The purpose of this study is to characterize the possible dioxin exposures of artists using
18 ball clay in ceramic art studios. This exploratory investigation makes preliminary exposure
19 estimates that can be used to evaluate whether more detailed follow up analyses are needed. The
20 limited resources available for this study required a strategy to base the analysis on existing data
21 to the fullest extent possible.
22 Dioxin exposure is primarily a function of the dioxin concentration in the clay and an
23 individual's level of exposure to the clay. Although studies in the literature provided
24 information about dioxin levels in clay, no information could be found on clay exposure levels in
25 ceramic art studios. Therefore, this study was designed to measure total clay exposures in a
26 ceramic art studio. No dioxin measurements were made in this study, rather the dioxin levels in
27 ball clay were assumed based on measurements from other studies. Three exposure pathways
28 were evaluated: inhalation, dermal contact, and incidental ingestion. The evaluations involved
29 measuring levels of clay particulates in air, clay residues on skin, and clay deposition on media
30 representing food and beverages. These data provided a basis for estimating potential dioxin
31 exposures and resulting doses, conducting an initial analysis of which exposure pathways
32 contribute most to total dose, and evaluating how individual behaviors affect exposure/dose.
33 Ultimately, the data helped develop distributions for input parameters for conducting a Monte
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 1 DRAFT—DO NOT CITE OR QUOTE
-------
1 Carlo analysis to estimate how dioxin exposure/dose may vary across a wide population of
2 artists.
3 An alternative way to evaluate dioxin exposures is by blood testing. While this provides
4 a direct measure of dioxin exposure, it represents exposures from all sources, not just work in an
5 art studio. Also, a blood study would not have provided any insights about how dioxin
6 exposures may occur in an art studio. Normal background exposures vary widely and factors
7 such as diet and age are known to have large impacts on dioxin body burden. Accordingly a
8 blood study would require a large number of subjects with controls to reduce the effects of these
9 factors. Also blood tests have very high analytical costs. On the basis of costs alone, blood
10 testing was beyond the scope of this effort. The clay exposure testing done here provided a low
11 cost way to explore the problem and gives future researchers an informed basis for deciding if
12 blood testing or other types of follow-up work are needed.
13 Dioxin concentrations and exposures are presented in terms of toxic equivalents (TEQs).
14 TEQs allow concentrations of dioxin mixtures to be expressed as a single value computed by
15 multiplying each congener concentration by a toxicity weight (toxic equivalency factor or TEF)
16 and summing across congeners. TEFs are expressed as a fraction equal to or less than 1 with 1
17 corresponding to the most toxic dioxin congener, 2,3,7,8-tetrachlorodibenzo-/>-dioxin
18 (2,3,7,8-TCDD). The TEQ data presented here are based on TEFs from the 1998 World Health
19 Organization (WHO) recommendations (Van den Berg et al., 1998). In 2005, WHO updated the
20 TEFs (Van den Berg et al., 2006). As discussed in Section 4, these updates had little impact on
21 the literature values used here, so no adjustments were made.
22 The term "dioxins" is used in this study to refer collectively to the tetra- through
23 octa-chlorinated dibenzo-p-dioxins (CDDs) and chlorinated dibenzofurans (CDFs) with chlorine
24 substitutions in all of the 2,3,7,8 positions. This term is commonly defined to include the 12 co-
25 planar pentachlorobiphenyls (PCBs) which also demonstrate dioxin-like toxicity. However,
26 PCBs are not addressed in this study. PCBs have been shown to make up a small fraction of the
27 total TEQs in a wide variety of background soils (U.S. EPA, 2007) and therefore are probably
28 not important contributors to TEQs in ball clay.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 2 DRAFT—DO NOT CITE OR QUOTE
-------
1 2. APPROACH OVERVIEW
2
3 While working in a ceramics studio, artists may be exposed to dioxin-contaminated clay
4 via three pathways: dermal contact, particle inhalation, and incidental ingestion. Exposure could
5 also occur via open cuts or eyes and this possibility is discussed in Section 9 on uncertainty. The
6 general strategy and procedures used to characterize each pathway are described below.
7
8 2.1. GENERAL STRATEGY
9 The site selected for this study was the Ceramics Area in Hopkins Hall at Ohio State
10 University (OSU) in Columbus, OH. The Ceramics Area, housed in the basement of Hopkins
11 Hall, has eight rooms, including classrooms, studios, a storage area, a glaze-mixing area, a clay
12 recycling area, and a furnace room. This facility was selected because it offered a convenient
13 location for assessing exposures during a variety of typical ceramic art activities.
14 The exposure measurements were carried out in two separate studies. The first study was
15 conducted in April 2003 and the second in July 2004. The results of both studies have been
16 combined in this report. Seven artisans and one nonartisan staff member in the OSU Ceramics
17 Department were recruited to serve as subjects for the first study, and two additional artisans
18 were recruited for the second study. An open solicitation was presented to the students and
19 departmental staff, and the first volunteers were selected. The subjects included three males and
20 seven females ranging in age from about 20 to 40 years. Approval for human subjects was
21 obtained via the Battelle Institutional Review Board (IRB) and EPA. Upon approval by the
22 Battelle IRB and EPA, OSU determined that review by their IRB was not necessary. The testing
23 was conducted while the subjects conducted a variety of unscripted tasks, including clay
24 mixing/preparation, sculpting, pottery wheel work, and molding.
25 To assess dioxin exposure levels, it is necessary to estimate dioxin levels in the various
26 exposure media (i.e., clay used by the artists, dust particles suspended in the studio air, and dust
27 settled onto surfaces). No actual dioxin measurements were made in this study. Rather, dioxin
28 levels were estimated using literature-reported concentrations of dioxins in ball clay and
29 information about the amount of ball clay in the clay mixtures used by the artists. Details about
30 this procedure are discussed in Section 4.
31 A questionnaire was administered to subjects during the first study to gather information
32 on their routines involving clay artwork. The questionnaire data are presented in Appendix A
33 and summarized in Section 6.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 3 DRAFT—DO NOT CITE OR QUOTE
-------
1 2.2. CHARACTERIZATION PROCEDURES
2 The following procedures were used to characterize each exposure pathway.
O
4 2.2.1. Dermal Contact
5 Dermal contact with clay can occur via direct handling of the clay, deposition from the
6 air onto exposed skin, transfer from surfaces, and splashing during wheel operations. The
7 amount of clay on skin was measured using rinsing procedures. Additionally, surface wipes
8 were collected in work areas to evaluate dermal exposures via transfers from surfaces. To
9 further evaluate dermal exposure, a dermatologist examined the condition of the stratum
10 corneum, the outermost layer of skin, before and after each subject worked with clay. The
11 primary focus of this examination was to determine if any damage to skin may have occurred
12 that would affect dermal absorption.
13
14 2.2.2. Inhalation
15 Both personal and area air-monitoring techniques were used to assess inhalation
16 exposures. Personal air samplers provide data most representative of an individual's exposure
17 because they sample the air in a person's breathing zone and reflect changes in concentration due
18 to their movement. An area sampler provides a general indication of exposure for people in its
19 vicinity and also can achieve lower detection levels. Both the personal and area-monitoring
20 techniques provided particle size-selective data, so that the deposition site of the particles in the
21 respiratory tract (nose/mouth, tracheobronchial airways, and alveolar region) could be
22 determined.
23 Two types of personal air samplers were used: real-time and time-integrating. Similarly,
24 two types of area air samplers were used: real-time and time-integrating. The real-time air
25 samplers provided data on particle levels on a nearly continuous basis (every minute). The
26 integrating samplers collected particles over the entire time period of a work activity, yielding a
27 time-weighted average (TWA) concentration. In this sampling design, the real-time exposure
28 monitoring was used to assess frequency, magnitude, and duration of peak exposures as well as
29 TWA across the entire sampling time, while the integrating samplers provided information on
30 average exposures.
31
32 2.2.3. Ingestion
33 Inadvertent ingestion of clay or dust can occur in several ways. Clay particles in the air
34 can deposit on food or in beverages. Deposition onto surrogate food samples (a quartz filter was
35 used to represent food and a beaker of water was used to represent a beverage, see Section 3.1.5
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 4 DRAFT—DO NOT CITE OR QUOTE
-------
1 for further details) was measured to evaluate this pathway. Ingestion can also occur via transfers
2 from hands to food or cigarettes and via transfers to the mouth resulting from wiping the hands
3 or licking the lips. These possibilities were evaluated qualitatively through observations about
4 individual behaviors. Finally, ingestion can also occur via particle deposition in the nose, mouth,
5 and tracheobronchial airways; clearance to the throat; and swallowing. This process was
6 evaluated using inhalation modeling (Appendix G).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 5 DRAFT—DO NOT CITE OR QUOTE
-------
1 3. SAMPLING METHODS
2
3 Methods used for collecting, preparing, and analyzing samples are described below.
4
5 3.1. SAMPLE COLLECTION
6 Samples were collected from personal air, area air, skin rinses, surface wipes, and
7 surrogate food and beverages.
8
9 3.1.1. Personal Air Sampling
10 The Respicon model 8522 particle sampler (TSI Incorporated, Shoreview, MN) is a two-
11 stage virtual impactor with a three-stage gravimetric filter sampler. The sampler sorts airborne
12 particulate matter into three size ranges. Each size range is collected on a 37-mm glass fiber
13 filter (GFF). The particle size collection ranges are as follows: stage 1, aerodynamic particle
14 diameter (Dae) < 4 |im; stage 2, 4 < Dae < 10 |im; and stage 3, 10 < Dae < 100 |im.
15 Before the start of sampling, three preweighed GFFs were removed from their protective
16 polystyrene containers (47-mm Millipore petri slides) and loaded into the Respicon using
17 nonmetallic filter forceps. A unique laboratory record book (LRB) identification number was
18 assigned to each GFF during tare weighing, and this weight was recorded onto the sampling data
19 sheet at that time. The Respicon was then assembled, and the total flow checker head was
20 installed. A personal sampling pump (SKC model no. 224-PCXR4, Eighty Four, PA) was
21 attached to the total flow head, and the flow rate through the Respicon was adjusted to 3.11 liters
22 per minute (L/min) ± 2%, according to the manufacturer's specifications. All flows were
23 verified by employing a calibrated National Institute of Standards and Technology (NIST)-
24 traceable Buck calibrator (Model M5, A.P. Buck, Orlando, FL). After confirmation of the
25 manufacturer's suggested flow rates at each stage of the sampler, the total flow checker was
26 replaced with the standard (100 jim) inlet head. A nylon chest harness (TSI Incorporated,
27 Shoreview, MN) was used to place the Respicon in each subject's breathing zone, approximately
28 15-20 cm below the chin. The personal sampling pump was attached to the subject's belt and
29 connected to the Respicon. Sampling was initiated by starting flow through the Respicon and
30 continued throughout a subject's entire work shift, typically 2 to 2.5 hours. The average
31 sampling volume was 387 L. Following sampling, the pump was turned off, the Respicon was
32 disassembled, and the filters were returned to their polystyrene petri dish containers for
33 transportation back to the laboratory for gravimetric analysis. Quality control samples, such as
34 field blank samples and matrix spike samples, were collected and analyzed for each sampling
35 technique (see Section 3.2.3).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 6 DRAFT—DO NOT CITE OR QUOTE
-------
1 The personal DataRAM-1000 (pDR-1000, Thermo Electron Corporation, Franklin, MA)
2 sampler was also used to measure personal particle exposure passively. No pump is required for
3 this instrument; instead, the air surrounding the sampler circulates freely through the open
4 sensing chamber by natural convection, diffusion, and background air motion. Particle
5 concentrations are measured using a light-scattering (nephelometry) technique. This instrument
6 responds optimally to particles with diameters in the range of 0.1 to 10 jim but will also respond
7 to a lesser extent to larger diameter particles. Via internal calibration, the sampler converted
8 particles/m3 to mg/m3 as final data units.
9 Before the start of sampling, the instrument sensor was zeroed by placing it in a
10 resealable bag into which particle-free (filtered) air was pumped. All zero operations were
11 performed successfully. To begin sampling, the instrument was clipped to the subject's waistline
12 (on the belt or strap holding the SKC pump) and the unit was activated. The pDR-1000 collected
13 data at 1 Hz and was programmed to record these data as 1-minute averages over the duration of
14 the sampling period. At the conclusion of sampling (typically 2-2.5 hours), data logging was
15 stopped and the instrument was turned off. The data were then uploaded to a personal computer
16 using software provided by the manufacturer and an RS-232 serial port connection.
17
18 3.1.2. Area Air Sampling
19 To assess the particle size and concentration in the ceramic studio's air, a six-stage
20 Delron cascade impactor (Delron Research Products, Powell, OH) was employed. Each stage
21 filters out successively smaller particles so that the following particle sizes are collected in
22 successive stages: >32 |im, 16-32 jim, 8-16 jim, 4-8 jim, 2-4 jim, and 0.5-2 jim; the final GFF
23 collects all particles smaller than 0.5 jim in diameter. Particles accumulate on glass slides
24 underneath each impactor orifice. To prevent particle loss due to bouncing, a small amount of
25 vacuum grease was applied to each glass slide. The area coverage of the grease on the slide was
26 determined by the approximate size of the impactor nozzle below which the slide was to be
27 placed. Correct airflow rate through the impactor ensures that the correct particle sizes are
28 collected on each stage. A carbon-vane pump (Gast Co., Benton Harbor, MI), with a critical
29 orifice that provides a pressure drop of at least 430 mm of mercury, was used to ensure the flow
30 rateof24L/min.
31 Before the start of sampling, preweighed glass slides were removed from their protective
32 polystyrene petri slide containers and loaded into the impactor using clean forceps or tweezers.
33 Unique LRB numbers, assigned to each slide during tare weighing, were recorded on sample
34 data forms. The impactor tower was then assembled and flow was initiated to verify the required
35 pressure drop. For each sample, the pressure drop was between 480 and 510 mm of mercury.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 7 DRAFT—DO NOT CITE OR QUOTE
-------
1 Flows were also verified using the Buck calibrator. Sampling times were approximately 2-2.5
2 hours, giving an average sample volume of approximately 2,900 L. Following sampling, the
3 impactor was disassembled and all slides were returned to their respective petri dish containers
4 for transportation back to the laboratory for gravimetric analysis.
5 The Climet CI-500 innovation laser particle counter (Redlands, CA) was a second
6 sampling device used to measure area particle concentrations. In a manner similar to the pDR-
7 1000, the Climet CI-500 measures particle number concentration using nephelometry. A self-
8 contained pump sampled air at a constant flow rate of approximately 3 L/min. In the count
9 mode, the Climet CI-500 measures particles in six particle size ranges: 0.3-0.5 jim, 0.5-1 jim,
10 1-2.5 |im, 2.5-5 jim, 5-10 jim, and >10 |im. The sampling frequency for the instrument is 1 Hz,
11 and the data were logged as 1-minute averages. The particle counts were converted from
12 particles/m3 to mg/m3 as final data units. The particle counts did not exceed the manufacturer's
13 recommended maximum (200-250 counts/cm3 at 3 L/min) at any time except for a few minutes
14 during two of the sampling periods. No instrument zero or span checks were necessary.
15 Following sampling, the data were uploaded to a computer using an RS-232 serial cable and
16 software provided by the manufacturer. The Climet CI-500 was located in close proximity to the
17 cascade impactor and generally very near the subject. For example, when the subject was
18 working with clay at a wheel, the two air samplers were placed on the side of the wheel opposite
19 the subject at a height and distance from the wheel similar to the subject's mouth and nose. The
20 inlet to the Climet was oriented in a vertical direction.
21
22 3.1.3. Skin Sampling
23 The total skin area of hands, arms, face, feet, and legs was estimated using a combination
24 of direct measurements and regression models based on body weight and height (U. S. EPA,
25 1997). The subject's exposed body parts were rinsed with a dilute soap solution (-2% soap in
26 deionized [DI] water, by weight). Approximately 100-150 mL of the soap solution was used to
27 rinse each exposed body part. After each body part was rinsed, the washbasin contents were
28 transferred to a polypropylene bottle with small amounts of deionized (DI) water rinses. The
29 bottle was labeled and sealed with a screw-top cap. The washbasin was then rinsed again, wiped
30 out, and reused. Between the first and second studies, the procedures differed as described
31 below.
32 April 2003. All subjects wore short-sleeved shirts, long pants, socks, and shoes.
33 Therefore, the only exposed skin areas were the hands and forearms, and the rinsing was limited
34 to these body parts. At three times during each subject's work session, the subject's exposed
35 skin was examined for clay residue. When clay was observed visually, the affected areas of the
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 8 DRAFT—DO NOT CITE OR QUOTE
-------
1 subject's body were rinsed. Rinses were performed at approximately equally spaced intervals,
2 and the last rinse usually coincided with the conclusion of the sampling period. The average of
3 the three measurements was used to represent the session.
4 July 2004. Both subjects wore short-sleeved shirts, short pants, and sandals. Therefore,
5 the exposed skin areas included the hands, arms, legs, and feet, and the rinsing was expanded
6 from the first tests to include all of these body parts. The subjects' faces were also rinsed during
7 these tests. Although no visible residues were apparent on the faces, this area was included for
8 the sake of completeness.
9 The rinse samples were collected in a washbasin using a squirt bottle of soap solution
10 while the subjects used their hands to gently wipe off the affected area. Rinses were conducted
11 in the following manner:
12
13 • Hands. Moving downward from the wrist, the technician rinsed the residual clay
14 off both sides of the artisans' hand; the residual clay from each hand was rinsed
15 into separate containers and analyzed separately.
16
17 • Arms. Moving downward from the elbow, the artisans rinsed the residual clay
18 from their arms.
19
20 • Feet. Moving downward from the ankle, the artisans rinsed the residual clay
21 from their feet.
22
23 • Legs. Moving downward from the top of the exposed area of the legs, the
24 artisans rinsed the residual clay from their legs.
25
26 • Face. The artisans rinsed the residual clay from their faces.
27
28 Skin rinse samples were collected at the close of each work session. In addition, if at any
29 point during the work session the subject indicated the need to wash an exposed body part, it was
30 rinsed into a sample container reserved for that body part.
31
32 3.1.4. Surface Wipe Sampling
33 A 20 cm by 20 cm horizontal surface near the subject's workspace was selected and
34 cleaned with dilute soap solution before the subject began working with any clay. Wipe samples
35 of this area were taken immediately after cleaning (to confirm that low levels were present
36 before starting the work session) and at the end of the work session. The wipe sampling
37 procedure consisted of the following steps. The selected area was wiped with 10 cm x 10 cm
38 rayon gauze wipes wetted with ~5 mL isopropanol using the following procedure. The wipe was
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 9 DRAFT—DO NOT CITE OR QUOTE
-------
1 secured between the thumb and forefinger of one hand, and the surface was wiped five times in
2 one direction using evenly applied pressure. The soiled side of the wipe was folded to the inside
3 and, in an orthogonal direction, the surface was wiped five more times. This soiled side of the
4 wipe was again folded to the inside and the wipe was placed into its prelabeled, resealable bag
5 for transportation back to the laboratory for gravimetric analysis. The entire wiping process
6 above was then repeated using one additional wipe.
7
8 3.1.5. Surrogate Food and Beverage
9 An 85-mm diameter quartz fiber filter and a 125-mL polypropylene jar filled with
10 100 mL DI water served as surrogates for food and beverage samples, respectively. Before clay
11 work began, both were placed in a location where the artisan indicated he or she might normally
12 place food or drink. In most cases, this location was away from the direct work area but still in
13 the same room. However, occasionally clay workers placed food and beverage directly adjacent
14 to their work. To begin sampling, the lid of the polycarbonate petri dish containing the food
15 surrogate and the screw-cap lid on the beverage surrogate were removed. Following the
16 conclusion of sampling, the lid to the petri dish was replaced and sealed with Teflon tape, and
17 the polypropylene jar was secured for transportation back to the laboratory for gravimetric
18 analysis.
19
20 3.2. SAMPLE PREPARATION AND ANALYSIS
21 Procedures used for sample preparation, analysis, and quality control are described
22 below.
23
24 3.2.1. Filtration and Drying
25 To collect the clay rinsed from the subject's skin during the skin rinse sampling
26 procedure and the clay deposited into the surrogate beverage sample, the clay-liquid suspensions
27 were filtered through a preweighed 85-mm diameter quartz fiber filter in a Buchner funnel using
28 vacuum filtration. Any remaining clay in the sample container was rinsed with several small
29 aliquots of DI water to ensure complete transfer of the clay to the filter. All filters from the
30 vacuum filtration procedure were subsequently placed on clean 10-cm watch glasses and dried
31 overnight at 100°C. The gauze wipes for surface residues were dried in this fashion as well. No
32 drying was required for the 37-mm Respicon filters or glass slides.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 10 DRAFT—DO NOT CITE OR QUOTE
-------
1 3.2.2. Gravimetric Analysis
2 The accuracy of the analytical balance (AT-20, Mettler-Toledo) used for all gravimetric
3 analyses was confirmed daily with weights approved by NIST. The calibration weights ranged
4 from 0.001 mg to 100 g. All 37-mm GFFs, 85-mm quartz fiber filter paper, 37-mm glass slides,
5 and gauze wipes were conditioned in a temperature- and humidity-controlled balance room
6 (temperature 22-23° C, relative humidity 46-56%) for a minimum of 24 hours before tare and
7 final weights were recorded. For conditioning, the lid of the container holding the filter or slide
8 was left slightly ajar, and the resealable bags containing the gauze wipes were left open. For
9 both kinds of filters and glass slides, three separate weights were recorded to the nearest
10 microgram. The weight was acceptable if the range of the three independent measurements was
11 less than 10 jig. For gauze wipes, the three separate weights were recorded to the nearest tenth
12 of a milligram and the acceptability criterion was that the range of the measurements be less than
13 1 milligram.
14
15 3.2.3. Quality Control Samples
16 At least one field blank sample was collected for each type of gravimetric sample,
17 including the Respicon, cascade impactor, food and beverage, and surface wipe samples. Such
18 samples were collected by transporting the sampling media to the field location and placing them
19 into their respective sampling device or position for sampling. As soon as the medium was ready
20 for sampling, it was collected as if the sampling time had come to a close and transported back to
21 the laboratory for gravimetric analysis. The detection limits for the gravimetric measurements
22 were determined by multiplying the standard deviation of the field blank net weights by 3. The
23 detection limits for each type of gravimetric measurement were as follows: 0.0025-0.015 mg/m3
24 for each stage of the cascade impactor, 0.878 mg/m3 for each stage of the Respicon, 10.6 mg for
25 the surface wipes, 0.6-1 mg for the food/beverage deposition samples, and 0.6-1.6 mg for the
26 dermal rinse samples.
27 As a quality control check, the skin rinse, surface wipe, and food and beverage sampling
28 and analysis methods were tested in a controlled laboratory setting. For the skin rinse method
29 evaluation, approximately 3 g of clay (obtained from one of the artisan subjects) was handled
30 carefully without dropping any until the entire sample was spread over the hands and forearms of
31 a Battelle researcher. The skin rinse and analysis method described above was performed, and
32 recoveries of 87 ± 3% of the clay applied were obtained. This compares favorably with Kissel et
33 al. (1996), who obtained 93% recovery when rinsing wet soil from the skin of human subjects
34 using a similar sampling method. Similarly, for the surface wipe method, approximately 1 g of
35 clay was deposited onto a precleaned laboratory bench, the wipe method described above was
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 11 DRAFT—DO NOT CITE OR QUOTE
-------
1 performed, and recoveries of 94 ± 5% were obtained. For the food and beverage samples,
2 approximately 50 mg of clay was added to those sampling matrices and recoveries of 90 and
3 95%, respectively, were obtained using the gravimetric analysis procedures described above.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 12 DRAFT—DO NOT CITE OR QUOTE
-------
1 4. DIOXIN CONTENT OF CLAY AND STUDIO RESIDUES
2
3 As discussed earlier, this study made no dioxin measurements in clays, dust residues, or
4 other materials from the Ohio State University ceramics studio. Instead, the possible levels were
5 estimated on the basis of other studies. A number of studies have measured dioxin levels in raw
6 and processed ball clay. Raw clay is the clay as it comes out of the ground. Processed clays are
7 the result of the initial processing, which is usually conducted at or near the mining site before
8 shipping. This processing typically involves drying with hot air at 120°C and pulverizing in a
9 series of milling stages (Ferrario and Byrne, 2002). The following studies describe dioxin levels
10 in raw and processed clay:
11
12 • Ferrario and Byrne (2000, 2002). Both papers present data for processed ball
13 clay used at one ceramics manufacturer. The mean of seven samples of processed
14 ball clay was 3,172 pg/g TEQ. Additional data are presented on dioxin levels in
15 clay mixtures and fired products. The authors noted that dioxin levels in the dust
16 samples collected at the facility were the same as those in the unfired clay
17 mixtures.
18
19 • Ferrario et al. (2000a). This study compared the mean levels in eight raw clay
20 samples from Mississippi (see Table 1) to the mean levels in four processed ball
21 clay samples. This comparison showed that the processed clays had much lower
22 levels of 2,3,7,8-TCDD and higher levels of 1,2,3,4,7,8-hexachlorodibenzo-/>-
23 dioxin (HxCDD), 1,2,3,4,6,7,8-heptachlorodibenzo-^-dioxin (HpCDD), and
24 octachlorodibenzo-p-dioxin (OCDD) than the raw clay. The mean total TEQ of
25 the processed clay (977 pg/g TEQ) was 37% lower than the raw clay (1,513 pg/g
26 TEQ).
27
28 • Ferrario et al. (2000b). This study also presents the data for raw and processed
29 clay described in Ferrario et al. (2000a). In addition, it presents dioxin levels in a
30 variety of other types of clays and discusses the evidence of a natural origin for
31 their presence.
32
33 • Ferrario et al. (2004, 2007). These studies collected processed ball clay directly
34 from four art-supply retailers. All ball clay types sold by these retailers were
35 purchased in 22.7 kg (50 pound) bags. One type of ball clay was sold by all four
36 retailers, five types were sold by two of the retailers and seven types were sold by
37 only one retailer. Thus a total of 21 bags representing 13 different types of ball
38 clays were purchased and sampled. A ceramics expert confirmed that the most
39 commonly used ball clays for making artware and pottery were represented in
40 these samples. These data are summarized in Table 2.
41
42
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 13 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 1. Raw ball clay dioxin concentrations
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
Total
PCDD concentration (pg/g dry weight)
Range
253-1,259
254-924
62-193
254-752
1,252-3,683
1,493-3,346
8,076-58,766
Median
617
492
134
421
1,880
2,073
4,099
Mean
711
508
131
456
2,093
2,383
20,640
Mean TEQ
711
508
13
46
209
24
2
1,513
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
TEQ = toxic equivalent
Source: Ferrario et al. (2000a).
Since the data from Ferrario et al. (2004, 2007) represented the types of clays most likely
used in ceramic art studios, these data were selected as the most representative ones to be used in
this study. Accordingly, it was assumed here that the dioxin TEQ levels in clay could range
from 289 to 1,470 pg/g with an average of 808 pg/g. As shown in Table 2, the TEQs from this
study were calculated on the basis of the WHO-98 Toxicity Equivalecy Factors or TEFs (Van
den Berg et al., 1998). In 2005, WHO updated the TEFs (Van den Berg et al., 2006). These
updates increased the TEF for OCDD from 0.0001 to 0.0003. None of the TEFs for the other six
congeners used to estimate the ball clay TEQs were changed by the WHO update. The increase
in the OCDD TEF would cause the overall average to increase by 6%. It was decided to use the
TEQ estimates for ball clay as originally reported instead of updating it on the basis of the 2005
WHO TEFs. This was based on two reasons, first the change would have been relatively minor
and second it would have complicated comparisons to exposure estimates which have not yet
been updated on the basis of the new TEFs.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 14 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 2. Processed ball clay dioxin concentrations (pg/g)
PCDDs
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
Total
TCDD
PeCDD
HxCDD
HpCDD
Total TEQsb
Average
76
374
335
526
1,480
9,780
254,000
1,450
4,600
13,500
25,000
808
Standard
deviation
60
144
141
204
608
4,480
88,200
606
1,890
5,710
11,700
318
Median
63.5
387
313
523
1,570
8,600
233,000
1,600
4,880
12,800
24,400
771
Minimum
21.8
125
142
167
394
3,940
118,000
412
1,560
4,800
9,320
289
Maximum
291
588
636
944
2,550
19,500
471,000
2,370
7,140
21,900
44,900
1,470
WHO-
TEFa
1
1
0.1
0.1
0.1
0.01
0.0001
Avg
TEQ
76.0
374
33.5
52.6
148
97.8
25.4
808
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
aWorld Health Organization Toxic Equivalency Factors (WHO-TEFs ) based on Van den Berg (1998)
bThe overall average presented by Ferrario et al. (2007) is based on averaging the mean congener levels across
samples. An alternative approach is to compute the average on the basis of the TEQ for each sample. This
approach yields an average of 819 pg/g (SD = 303 pg/g). Similarly the median TEQ is 810 pg/g based on the
individual samples. The minimum and maximum TEQ values are reported on the basis of the individual samples.
TEQ = toxic equivalent
Source: Ferrario et al. (2004, 2007).
All of these studies indicate that ball clay has relatively high levels of CDDs and very
low levels of CDFs. Based on Ferrario et al. (2004, 2007), about 95% of the TEQs in processed
clay are contributed by four congener groups: TCDDs (9%), pentachlorodibenzo-p-dioxin
(PeCDDs) (46%), HxCDDs (28%), and HpCDDs (12%).
Artists commonly use a mixture of clays to achieve various physical properties and visual
effects. The percentage of ball clay in the mixture can vary widely. The amount of ball clay in
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
the mixtures used on days when the testing occurred ranged from 0 to 100% with an average of
21.5% (Table 3). Although 4 of the 10 subjects used mixtures containing no ball clay on the test
days, on other days these subjects would likely use mixtures that do contain ball clay. This is
because students are required to conduct a variety of projects, and some of these are better suited
to using ball clay and others are not. Accordingly, it was assumed here that the ball clay portion
of clay mixtures used by artists can range from 0 to 100% with an average of 20%. Furthermore,
it was assumed that the dioxin levels in the non-ball clays were negligible. This is supported by
Ferrario et al. (2000b), who analyzed 15 different mined clays and concluded their dioxin levels
were significantly lower than levels in ball clay.
Table 3. Percentage ball clay in the clay mixtures used during this study
Subject
1
2
O
4
5
6
7
8
9
10
Percentage ball clay
0
27
48
0
20
0
0
15
100
5
Finally, it was assumed that the dusts suspended in the air and settled onto food or skin
would have the same dioxin levels as the clay. Material other than clay may contribute to these
dusts, further diluting dioxin concentrations. This possibility was evaluated using scanning
electron microscopy (SEM) with energy dispersive spectroscopy (EDS). These techniques were
applied to four types of samples:
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 16 DRAFT—DO NOT CITE OR QUOTE
-------
1 • Blank OFF.
2
3 • Dust on a OFF collected from a storeroom at the Battelle Laboratory (not
4 impacted by clay).
5
6 • Air particles on a Respicon OFF collected in the studio.
7
8 • Clay used by subjects.
9
10 SEM photographs and elemental spectra of samples associated with Subject 6 are shown
11 in Figure 1. A visual comparison of the SEM photographs suggests that the particles on the
12 Respicon filter appear to differ from those in the storeroom dust. Also, the spectra of the
13 particles on the Respicon filters resemble clay more than those of storeroom dust. The clay
14 samples and Respicon filter samples had high abundances of titanium, iron, and aluminum,
15 which were not seen in the GFF blank or in the storeroom dust sample. Similar results were
16 found for all eight subjects in the April 2003 tests, as shown in Appendix E. The analysis was
17 not repeated in the July 2004 tests. These observations suggest that clay dominates the air
18 particles collected in the studio. On this basis, it was assumed that the studio dust was
19 dominated by clay and no further dilution factor was needed to adjust dioxin concentrations.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 17 DRAFT—DO NOT CITE OR QUOTE
-------
to
I
-------
1 5. DOSE ESTIMATION PROCEDURES
2
3 This section presents the procedures used to estimate the dioxin dose to artisans from all
4 three routes of exposure: dermal contact, inhalation, and ingestion. Because the dermal dose is
5 expressed on an absorbed basis, the dose by other pathways must also be expressed on an
6 absorbed dose basis. This provides an equivalent basis for comparison and addition across
7 pathways. All doses are presented as daily estimates. No adjustments are made for the
8 frequency with which artists work with clay. Therefore, these dose estimates should be
9 interpreted as the dose that could occur on a day that clay work is conducted, rather than as a
10 long-term average.
11
12 5.1. DERMAL CONTACT
13 A fraction absorbed approach is used to estimate dermal absorption. This method has
14 been widely used to assess dermal exposures to solid residues and is endorsed in current Agency
15 guidance (U.S. EPA, 2004, 1992). Bunge and Parks (1998) have proposed an alternative
16 approach based on a more mechanistic model. This model has had only limited testing and is not
17 addressed in Agency guidance. Therefore, it was not chosen as the primary basis for this
18 assessment, but Appendix I discusses how it could be applied to this situation. This new model
19 suggests similar estimates of absorbed dose to those presented here using the traditional
20 absorption fraction approach.
21
22 5.1.1. Estimating Particle Loading on Skin
23 As described earlier, rinsing procedures were used to determine the total amount of clay
24 on exposed skin. This mass was divided by the exposed skin area to derive a loading in units of
25 mg/cm2.
26
27 5.1.2. Estimating Monolayer Load
28 The monolayer is the layer of particles immediately adjacent to the skin. According to
29 the monolayer theory, the only significant dermal absorption comes from chemicals contained in
30 this first layer (U.S. EPA, 2004, 1992). Experimental evidence supporting the monolayer theory
31 has been published by Duff and Kissel (1996) and Touraille et al. (2005). To properly apply the
32 dermal absorption fractions, it was necessary to determine whether residue loads on skin
33 exceeded monolayer loads. The monolayer load for a specific soil can be estimated on the basis
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 19 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
of the median particle size.
Assuming spherical
particles
monolayer loads can be calculated as follows (U.S. EPA,
where:
L'mono
pdp/6
and face-centered packing, the
2004):
(1)
Lmono = monolayer load (mg/cm2)
p = particle density (mg/cm3)
dp = physical particle diameter (cm)
The average particle density of the processed clays analyzed by Ferrario et al. (2004) was
2.64 g/cm3. Clays typically
have very
small particles relative to other components of soil. The
U.S. Department of Agriculture (USD A) defines clays as
having less than 2 jim diameter
particles (Brady, 1984). The particle size specifications for a Tennessee ball clay is shown in
Table 4 (Ceramics Materials Info, 2003). Reviewing the specifications for a variety of
commercial ball clays, median particle
Materials Info, 2003).
sizes ranged from
Table 4. Particle size distribution of Tennessee
Particle diameter (um)
% finer than
Source: Ceramics Materials
20
99
10
97
5
93
about 0.5
to 1.0 jim
(Ceramics
ball clay
2
81
1
72
0.5
56
0.2
35
Info (2003).
The particle sizes found in the studio air
had median physical diameters ranging across
subjects from 8 to 27 jim (this is derived from the mass median aerodynamic
diameter
[MMAD]
range of 13 to 44 |im described in Appendix G and converted to physical diameters using the
procedure in Appendix G, footnote 1).
These airborne particles appear larger
than what would
be expected from the original clay product. This may be explained by the bonding of particles
caused by the addition of water to the clay or the firing process, which fuses particles.
that accumulate on the skin
primarily from air deposition
are likely
to resemble the air
Particles
particles
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 20 DRAFT—DO NOT CITE OR QUOTE
-------
1 more than the original clay particles. Particles that transfer to skin primarily from direct
2 handling of the clay should more closely resemble the original clay product than the airborne
3 particles. Accordingly, the particle sizes of the clay residues on skin could vary widely, with
4 medians ranging from 0.75 to 27 jim. For purposes of the central exposure estimates, the
5 geometric mean of this range is assumed, i.e., 4.5 jim. This implies a monolayer load of
6 0.62 mg/cm2. The uncertainty resulting from this assumption is discussed further in Section 9.
7
8 5.1.3. Estimating Fraction Absorbed
9 As discussed in U.S. EPA (1992), three teams of investigators have examined dermal
10 absorption of TCDD from soil (Roy et al., 1990; Shu et al., 1988; Poiger and Schlatter, 1980).
11 The Roy et al. (1990) data (also described in U.S. EPA, 1991) were selected as the best basis for
12 estimating dermal absorption fractions applicable to the ceramics studio. This was because the
13 test soil was most fully described allowing comparisons to the clay, and multiple exposure times
14 were used allowing evaluation of how dose varies with time.
15 Roy et al. (1990) conducted a variety of experiments in which TCDD was applied to soil
16 on human skin in vitro, rat skin in vitro, and rat skin in vivo. The experiments were conducted
17 with both a low organic carbon soil and a high organic carbon soil. Ferrario et al. (2004, 2007)
18 studied 21 samples of processed ball clay used in ceramics studios. They found that the organic
19 carbon content of these samples ranged from 0.06% to 1.1% with a median and geometric mean
20 of approximately 0.4%. This level is very similar to the level in the low organic carbon soil used
21 by Roy et al. (0.45%). Accordingly, this discussion focuses on the Roy et al. results for the low
22 organic carbon soil.
23 Roy et al. (1990) calculated the percentage absorbed at various times over the 96-hour
24 experiment (Table 5). The second column shows the results for the human skin in vitro
25 experiments. The percentage absorbed includes the amount measured in the skin at the end of
26 the experiment. These values were adjusted in two ways. First, as recommended in U.S. EPA
27 (1992), they were multiplied by the ratio of the percentage absorbed for rat skin in vivo (16.3%)
28 to percentage absorbed for rat skin in vitro (7.7%). Second, they were adjusted to reflect the
29 assumption that the absorption occurs exclusively from the monolayer. In the low organic
30 carbon soil tests, Roy et al. (1990) used "Chapanoke" soil, which is composed of 15.1% sand,
31 68.2% silt, and 16.7% clay. Chapanoke soil has an organic matter content of 0.77% (0.45%
32 organic carbon). Based on the USDA soil classification system, this composition is a silty loam.
33 Silty loams have a median particle size of about 10 jim (Brady, 1984), which corresponds to a
34 theoretical monolayer load of 1.3 mg/cm2. Roy et al. applied a soil load of 6 mg/cm2, exceeding
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 21 DRAFT—DO NOT CITE OR QUOTE
-------
1 the monolayer load by a factor of 4.6. Accordingly the percentage absorbed was also multiplied
2 by this factor. The results of these two adjustments are shown in the third column of Table 5.
4
5
Table 5. Adjustments to Roy et al. (1990) dermal absorption data
Time (hr)
1
2
4
8
24
48
72
96
Percentage absorbed -
human in vitro
0.19
0.25
0.24
0.19
0.45
1.08
1.71
2.42
Percentage absorbed -
adjusted"
1.85
2.43
2.34
1.85
4.38
10.52
16.65
23.57
Percentage absorbed -
best fitb
1.01
1.24
1.69
2.59
6.19
11.59
16.99
22.39
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
aThese values were adjusted first by multiplying by the ratio of the percentage absorbed for rat skin in vivo (16.3%)
to percentage absorbed for rat skin in vitro (7.7%) and second by multiplying by 4.6 to reflect the assumption that
the absorption occurs exclusively from the monolayer.
bThese values were derived using eq. 2 and converting to percent.
The Roy et al. (1990) data show a strong linear correlation between percent absorbed and
time (r2 = 0.98). The scatter plot for these data and the best fit line are shown in Figure 2. The
equation for this line is as follows (converting percent to fraction):
AFdermai = 0.00225^ + 0.00787, t < 96hr
(2)
where:
= dermal absorption fraction
t = time (hr)
This equation was adopted in this study for purposes of estimating dermal absorption of
dioxin. The percentage absorbed values based on this equation are shown in the last column of
Table 5.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 22 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Percentage Absorbed vs. Time
(Based on adjusted data from Roy et al., 1990)
25.00
20.00
15.00
8
Si
10.00
5.00
0.00
20
40
60
Time (hr)
80
100
120
Figure 2. Scatter plot of adjusted absorption data versus time with
linear trend line.
Source: Adapted from Roy et al. (1990).
5.1.4. Calculating Dermal Dose
The rinsing experiments indicated that clay loading exceeded the monolayer load in
some, but not all, cases. The dermal absorption fractions presented above were applied to the
measured loads where these were less than or equal to monolayer loads. At soil loadings greater
than monolayer, the dermal absorption fraction was applied to only the monolayer load.
Accordingly, the dose of dioxins absorbed through the skin of the artisan subjects during this
study was estimated using the following equation for each body part and then summed:
Ddermal = SAL CAP,
dermal
where:
Ddermai = dermally absorbed dose (pg TEQ/d)
SA = skin area exposed (cm2)
L = daily clay loading on skin (measured or monolayer, whichever is less) (mg/cm2-d)
C = dioxin concentration in clay (pg TEQ/g)
AFdermai = dermal absorption fraction
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 23 DRAFT—DO NOT CITE OR QUOTE
-------
1 5.2. INHALATION
2 The portion of particles that enter the respiratory tract through the nose or mouth
3 (inhalability) depends mainly on particle size, route of breathing (through the nose or mouth),
4 wind speed, and a person's orientation with respect to wind direction. Inhaled particles may be
5 either exhaled or deposited in the extrathoracic (ET), tracheobronchial (TB), or pulmonary (PU)
6 airway. The deposition of particles in the respiratory tract depends primarily on inhaled particle
7 size, route of breathing, tidal volume, and breathing frequency (ACGIH, 2004; ICRP, 1994).
8 Appendix G presents a detailed discussion of how to consider these factors and estimate the
9 amount of particulate that deposits in various regions of the respiratory tract.
10 The absorbed inhalation dose is estimated as follows:
11
12 Dmhalatlon= Dr CAFr (lg/1000 mg) (4)
13
14 where:
15 Dinhaiation = inhalation dose (pg TEQ/d)
16 Dr = dose of particles to region r of the respiratory tract (mg/d)
17 C = dioxin concentration on particles (pg/g)
18 AFr = absorption fraction for region r of the respiratory tract
19
20 This equation is used to estimate the absorbed dose to the three regions of the respiratory
21 tract (ET, TB, and PU) and then summed to derive total inhalation dose. In general, particles
22 deposited in the ET and TB regions clear rapidly (within 1-2 days) to the throat and are
23 swallowed. Accordingly, the absorption of dioxin from particles deposited in these regions is
24 treated as if the particles had been ingested with an absorption fraction of 0.3 (U.S. EPA, 2003).
25 The particles depositing in the PU region remain there a long time, and most of them are
26 ultimately absorbed directly into the body (assumed absorption fraction of 0.8 based on U.S.
27 EPA, 2003).
28
29 5.3. INGESTION
30 The ingestion dose is estimated by assuming that all particles deposited on the surrogate
31 food and beverage samples are ingested. For both types of samples, the dose was calculated
32 using the equation below:
33
34 Dmgestion = (F + B)C AFmgestion (5)
35
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 24 DRAFT—DO NOT CITE OR QUOTE
-------
1 where:
2 Dingestion = ingestion dose (pg TEQ/d)
3 F = deposited clay on food (g/d)
4 B = deposited clay on beverage (g/d)
5 C = dioxin concentration in clay (pg TEQ/g)
6 AFingestion = absorption fraction for ingestion
7
8 AFingestion was assumed to equal 0.3 based on recommendations in U.S. EPA (2003) for
9 ingestion of dioxin in soil. The ingestion of dioxin from inhaled particles is included in the
10 inhalation dose as discussed above.
11
12 5.4. TOTAL DOSE
13 The total absorbed dose was estimated to be the sum of the dermal absorption, inhalation,
14 and ingestion doses as shown below:
15
J-1> J-J total J-J dermal ' J-J inhalation ' J-J ingestion (y
17
18 where:
19 Dtotai = total dose (pg TEQ/d)
20 Ddermai = dermally absorbed dose (pg TEQ/d)
21 Delation = inhalation dose (pg TEQ/d)
22 Digestion = ingestion dose (pg TEQ/d)
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 25 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
6. QUESTIONNAIRE RESULTS
The complete questionnaire and all responses are presented in Appendix A. The
questionnaire focused on characterizing each subject's work with clay in terms of
frequency/duration, type of activity, clothing worn, and impact on skin. Table 6 summarizes the
questionnaire results for the amount of time that the subjects spent working directly with clay.
The subjects worked with clay, on average, for 30 hours per week and 38 weeks per year over a
6-year period. The times varied widely, however, reflecting the types of students involved. A
student obtaining an advanced degree in ceramics is likely to work with clay daily over many
years. In contrast, a student who takes a pottery class to fulfill a general education requirement
is likely to experience similar exposures, but only for 1-3 hours per day over the duration of the
class (9 months or less).
Table 6. Questionnaire questions on duration and frequency of subject's
clay work
Question (n = 8)
Approximately how many hours per
week do you work with clay?
Approximately how many weeks per
year do you work with clay?
How long (years) have you been
doing clay work with this level of
intensity?
Mean (SD)
30(21)
38(10)
6(8)
Median
23
38
3
Max
70
52
24
Min
10
20
1
18
19
20
21
22
23
24
25
26
27
28
29
Table 7 summarizes the participants' answers to several questions about their clay work.
Some of the questions address the types of clothing worn, how often the subjects wash their
hands, and whether the subjects could correlate any skin health effects with working with clay.
All eight subjects answered that they have dry skin because of the clay work. In general, the
subjects wash their hands soon after working with clay, their face and arms within a few hours,
and the rest of their body within 24 hours. The responses indicated that one subject gets a rash
when using the wheel for throwing, another subject has nasal congestion due to clay work, and
another subject's fingernails do not grow well.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 26 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 7. Questionnaire questions about clay work
Question (n = 8)
Summary of answers
(number of subjects with similar answers)
What type of clay artwork do you do?
What types of clothing do you wear
while you work?
What areas of skin typically are exposed
to the clay while you work?
In relation to the time you complete
working with clay, when do you wash
parts of your body that have been
exposed to clay?
How do you wash your skin after you
work with clay?
Do you correlate any skin health issues
with how much you work with clay? If
yes, what?
Hand building/sculptural work (7), throwing on wheel (3),
mixing clay and maintenance work (1)
In general, long sleeves and pants in cool weather and short
sleeves and pants or shorts in warm weather; both closed-
toe shoes and sandals are worn at times
Always face and hands; arms, legs, and feet when exposed
Soon after: hands (8), arms (1), face (1)
Within a few hours: arms (2), face (6)
Within 24 hours: face (1), rest of body (4)
Soap and water or just water (8)
Dryness (8), rash on hands when using wheel (1), nasal
congestion (1), fingernails do not grow well (1)
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 27 DRAFT—DO NOT CITE OR QUOTE
-------
1 7. COMPARING EXPOSURES ACROSS SUBJECTS
2
3 In this section, a hypothetical dioxin dose is estimated for each subject and used to
4 evaluate which pathways and activities contribute most to total dose. This is done by assuming
5 that each subject uses clay with the same level of dioxin. More specifically, it is assumed that
6 each subject uses a clay mixture with 20% ball clay and that the ball clay contains 808 pg TEQ/g
7 (these are typical values as discussed in Section 4). Accordingly, the dioxin levels in the clay
8 were assumed to be 20% of 808 pg TEQ/g or 162 pg TEQ/g. This concentration was also
9 assumed to apply to inhaled dust and dust settled onto food. A variety of other factors were also
10 held constant across subjects to facilitate this analysis:
11
12 • Exposure duration. The questionnaire results presented in Section 6 indicate a
13 median weekly time for clay work of 23 hours. Assuming a 5-day work week,
14 this would correspond to about 4 hours/day. This value was applied to all
15 subjects.
16
17 • Monolayer load. The monolayer load varies depending on particle size but is
18 assumed here to be 0.62 mg/cm2 for all subjects. This is based on the geometric
19 mean of the range of possible median particle sizes, i.e., 0.75 to 27 jim (see
20 Section 5.1 for further discussion of this issue).
21
22 • Dermal absorption fraction. This will depend on exposure time, as discussed in
23 Section 5.1. The time that the skin is exposed to clay will vary with individual
24 behaviors and body parts. Some body parts (such as hands and faces) are likely to
25 be washed more frequently than others (such as feet, legs, and arms), resulting in
26 longer exposure times. The questionnaire data collected during this study (see
27 Section 6) suggest that the artists generally wash their hands soon after working
28 with clay, wash their faces and arms within a few hours, and wash the rest of their
29 body within 24 hours. Accordingly, the exposure time for feet and legs was
30 assumed to be 24 hours, and the absorption fraction corresponding to 24 hours
31 was applied (6.2%). The exposure time for hands, arms, and face was assumed to
32 be 4 hours with a corresponding 1.7% absorption.
33
34 • Ingestion absorption fraction. This was set to 0.3 based on recommendations in
35 U.S. EPA (2003) for ingestion of dioxin in soil.
36
37 • Inhalation absorption fraction. This was set to 0.3 for ET and TB regions based
38 on the assumption that the area is rapidly cleared to the gastrointestinal tract. It
39 was set to 0.8 for the PU region based on recommendations in U.S. EPA (2003)
40 for inhalation of dioxin in air.
41
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 28 DRAFT—DO NOT CITE OR QUOTE
-------
1 The hypothetical dioxin dose for each subject is calculated using the constant values
2 described above and their individual exposure conditions (e.g., dust level in air, clay load on
3 skin, clay load on food). The dose estimates are considered to be hypothetical because they are
4 based on assumed dioxin levels in the various exposure media rather than on studio-specific
5 measurements. Section 8 presents an analysis of the possible variability in dose resulting from a
6 range of dioxin levels in clay, ball clay mixtures, and exposure factors (Monte Carlo
7 simulations).
8 This section first addresses each pathway separately (dermal contact, inhalation, and
9 ingestion) and then addresses total dose. Individual exposures vary widely, and it is important to
10 consider the subject's activity and clothing in evaluating the results. Table 8 is provided as a
11 reference for this purpose with summaries of each participant's activities and clothing.
12
13 7.1. DERMAL CONTACT
14 As described in Section 5.1, the mass of clay rinsed from the skin was used to estimate
15 clay loadings on the skin for each exposed body part. The rinsing data are presented in
16 Appendix H. Section 5.1 also explains that the skin loading is compared to the monolayer load,
17 and the absorption fraction is applied to the lower amount. The dermal absorption estimate for
18 each subject is shown in Table 9. Subjects 1 through 8 wore clothing that limited their exposures
19 to only hands and arms (although arm exposure was detected on only Subjects 1 and 6). The
20 estimates for Subjects 9 and 10 include hands, arms, legs, and feet because they wore clothing
21 allowing exposure to these areas. All subjects could have had exposure to the face, but this was
22 evaluated only for Subjects 9 and 10. Pictures of the clay residues on skin are shown in
23 Appendix B. Table 9 shows that 5 of the 10 subjects had skin exposures exceeding the
24 monolayer. The absorbed dose ranged from 0.41 to 20.80 pg TEQ/d with a mean of 3.37 pg
25 TEQ/d(SD = 6.18).
26 The relationships between the activities of the subjects and their dermal exposure, as
27 presented in Table 9, are discussed below:
28
29 • Wheel work (Subjects 6 and 9). This activity led to the highest dermal
30 exposures. The high exposures were caused by the close proximity of the subjects
31 to the wheel, the splashing of wet clay onto their bodies, and the use of both hands
32 to mold the clay. The total dermal dose for Subject 9 was about 6 times greater
33 than that for Subject 3, resulting primarily from their clothing difference. Both
34 had similar hand and arm exposure, but Subject 9 had high exposure to legs and
35 feet and Subject 6 had no exposure in these areas.
36
37
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 29 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 8. Artisan activities of each subject
Artisan/staff
(minutes sampled)
Description of activity
Clothing
Test 1, April 2003
Subject I/male
(153 min)
Subject 2/male,
nonartisan staff
(84 min)
Subject 3/female
(124 min)
Subject 4/female
(121 min)
Subject 5/male
(136 min)
Subject 6/female
(123 min)
Subject 7/female
(124 min)
Subject 8/female
(138 min)
Wedged clay on a wedging board to remove air from the
clay before kneading and shaping clay by hand. Used a
wooden press to press the clay into flat, approximately 2.5
cm thick sheets. Also, pounded semi-dry clay into balls,
placed in ball mill for smoothing rough edges.
Poured powdered components into large mixer for clay
manufacture while wearing dust mask and while the dust
removal system was operational. Weighed out portions of
clay, and bagged and stored them. Subject moved to gas
kiln room, where he cut blocks, built the kiln up a bit, and
vacuumed. Finally, subject used compressed air to clean
the dust off himself.
Subject wedged clay and covered a prefabricated mold
with clay using her hands to mold and shape the clay.
Subject cut pre-wedged and formed blocks of clay into 5
cm thick pieces, loaded the blocks into a pneumatic press,
pressed a pattern into each and cut blocks to the proper
shape, and then stacked the finished pieces to be fired.
Subject hand rolled clay into 60 cm long "snake-like"
cylinders, which he then hand-formed into conical pots.
Subject threw a variety of clay items, including a pitcher,
a vase, pots, and bowls on the pottery wheel.
Subject wedged, rolled, cut, and hand-built a variety of
items.
Subject wedged, rolled, shaped, cut, and hand-built large
pieces of clay and placed them on a mold.
Short-sleeved shirt, long
pants, socks, shoes
Short-sleeved shirt, long
pants, socks, shoes
Short-sleeved shirt, long
pants, socks, shoes
Long-sleeved shirt (rolled
up), long pants, socks,
shoes
Short-sleeved shirt, long
pants, socks, shoes
Short-sleeved shirt, long
pants, socks, shoes
Short-sleeved shirt, long
pants, socks, shoes
Short-sleeved shirt, long
pants, socks, shoes
Test 2, July 2004
Subject 9/female,
five sessions
(295-476 min)
Subject 10/female,
three sessions
(406-438 min)
Subject threw a variety of clay items, including plates,
bowls, vases, and cups, on the pottery wheel.
Subject sculpted detailed designs into clay tiles and
plaques; also chipped small bits of excess clay off pieces
of art that had already been fired.
Short-sleeved shirt, short
pants, sandals
Short-sleeved shirt, 3/4
length pants, sandals
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 30 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 9. Hypothetical estimates of dermal dose
Body part
Subject 1
Hands
Arms
Total
Subject 2
Hands
Subject 3
Hands
Subject 4
Hands
Subject 5
Hands
Subject 6
Hands
Arms
Total
Subject 7
Hands
Subject 8
Hands
Clay load on skin Skin area
(mg/cm2) c (cm2)6
0.38 970
0.15 2,406
[2.01] 970
0.51 865
0.17 855
[2.61] 1,005
[9.25] 790
[2.99] 2,005
0.26 785
[1.90] 715
Fraction
uncovered
1.0
0.5
1.0
1.0
1.0
1.0
1.0
0.6
1.0
1.0
Absorbed dioxin
dose
(pg TEQ/day)a'M
1.00
0.49
1.50
1.65
1.2
0.41
1.71
1.34
2.04
3.38
0.57
1.21
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 31 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 9. Hypothetical estimates of dermal dose (continued)
Body part
Subject 9
Hands
Arms
Lower legs
Feet
Face
Total
Subject 10
Hands
Arms
Lower legs
Feet
Face
Total
Clay load on skin
(mg/cm2) c
[10.12]
[1.50]
[0.72]
0.26
0.03
0.20
0.04
0.11
0.03
0.04
Skin area
(cm2)6
857
2,265
2,161
1,151
374
783
2,271
2,095
1,109
368
Fraction
uncovered
1.0
0.75
1.0
1.0
1.0
1.0
0.9
0.1
1.0
1.0
Absorbed dioxin
dose
(pg TEQ/day)a'M
1.45
2.88
13.44
2.99
0.03
20.80
0.42
0.22
0.23
0.30
0.04
1.22
4
5
6
7
8
9
10
11
12
13
14
15
16
17
"Absorption = skin load (mg/cm2-day) x skin area (cm2) x fraction uncovered x dioxin concentration in clay (pg
TEQ/g) x 10~3 mg/g x absorption fraction.
bAll calculations assume dioxin concentration in clay = 162 pg TEQ/g and absorption fraction is 6.19% for feet and
legs, and 1.69% for hands, arms, and face.
°A11 bracketed loads exceed monolayer of 0.62 mg/cm2 and were reduced to this value in absorption calculation.
dResults from Subjects 1 through 8 are based on one work session, from Subject 9 are based on average of five
sessions, and from Subject 10 are based on average of three sessions.
eSkin area is for total body parts; for two-sided parts, it is the sum of right and left sides.
TEQ = toxic equivalent
• Mixing (Subject 2). Subject 2 was involved in the mixing and handling of dry
clays and furnace/kiln maintenance during the work session. This activity
produced relatively large hand loadings.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 32 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
• Wedging and molding (Subjects 1, 3, 4, 5, 7, and 8). Wedging clay involves
kneading and hitting clay against a tabletop to purge air pockets from the clay.
During the wedging process, the clay is firm and dry as compared with clay used
on the wheel. This activity produced a wide range of hand loadings (from 0.17 to
2.61 mg/cm2).
• Sculpting (Subject 10). This involved sculpting activities on dry clay. At times,
fine detailing tools were used that involved very little contact with the clay,
resulting in low hand loading.
Table 10 shows the percent contribution to the dermal dose by body part for Subjects 9
and 10. Subjects 9 and 10 were tested in July 2004 and wore summer clothing, which allowed
exposure to their legs and feet. Leg and foot exposure accounted for 79% of the total dose for
Subject 9 and 44% of the total dose for Subject 10. This reflects the relatively large surface
areas and higher absorption fraction (due to longer exposure time) for these parts. The
uncovered portion of Subject 10's lower legs was only 10%, so the leg contribution to total dose
was much less than that of Subject 9. Facial exposures were low, accounting for only 0.1-3% of
total dose.
Table 10. Percent contribution to dermal dose by body part
Body part
Hands
Arms
Legs
Feet
Face
Percentage of dose
Subject 9 (wheel)
7
14
65
14
0.1
Subject 10 (sculpture)
34
18
19
25
3
23
24
25
26
27
28
7.1.1. Clay Loads on Surfaces
The horizontal surfaces in ceramic art studios can have high dust loads resulting from air
deposition. Most clay on the hands of artisans probably results from direct contact with clay, but
some could also result from contact with surfaces. In the interest of exploring this issue, wipe
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 33 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
samples were collected from the work surface of each subject. These results are shown in
Table 11. The surface dust loads ranged from 0.2 to 7 mg/cm2, which are high compared with
dust loads on floors in residences (i.e., 0.005 to 0.7 mg/cm2) (Lioy et al., 2002). The efficiency
of transfers from surfaces to hands will vary depending on the type of surface, type of residue,
hand condition, force of contact, etc. Rodes et al. (2001) conducted hand press experiments on
particle transfer to dry skin and measured transfers with central values of about 50% from hard
surfaces. Several of the ratios of hand loads to surface loads given in Table 11 exceed 50% by a
wide margin. Subject 6 was working on a wheel and clearly had hand loads resulting from direct
contact with clay. Similarly, Subjects 5 and 8 had very high hand loads that must have resulted
from direct clay contact. The other subjects had ratios ranging from 0.05 to 0.30, which are in
the range that could result from surface transfers. Observation of the subjects indicated that
almost all contact with the work surface also involved some contact with the clay. Therefore, the
hand residues are most likely derived from a combination of direct clay contact and transfers
from surfaces.
Table 11. Comparing clay loads on surfaces to clay loads on hands
Subject
1
2
3
4
5
6
7
8
Clay loading on surface
(mg/cm2)
7.002
NA
2.966
0.572
0.774
0.238
1.206
0.419
Clay load on hand
(mg/cm2)
0.38
2.01
0.51
0.17
2.61
9.25
0.26
1.90
Ratio of hand load to
surface load
0.05
NA
0.17
0.30
3.4
38.9
0.22
4.5
19
20
21
22
23
NA = Nonartisan subject was not working at a surface during sampling, so this type of sample
was not collected.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 34 DRAFT—DO NOT CITE OR QUOTE
-------
1 7.1.2. Dermatologist Report
2 The dermatologist did not diagnose any serious skin health problems among the subjects.
3 Small abrasions and common skin conditions such as dryness and cracking, as the subjects
4 reported on the questionnaires, were noted, but changes in these conditions could not be detected
5 based on before and after observations.
6
7 7.2. INHALATION
8 Estimating the inhalation dose involved measuring particle concentrations in air and
9 modeling deposition to various regions of the respiratory system. Classroom exposures were not
10 estimated.
11
12 7.2.1. Particle Levels in Air
13 As described in Section 3, four different sampling techniques were used during the April
14 2003 tests to measure clay particle concentrations in air: two personal monitors and two area
15 monitors. The data from all four devices are shown in Appendixes C and D. The Respicon
16 personal air sampler normally would have been the best indicator of individual exposures, but
17 the blanks were high, resulting in a high detection limit and a high frequency of nondetects in the
18 data. Instead, the cascade impactor was chosen as the best indicator of daily exposure. Although
19 this is an area sampler, it was located near the subjects and the subjects were generally stationary
20 during the test. Thus, it should have been a reasonable indicator of individual exposures. Also,
21 the cascade impactor uses deposition collectors and gravimetric techniques to estimate air
22 concentrations; consequently, it is a more direct measurement technique than the other two
23 instruments (pDR-1000 and Climet), which use light scattering to estimate particle
24 concentration. These optical devices provide a nearly continuous readout of concentration
25 levels, making them better suited to evaluating short-term fluctuations in particle levels rather
26 than long-term concentrations.
27 Only the cascade and Climet monitors were used in the July 2004 tests. The instruments
28 were located even closer to the individuals, i.e., within 30 cm of their breathing zones. The data
29 were used in a fashion consistent with the April 2003 tests, i.e., daily exposures were based on
30 the cascade data and the Climet was used to evaluate short-term fluctuations.
31 Table 12 presents the air data for each subject on the basis of the cascade measurements.
32 The MMADs were estimated by fitting the data to log-normal distributions (see the discussion in
33 Appendix G). Table 12 indicates that the range for total particulate matter is 0.084 to 0.99
34 mg/m3. Note that the upper end of this range is less than the Occupational Safety and Health
35 Administration (OSHA) standard for total particulates of 15 mg/m3 (OSHA, 2004). Subject 3' s
36 concentration was the highest because students were cleaning the floor near the area samplers
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 35 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
(see the discussion below). Subject 9's concentration was the lowest as a result of a relatively
low activity level during the testing. Subject 5's concentration was also low, likely because a
steady breeze entered through an open window in the room in which sampling was occurring.
All of the other subjects had fairly similar concentrations.
Table 12. Particle concentrations in air and mass median aerodynamic
diameter (MMAD) based on cascade impactor
Subject
1
2
o
J
4
5
6
7
8
9
10
MMAD (urn)
26.9
44.6
18.5
25.0a
25.0a
20.2
13.0
26.7
32.6
16.0
Total concentration (mg/m3)
0.35
0.47
0.99
0.37
0.13
0.61
0.51
0.64
0.084
0.24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
"Nondetects prevented calculation of the MMAD for these subjects; they were assumed equal to the average over the
remaining first eight subjects.
The two subjects using wheels (Subjects 6 and 9) had very different air exposures.
Because a great deal of water is used to moisten clay during wheel molding (the clay was
saturated with water and a pan of water was placed directly next to the artisans for their use), this
setting would not be expected to produce much clay dust, which was observed for Subject 9.
Subject 6, however, had fairly high air levels. Subject 6 was located near a classroom that, as
discussed below, had high activity levels. Therefore, this subject's high air levels may have been
associated more with the classroom activities than the wheel activities.
Figure 3 shows the plot of concentration versus time (based on the Climet CI-500 area
particle counter) for Subject 3, who worked in an area designated for graduate student work
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 36 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
adjacent to a large classroom. Approximately 50 minutes into the sampling session, about 20
students from the adjacent classroom began sweeping and wiping down the surfaces. This
activity continued for approximately 15 minutes and generated a significant cloud of dust. As
shown in Figure 3, particle levels began rising at about 50 minutes, peaked sharply at 60-70
minutes, and declined to low levels at about 80 minutes.
Area Particle Concentration
using the CI-500 Particle Counter
25
50
75
100
125
Time (min)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Figure 3. Real-time particle concentration for Subject 3 using the CI-500
particle counter.
During two of Subject 10's sculpture work sessions, a small dog was present. The dog's
movement disturbed dust on the floor of the ceramics studio and, in turn, increased the particle
concentration. Figures 4 and 5 are the real-time traces for the Climet monitor for the sculpting
work sessions during which the dog was present. The dog was present for the entire first
sculpting work session. This was reflected in the relatively constant variation in the particle
concentration throughout the work session. During the second sculpting work session, the dog
did not arrive until 138 minutes into sampling. Note the increase in overall particle
concentration and increase in variability of particle concentration after arrival of the dog. The
presence of a dog in the studios and classrooms is not likely to be a common occurrence,
especially during the regular school year. Therefore, the particle concentrations during the work
sessions when the dog was present (1 and 2) were not used to estimate the exposures for this
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 37 DRAFT—DO NOT CITE OR QUOTE
-------
1 subject. It should be noted, however, that pets, which may be present in many ceramic art
2 studios, can have a large influence on the suspended dust levels and spread dust to other areas.
Sculpture Session 1
4
5
6
7
9
10
60
120
180 240
Time (min)
300
360
420
Figure 4. Sculpture session 1 with dog present.
Sculpture Session 2
60
120
180 240
Time (min)
300
360
420
Figure 5. Sculpture session 2 with dog present.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 38 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
7.2.2. Inhalation Dose
Table 13 shows the absorbed dose in various regions of the respiratory system for all 10
subjects. The total inhalation doses ranged from 0.006 to 0.09 pg TEQ/d with an average of
0.04 pg TEQ/d. Most particle deposition was found to occur in the extrathoracic region. The
modeling to support these estimates is presented in Appendix G.
Table 13. Hypothetical estimates of inhalation dose
Subject
1
2
o
J
4
5
6
7
8
9
10
Absorbed dose (pg TEQ)a
ETb
0.032
0.033
0.082
0.028
0.012
0.054
0.049
0.048
0.005
0.022
TBb
0.001
0.001
0.002
0.001
0.000
0.001
0.001
0.001
0.000
0.001
PIT
0.003
0.003
0.010
0.002
0.001
0.004
0.006
0.003
0.001
0.002
Total
0.035
0.036
0.094
0.031
0.014
0.059
0.057
0.052
0.006
0.025
10
11
12
13
14
15
16
17
18
19
20
21
22
aDose calculated using procedures in Appendix G for nasal breathing; subject exposure concentrations from
Appendix D; 4-hour exposure duration and dioxin concentration of 162 pg TEQ per gram clay.
bAbsorption fraction of 0.3 assumed, since these regions rapidly clear into the gastrointestinal tract.
0 Absorption fraction of 0.8 assumed, in part, due to slow particle clearance from this region.
TEQ = toxic equivalent; ET = extrathoracic; TB = tracheobronchial; PU = pulmonary
The inhalation exposure estimates assume that no respiratory protection was used.
Generally this was true, however, Subject 2 used a dust mask while pouring powdered clay into a
mixer for clay preparation. This reduced his inhalation exposures relative to levels reported here.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 39 DRAFT—DO NOT CITE OR QUOTE
-------
1 7.2.3. Classroom Exposure
2 Estimating student exposures in a classroom setting was not an objective of this study.
3 However, some insight on this issue can be gained from the data for Subjects 1, 3, and 6. These
4 subjects performed their clay activities adjacent to the undergraduate classroom during times
5 when undergraduate classes of 20-25 students were participating in clay-related activities. The
6 area particle samples collected for these subjects are generally representative of the inhalation
7 exposure of students in those classes. As discussed above, students in this class swept the floor
8 during Subject 3's testing period, producing elevated particle concentrations for about
9 30 minutes.
10
11 7.3. INGESTION
12 The ingestion dose was calculated by assuming that all deposited material on the
13 surrogate food and beverage samples was ingested. As Table 14 shows, clay deposition onto the
14 food and beverage samples reached detectable levels in only 5 out of 16 total samples. The
15 deposition amounts for the nondetects were assumed to equal half the detection limit. The
16 resulting ingestion doses ranged from 0.03 to 0.1 pg TEQ/d. The field technicians did not
17 observe hand-to-mouth activities for any of the subjects. Also, none of the subjects ate food or
18 smoked without first washing the clay from their hands. No deposition samples were collected
19 for Subjects 9 and 10.
20
21 7.4. TOTAL DOSE
22 Table 15 lists the hypothetical estimates of total dioxin dose derived by summing across
23 exposure pathways for each subject. The total doses ranged from 0.49 to 20.81 pg TEQ/d with
24 an average of 3.45 pg TEQ/d. Table 16 shows the percentage contribution of each exposure
25 pathway to the total dose of each subject. Dermal absorption is the major contributor to total
26 dose for all subjects, exceeding 78% for all subjects. Ingestion and inhalation contribute similar
27 amounts, generally in the range of 1-10%.
28 Table 17 shows the dose estimates by activity. The highest total doses were associated
29 with wheel activities.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 40 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 14. Clay deposition and hypothetical estimates of ingestion dose
Subject
1
2
3
4
5
6
7
8
Clay deposited onto
food (mg)
0.71
-------
1
2
Table 15. Hypothetical estimates of total dioxin dose (pg TEQ/day)
Subject
1
2
3
4
5
6
7
8
9
10
Mean (SD)
Median
Minimum
Maximum
Estimated dioxin dose (pg TEQ/day)
Inhalation
0.035
0.036
0.094
0.031
0.014
0.059
0.057
0.052
0.006
0.025
0.041 (0.025)
0.036
0.006
0.094
Ingestion
0.07
0.03
0.03
0.05
0.03
0.03
0.1
0.09
NM
NM
0.05 (0.03)
0.04
0.03
0.10
Dermal absorption
1.50
1.65
1.20
0.41
1.71
3.38
0.57
1.21
20.80
1.22
3.37(6.18)
1.36
0.41
20.80
Total
1.61
1.72
1.32
0.49
1.75
3.47
0.73
1.35
20.81
1.25
3.45(6.15)
1.48
0.49
20.81
3
4
TEQ = toxic equivalent; NM = not measured; SD = standard deviation
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 42 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 16. Percent contribution to total dioxin dose
Subject
1
2
3
4
5
6
7
8
9
10
Percentage of dose
Inhalation
2.2
2.1
7.1
6.3
0.8
1.7
7.8
3.9
0.0
2.0
Ingestion
4.4
1.7
2.3
10.2
1.7
0.9
13.8
6.7
NM
NM
Dermal absorption
93.4
96.2
90.7
83.5
97.5
97.4
78.4
89.5
100.0
98.0
3
4
NM = not measured
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 43 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 17. Dose estimates by activity
Activity
Wedging and
molding
Mixing
Wheel
Sculpting
Subject
1
3
4
5
7
8
2
6
9
10
Inhalation
dose
(pg TEQ/day)
0.035
0.094
0.031
0.014
0.057
0.052
0.036
0.059
0.006
0.025
Ingestion
dose
(pg TEQ/day)
0.07
0.03
0.05
0.03
0.1
0.09
0.03
0.03
NM
NM
Dermal
dose (pg
TEQ/day)
1.50
1.20
0.41
1.71
0.57
1.21
1.65
3.38
20.80
1.22
Total dose
(Pg
TEQ/day)
1.61
1.32
0.49
1.75
0.73
1.35
1.72
3.47
20.81
1.25
4 NM = not measured; TEQ = toxic equivalent
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 44 DRAFT—DO NOT CITE OR QUOTE
-------
1 8. MONTE CARLO SIMULATION OF THE EXPOSURE DATA
2
3 Section 7 presented hypothetical dose estimates for each subject, assuming that all were
4 using typical amounts of ball clay with average dioxin levels. In this section, Monte Carlo
5 simulations are used to explore the doses that could occur in a broad population of artists with a
6 wide range of behaviors using ball clay with differing levels of dioxin.
7 The general strategy for selecting input value distributions was as follows. The
8 distribution of skin surface areas across adults in the general population was assumed to be log-
9 normal with mean and standard deviation from the Exposure Factors Handbook (U.S. EPA,
10 1997). Similarly, the dioxin concentration in clay was assumed to have a log-normal distribution
11 with mean and standard deviation from Ferrario et al (2004, 2007). The rationale for choosing
12 log-normal distributions was that physiological parameters and environmental media
13 concentrations are commonly found to have these types of distributions. The distributions were
14 truncated at the minimum and maximum data points to eliminate the chance that some simulation
15 trials could use unreasonable values. The remaining exposure factor parameters were based on
16 observations from this study. These were generally assumed to have triangular distributions with
17 ranges based on minimum and maximum values and peaks based on means. The rationale for
18 choosing a triangular distribution was that (1) the small sample sizes associated with the study
19 observations prevented fitting the data to standard distributions and (2) it reflected the likelihood
20 that a central value would occur most often. In some cases (e.g., clay load on face), only two
21 data points were available and a uniform distribution was assumed. The distributions assumed
22 for all input variables are listed in Table 18.
23 Crystal Ball 7 software was used to conduct 1,000 trial simulations. For each simulation
24 trial, a set of parameter values was obtained by randomly sampling the parameter distributions as
25 listed in Table 18 and then computing the dioxin dose. The dose was calculated using the
26 equations presented in Section 5. All simulation trials first select a set of values for the dioxin
27 concentration in ball clay, the fraction of ball clay in the blend used by the artist, and the
28 exposure duration. These are shown as general parameters in Table 18. The simulation then
29 calculates the dose from the dermal, inhalation, and ingestion pathways, as discussed below:
30
31 • Dermal. The simulation was designed to first select a total body surface area
32 from a log-normal distribution. Subsequently, skin surface areas for individual
33 body parts were calculated by multiplying the total surface area by the average
34 percentage of total surface area. These percentages were obtained from U.S. EPA
35 (1997): hands, 5.2%; arms, 14%; legs, 31.8%; feet, 6.8%; and face, 2.5%
36 (assumes face area equals one-third of head area). This approach ensures that
37 simulation trials have realistically matched body part areas. Since the body part
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 45 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 18. Monte Carlo simulation input parameters and sampling distributions
Parameter
Distribution
Basis
General parameters
Dioxin concentration in ball clay
(pg TEQ/g)
Fraction of ball clay in blend
Exposure duration (hr/d)
Log-normal (mean = 808,
SD = 318)
Triangular (0, 0.2, 1.0)
Triangular (1,4, 10)
Ferrario et al. (2004, 2007) (n = 21); truncated
at range limits
Data in this study (n = 10)
Judgment and data from this study (n = 8)
Dermal absorption parameters
Total body surface area (cm2)
Clothing selector
Clay load on hand (mg/cm2)
Clay load on arm (mg/cm2)
Clay load on leg (mg/cm2)
Clay load on feet (mg/cm2)
Clay load on face (mg/cm2)
Log-normal (mean= 18,000,
SD = 37.4)
Uniform (0, 1.0)
Triangular (0.1, 3.0, 10)
Triangular (0.04, 0.35, 3.0)
Uniform (0.1, 0.70)
Uniform (0.03, 0.3)
Uniform (0.03, 0.04)
Exposure Factors Handbook (U.S. EPA,
1997); truncated at range limits (n = 32)
Judgment and data from this study (n = 8)
Range and mean based on observations from
this study (n = 10)
Data in this study (n = 4)
Data in this study (n = 2)
Data in this study (n = 2)
Data in this study (n = 2)
Ingestion parameters
Clay load on food (mg)
Clay load on beverage (mg)
Triangular (0.3, 0.7, 1.66)
Triangular (0.3, 0.5, 0.72)
Range and mean based on observations from
this study (n = 8)
Range and mean based on observations from
this study (n = 8)
Inhalation parameters
Particle concentration in air
(mg/m3)
Median particle size (um)
Lung parameters
Fraction of time engaged in light
vs. moderate exertion.
Breathing type
Triangular (0.08, 0.44, 0.99)
Triangular (13, 25, 45)
Male, 30%; female, 70%
Uniform (0, 1.0)
Oronasal, 13%; nasal, 87%
Range and mean based on observations from
this study (n = 10)
Judgment and data from this study (n = 10)
Male/female split based on data in this study
(n=10)
Judgment
Brown (2005)
3
4
5
6
7
area calculations give total areas, a fraction unclothed was used to reduce this to
the exposed area. These fractions were based on four clothing scenarios as shown
in Table 19. These clothing scenarios were based on questionnaire responses and
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 46 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
judgment about typical apparel for a moderate climate. A clothing scenario was
selected randomly for each simulation trial according to the time fractions shown
in Table 19. Distributions were also assumed for the clay loads on skin. These
were assumed to be spread uniformly over the entire unclothed area. As
discussed in Section 5.1, dermal absorption was assumed to be limited to the
monolayer that was held constant at the median value of 0.62 mg/cm2 (the impact
of changing this value is discussed as an uncertainty issue in Section 9). Finally,
the absorption fractions (as presented in Section 5.1) were applied to derive the
absorbed dose from exposed body parts and then summed to derive total dermal
dose.
Table 19. Clothing scenarios based on questionnaire responses
Clothing scenario
Long-sleeved shirt, long pants, shoes
Short-sleeved shirt, long pants, shoes
Short-sleeved shirt, short pants, shoes
Short-sleeved shirt, short pants, sandals
Time fraction
0.2
0.6
0.1
0.1
Fraction unclothed
Arms
0
0.67
0.67
0.67
Legs
0
0
0.67
0.67
Feet
0
0
0
1.0
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Inhalation. The inhalation dose was calculated using the procedures summarized
in Section 5.2 and presented in detail in Appendix G. Distributions were used to
represent the variability in total particulate concentration in air and median
particle size (see Table 18). Breathing was assumed to be either oronasal (13%)
or nasal (87%), based on Brown (2005). Inhalation parameters (see Appendix G)
were based on an average female for 70% of the trials and an average male for
30% of the trials. The rate of breathing was determined by the fraction of time
engaged in light versus moderate exertion. These fractions were varied randomly
from 0 to 1.0 using a uniform distribution. Depositions to various parts of the
respiratory system were modeled as described in Appendix G, multiplied by the
absorption fraction, and summed to derive the total inhalation dose.
Ingestion. The variability in ingested dose was simulated using distributions for
the levels of clay in the food and beverages as shown in Table 18. As discussed
in Section 5.3, all deposited material was assumed to be ingested.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 47 DRAFT—DO NOT CITE OR QUOTE
-------
1 Two Monte Carlo stimulations were conducted. The first simulation was designed to
2 evaluate the influence of clay use only. Accordingly, it was conducted using the distributions for
3 dioxin concentration in the clay and the fraction of ball clay in the blend used by the artists. All
4 other inputs were held constant at their central values. The summer clothing scenario was used
5 (i.e., short-sleeved shirt, short pants, sandals). This simulation produced a mean total dose of
6 39 pg/d, median of 33 pg/d, and 90th percentile of 73 pg/d. These results are best compared to
7 the hypothetical dose estimate for Subjects 9 and 10 (presented in Section 7) because they wore
8 summer clothing matching the simulation assumption. Subject 9 had a dose estimate of 21 pg/d,
9 corresponding to about the 30th percentile of the simulation. Subject 10 had a dose of 1.5 pg/d,
10 corresponding to about the 2nd percentile of the simulation. This simulation suggests that clay
11 choice alone can account for a wide range of exposures with the potential to elevate exposures
12 above the hypothetical estimates for the 10 subjects.
13 The second simulation used the distributions for all parameters as shown in Table 18.
14 This simulation produced a mean total dose of 16 pg/d, median of 8 pg/d, and 90th percentile of
15 37 pg/d. The standard deviation exceeds the mean indicating that the results have a wide spread
16 as shown in Figure 6. The hypothetical dose estimates of most subjects would have
17 corresponded to low percentiles of this simulation except Subject 9 (80th percentile). Table 20
18 shows the simulation results for each pathway. The simulation means for each pathway
19 exceeded by 3 to 4 times the means of the hypothetical dose estimates for the 10 subjects. As
20 observed during the field study, the ingestion and inhalation doses are much smaller than the
21 dermal dose. The frequency diagram for total dose is shown in Figure 6. This figure shows a
22 highly skewed distribution with a peak around 3 pg TEQ/d and a long tail to the right extending
23 to about 70 pg TEQ/d. A detailed report showing all inputs and outputs for this simulation is
24 presented in Appendix F.
25 A sensitivity analysis was performed using the Crystal Ball 7 software. Each input
26 parameter was evaluated using contribution to variance and rank order correlation (Figures 7 and
27 8). These analyses showed that clothing selected contributed most to variance (37.9%), followed
28 closely by fraction of ball clay in blend (37.7%), dioxin concentration (16.6%), and exposure
29 duration (5%).
30 Overall, the simulation suggests that higher exposures than those reflected in the
31 hypothetical dose estimates of the 10 subjects may occur. This results from the skewed input
32 distributions, which generally have long right-hand tails. Also 6 of the 10 subjects had hand
33 exposure only, and the simulation uses a range of clothing that will result in more skin exposure
34 in most trials.
35
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 48 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
Table 20. Descriptive statistics of dioxin doses from ball clay use, based on a
Monte Carlo simulation
Pathway
Dermal dose (pg
TEQ/d)
Ingestion dose (pg
TEQ/d)
Inhalation dose (pg
TEQ/d)
Total dose
(pg TEQ/d)
Mean
15.5
0.14
0.12
15.76
Standard deviation
22.91
0.10
0.13
23.01
Median
7.92
0.11
0.08
8.12
90th Percentile
36.15
0.28
0.27
36.63
4
5
6
7
8
9
10
11
12
13
14
15
16
Total Dose
140
I. I 4'
0.00
10.00
20.00
30 00
40.00
pg/d
50.00
60 .00
70.00
80 .00
Figure 6. Frequency distribution of total dose (pg TEQ/day) based on Monte Carlo
simulation.
Many of the input distributions used in this simulation were based on very limited data or
judgment. A number of the distributions were based on data from this study, and the degree to
which the study subjects represented a broader population of artists is unknown. Similarly, the
degree to which the studio conditions observed in this study represent a broader set of studios is
unknown. The simulation should be interpreted as a preliminary indication of how to extrapolate
the study results to a broader population of artists.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 49 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
16.6%
Sensitivity: Total Dose
0.0% 10.0% 200% 300% 40.0%
clothing selector
Fraction o1 ball clay in blend
Dioxin cone in ball clay (p...
Exposure Duration (hrlcTi
Clay load on arm (mg/cm2)
Clay load on feet (mg/cm2)
Clay load on hand (mg/cm2)
M/F Selector
Breath Selector
MMAD (urn)
Activity selector
Clay load on beverage (nig)
5.0%
,c
0.9%
-0.2%
-0 %
-0. %
0.1%
-0. %
0.1%
Figure 7. Sensitivity analysis based on percent
contribution to variance.
0.00
0.20
0.40
0.60
clothing selector
Fraction ot ball clay in blend
Dioxin cone in ball clay (p...
Exposure Duration (hr/d)
Clay load on arm (mg/cm2)
Clay load on feet (mglcrr\2)
Clay load on hand (mg.tm2)
MF Selector
Breath Selector
MMAD (urn)
Activity selector
Clay load on beverage (mg)
0.21
0.09
0.03
Figure 8. Sensitivity analysis based on rank
correlation.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 50 DRAFT—DO NOT CITE OR QUOTE
-------
1 9. UNCERTAINTY
2
3 This section discusses general uncertainty issues and uncertainties related to the three
4 exposure pathways: dermal, inhalation, and ingestion.
5
6 9.1 GENERAL UNCERTAINTY ISSUES
7
8 The sensitivity analyses showed that the dioxin concentrations in clay and the fraction of
9 ball clay used account for a large part of the overall variance in the exposure estimates. Thus it
10 is important to consider the uncertainty in the assumptions regarding these two parameters.
11 The dioxin levels in ball clay were assumed on the basis of the study by Ferrario et al.
12 (2004, 2007). An important uncertainty issue is whether the ball clay sampled by Ferrario is
13 representative of the ball clay used in the studio and by the broader community of ceramic
14 artists. Ferrario et al. (2004, 2007) explained that the major mining companies market a total of
15 32 ball clay products of which 13 were sampled. Although marketing data were not available to
16 do true statistical sampling, a ceramics expert confirmed that the most commonly used ball clays
17 were included in this study. The samples were collected from 22.7 kg (50 pound) bags in the
18 same form as delivered to ceramic studios. Four of the 21 samples analyzed by Ferrario et al.
19 matched exactly the primary type of ball clay used in the OSU ceramics studio.
20 As explained earlier, ceramic artists use a wide range of clay blends with ball clay
21 contents ranging from 0 to 100%. The hypothetical dose estimates were based on the assumption
22 of 20% ball clay in the blend, which is the average fraction used by the 10 subjects in this study.
23 It is unknown how representative this is of the wider population of ceramic artists. The ball clay
24 fraction assumption may also affect other exposure factors. For example, it could affect how
25 much clay adheres to skin. Soil adherence to skin has been shown to be influenced by moisture
26 content and particle size. Ball clay is similar to other clays in terms of these properties. The
27 primary way that ball clay is unique from other clays is its high plasticity. It is not known how
28 this property would affect skin adherence.
29
30 9.2. DERMAL EXPOSURE UNCERTAINTIES
31 A fraction absorbed approach is used to estimate dermal absorption based on current
32 Agency guidance. As discussed in Section 5.1, this method has acknowledged weaknesses, but
33 the uncertainties are difficult to assess. Appendix I presents an alternative approach using a
34 more mechanistic model. This model predicts an absorbed dose that is similar to the fraction
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 51 DRAFT—DO NOT CITE OR QUOTE
-------
1 absorbed approach. The mechanistic model has had limited testing, and it is not yet clear
2 whether it provides more reliable estimates.
3 The exposures in the studio are caused by clay, but the dermal absorption fraction is
4 derived from soil experiments. An important uncertainty issue is whether clay has properties
5 that differ significantly from soil and consequently make the soil-derived absorption estimates
6 invalid for clay. The soil used by Roy et al. (1990) was 16.7% clay. This fraction of the soil
7 should have properties similar to those of the studio clay. The organic carbon content of the clay
8 is approximately the same as that of the low organic soil used by Roy et al. In terms of particle
9 size, clays typically have lower particle sizes than soil and would be expected to more strongly
10 sorb organic contaminants (e.g., dioxins) as compared with normal soils, all other factors being
11 equal. As discussed in Section 5, commercial ball clay specifications report a median particle
12 size of about 0.75 jim, which is smaller than that of the Roy et al. soil (median diameter of about
13 10 jim). The particle sizes measured in the studio air had median diameters ranging from 8 to
14 27 |im, which are larger than those of the soils used by Roy et al. This may be explained by the
15 bonding of particles caused by the addition of water to the clay or the firing process, which fuses
16 particles. Thus, it appears that the particle size of the soil used by Roy et al. falls within the
17 range present in the studio.
18 The studies on dermal absorption of dioxin from soil by Roy et al. and other investigators
19 have exclusively used TCDD. It is important to consider whether results for TCDD can be
20 extrapolated to the other dioxin congeners found in clay. As mentioned previously, the
21 compounds of concern in the clay are the tetra- through octa-CDD congener groups, as listed in
22 Table 21. This table indicates that molecular weight and the octanol-water partition coefficient
23 (Kow) increase with chlorine substitution. Molecular weight and Kow have been identified as key
24 chemical properties affecting dermal absorption (U.S. EPA, 1992). These properties also relate
25 to how tightly bound chemicals are to soils and their release kinetics. The higher chlorinated
26 congeners would be released from soils more slowly and permeate skin more slowly than TCDD.
27 Thus, use of TCDD experiments to represent the penta - octa dioxin congeners found in clay
28 probably leads to some overestimates of dermal absorption, but it is uncertain to what degree.
29 A related question is whether TCDD-derived dermal absorption values can be applied to
30 TEQs. As shown in Table 21, only about 9% of the TEQ in processed clay is derived from
31 TCDD. The TEFs used to determine TEQs discount the hepta- and octa- congeners much more
32 than the tetra- and penta- groups. The overestimates of dermal absorption for the higher
33 chlorinated congeners due to their higher molecular weights and Kow values will be compensated
34 to some extent by the large discounts during the TEQ calculation and thus make extrapolation of
35 dermal absorption data from TCDD to TEQs more reasonable.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 52 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
Table 21. Physical properties of dioxin congeners and concentration in processed
clay
Congener
TCDD
PeCDD
HxCDD
HpCDD
OCDD
Total
Molecular
weight
322
356.4
390.9
425.3
459.8
Log KoWa
6.1 to 7.1
6.2 to 7.4
6.85 to 7.8
8.0
8.2
Concentration
in processed
clay"
(Pg/g)
76
374
2,341
9,780
254,000
Concentration in
processed clayb
(pgTEQ/g)
76
374
234
97.8
25.4
808
% of total
TEQ
9
46
28
12
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
aU.S. EPA (2000)
bAverage values from Ferrario et al. (2004, 2007)
The amount of chemical that is dermally absorbed has been shown to be related to skin
thickness and whether the skin is dead or alive (U.S. EPA, 1992). Skin thickness varies across
body parts and across individuals. No information was found that could be used to account for
these factors in this analysis.
As discussed in Section 5.1, the monolayer calculation is also an important source of
uncertainty for the dermal absorption estimates. The monolayer load is estimated on the basis of
the median particle size and assumption of ideal packing. Actual monolayers will be composed
of a mix of sizes with complex packing that could result in loadings higher or lower than this
theoretical estimate. It is also uncertain how to best characterize the size distribution of particles
on the skin. The particles in the original clay product have a median particle size of about
0.75 jim, and the airborne particles have medians ranging from 8 to 27 jim. The particles on the
skin could more closely resemble either the airborne particles or the clay particles, depending on
the deposition mechanism. Accordingly, particle sizes of the clay residues on skin could vary
widely, with medians ranging from 0.75 to 27 jim. For purposes of the central exposure
estimates, the geometric mean of this range was assumed, i.e., 4.5 jim. This implies a monolayer
load of 0.62 mg/cm2. The monolayer loads corresponding to the upper and lower ends of the
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 53 DRAFT—DO NOT CITE OR QUOTE
-------
1 particle size range are 0.1 to 3.7 mg/cm2. This uncertainty is dampened in the dose estimate as a
2 result of the assumption that absorption occurs from only the monolayer. This dampening is
3 especially strong for low-exposure subjects. For example, the dose estimates for Subject 4 (who
4 had the lowest dermal exposure) corresponding to the low and high ends of the monolayer load
5 range would be 0.23 and 0.41 pg TEQ/day. Thus, a 37-fold variation in monolayer load resulted
6 in only a 1.8-fold variation in dose. The dampening is less (but still significant) for Subject 9
7 (who had the highest dermal exposures). For this subject, the doses corresponding to the low and
8 high ends of the monolayer load range would be 4.1 and 34.2 pg TEQ/day, respectively.
9 Another source of uncertainty in the dermal absorption estimates concerns the condition
10 of the skin. Some of the artists reported dryness and cracking of skin due to clay activities.
11 These conditions were observed by the dermatologist, but correlation with clay activities could
12 not be confirmed. Wheel operations involve work with wet clay which would hydrate the skin.
13 The abrasive nature of this work could also reduce the thickness of the stratum corneum which is
14 considered the primary barrier to permeation (U.S. EPA, 1992). It is possible that these
15 conditions would allow more dermal permeation than normal intact skin. However, any
16 increased permeation would be limited to the surface areas associated with the damaged skin.
17 Exposure could also occur through the eyes where absorption would likely be greater than intact
18 skin. This would be limited to particles that contact the eye surface which is probably minimal.
19
20 9.3. INHALATION UNCERTAINTIES
21 Data from the cascade sampler were used to estimate inhalation exposures. These data
22 were considered to be the most reliable because no samples were below detection limits and the
23 sampler uses a direct measurement method. The cascade, an area sampler, was located as near
24 the subject as possible but normally would not represent an individual's exposure as accurately
25 as a personal air monitor. Unfortunately, the data from the Respicon personal monitor were
26 dominated by nondetects and could not be used. The limited Respicon data that were above
27 detection limits generally indicated higher levels than the cascade, suggesting that personal
28 exposures may have been higher than those detected by the area monitor. Accordingly, use of
29 the cascade data may have resulted in underestimates of inhalation exposures.
30
31 9.4. INGESTION UNCERTAINTIES
32 The only ingestion pathway quantitatively evaluated in this study was direct ingestion of
33 clay deposited from the air onto food items. The measured deposition onto surrogate
34 food/beverage samplers may not match that of actual foods/beverages. Also, other pathways of
35 ingestion may occur. For example, clay could be transferred from hands directly to food.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 54 DRAFT—DO NOT CITE OR QUOTE
-------
1 Although this transfer was not observed in this study, it could be a fairly common occurrence
2 and has the potential for significant transfers to handheld food items (e.g., sandwiches, chips,
3 cookies). Clay ingestion could also occur from wiping the mouth or licking the lips. The
4 maximum ingestion levels estimated in this study involved about 2 mg of clay. This appears to
5 be low when compared to the 50 mg/day adult soil ingestion rate specified as a default
6 assumption in EPA guidance (U.S. EPA, 1997, 1989). This value is for residential scenarios and
7 includes both outdoor soils and indoor dusts. While it is logical that dust ingestion alone would
8 be less than ingestion of both soil and dust, a residence is likely to be much less dusty than a
9 ceramics studio. Ingestion of 69 mg of clay would be required to result in an absorbed dose
10 equal to the average dermal dose of 3.37 pg TEQ/d (this assumes the clay has an average
11 concentration of 162 pg TEQ/g and 30% of the dioxin is absorbed during ingestion).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 55 DRAFT—DO NOT CITE OR QUOTE
-------
1 10. CONCLUSIONS
2
3 Hypothetical dioxin dose estimates were calculated for each subject assuming that all
4 used a 20% ball clay blend with 162 pg TEQ/g. The single-day total doses across the 10 subjects
5 ranged from 0.49 to 20.81 pg TEQ/d, with an average of 3.45 pg TEQ/d. The dermal dose was
6 the major contributor to total dose, exceeding 78% for all subjects. Ingestion and inhalation
7 contributed similar amounts, generally in the range of 1 to 10% of total dose. Hand and arm
8 exposure accounted for much of the dermal dose for all subjects. The two subjects who wore
9 summer clothing had foot and leg exposures accounting for about 44 to 79% of the dermal dose.
10 Facial exposures were low accounting for less than 3% of total dermal dose.
11 Clay exposure was found to be highly dependent on the type of work being performed.
12 Throwing clay on the wheel resulted in much higher clay exposures than did any other clay
13 activities. This is due to the increased contact with clay while working on the wheel and the wet,
14 sticky consistency of the clay needed for that work. Emptying bags and mixing dried clays also
15 led to high exposures.
16 A Monte Carlo simulation was performed to model how doses could vary in a broad
17 population of artists with exposures outside the hypothetical scenario evaluated in this study.
18 The simulation, using a variety of assumed input distributions, suggests that doses could extend
19 to levels higher or lower than those estimated for the hypothetical scenario. Also, it indicated
20 that clothing, the fraction of ball clay in the blend and dioxin concentration contributed most to
21 variance in total dose. Many of the input distributions used in this simulation were based on very
22 limited data or judgment. Therefore, the simulation results are best interpreted as preliminary
23 indications of how to extrapolate the observations of this study to a broader population, and
24 further study is recommended to confirm these predictions.
25 In the general population, adult daily intakes of CDD/CDFs and dioxin-like
26 polychlorinated biphenyls (PCBs) are estimated to average 43 and 23 pg TEQ, respectively, for a
27 total intake of 66 pg TEQ/day (U.S. EPA, 2003). More than 90% of this intake is derived from
28 food ingestion. These intake values are based on the "administered" dose or the amount taken
29 into the body before absorption. The hypothetical doses presented in this report are on an
30 absorbed dose basis. Thus, the background dose must be converted to an absorbed basis to
31 compare it to the values presented here. U.S. EPA (2003) reports that about 80% of dioxins in
32 foods are absorbed into the body. Applying this factor, the background dose on an absorbed
33 basis is 34.4 and 18.4 pg TEQ/day for CDD/CDFs and dioxin-like PCBs, respectively, for a total
34 intake of 52.8 pg TEQ/day. Comparing these values to the average of the hypothetical doses for
35 the 10 subjects estimated here (3.45 pg TEQ/day) indicates that the ball clay dose is 10% of the
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 56 DRAFT—DO NOT CITE OR QUOTE
-------
1 background CDD/CDF dose and about 7% of the total CDD/CDF/PCB dose (on a TEQ basis).
2 Note that the general population dioxin dose is a long-term average and the hypothetical ball clay
3 dioxin dose is an estimate for a single day when exposure occurs. Accordingly, this comparison
4 implies that ball clay use is a frequent event, so that the long-term daily average ball clay dose is
5 similar to the single-day dose. If ball clay use is infrequent, then the long-term average dose
6 from ball clay will be reduced and adjustments would be needed to make a valid comparison to
7 the background dioxin dose.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 57 DRAFT—DO NOT CITE OR QUOTE
-------
REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). (2004) TLVs and BEIs: based on the
documentation of the threshold limit values for chemical substances and physical agents and biological exposure
indices. ACGIH, Cincinnati, OH.
Brady, NC. (1984) The nature and properties of soils. 9th ed. New York, NY: Macmillan Publishing.
Brown, JS. (2005) Particle inhalability at low wind speeds. Inhal Toxicol 17(14):831-837.
Bunge, AL; Parks, JM. (1998) Soil contamination: theoretical descriptions. In: Roberts, MA; Walters, KA, eds.
Dermal absorption and toxicity assessment. New York, NY: Marcel Dekker; pp. 669-696.
Ceramics Materials Info. (2003) Digitalfire Corporation, Canada. Available online at: http://ceramic-
materials.com/cermat/index.php?
Duff, R.M. and Kissel, J.C. (1996) Effect of soil loading on dermal absorption efficiency from contaminated soil. J
Toxicol Env Health, 48:93-106.
Ferrario, J; Byrne, C. (2000) Fob/chlorinated dibenzo-p-dioxins in the environment from ceramics and pottery
produced from ball clay. Organohalogen Compd 46:268-271.
Ferrario, J; Byrne, C. (2002) Dibenzo-p-dioxins in the environment from ceramics and pottery produced from ball
clay mined in the United States. Chemosphere 46(9-10):1297-1301.
Ferrario, J; Byrne, C; Cleverly, D. (2000a) Summary of evidence for the possible natural formation of dioxins in
mined clay products. Organohalogen Compds 46:23-26.
Ferrario, J; Byrne, C; Cleverly, D. (2000b) 2,3,7,8-Dibenzo-p-dioxins in mined clay products from the United
States: evidence for possible natural origin. Environ Sci Technol 34:4524-4532.
Ferrario, J; Byrne, C; Schaum, J. (2004) An assessment of dioxin levels in processed ball clay from the United
States. Organohalogen Compds 66:1639-1644.
Ferrario J; Byrne, C; Schaum, J. (2007) Concentrations of polychlorinated dibenzo-p-dioxins in processed ball clay
from the United States. Chemosphere 67(9):1816-1821.
Ferrario, J; Byrne, C; Lorber, M.; et al. (1997) A statistical survey of dioxin-like compounds in U.S. poultry fat.
Organohalogen Compd 32:245-251.
ICRP (International Commission on Radiological Protection). (1994) Human respiratory tract model for radiological
protection. A report of a Task Group of the International Commission on Radiological Protection; ICRP Publication
66. Ann ICRP 24(1-3): 1-482.
Kissel, JC; Richter, KY; Fenske, RA. (1996) Factors affecting soil adherence to skin in hand press trials. Bull
Environ Contam Toxicol 56(5):722-728.
Lioy, JL; Freeman, CG; Millette, JR. (2002) Dust: a metric for use in residential and building exposure assessment
and source characterization. Environ Health Persp 110(10):969-983.
OSHA (Occupational Safety and Health Administration). (2004) Part 1910, Occupational Safety and Health
Standards, subpart Z. Toxic and hazardous substances. Table Z-l, Limits for Air Contaminants. Washington, DC:
U.S. Government Printing Office. Available online at
http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=9992
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 58 DRAFT—DO NOT CITE OR QUOTE
-------
Poiger, H; Schlatter, CH. (1980) Influence of solvents and adsorbents on dermal and intestinal absorption of TCDD.
Food Cosmet Toxicol 18(5):477-481.
Rodes, CE; Newsome, JR; Vanderpool, RW; et al. (2001) Experimental methodologies and preliminary transfer
factor data for estimation of dermal exposure to particles. J Expos Anal Environ Epidemiol 11:123-139.
Roy, TA; Yang, JJ; Krueger, AJ; et al. (1990) Percutaneous absorption of neat 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD) and TCDD sorbed on soils. Toxicology 10(1):308.
Shu, H; Teitelbaum, P; Webb, AS; et al. (1988) Bioavailability of soil-bound TCDD: dermal bioavailability in the
rat. Fundam Appl Toxicol 10(2):335-343.
Touraille, GD; KD Mccarley, AL Bunge, JP Marty and RH Guy (2005). Percutaneous absorption of 4-cyanophenol
from freshly contaminated soil in vitro: Effects of soil loading and contamination concentration. Environ. Sci.
Technol. 39:3723-3731.
U.S. EPA (Environmental Protection Agency). (1989) Risk assessment guidance for Superfund. Vol. 1. Human
health evaluation manual (part A) [interim final]. U.S. Environmental Protection Agency, Office of Emergency and
Remedial Response, Washington, DC; EPA/540/1-89/002. Available online at
http://www.epa.gov/oswer/riskassessment/ragsa/index.htm.
U.S. EPA (Environmental Protection Agency). (1991) Percutaneous absorption of 2,3,7,8-TCDD and 3,3',4,4'-TCB
applied in soil [review draft]. U.S. Environmental Protection Agency, Office of Health and Environmental
Assessment, Exposure Assessment Group, Washington, DC; OHEA-E-453.
U.S. EPA (Environmental Protection Agency). (1992) Dermal exposure assessment: principles and applications.
U.S. Environmental Protection Agency, Office of Health and Environmental Assessment, Exposure Assessment
Group, Washington, DC; EPA/600/8-91/01 IB. Available online at http://www.epa.gov/NCEA/pdfs/derexp.pdf.
U.S. EPA (Environmental Protection Agency). (1997) Exposure factors handbook. U.S. Environmental Protection
Agency, Office of Health and Environmental Assessment, Exposure Assessment Group, Washington, DC;
EPA/600/P-95/002. Available online at http://nepis.epa.gov/EPA/html/Pubs/pubtitleORD.htm.
U.S. EPA (Environmental Protection Agency). (2000) Exposure and human health reassessment of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Parts I-III. U.S. Environmental Protection Agency,
National Center for Environmental Assessment, Washington, DC; EPA/600/P-00/001. Available online at
http://www.epa.gov/ncea.
U.S. EPA (Environmental Protection Agency). (2003) Exposure and human health reassessment of 2,3,7,8-
tetrachlorodibenzo-p-dioxin (TCDD) and related compounds [NAS review draft]. U.S. Environmental Protection
Agency, National Center for Environmental Assessment, Washington, DC; EPA/600/P-00/001 Cb. Available online
at http://www.epa.gov/ncea/pdfs/dioxin/nas-review/.
U.S. EPA (Environmental Protection Agency). (2004) Risk assessment guidance for Superfund. Vol. I. Human
health evaluation manual (part E, supplemental guidance for dermal risk assessment). Office of Superfund
Remediation and Technology Innovation, Washington, DC; EPA/540/R/99/005. Available online at
http://www.epa.gov/oswer/riskassessment/ragse/index.htm.
U.S. EPA (Environmental Protection Agency). (2007) Pilot survey of levels of poly chlorinated dibenzo-p-dioxins,
polychlorinated dibenzofurans, polychlorinated biphenyls, and mercury in rural soils of the United States. National
Center for Environmental Assessment, Washington, DC; EPA/600/R-05/048F. Available online at
http://www.epa.gov/ncea.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 59 DRAFT—DO NOT CITE OR QUOTE
-------
U.S. FDA (Food and Drag Administration). (1997) Letter from L. Tollefson, Director, Office of Surveillance and
Compliance, Center for Veterinary Medicine to producers and users of clay products in animal feeds. October 7,
1997. Available online at http://www.fda.gov/cvm/Documents/ballclay.pdf.
U.S. FDA (Food and Drag Administration). (2000) Guidance for industry: dioxin in anti-caking agents used in
animal feed and feed ingredients. Guidance #98 [Revised 04/14/2000]. Center for Veterinary Medicine, U.S. Food
and Drag Administration. Available online at http://www.fda.gov/cvm/Guidance/guida98.PDF.
USGS (U.S. Geological Society). (2007) 2005 Minerals Yearbook - Clay and Shale. February 2007. Available
online at http://minerals.usgs.gov/minerals/pubs/commodity/clays/claysmyb05.pdf.
Van den Berg, M; Birnbaum, ML; Bosveld, AT; et al. (1998) Toxic equivalency factors (TEFs) for PCBs, PCDDs,
PCDFs for humans and wildlife. Environ Health Perspect 106(12):775-792.
Van den Berg, M; Birnbaum, LS; Denison, M; et al. (2006) The 2005 World Health Organization re-evaluation of
human and mammalian toxic equivalency factors for dioxins and dioxin-like compounds. Toxicol Sci 93(2):223-
241.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 60 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix A
Subject Questionnaire Results
-------
Table A-l. Subject 1
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
50 hours
40 weeks
1 year
Hand building, sculptural work. Largely
consists of rolling out slabs and assembling
clay parts.
Short sleeve t-shirt and jeans and closed toe
shoes.
Hands and forearms.
Yes. Dryness. No cracking/bleeding. I use
lotion 3-4 times through the day.
Hands:
when rolling slabs - once per hour
when assembling clay - 3 or more times per
hour
Face: 1-2 times per day
Water only.
Yes, Aveeno brand lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-2 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-2. Subject!
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
10-15 hours
15-25 weeks
24 years
Mixing clay and maintenance activities
associated with the OSU Ceramics area.
Long and short sleeves, long pants, work
shoes.
Hands, arms, and face.
Dryness and cracking.
Hands: 2 minutes
Face: 5 hours
Soap and water.
Lotion during winter, but when my hands are
very dry a product called Satin Hands is
used.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-3 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-3. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
25 hours
30 weeks
14 months
Functional - thrown on wheel
Structural - hand built
Jeans with t-shirt and sandals (summer) or
long sleeves and closed toe shoes (winter).
Hands, arms, face, neck, and feet.
Dry cracking skin and cuticles on hands, red
small-bump rash on backs of hands and
inner forearms when using wheel, nasal
congestion.
Arms and hands: 3 to 5 minutes
Feet, face, and neck: 1-10 hours
Water only if returning to work, soap and
water when finished.
Aveda hand creme, Neutrogena Swiss
therapy lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-4 DRAFT—DO NOT CITE OR QUOTE
-------
Table A-4. Subject 4
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
More than 70 hours
50 weeks
2 years
Functional pots, cups, bowls, etc.
Overalls, long/short sleeve shirts and
sneakers.
Face, hands, sometimes arms and legs.
Extremely dry with cracking on fingertips.
Hands: 10 minutes
Face and body: 10-24 hours
Water only if returning to work, soap and
water when finished.
Heavy cream lotion or bag balm at the end
of the day and at intervals throughout the
day.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-5 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-5. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
More than 14 hours
35 weeks
6 years
Hand building objects about 1.5 feet tall.
Short sleeves/pants and shoes.
Hands, lower arms, face.
Yes, dryness, sometimes cracking.
Hands: <5 minutes
Arms: 8 hours
Face: 0.5-8 hours
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-6 DRAFT—DO NOT CITE OR QUOTE
-------
Table A-6. Subject 6
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
30-40 hours
30-40 weeks
25 weeks
Throwing objects using wheel, hand
building, and sculptural work.
Short sleeves, pants, shorts, and flip flops
shoes.
Arms, hands, feet, face.
Yes, dry skin on feet and hands and nails
being unable to grow healthily.
Hands: 30 minutes
Legs, feet, and face: 3-5 hours.
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-7 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-7. Subject?
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
10 hours
40 weeks
4 years
Clay sculpture. Rolling out slabs, pressing
them into molds. Limited work throwing
objects using wheel.
Short sleeves and pants (winter/spring/fall)
and shorts (summer).
Arms, hands, and face.
Dryness and cracking.
Hands: 1-2 minutes
Face and legs 1-2 minutes (powdered clay)
or end of day (wet clay).
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-8 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-8. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
20 hours
52 weeks
6 years
Large clay sculpture. Rolling out slabs, cut
and bend them and then press them together.
Pants or shorts, short sleeves or tank tops,
sneakers or sandals.
Arms, neck, hands, calves, and shins.
Dryness and cracking.
Hands: 5 minutes
Face and legs: 4-24 hours
Soap and water or just water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-9 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix B
Pictures of Artisans Prior to Skin Rinse Procedure
-------
B
D
Figure B-l. Subjects 1-4.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-2 DRAFT—DO NOT CITE OR QUOTE
-------
B
D
Figure B-2. Subjects 5-8.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-3 DRAFT—DO NOT CITE OR QUOTE
-------
Figure B-3. Subject 9.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-4 DRAFT—DO NOT CITE OR QUOTE
-------
Figure B-4. Subject 10.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-5 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix C
Real-time Particle Concentration Data
-------
E
"3>
E
d
o
Q.
10
8
6
2
0
B
Personal Particle Concentration
using the pDR-1000 Particle Counter
0 20 40 60 80 100 120 140 160
Time (min)
Area Particle Concentration
using the CI-500 Particle Counter
0 20 40 60 80 100 120 140 160
Time (min)
Figure C-l. Subject 1.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-2 DRAFT—DO NOT CITE OR QUOTE
-------
10
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
Mixing and bagging
of powdered clay
Subject cleaned off
using compressed air
20 40 60
Time (min)
80
100
B
Area Particle Concentration
using the CI-500 Particle Counter
"c- 1 9
E I ^
"3) ,•/->
8 R
° D
o
"o A
~ 4
a 9
Q. Z
n
A I
i_
Mixing and bagging
of powdered clay
t
K
^ n ft. A .» .
^V uv^^w^
Subject cleaned off
using compressed air
/
0 20 40 60
Time (min)
80
100
Figure C-2. Subject 2.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-3 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
25 50 75
Time (min)
100
125
B
0
Area Particle Concentration
using the CI-500 Particle Counter
25
50
75
100
125
Time (min)
Figure C-3. Subjects.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-4 DRAFT—DO NOT CITE OR QUOTE
-------
SO.
g?0.
80
io.
Jo.
o.
Personal Particle Concentration
using the pDR-1000 Particle Counter
0 20 40 60 80
Time (min)
100 120
B
Area Particle Concentration
using the CI-500 Particle Counter
£0.5
0
0 20 40 60 80 100 120
Time (min)
Figure C-4. Subject 4.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-5 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
20
40
60 80
Time (min)
100 120 140
B
Area Particle Concentration
using the CI-500 Particle Counter
20
40
60 80
Time (min)
100 120 140
Figure C-5. Subjects.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-6 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
30 60 90
Time (min)
120
B
o
Area Particle Concentration
using the CI-500 Particle Counter
20
40
60 80
Time (min)
100 120
Figure C-6. Subject 6.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-7 DRAFT—DO NOT CITE OR QUOTE
-------
"E4
"3)
o
'+*
O -I
Q. I
0
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
10 20
Time (min)
40
B
Area Particle Concentration
using the CI-500 Particle Counter
0 20 40 60 80 100 120
Time (min)
Figure C-7. Subject 7.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-8 DRAFT—DO NOT CITE OR QUOTE
-------
Personal Particle Concentration
using the pDR-1000 Particle Counter
u
„ c
1
1*4
c t
u
= ^
O **>
I2
u ^
t
n 1
U. |
n
ry
^TAjv^AAAl^A^^^/v^^
W I I I I I
0 20 40 60 80 100 120 140
Time (min)
B
14
12
10
8
o
o
-------
Wheel Session 1
120
180 240
Time (min)
300
360
420
B
CO
E
O)
s.
Wheel Session 2
120 180
Time (min)
240
300
Figure C-9. Subject 9.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-10 DRAFT—DO NOT CITE OR QUOTE
-------
Wheel Session 3
4.5
£ 3.5
o
c
2.5
1.5
0.5
o ;=*£•
60 120
180 240 300
Time (min)
360 420 480
Wheel Session 4
60 120 180 240 300 360 420
Figure C-9. Subject 9 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-11 DRAFT—DO NOT CITE OR QUOTE
-------
E
CO
E
0.6
0.5
04
c 0.3
.2
o
8.
0.2
Wheel Session 5
60 120 180 240 300 360 420 480
Time (min)
Figure C-9. Subject 9 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-12 DRAFT—DO NOT CITE OR QUOTE
-------
Sculpture Session 1
30
25
20
o
60 120 180 240
Time (min)
Sculpture Session 2
300
360
420
120
180 240
Time (min)
300
300
420
Figure C-10. Subject 10.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-13 DRAFT—DO NOT CITE OR QUOTE
-------
Sculpture Session 3
Particle cone, (mg m-3)
D
Background Particle Concentration
OA _
ri o _
n 9
n 1 ^ -
01-
n n c
n -
I 1
V M ^ft ^^
T1VV'>Wln--^*-^V(^'^^ V
0 60 120 180 240 300 360 42
Time (min)
Figure C-10. Subject 10 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-14 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix D
Respicon, Cascade Impactor, pDR-1000, and Climet CI-500
Data for Each Individual Subject
-------
Table D-l. Concentration by particle diameter (um) as measured by the
Respicon Air Sampler (mg/m3)a'b
Aerodynamic
Diameter
Subject 1
Subject 2
Subject 3
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Background0
<4
32 jim
0.18
0.31
0.39
0.22
0.10
0.23
0.15
0.31
0.085
Total
0.35
0.47
0.99
0.37
0.13
0.61
0.51
0.64
0.13
aDL (Detection Limit) = 0.015 mg/m3.
b!/2 DL was used in place of the
-------
Table D-3. Particle concentration as measured by the pDR-1000 Air
Sampler (mg/m3)
Subject 1
Subject 2
Subjects
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Mean
0.75
0.57
0.30
0.14
0.049
1.22
0.32
0.34
Maximum
8.42
8.33
0.84
0.81
0.27
7.70
3.51
5.14
Minimum
0.047
0.016
0.093
0.027
0.019
0.078
0.080
0.015
Table D-4. Concentration by particle diameter (um) as measured by the
3\a
Climet CI-500 Air Sampler (mg/m3)
Physical
Diameter
Subject 1
Subject 2
Subject 3
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Background13
0.3-0.5
0.001
0.001
0.002
0.002
0.008
0.011
0.005
0.006
0.009
0.5-1.0
0.005
0.002
0.009
0.003
0.002
0.006
0.010
0.004
0.005
1.0-2.5
0.026
0.016
0.058
0.013
0.003
0.029
0.054
0.021
0.002
2.5-5.0
0.222
0.166
0.411
0.124
0.025
0.260
0.377
0.186
0.010
5.0-10
0.560
0.535
1.214
0.323
0.055
0.679
0.631
0.578
0.010
>10.0
1.499
1.747
3.756
0.964
0.167
1.746
0.817
1.878
0.019
Total
2.313
2.467
5.450
1.429
0.260
2.731
1.895
2.672
0.055
aConcentration calculations assume particle density of 2.6 g/cm3.
bBased on measurements taken late at night when no students were present in building.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 D-3 DRAFT—DO NOT CITE OR QUOTE
-------
Table D-5. Average concentrations by particle diameter ranges (um)
measured by the Cascade Impactor Air Sampler (mg/m3)a'b
Aerodynamic
Diameter
Subject 9
Session 1
Subject 9
Session 2
Subject 9
Session 3
Subject 9
Session 4
Subject 9
Session 5
Subject 10
Session lc
Subject 10
Session 2C
Subject 10
Session 3
Background11
0.5-2
0.004
32
0.024
0.024
0.044
0.053
0.081
0.198
0.092
0.079
0.005
Total
0.049
0.046
0.102
0.073
0.152
0.480
0.237
0.241
0.023
aDL (Detection Limit) = 0.0025 mg/m3.
b!/2 DL was used in place of the
-------
Table D-6. Concentration by particle diameter ranges (urn) measured by the
Climet CI-500 Air Sampler (mg/m3)a
Physical
Diameter
Subject 9
Session 1
Subject 9
Session 2
Subject 9
Session 3
Subject 9
Session 4
Subject 9
Session 5
Subject 10
Session lb
Subject 10
Session 2b
Subject 10
Session 3
Background0
0.3-0.5
0.008
0.010
0.006
0.012
0.011
0.018
0.003
0.006
0.012
0.5-1.0
0.003
0.005
0.004
0.007
0.008
0.015
0.005
0.008
0.009
1.0-2.5
0.005
0.003
0.005
0.011
0.004
0.067
0.031
0.039
0.003
2.5-5.0
0.026
0.014
0.026
0.055
0.018
0.353
0.172
0.181
0.011
5.0-10
0.042
0.027
0.054
0.113
0.026
0.746
0.367
0.341
0.012
>10.0
0.070
0.058
0.124
0.240
0.048
1.430
0.700
0.656
0.016
Total
0.155
0.117
0.220
0.439
0.115
2.629
1.278
1.231
0.064
""Concentration calculations assume particle density of 2.6 g/cm3.
bConcentration not adjusted for presence of dog.
°Based on measurements taken late at night when no students were present in building.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 D-5 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix E
SEM and EDS Data by Subject
-------
&5
rs
o-
^
a
rs
I
1 2
Fu\\ Scale 1314 els
1 2
Full Scale 3934 els Cursor: 1.622 keV (125 cts)
"1
--,-
'TS
I
o
W \
Figure E-la. Sample of clay used by Subject 1.
to
i
a,
gl
Hl
O 3
H
Figure E-lb. Clay particles on Subject 1's Respicon
Filter.
1 2
-ull Scale 8SSO cts Cursor: O.OOO keV
1 2
rLill Scale -1 3229 cts Cursor: 9.533 keV (1 7 ctsl
o
d
o
H
W
Figure E-2a. Sample of clay used by Subject 2.
Figure E-2b. Clay particles on Subject 2's Respicon
Filter.
-------
1 2
Full Scale 251 23 cts
•• i •'
8
keV
123
Full Scale 31 955 cts
• • • i •
6
keV
Figure E-3a. Sample of clay used by Subject 3.
Figure E-3b. Clay particles on Subject 3's Respicon
Filter.
• i • • •
2
Full Scale 25579 cts
1 • I •
•10
keV
1 2
Full Scale 14147 cts
keV
Figure E-4a. Sample of clay used by Subject 4.
Figure E-4b. Clay particles on Subject 4's Respicon
Filter.
-------
o
i
^s
^s
o
w !
-U °
•I"
I
of
>i
O i
o I
^ ^
§1
H
H
W
O
5^
O
d
o
H
W
123
Full Scale 14395 cts
Figure E-5a. Sample of clay used by Subject 5.
1 2
Full Scale 21 36 cts
Figure E-6a. Sample of clay used by Subject 6.
8
keV
keV
1 2
Full Scale 930 cts
keV
Figure E-5b. Clay particles on Subject 5's Respicon
Filter.
123
Full Scale 795 cts
10
keV
Figure E-6b. Clay particles on Subject 6's Respicon
Filter.
-------
VO
o
^1
§•
o
e
5
?
I"
'TS
J
o
a
1
H
H S.
W o'
o
a
o
H
W
123
Full Scale 5201 cts
10
keV
Figure ~E-7a. Sample of clay used by Subject 7.
10
Full Scale 751 6 cts
keV
Figure E-8a. Sample of clay used by Subject 8.
Full Scale "1344 cts
keV
Figure E-7b. Clay particles on Subject 7's Respicon
Filter.
2 4
Full Scale 9593 ds
10
keV
Figure E-8b. Clay particles on Subject 8's Respicon
Filter.
-------
Appendix F
Monte Carlo Simulation Result Graphics
-------
Append ixF
Do Not Quote or Cite
Crystal Ball Report - Full
Simulation started on 3/31/2006 at 7:15:34
Simulation stopped on 3/31/2006 at 7:23:41
Run preferences:
Number of trials run 1,000
Monte Carlo
Random seed
Precision control on
Confidence level 95.00%
Run statistics:
Total running time (sec) 487.37
Trials/second (average) 2
Random numbers per sec 35
Crystal Ball data:
Assumptions 17
Correlations 0
Correlated groups 0
Decision variables 0
Forecasts 4
F-1
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecasts
Worksheet: [VarDp-Dep monte5.xls]Monte
Forecast: Ingestion Dose
Summary:
Entire range is from 0.003 to 0.730
Base case is 0.058
After 1,000 trials, the std. error of the mean is 0.003
Cell: CSS
Ingestion Dose
0.06
0420
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
0.141
0.115
0.104
0.011
1.56
6.04
0.74
0.003
0.730
0.727
0.003
F-2
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Ingestion Dose (cont'd) Cell: C53
Percentiles: Forecast values
0% 0.003
10% 0.039
20% 0.059
30% 0.077
40% 0.097
50% 0.115
60% 0.135
70% 0.161
80% 0.207
90% 0.284
100% 0.730
F-3
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Inhalation Dose
Summary:
Entire range is from 0.00 to 1.05
Base case is 0.04
After 1,000 trials, the std. error of the mean is 0.00
Cell: C83
Inhalation Dose
000 005 0.10 0.15
0,20 025
pg/g
030 035 Q.«0 045
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
0.12
0.08
0.13
0.02
2.51
11.75
1.07
0.00
1.05
1.05
0.00
F-4
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Inhalation Dose (cont'd) Cell: C83
Percentiles: Forecast values
0% 0.00
10% 0.02
20% 0.03
30% 0.04
40% 0.06
50% 0.08
60% 0.10
70% 0.14
80% 0.18
90% 0.27
100% 1.05
F-5
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dermal Dose
Cell: C45
Summary:
Entire range is from 0.27 to 217.51
Base case is 10.91
After 1,000 trials, the std. error of the mean is 0.72
Total Dermal Dose
0.14
140
a.00
1000 2000 30,00 40.00 5000
6000 7000
v< on
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
15.50
7.92
22.91
524.87
3.67
20.69
1.48
0.27
217.51
217.24
0.72
F-6
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dermal Dose (cont'd) Cell: C45
Percentiles: Forecast values
0% 0.27
10% 2.02
20% 3.16
30% 4.29
40% 5.90
50% 7.92
60% 10.08
70% 14.09
80% 20.03
90% 36.15
100% 217.51
F-7
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dose
Cell: C86
Summary:
Entire range is from 0.28 to 219.14
Base case is 11.01
After 1,000 trials, the std. error of the mean is 0.73
Total Dose
0.00
10 00 20 DO
40.00
pg/d
50.00 6000 70.00
50 on
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
15.76
8.12
23.01
529.38
3.66
20.67
1.46
0.28
219.14
218.86
0.73
F-8
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dose (cont'd) Cell: C86
Percentiles: Forecast values
0% 0.28
10% 2.15
20% 3.32
30% 4.51
40% 6.15
50% 8.12
60% 10.39
70% 14.44
80% 20.58
90% 36.63
100% 219.14
End of Forecasts
F-9
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumptions
Worksheet: [VarDp-Dep monte5.xls]Monte
Assumption: Activity selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C56
UTO BID 0-» 0X1 040 0» OM Ore OtO OBO 100
Assumption: Breath Selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C61
Dieafh Selector
Assumption: Clay load on arm (mg/cm2)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.04
0.35
3.00
Cell: C22
UWJ OJO OKI 0*0 I JO ' W l«0 2 <0 1*0 ."U
Assumption: Clay load on beverage (mg)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
Cell: C51
0.30
0.50
0.72
F-10
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Clay load on face (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
0.030
0.040
Cell: C40
Assumption: Clay load on feet (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
0.03
0.30
Cell: C34
Assumption: Clay load on food (mg)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.30
0.70
1.66
Cell: C49
d«y load on food (mq)
OM CM WO 100 1J> t«0 1*0
Assumption: Clay load on hand (mg/cm2)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.10
3.00
10.00
Cell:C17
Cta, lo.d „, h.nd
F-11
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Clay load on leg (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
Cell: C28
Assumption: clothing selector
Uniform distribution with parameters:
Minimum
Maximum
0.10
0.70
0.00
1.00
Assumption: Dioxin cone in ball clay (pg/g)
Lognormal distribution with parameters:
Mean 808.00
Std. Dev. 318.00
Cell: C5
Assumption: Exposure Duration (hr/d)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
1.00
4.00
10.00
Cell: C7
Fxpcilif e Duration (fir/if)
F-12
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Fraction of ball clay in blend
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
Cell: C6
0.00
0.20
1.00
Assumption: M/F Selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C62
Assumption: MMAD (um)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
13.00
25.00
45.00
Cell: C60
xoo 1*00 xoo 1300 aoo *ooo
Assumption: Particle Concentration in Air(mg/m3)
Triangular distribution with parameters:
Minimum 0.08
Likeliest 0.44
Maximum 0.99
Cell: C59
010 930 030 9*9 9.9) Of) 0,70 090 OUO t TO
F-13
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Total Body Surface Area (cm2)
Lognormal distribution with parameters:
Mean 18,000.00
Std. Dev. 37.40
Cell: C8
End of Assumptions
F-14
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Appendix G
Evaluation of Clay Dust Inhalation
-------
1 APPENDIX G. EVALUATION OF CLAY DUST INHALATION
2
3 The methodology used to evaluate the dose of clay dust and associated dioxin received
4 via inhalation is discussed in this appendix. The appendix is divided into four sections: clay dust
5 size distribution, particle inhalability, respiratory deposition of clay dust, and delivered dose
6 estimates.
7
8 CLAY DUST SIZE DISTRIBUTION
9 As discussed in the main body of this report, the size distribution of clay dust was
10 measured using a Delron cascade impactor and a Climet during regular daily activities in the art
11 studio. The Climet optically determines particle concentration for six size bins with the
12 associated physical particle diameter (dp) of 0.3-0.5, 0.5-1, 1-2.5, 2.5-5, 5-10, and >10 |im.
13 Aerodynamic particle diameter (dae) can be estimated for the Climet's size bins by assuming that
14 the airborne clay dust has a density of 2.6 g/cm3, similar to that of bulk clay.1 Using this
15 approach, a clay particle with a dp of 10 jim has a dae of 16 jim. The Delron cascade impactor
16 fractionates particles directly, based on their dae, into the seven ranges of <0.5, 0.5-2, 2-4, 4-8,
17 8-16, 16-32, and >32|im.
18 During normal artisan activities (Subjects 1-8), 64 ± 9% (mean ± SD) of the aerosol is
19 associated with particles having a dae > 16 jam based on average Climet data. Based on average
20 impactor data, 63 ± 13% of the aerosol is associated with a dae > 16 jim (Subjects 1-8). The
21 particle size distributions to which the artisans were exposed was assumed to be log-normally
22 distributed.2 The cascade impactor data were selected for estimating particle size distributions
23 for the following reasons: (1) the impactor measures particle size based on the aerodynamic
24 behavior of particles, whereas the Climet uses light scattering to estimate a physical particle size;
25 (2) the impactor affords a better characterization of the large particles than does the Climet
26 because it contains an additional size bin of 16-32 jim; and (3) particle deposition in the
27 respiratory tract is a function of dae. Thus, uncertainty in estimates of respiratory deposition is
28 reduced by the direct measurement of dae by the impactor. The clay dust size distribution was
29 not estimated for runs where two or more of the impactor stages were below the nondetect level.
30 When engaged in normal artisan activities, the mass median aerodynamic diameter
31 (MMAD) of clay dust to which artisans were exposed ranged from 13 to 45 |im. Table G-l
Mae = dp {(clay density * Cc(dp) )/(H2O density * Cc(dae) )}05, where: Cc(dp) and Cc(dae) are the Cunningham slip
correction factor for the physical and aerodynamic particle size, respectively. For more information, the reader is
referred to ICRP (1994), page 239.
2For more information about particle sizing and the log-normal distribution, the reader is referred to Hinds (1999).
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-2 DRAFT-DO NOT CITE OR QUOTE
-------
1 provides a characterization of clay dust exposures for each subject. Figure G-l illustrates a log-
2 probability plot of a typical (i.e., near the average MMAD) clay dust particle size distribution
3 and a background sample from the studio. The prevalence of fewer large particles in the
4 background aerosol can be explained easily, based on particle-settling velocities. The settling
5 velocities for the dae of 1-, 10-, and 20-|im particles are 3.5 x 10"3, 0.3, and 1.2 cm/s,
6 respectively. Due to their rapidly settling velocities, large particles (dae > 10 jim) are maintained
7 in the air only by active generation or resuspension from surfaces. The substantive presence of
8 large particles (52% of mass associated with a dae > 10 jim) in the background sample is
9 suggestive of particle resuspension due to movement (e.g., walking and setting up sampling
10 equipment in the studio).
Table G-l. Clay dust size distribution and concentration during normal
activities
Subject
1
2
3
4
5
6
7
8
Mean ± SD
Size distribution3
MMAD (urn) og
26.9
44.6
18.5
n.a.
n.a.
20.2
13.0
26.7
25.0±11 3.*
3.9
4.8
4.3
n.a.
n.a.
3.0
3.6
3.3
I ±0.7
Total concentration
(mg/m3)
0.35
0.47
0.99
0.37
0.13
0.61
0.51
0.64
0.51 ±0.25
aThe aerosol size distribution is described in terms of the mass median aerodynamic diameter (MMAD) and
geometric standard deviation (og).
n.a. = not available
11 Data were also available for two subjects during specific activities (i.e., when sculpting
12 and using a pottery wheel) (see Table G-2). During pottery wheel operations, an average
13 MMAD of 33 jim with a geometric standard deviation (og) of 5.4 was observed. A dog was
14 present during two of the sculpting runs. The MMAD with the dog present was 21 |im versus
15 only 16 |im without the dog. The shift toward larger particles when the dog was present appears
16 to be consistent with particle resuspension due to the dog's movement around the studio.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-3 DRAFT-DO NOT CITE OR QUOTE
-------
i uu —
-
E
ZL
kT
1
Q
J>
o -in
•— 1 U
Is -
Q.
O
(0
c
T3
O
L.
n>
1
0
/
f
f
1
,
y
/
/
2
£
/
a'
/
/
-*1
jfl
/'/
/
/
/°
10
i 2
/
f
¥
/
P
O
/
/
30
0 4
/
f
X'
<
50
0 6
ji
/
f
/
i
f
/
/
70
0 8
/
' /
/
/
/
0 9°9
98
5 9
99.9
9
Cumulative Percent Less Than Indicated Size
Figure G-l. Clay dust particle size distribution during normal artisan activities
from analysis of cascade impactor data. Illustrated are the data for Subject 8 ( ) and
a background sample when work was not being done in the studio (o). The dashed and
solid lines illustrate the log-normal distribution for these respective data. The mass
median aerodynamic diameter (MMAD) of clay dust was 27 jim (og= 3.3) for Subject
8, whereas the background sample had an MMAD of 11 jim (og = 4.6).
1 PARTICLE INHALABILITY
2 For a given particle size, inhalability is the ratio of the particle concentration that enters
3 the respiratory tract through the nose or mouth to the concentration of these particles in the
4 ambient air. Inhalability depends mainly on particle size (i.e., dae), route of breathing, wind
5 speed, and a person's orientation with respect to wind direction. Wind speeds in the art studio
6 were assumed to be 0.3 m/s or less (Baldwin and Maynard, 1998). The artisans were presumed
7 to move about the studio such that their orientation was random with respect to wind direction.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-4 DRAFT-DO NOT CITE OR QUOTE
-------
Table G-2. Clay dust size distribution and concentration during specific activities
Subject
Subject 9
(Pottery wheel)
Run 1
Run 2
Run 3
Run 4
Run5
Mean ± SD
Subject 10b
(Sculpting work)
Run 1
Run 2
Run 3
Size
distribution3
MMAD urn og
33.7
n.a.
24.8
n.a.
39.3
32.6 ±7.3
21.2
20.4
16.0
6.2
n.a.
4.3
n.a.
5.6
5.4 ±0.9
3.9
3.2
3.5
Total
concentration
(mg/m3)
0.049
0.046
0.102
0.073
0.152
0.085 ± 0.044
0.48
0.24
0.24
aThe aerosol size distribution is described in terms of the mass median aerodynamic diameter (MMAD) and
geometric standard deviation (og).
bA dog was present during Runs 1 and 2 but not during Run 3. Therefore, these three runs were not averaged as
was done in the case of the pottery wheel work.
n.a. = not available
1 The clay dust aerosol present under normal activities in the art studio was observed to
2 have an average MMAD of 25 jim and og of 3.8. Hence, 50% (on average, by mass) of the
3 airborne clay dust is composed of particles having a dae of >25 jim, a size that is generally
4 considered to be unable to penetrate the thorax (ACGIH, 2004). These large particles
5 (dae >25 |im), if inhaled, will deposit almost completely and exclusively in the extrathoracic (ET)
6 airways. Thus, determining inhalability is key to estimating the delivered dose of these large
7 particles. For smaller particles, inhalability still describes the fraction of airborne particles that
8 may enter the respiratory tract and thereby the availability of these particles for deposition in the
9 lung.
10 Only limited data are available on the inhalability of particles from calm air (wind speeds
11 of 0.3 m/s and less). Inhalability from calm air depends on the route of breathing. Logistic
12 functions describing particle inhalability during nasal [P(!N)] and oral [P(Io)] breathing are given
13 by Menache et al. (1995) and Brown (2005):
14
1
exp(10.32-3.1141n(dae))
(G-l)
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-5 DRAFT-DO NOT CITE OR QUOTE
-------
1 44
P(I0) = : (G-2)
OJ l + 0.44exp(0.0195dae)
1
2 Note that these equations depend only on aerodynamic particle diameter, dae. Given by Eq G-l,
3 P(IN) begins a rapid decline from 0.95 at dae = 11 |im, to 0.5 at dae = 27.5 |im, and 0.1 at
4 dae = 56 |im. Equation G-2 predicts a slow decline in P(Io) from 0.95 at dae = 8 |im, to 0.5 at
5 dae = 74 jam, and 0.1 at dae =175 jim.
6 Figure G-2 illustrates particle inhalability predicted by Eqs G-l and G-2 (shown by solid
7 lines) along with relevant experimental data. Based on high wind speeds (1-8 m/s), the
8 American Conference of Governmental Industrial Hygienists (ACGIH) inhalability criterion is
9 also illustrated (shown by dashed lines) for comparative purposes. Equation G-l for P(!N)
10 describes the experimental nasal inhalability data well with an r2 of 0.86 (model sum of squares
11 divided by the total corrected sum of squares). A negative r2 is obtained for the fit of the
12 ACGIH (2004) criterion to these data.3 Equation G-2 describes the experimental oral
13 inhalability data with an r2 of 0.69, whereas the ACGIH criterion fit with an r2 of 0.32.
14
15 RESPIRATORY DEPOSITION OF CLAY DUST
16 Inhaled particles may be either exhaled or deposited in the ET, tracheobronchial (TB), or
17 pulmonary (PU) airways. The deposition of particles in the respiratory tract depends primarily
18 on inhaled particle size (i.e., dae), route of breathing (through the nose or mouth), tidal volume
19 (Vx), and breathing frequency (/"). Reference respiratory values for males and females were
20 adopted from the International Commission on Radiological Protection (ICRP, 1994). In
21 addition to breathing patterns (Table G-3) necessary for deposition calculations, males and
22 females were assumed to have a functional residual capacity of 3,300 mL and 2,680 mL,
23 respectively. The majority (70%) of the subjects were female; only Subjects 1, 2, and 5 were
24 male.
25 Particle deposition in the respiratory tract was predicted using the publicly available
26 Multiple Path Particle Dosimetry (MPPD) model.4 The MPPD model was developed by the
27 CUT
3An r2 is calculated as the model sum of squares (MSS) divided by the total corrected sum of squares (TSS). The
MSS equals the TSS minus the residual sum of squares (RSS). In typical linear regressions, when a model is fitted
to a data set, the resulting r2 must be non-negative because the least square fitting procedure assures RSS < TSS.
When r2 is computed on excluded data, i.e., data not used to fit the model, the RSS can exceed the TSS. In this
case, r2 (which is not the square of r) can be negative, indicating that the mean of the data is a better predictor than
the model.
4 The MPPD program is available on request from the CUT Centers for Health Research ().
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-6 DRAFT-DO NOT CITE OR QUOTE
-------
10
dae (Mm)
100
Figure G-2. Particle inhalability from calm air for nasal [P(!N)] and oral [P(Io)]
breathing as a function of aerodynamic particle diameter (dae) Left panel [
Equation G-l, • Breysse and Swift (1990), + Hinds et al. (1998), o Hsu and Swift
(1999), - - - ACGIH (2004)]. Right panel [ Equation G-2, o Aitken et al. (1999), •
Kennedy and Hinds (2002), - - - ACGIH (2004)].
Table G-3. Breathing patterns used in particle deposition calculations"
Activity
Sitting
Light exercise
VT (mL)
/(min1)
VT (mL)
/(min1)
Males
750
12
1,250
20
Females
464
14
992
21
1
2
3
4
5
Source: ICRP (1994), Table 8.
Centers for Health Research (CUT), United States, in collaboration with the National Institute of
Public Health and the Environment (RIVM), the Netherlands, and the Ministry of Housing,
Spatial Planning and the Environment, the Netherlands. The MPPD model may be used to
predict the deposition in the human respiratory tract for particles between 0.01 and 20 jim in
diameter. In the lung, the model considers deposition by the mechanisms of impaction,
sedimentation, and diffusion. Additional model details are available elsewhere (DeWinter-
Sorkina and Cassee, 2002). For the size of the clay dust, only impaction and sedimentation are
of concern.
Using the MPPD model, deposition was predicted for the ET, TB, and PU regions of the
respiratory tract. Particle deposition was estimated individually for oral and nasal breathing.
During oral breathing, deposition in the TB airways did not always reach zero by a dae of 20 jim
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-7 DRAFT-DO NOT CITE OR QUOTE
-------
1 (the upper limit for the MPPD model). For dae > 20 |im, deposition in the TB airways was
2 estimated by a best fit polynomial (3rd or 4th degree) determined using CurveExpert 1.3 (112B
3 Crossgate St., Starkville, MS 39759). This polynomial function was fitted to TB deposition
4 fractions for dae from 10 to 20 jim. The predicted ET deposition during oral breathing for a dae
5 >20 |im was taken as one minus the TB deposition fraction for oral breathing. For nasal
6 breathing, these additional steps were unnecessary because TB deposition was well under 1% at
7 a dae of 20 |im.
8 External to the MPPD model, all of the predicted deposition fractions were corrected for
9 particle inhalability using Eqs G-l and G-2. The current version of MPPD model offers an
10 inhalability correction for nasal breathing only. For a given dae, an inhalability corrected
11 deposition fraction is the product of the uncorrected deposition fraction and the predicted
12 inhalability for that dae. Unless otherwise specified, all mention of particle deposition fractions
13 in the main body of this report and subsequently in this appendix refer explicitly to inhalability
14 corrected deposition fractions.
15 The deposition fraction (DFr) of an aerosol in a region of the respiratory tract is the
16 integral of the deposition fractions across all particle sizes in the aerosol:
17
00
DFr (MMAD, ag) = J DFr (d, )p(d1 )5d, (G-3)
o
18
19 where:
20 DFr(di) = the deposition fraction in region, r, of particles having an aerodynamic
21 diameter of d;
22 p(di) = the mass fraction associated with the interval 5d;
23
24 The total deposition fraction for the respiratory tract is the sum of DFr for the ET, TB, and
25 PU regions. Equation G-3 can be approximated by summing the particle deposition fractions at
26 known intervals or percentiles of the particle size distribution. Here, the interval of 1% was used
27 and the approximation is:
28
i 0.99
DFr(MMAD,ag)* —^DF^) (G-4)
P=0.oi
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-8 DRAFT-DO NOT CITE OR QUOTE
-------
1 where:
2 DFr(di) = the deposition fraction in region, r, of particles having an aerodynamic diameter
3 d; (the particle size associated with a given percentile, P, of the size
4 distribution).
5
6 For a log-normal distribution, d; is given by:
7
(G-5)
8 where:
9 z(P) = the normal standard deviate for a given probability
10
1 1 Table G-4 provides the predicted regional deposition fractions for the clay dust in the
12 respiratory tract of each subject for oral and nasal breathing at two activity levels. These
13 deposition fraction estimates were based on each subject's measured aerosol exposure size
14 distribution (see Tables G-l and G-2). Subjects 4 and 5 lacked aerosol size distribution data and
15 were assumed exposed to an aerosol with an MMAD of 25 jim and og of 3.8, this being the
16 average for artisans during normal activities (see Table G-l). The deposition fraction estimates
17 for Subject 10 were based on Run 3, when the dog was not present in the studio.
18
19 DELIVERED DOSE ESTIMATES
20 The rate of particle deposition in a region of the respiratory tract may be expressed as:
21
Dr(t) = C(t) (t)VT(t)DFr(t) (G-6)
22
23 where:
24 br = the rate of deposition per unit time in region r
25 C = the exposure concentration
26 /= breathing frequency
27 VT = tidal volume
28 DFr = the deposition fraction in region r
29
30 Note that all of the variables in Eq G-6 may vary with time. The dose to a respiratory region is
3 1 determined by integrating Eq G-6 over the exposure duration.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-9 DRAFT-DO NOT CITE OR QUOTE
-------
1
2
Table G-4. Regional deposition fractions (corrected for inhalability) for clay
dust in the respiratory tract
Subject
1
2
3
4
5
6
7
8
9
10
Sitting
Nasal breathing
ET TB PU
0.441 0.015 0.022
0.336 0.011 0.016
0.472 0.028 0.033
0.447 0.021 0.022
0.458 0.016 0.023
0.526 0.023 0.022
0.549 0.035 0.041
0.451 0.018 0.017
0.368 0.020 0.023
0.533 0.030 0.033
Oral breathing
ET TB PU
0.473 0.082 0.058
0.412 0.059 0.042
0.431 0.104 0.067
0.471 0.091 0.050
0.479 0.086 0.061
0.521 0.108 0.053
0.432 0.128 0.085
0.507 0.087 0.041
0.396 0.077 0.047
0.462 0.118 0.072
Light exercise
Nasal breathing
ET TB PU
0.473 0.006 0.011
0.360 0.004 0.008
0.531 0.010 0.020
0.487 0.007 0.013
0.492 0.006 0.011
0.566 0.007 0.012
0.622 0.013 0.025
0.483 0.005 0.010
0.410 0.007 0.014
0.593 0.010 0.020
Oral breathing
ET TB PU
0.516 0.060 0.052
0.442 0.044 0.037
0.486 0.074 0.075
0.521 0.064 0.056
0.523 0.063 0.054
0.581 0.075 0.059
0.498 0.090 0.095
0.557 0.061 0.046
0.437 0.054 0.053
0.525 0.083 0.081
ET = extrathoracic; PU = pulmonary; TB = tracheobronchial
4
5
6
By assuming that aerosol characteristics and an individual's activity levels are fairly
constant over discrete periods of time, the dose to a respiratory region may be approximated by:
Dr =0.06X(VT/)J(CT)J[FmDFm^r = FNDFN>r]j
(G-7)
7 where:
8
9
10
11
12
13
14
15
16
Dr = the dose (jig) to region r of the respiratory tract
VT and/= tidal volume (mL) and breathing frequency (min"1) for a specified activity j
C and T = exposure concentration (mg/m3) and duration (hr) during activity j
Fm and FN = the fraction of a breath entering the respiratory tract through the mouth and
nose, respectively, during activity j
DFm>r and DFN>r = the deposition fraction for oral and nasal breathing, respectively, in
region r of the respiratory tract while performing activity]
Constant 0.06 = a unit conversion parameter
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-10 DRAFT-DO NOT CITE OR QUOTE
-------
1 As expressed, an "activity" in Eq G-7 could be associated with changes in exposure
2 concentration, the particle size distribution, and/or an individual's exertion level. For simplicity,
3 only two exertion levels (sitting and light exercise) and a single particle size distribution (see
4 Tables G-l and G-2) were considered for each subject.
5 The fraction of flow through the mouth (Fm in Eq G-7) increases with activity level and
6 varies between individuals. For the two activity levels considered here, most people (87%) will
7 breathe through their nose (Niinimaa et al., 1981). Hence, for these people, Fm= 0 and FN= 1 in
8 Eq G-7. However, 13% of people will be oronasal breathers even at rest, i.e., they will breathe
9 simultaneously through the nose and mouth (Niinimaa et al., 1981). This latter group is
10 commonly referred to in the literature as "mouth breathers" (e.g., ICRP, 1994). Derived from
11 Niinimaa et al. (1981), the fraction of air respired through the mouth (Fm) is well described by a
12 modified exponential function in the form of:
13
Fm=aexp
_J_
vVey
(G-8)
14 where:
15 Ve = minute ventilation
16 a = 0.748 and y=-7.09 (r2=0.997) in mouth breathers for 1QV£80 L/min and
17 35.3Ve80 L/min, a = 0.744, and y=-18.3 (r2=0.998) in normal augmenters
18
19 For Ve <35.3 L/min, normal augmenters breathe entirely through the nose, i.e., Fm = 0. FN is one
20 minus Fm regardless of the activity.
21 Table G-5 gives the estimated clay dust doses to regions of the respiratory tract for each
22 subject during nasal and oronasal breathing. Estimates are for a 4-hour exposure assuming that
23 the exposed individual spent 50% of his or her time sitting and 50% engaged in light exercise.
24 For oronasal breathing in Table G-5, there is a small positive bias in ET doses and a
25 corresponding negative bias in TB doses calculated by Eq G-7. In other words, this method of
26 calculating ET and TB doses shifts the pattern of deposition toward the head relative to the real-
27 life pattern of deposition. This shift occurs due to deposition being calculated at a higher airflow
28 rate through the nose and mouth than actually occurs during oronasal breathing. The deposition
29 calculations presumed that all inhaled airflow was through the nose or mouth. In reality, inhaled
30 air is partitioned between the nose and the mouth, and the actual flows (for sitting and light
31 exercise) are roughly half of that used in the deposition calculations. For breathing by a single
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-ll DRAFT-DO NOT CITE OR QUOTE
-------
1 route (nasal or oral), changing activity from sitting to light exercise approximately triples flow
2 rates but only slightly increases ET deposition and modestly decreases TB deposition (see Table
3 G-4). The effect of using Eq G-7 for calculating doses during oronasal breathing should
4 similarly affect the pattern of deposition. Ultimately, particles deposited in the ET and TB
5 regions will typically be cleared to the throat and swallowed within 24 to 48 hours
6 postdeposition (ICRP, 1994). Hence, the exact site of deposition (i.e., ET versus TB) is of little
7 significance because both regions effectively contribute to ingested doses.
8 Table G-6 provides estimates of the dioxin absorption in each subject for nasal and
9 oronasal breathing. Particles deposited in the ET and TB regions clear rapidly (within 1-2 days)
10 to the throat and are swallowed. The absorption of dioxin from particles deposited within the ET
11 and TB regions was treated as if the particles had been ingested. Dose estimates for oronasal
12 breathing are slightly more conservative from a safety or risk perspective than presuming nasal
13 breathing. However, nasal breathing may be considered as representative of the majority of the
14 population (87%). Oronasal breathing is thought to represent 13% of healthy individuals
15 (Niinimaa et al., 1981). In contrast to healthy subjects, Chadha et al. (1987) found that the
16 majority (11 of 12) of patients with asthma or allergic rhinitis breathe oronasally even at rest.
17 On average across all the subjects, dioxin doses are about 1.2 times greater for oronasal than for
18 nasal breathing.
19
Table G-5. Regional doses (jig) of clay dust in the respiratory tract3
Subject
1
2
3
4
5
6
7
8
9
10
Mean
SD
Nasal breathing
ET
664
678
1,677
580
256
1,114
1,011
997
110
455
754
460
TB
12
11
47
13
4.6
22
30
18
2.9
12
17
13
PU
20
19
75
19
7.7
29
49
24
4.5
18
27
21
Oronasal breathing
ET
693
757
1,612
598
264
1,126
917
1,067
114
431
758
445
TB
53
52
143
45
21
85
90
72
8.8
39
61
39
PU
48
47
154
41
19
70
100
57
9.2
39
58
42
a Doses calculated by Eq G-7 as described in the text.
ET = extrathoracic; PU = pulmonary; TB = tracheobronchial
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-12 DRAFT-DO NOT CITE OR QUOTE
-------
Table G-6. Estimates of dioxin absorption" (pg TEQ)
Subject
1
2
3
4
5
6
7
8
9
10
Mean
SD
ET and TB
0.033
0.034
0.084
0.029
0.013
0.055
0.051
0.049
0.005
0.023
0.038
0.023
Nasal breathing
b puc
0.003
0.003
0.010
0.002
0.001
0.004
0.006
0.003
0.001
0.002
0.004
0.003
Total
0.035
0.036
0.094
0.031
0.014
0.059
0.057
0.052
0.006
0.025
0.041
0.026
Oronasal breathing
ET and TBb PUC
0.036
0.039
0.085
0.031
0.014
0.059
0.049
0.055
0.006
0.023
0.040
0.023
0.006
0.006
0.020
0.005
0.002
0.009
0.013
0.007
0.001
0.005
0.007
0.006
Total
0.043
0.045
0.105
0.037
0.016
0.068
0.062
0.063
0.007
0.028
0.047
0.029
a Dioxin concentration was assumed to be 162 pg toxic equivalent (TEQ) per gram clay.
b Absorption fraction of 0.3 assumed, extrathoracic (ET) and tracheobronchial (TB) rapidly clear into the
gastrointestinal tract.
0 Absorption fraction of 0.8 assumed, due to slow clearance from pulmonary (PU) region.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-13 DRAFT-DO NOT CITE OR QUOTE
-------
1 REFERENCES
2
3 ACGIH (American Conference of Governmental Industrial Hygienists). (2004) TLVs and BEIs: based on the
4 documentation of the threshold limit values for chemical substances and physical agents and biological exposure
5 indices. Cincinnati, OH: ACGIH Worldwide.
6
7 Aitken, RJ; Baldwin, PEJ; Beaumont, GC; et al. (1999) Aerosol inhalability in low air movement environments. J
8 Aerosol Sci 30:613-626.
9
10 Baldwin, PEJ; Maynard, AD. (1998) A survey of wind speeds in indoor workplaces. AnnOccup Hyg 42:303-313.
11
12 Breysse, PN; Swift, DL. (1990) Inhalability of large particles into the human nasal passage: in vivo studies in still
13 air. Aerosol Sci Technol 13:459-464.
14
15 Brown, JS. (2005) Particle inhalability at low wind speeds. Inhal Toxicol 17:831-837.
16
17 Chadha, TS; Birch, S; Sacker, MA. (1987) Oronasal distribution of ventilation during exercise in normal subjects
18 and patients with asthma and rhinitis. Chest 92(6): 1037-1041.
19
20 De Winter-Sorkina, R; Cassee, FR. (2002) From concentration to dose: factors influencing airborne paniculate
21 matter deposition in humans and rats. Bilthoven, The Netherlands: National Institute of Public Health and the
22 Environment (RIVM); report no. 650010031/2002. Available online at
23 http://www.rivm.nl/bibliotheek/rapporten/650010031 .html.
24
25 Hinds, WC. (1999) Aerosol technology: properties, behavior, and measurement of airborne particles (2nd ed.). New
26 York, NY: Wiley-Interscience.
27
28 Hsu, DJ; Swift, DL. (1999) The measurement of human inhalability of ultralarge aerosols in calm air using
29 manikins. J Aerosol Sci 30:1331-1343.
30
31 ICRP (International Commission on Radiological Protection). (1994) Human respiratory tract model for
32 radiological protection: a report of a task group of the International Commission on Radiological Protection.
33 Oxford, United Kingdom: Elsevier Science Ltd. ICRP publication 66; Annals of the ICRP. Vol. 24, pp. 1-482.
34
3 5 Kennedy, NJ; Hinds, WC. (2002) Inhalability of large solid particles. J Aerosol Sci 33:237-255.
36
37 Menache, MG; Miller, FJ; Raabe, OG. (1995) Particle inhalability curves for humans and small laboratory animals.
38 AnnOccup Hyg 39:317-328.
39
40 Niinimaa, V; Cole, P; Mintz, S; et al. (1981) Oronasal distribution of respiratory airflow. Respir Physiol 43:69-75.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-14 DRAFT-DO NOT CITE OR QUOTE
-------
Appendix H
Skin Rinsing Data
-------
Table H-l. Weight of clay rinsed from skin of each subject during each
individual skin rinse (g)
Subject
1
2
3
4
5
6
7
8
Rinse 1
0.321
2.957
0.558
0.139
2.908
9.893
0.158
0.443
Rinse 2
NAa
2.804
0.427
0.126
1.919
12.522
0.149
1.018
Rinse 3
0.773
0.083
0.333
0.18
3.042
10.319
0.313
2.618
aSample lost during analysis.
Table H-2. Residual clay (nig)
Subject
Subject 9
Wheel
Subject 10
Sculpture
Right Hand
9,750
1,874
4,059
1,536
1,367
70
83
74
Left Hand
11,243
2,352
4,270
2,845
3,426
14
65
98
Arms
398.55
790.25
388.60
5,005.35
8,630.60
33.50
58.50
131.80
Legs
509.80
596.25
1,276.70
958.50
273.95
8.40
42.85
9.20
Feet
214.40
144.00
267.20
220.65
2,991.50
17.40
42.65
14.10
Face
16.70
0.00
4.35
9.60
524.60
0.00
9.80
25.70
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/2007 H-2 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix I
Alternative Method for Estimating Dermal Absorption
-------
1 APPENDIX I. ALTERNATIVE METHOD FOR ESTIMATING DERMAL
2 ABSORPTION
O
4 This document uses the fraction absorbed approach to estimate dermal absorption, which
5 is the method recommended in current U.S. Environmental Protection Agency guidance (U.S.
6 EPA, 2004, 1992). This method does not accurately represent the mechanisms of dermal
7 absorption and presents difficulties in extrapolating experimental results to actual exposure
8 conditions. The discussion below presents an alternative approach using a more mechanistic
9 model. This method is based on work by Dr. Annette Bunge, as published in Bunge and Parks
10 (1998).
11
12 BASIC MODEL
13 Bunge and Parks (1998) present three approaches for estimating dermal dose from soil,
14 depending on whether absorption is small, large, or based on slow soil-release kinetics (i.e.,
15 desorption from soil is slow relative to dermal permeation). The slow-release kinetics approach
16 was selected as the best one to use because the high lipophilicity of dioxin, presence of organic
17 carbon in the clay, and small particles associated with clay all suggest that dioxin will be tightly
18 bound to the particles and slowly released. On this basis, the absorbed dermal dose (pg) is
19 estimated as follows:
AbsDose = CsmLoMsml [l-exp(-ksmlpsmlfareaAexptexp /Msml)\ (1-1)
2Q where:
21 Csoii,o = concentration of dioxin in soil at = 0 (pg/mg)
22 Msoii = mass of soil on exposed skin (mg)
23 ksoii = rate constant for first-order soil release kinetics (cm/s)
24 psoii = soil bulk density (mg/cm3)
25 farea = fraction of exposed area in contact with soil
26 Aexp = exposed skin area (cm2)
27 texp = exposure time (hr)
28
29 The rate constant and soil density terms can be combined into one term representing the
30 transfer rate from soil (k) with units of mg cm"2 hr"1. If the amount of dioxin absorbed is less
31 than about 10% of the original amount on the skin (i.e., Csoii,o x Mso;i), then Eq 1-1 simplifies to:
kf_AestexCsm,0 (1-2)
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 1-2 DRAFT-DO NOT CITE OR QUOTE
-------
1 ESTIMATING k
2 As discussed above, Eq 1-2 is based on the assumption of slow soil-release kinetics.
3 Assuming that desorption from soil is slow relative to dermal permeation, the rate of dermal
4 permeation can be used to estimate the rate of desorption from soil. This approach is used here.
5 As discussed in Section 5, this report derives the dermal absorption properties of dioxin
6 from Roy et al. (1990), who measured dermal absorption of tetrachlorodibenzo-/?-dioxin (TCDD)
7 in soil with an organic carbon content of 0.45% and applied at supermonolayer coverage
8 (monolayer estimated as 1.3 mg/cm2 and amount applied was 6 mg/cm2). The saturation limit
9 for TCDD in this soil was estimated as follows:
Csat — Foe Koc Sw (I~3)
10 where:
11 Csat = saturation limit for TCDD in soil (mg/kg)
12 Foc = fraction organic carbon in soil = 0.0045
13 Koc = organic carbon-to-water partition coefficient = 107 L/kg (U.S. EPA, 2003)
14 Sw = solubility of TCDD in water = 2 x 10"5 mg/L (U.S. EPA, 2003)
15
16 On this basis, the soil used by Roy et al. would have a saturation limit for TCDD of 0.8 mg/kg.
17 Roy et al. used soils with TCDD concentration of 1 mg/kg (1 ppm). Thus, the testing was
18 conducted at levels slightly above the saturation limit, which should yield maximum flux rates
19 through the skin.
20 The 24-hour average flux rate from Roy et al. was calculated as follows:
J = AbsDose l(Aexp texp) (1-4)
where:
22 J = flux through the skin (ng cm"2 hr"1)
23 AbsDose = 0.048 ng (includes amount in skin)
24 Aexp=1.77cm2
25 texp = 24 hr
26
27 This yields a flux estimate of 0.0011 ng cm"2 hr"1. Now, an absorption rate constant (ka) can be
28 calculated as follows:
ka = JSMICsat (1-5)
29 where:
30 JSM = maximum flux for supermonolayer coverage = 0.0011 ng cm"2 hr"1
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 1-3 DRAFT—DO NOT CITE OR QUOTE
-------
1 Csat = 0.8 mg/kg = 0.8 ng/mg
2
3 On this basis, ka is estimated to be 0.0014 mg cm"2 hr"1 and assumed equal to k.
4
5 ESTIMATING THE ABSORBED DOSE
6 Finally, the absorbed dose can be calculated using Eq 1-2. As an example, the parameter
7 values for Subject 2 were used:
8
9 Csoli,o = 1 62 pg/g = 0. 1 62 pg/mg
xp
10 Aexp = 970cm2
12 farea =1.0 (actual load exceeded monolayer)
13
14 This yields an absorbed dose of 0.88 pg. The absorbed dose calculation presented in Section 7
1 5 included an adjustment to reflect the observed difference between rat in vivo testing and rat in
16 vitro testing. These tests indicated that the absorbed dose in vivo was about twice as high as the
17 absorbed dose in vitro. Applying that factor to the dose estimate derived above yields an
18 absorbed dose of 1.8 pg. This is very similar to the value reported in Table 9 (1.65 pg) based on
19 the fraction absorbed approach. Note that the amount of dioxin in the monolayer can be
20 estimated as 97 pg (0. 162 pg/mg x 0.62 mg/cm2 x 970 cm2). This means that the absorbed dose
21 is less than 10% of the applied dose and Eq 1-2 is valid to use.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 1-4 DRAFT—DO NOT CITE OR QUOTE
-------
1 REFERENCES
2
3 Bunge, AL; Parks, JM. (1998) Soil contamination: theoretical descriptions. In: Roberts, MA; Walters, KA, eds.
4 Dermal absorption and toxicity assessment. New York, NY: Marcel Dekker; pp. 669-696.
5 Roy, TA; Yang, JJ; Krueger, AJ; et al. (1990) Percutaneous absorption of neat 2,3,7,8-tetrachlorodibenzo-p-dioxin
6 (TCDD) and TCDD sorbed on soils. Toxicology 10(1):308.
7 U.S. EPA (Environmental Protection Agency). (1992) Dermal exposure assessment: principles and applications.
8 Office of Science Policy, Office of Research and Development, Washington, DC; EPA/600/8-91/01 IB. Available
9 online at http://www.epa.gov/osa/spc.
10 U.S. EPA (Environmental Protection Agency). (2004) Risk assessment guidance for Superfund. Vol. I: human
11 health evaluation manual (part E, supplemental guidance for dermal risk assessment). Office of Superfund
12 Remediation and Technology Innovation, Washington, DC; EPA/540/R/99/005. Available online at
13 http://www.epa.gov/superfund/programs/risk/ragse/index.htm.
14
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 1-5 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix A
Subject Questionnaire Results
-------
Table A-l. Subject 1
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
50 hours
40 weeks
1 year
Hand building, sculptural work. Largely
consists of rolling out slabs and assembling
clay parts.
Short sleeve t-shirt and jeans and closed toe
shoes.
Hands and forearms.
Yes. Dryness. No cracking/bleeding. I use
lotion 3-4 times through the day.
Hands:
when rolling slabs - once per hour
when assembling clay - 3 or more times per
hour
Face: 1-2 times per day
Water only.
Yes, Aveeno brand lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-2 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-2. Subject!
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
10-15 hours
15-25 weeks
24 years
Mixing clay and maintenance activities
associated with the OSU Ceramics area.
Long and short sleeves, long pants, work
shoes.
Hands, arms, and face.
Dryness and cracking.
Hands: 2 minutes
Face: 5 hours
Soap and water.
Lotion during winter, but when my hands are
very dry a product called Satin Hands is
used.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-3 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-3. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
25 hours
30 weeks
14 months
Functional - thrown on wheel
Structural - hand built
Jeans with t-shirt and sandals (summer) or
long sleeves and closed toe shoes (winter).
Hands, arms, face, neck, and feet.
Dry cracking skin and cuticles on hands, red
small-bump rash on backs of hands and
inner forearms when using wheel, nasal
congestion.
Arms and hands: 3 to 5 minutes
Feet, face, and neck: 1-10 hours
Water only if returning to work, soap and
water when finished.
Aveda hand creme, Neutrogena Swiss
therapy lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-4 DRAFT—DO NOT CITE OR QUOTE
-------
Table A-4. Subject 4
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
More than 70 hours
50 weeks
2 years
Functional pots, cups, bowls, etc.
Overalls, long/short sleeve shirts and
sneakers.
Face, hands, sometimes arms and legs.
Extremely dry with cracking on fingertips.
Hands: 10 minutes
Face and body: 10-24 hours
Water only if returning to work, soap and
water when finished.
Heavy cream lotion or bag balm at the end
of the day and at intervals throughout the
day.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-5 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-5. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
More than 14 hours
35 weeks
6 years
Hand building objects about 1.5 feet tall.
Short sleeves/pants and shoes.
Hands, lower arms, face.
Yes, dryness, sometimes cracking.
Hands: <5 minutes
Arms: 8 hours
Face: 0.5-8 hours
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-6 DRAFT—DO NOT CITE OR QUOTE
-------
Table A-6. Subject 6
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
30-40 hours
30-40 weeks
25 weeks
Throwing objects using wheel, hand
building, and sculptural work.
Short sleeves, pants, shorts, and flip flops
shoes.
Arms, hands, feet, face.
Yes, dry skin on feet and hands and nails
being unable to grow healthily.
Hands: 30 minutes
Legs, feet, and face: 3-5 hours.
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-7 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-7. Subject?
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
10 hours
40 weeks
4 years
Clay sculpture. Rolling out slabs, pressing
them into molds. Limited work throwing
objects using wheel.
Short sleeves and pants (winter/spring/fall)
and shorts (summer).
Arms, hands, and face.
Dryness and cracking.
Hands: 1-2 minutes
Face and legs 1-2 minutes (powdered clay)
or end of day (wet clay).
Soap and water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-8 DRAFT—DO NOT CITE OR QUOTE
-------
TableA-8. Subjects
Question
Approximately how many hours per week do
you work with clay?
Approximately how many weeks per year?
How long have you been doing clay work with
this level of intensity?
What type of clay artwork do you do?
What types of clothing do you wear while you
work?
What areas of skin typically are exposed to the
clay while you work?
Do you correlate any skin health issues with
how much you work with clay? If yes, what?
In relation to the time you complete working
with clay, when do you wash parts of your
body that have been exposed to clay?
How do you wash your skin after you work
with clay?
Do you treat your skin with anything in
particular after working with clay?
Answer
20 hours
52 weeks
6 years
Large clay sculpture. Rolling out slabs, cut
and bend them and then press them together.
Pants or shorts, short sleeves or tank tops,
sneakers or sandals.
Arms, neck, hands, calves, and shins.
Dryness and cracking.
Hands: 5 minutes
Face and legs: 4-24 hours
Soap and water or just water.
Lotion.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 A-9 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix B
Pictures of Artisans Prior to Skin Rinse Procedure
-------
B
D
Figure B-l. Subjects 1-4.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-2 DRAFT—DO NOT CITE OR QUOTE
-------
B
D
Figure B-2. Subjects 5-8.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-3 DRAFT—DO NOT CITE OR QUOTE
-------
Figure B-3. Subject 9.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-4 DRAFT—DO NOT CITE OR QUOTE
-------
Figure B-4. Subject 10.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 B-5 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix C
Real-time Particle Concentration Data
-------
E
"3>
E
d
o
Q.
10
8
6
2
0
B
Personal Particle Concentration
using the pDR-1000 Particle Counter
0 20 40 60 80 100 120 140 160
Time (min)
Area Particle Concentration
using the CI-500 Particle Counter
0 20 40 60 80 100 120 140 160
Time (min)
Figure C-l. Subject 1.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-2 DRAFT—DO NOT CITE OR QUOTE
-------
10
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
Mixing and bagging
of powdered clay
Subject cleaned off
using compressed air
20 40 60
Time (min)
80
100
B
Area Particle Concentration
using the CI-500 Particle Counter
"c- 1 9
E I ^
"3) ,•/->
8 R
° D
o
"o A
~ 4
a 9
Q. Z
n
A I
i_
Mixing and bagging
of powdered clay
t
K
^ n ft. A .» .
^V uv^^w^
Subject cleaned off
using compressed air
/
0 20 40 60
Time (min)
80
100
Figure C-2. Subject 2.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-3 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
25 50 75
Time (min)
100
125
B
0
Area Particle Concentration
using the CI-500 Particle Counter
25
50
75
100
125
Time (min)
Figure C-3. Subjects.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-4 DRAFT—DO NOT CITE OR QUOTE
-------
SO.
g?0.
80
io.
Jo.
o.
Personal Particle Concentration
using the pDR-1000 Particle Counter
0 20 40 60 80
Time (min)
100 120
B
Area Particle Concentration
using the CI-500 Particle Counter
£0.5
0
0 20 40 60 80 100 120
Time (min)
Figure C-4. Subject 4.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-5 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
20
40
60 80
Time (min)
100 120 140
B
Area Particle Concentration
using the CI-500 Particle Counter
20
40
60 80
Time (min)
100 120 140
Figure C-5. Subjects.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-6 DRAFT—DO NOT CITE OR QUOTE
-------
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
30 60 90
Time (min)
120
B
o
Area Particle Concentration
using the CI-500 Particle Counter
20
40
60 80
Time (min)
100 120
Figure C-6. Subject 6.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-7 DRAFT—DO NOT CITE OR QUOTE
-------
"E4
"3)
o
'+*
O -I
Q. I
0
0
Personal Particle Concentration
using the pDR-1000 Particle Counter
10 20
Time (min)
40
B
Area Particle Concentration
using the CI-500 Particle Counter
0 20 40 60 80 100 120
Time (min)
Figure C-7. Subject 7.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-8 DRAFT—DO NOT CITE OR QUOTE
-------
Personal Particle Concentration
using the pDR-1000 Particle Counter
u
„ c
1
1*4
c t
u
= ^
O **>
I2
u ^
t
n 1
U. |
n
ry
^TAjv^AAAl^A^^^/v^^
W I I I I I
0 20 40 60 80 100 120 140
Time (min)
B
14
12
10
8
o
o
-------
Wheel Session 1
120
180 240
Time (min)
300
360
420
B
CO
E
O)
s.
Wheel Session 2
120 180
Time (min)
240
300
Figure C-9. Subject 9.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-10 DRAFT—DO NOT CITE OR QUOTE
-------
Wheel Session 3
4.5
£ 3.5
o
c
2.5
1.5
0.5
o ;=*£•
60 120
180 240 300
Time (min)
360 420 480
Wheel Session 4
60 120 180 240 300 360 420
Figure C-9. Subject 9 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-11 DRAFT—DO NOT CITE OR QUOTE
-------
E
CO
E
0.6
0.5
04
c 0.3
.2
o
8.
0.2
Wheel Session 5
60 120 180 240 300 360 420 480
Time (min)
Figure C-9. Subject 9 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-12 DRAFT—DO NOT CITE OR QUOTE
-------
Sculpture Session 1
30
25
20
o
60 120 180 240
Time (min)
Sculpture Session 2
300
360
420
120
180 240
Time (min)
300
300
420
Figure C-10. Subject 10.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-13 DRAFT—DO NOT CITE OR QUOTE
-------
Sculpture Session 3
Particle cone, (mg m-3)
D
Background Particle Concentration
OA _
ri o _
n 9
n 1 ^ -
01-
n n c
n -
I 1
V M ^ft ^^
T1VV'>Wln--^*-^V(^'^^ V
0 60 120 180 240 300 360 42
Time (min)
Figure C-10. Subject 10 (continued).
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 C-14 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix D
Respicon, Cascade Impactor, pDR-1000, and Climet CI-500
Data for Each Individual Subject
-------
Table D-l. Concentration by particle diameter (um) as measured by the
Respicon Air Sampler (mg/m3)a'b
Aerodynamic
Diameter
Subject 1
Subject 2
Subject 3
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Background0
<4
32 jim
0.18
0.31
0.39
0.22
0.10
0.23
0.15
0.31
0.085
Total
0.35
0.47
0.99
0.37
0.13
0.61
0.51
0.64
0.13
aDL (Detection Limit) = 0.015 mg/m3.
b!/2 DL was used in place of the
-------
Table D-3. Particle concentration as measured by the pDR-1000 Air
Sampler (mg/m3)
Subject 1
Subject 2
Subjects
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Mean
0.75
0.57
0.30
0.14
0.049
1.22
0.32
0.34
Maximum
8.42
8.33
0.84
0.81
0.27
7.70
3.51
5.14
Minimum
0.047
0.016
0.093
0.027
0.019
0.078
0.080
0.015
Table D-4. Concentration by particle diameter (um) as measured by the
3\a
Climet CI-500 Air Sampler (mg/m3)
Physical
Diameter
Subject 1
Subject 2
Subject 3
Subject 4
Subject 5
Subject 6
Subject 7
Subject 8
Background13
0.3-0.5
0.001
0.001
0.002
0.002
0.008
0.011
0.005
0.006
0.009
0.5-1.0
0.005
0.002
0.009
0.003
0.002
0.006
0.010
0.004
0.005
1.0-2.5
0.026
0.016
0.058
0.013
0.003
0.029
0.054
0.021
0.002
2.5-5.0
0.222
0.166
0.411
0.124
0.025
0.260
0.377
0.186
0.010
5.0-10
0.560
0.535
1.214
0.323
0.055
0.679
0.631
0.578
0.010
>10.0
1.499
1.747
3.756
0.964
0.167
1.746
0.817
1.878
0.019
Total
2.313
2.467
5.450
1.429
0.260
2.731
1.895
2.672
0.055
aConcentration calculations assume particle density of 2.6 g/cm3.
bBased on measurements taken late at night when no students were present in building.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 D-3 DRAFT—DO NOT CITE OR QUOTE
-------
Table D-5. Average concentrations by particle diameter ranges (um)
measured by the Cascade Impactor Air Sampler (mg/m3)a'b
Aerodynamic
Diameter
Subject 9
Session 1
Subject 9
Session 2
Subject 9
Session 3
Subject 9
Session 4
Subject 9
Session 5
Subject 10
Session lc
Subject 10
Session 2C
Subject 10
Session 3
Background11
0.5-2
0.004
32
0.024
0.024
0.044
0.053
0.081
0.198
0.092
0.079
0.005
Total
0.049
0.046
0.102
0.073
0.152
0.480
0.237
0.241
0.023
aDL (Detection Limit) = 0.0025 mg/m3.
b!/2 DL was used in place of the
-------
Table D-6. Concentration by particle diameter ranges (urn) measured by the
Climet CI-500 Air Sampler (mg/m3)a
Physical
Diameter
Subject 9
Session 1
Subject 9
Session 2
Subject 9
Session 3
Subject 9
Session 4
Subject 9
Session 5
Subject 10
Session lb
Subject 10
Session 2b
Subject 10
Session 3
Background0
0.3-0.5
0.008
0.010
0.006
0.012
0.011
0.018
0.003
0.006
0.012
0.5-1.0
0.003
0.005
0.004
0.007
0.008
0.015
0.005
0.008
0.009
1.0-2.5
0.005
0.003
0.005
0.011
0.004
0.067
0.031
0.039
0.003
2.5-5.0
0.026
0.014
0.026
0.055
0.018
0.353
0.172
0.181
0.011
5.0-10
0.042
0.027
0.054
0.113
0.026
0.746
0.367
0.341
0.012
>10.0
0.070
0.058
0.124
0.240
0.048
1.430
0.700
0.656
0.016
Total
0.155
0.117
0.220
0.439
0.115
2.629
1.278
1.231
0.064
""Concentration calculations assume particle density of 2.6 g/cm3.
bConcentration not adjusted for presence of dog.
°Based on measurements taken late at night when no students were present in building.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 D-5 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix E
SEM and EDS Data by Subject
-------
&5
rs
o-
^
a
rs
I
1 2
Fu\\ Scale 1314 els
1 2
Full Scale 3934 els Cursor: 1.622 keV (125 cts)
"1
--,-
'TS
I
o
W \
Figure E-la. Sample of clay used by Subject 1.
to
i
a,
gl
Hl
O 3
H
Figure E-lb. Clay particles on Subject 1's Respicon
Filter.
1 2
-ull Scale 8SSO cts Cursor: O.OOO keV
1 2
rLill Scale -1 3229 cts Cursor: 9.533 keV (1 7 ctsl
o
d
o
H
W
Figure E-2a. Sample of clay used by Subject 2.
Figure E-2b. Clay particles on Subject 2's Respicon
Filter.
-------
1 2
Full Scale 251 23 cts
•• i •'
8
keV
123
Full Scale 31 955 cts
• • • i •
6
keV
Figure E-3a. Sample of clay used by Subject 3.
Figure E-3b. Clay particles on Subject 3's Respicon
Filter.
• i • • •
2
Full Scale 25579 cts
1 • I •
•10
keV
1 2
Full Scale 14147 cts
keV
Figure E-4a. Sample of clay used by Subject 4.
Figure E-4b. Clay particles on Subject 4's Respicon
Filter.
-------
o
i
^s
^s
o
w !
-U °
•I"
I
of
>i
O i
o I
^ ^
§1
H
H
W
O
5^
O
d
o
H
W
123
Full Scale 14395 cts
Figure E-5a. Sample of clay used by Subject 5.
1 2
Full Scale 21 36 cts
Figure E-6a. Sample of clay used by Subject 6.
8
keV
keV
1 2
Full Scale 930 cts
keV
Figure E-5b. Clay particles on Subject 5's Respicon
Filter.
123
Full Scale 795 cts
10
keV
Figure E-6b. Clay particles on Subject 6's Respicon
Filter.
-------
VO
o
^1
§•
o
e
5
?
I"
'TS
J
o
a
1
H
H S.
W o'
o
a
o
H
W
123
Full Scale 5201 cts
10
keV
Figure ~E-7a. Sample of clay used by Subject 7.
10
Full Scale 751 6 cts
keV
Figure E-8a. Sample of clay used by Subject 8.
Full Scale "1344 cts
keV
Figure E-7b. Clay particles on Subject 7's Respicon
Filter.
2 4
Full Scale 9593 ds
10
keV
Figure E-8b. Clay particles on Subject 8's Respicon
Filter.
-------
Appendix F
Monte Carlo Simulation Result Graphics
-------
Append ixF
Do Not Quote or Cite
Crystal Ball Report - Full
Simulation started on 3/31/2006 at 7:15:34
Simulation stopped on 3/31/2006 at 7:23:41
Run preferences:
Number of trials run 1,000
Monte Carlo
Random seed
Precision control on
Confidence level 95.00%
Run statistics:
Total running time (sec) 487.37
Trials/second (average) 2
Random numbers per sec 35
Crystal Ball data:
Assumptions 17
Correlations 0
Correlated groups 0
Decision variables 0
Forecasts 4
F-1
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecasts
Worksheet: [VarDp-Dep monte5.xls]Monte
Forecast: Ingestion Dose
Summary:
Entire range is from 0.003 to 0.730
Base case is 0.058
After 1,000 trials, the std. error of the mean is 0.003
Cell: CSS
Ingestion Dose
0.06
0420
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
0.141
0.115
0.104
0.011
1.56
6.04
0.74
0.003
0.730
0.727
0.003
F-2
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Ingestion Dose (cont'd) Cell: C53
Percentiles: Forecast values
0% 0.003
10% 0.039
20% 0.059
30% 0.077
40% 0.097
50% 0.115
60% 0.135
70% 0.161
80% 0.207
90% 0.284
100% 0.730
F-3
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Inhalation Dose
Summary:
Entire range is from 0.00 to 1.05
Base case is 0.04
After 1,000 trials, the std. error of the mean is 0.00
Cell: C83
Inhalation Dose
000 005 0.10 0.15
0,20 025
pg/g
030 035 Q.«0 045
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
0.12
0.08
0.13
0.02
2.51
11.75
1.07
0.00
1.05
1.05
0.00
F-4
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Inhalation Dose (cont'd) Cell: C83
Percentiles: Forecast values
0% 0.00
10% 0.02
20% 0.03
30% 0.04
40% 0.06
50% 0.08
60% 0.10
70% 0.14
80% 0.18
90% 0.27
100% 1.05
F-5
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dermal Dose
Cell: C45
Summary:
Entire range is from 0.27 to 217.51
Base case is 10.91
After 1,000 trials, the std. error of the mean is 0.72
Total Dermal Dose
0.14
140
a.00
1000 2000 30,00 40.00 5000
6000 7000
v< on
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
15.50
7.92
22.91
524.87
3.67
20.69
1.48
0.27
217.51
217.24
0.72
F-6
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dermal Dose (cont'd) Cell: C45
Percentiles: Forecast values
0% 0.27
10% 2.02
20% 3.16
30% 4.29
40% 5.90
50% 7.92
60% 10.08
70% 14.09
80% 20.03
90% 36.15
100% 217.51
F-7
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dose
Cell: C86
Summary:
Entire range is from 0.28 to 219.14
Base case is 11.01
After 1,000 trials, the std. error of the mean is 0.73
Total Dose
0.00
10 00 20 DO
40.00
pg/d
50.00 6000 70.00
50 on
Statistics:
Trials
Mean
Median
Mode
Standard Deviation
Variance
Skewness
Kurtosis
Coeff. of Variability
Minimum
Maximum
Range Width
Mean Std. Error
Forecast values
1,000
15.76
8.12
23.01
529.38
3.66
20.67
1.46
0.28
219.14
218.86
0.73
F-8
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Forecast: Total Dose (cont'd) Cell: C86
Percentiles: Forecast values
0% 0.28
10% 2.15
20% 3.32
30% 4.51
40% 6.15
50% 8.12
60% 10.39
70% 14.44
80% 20.58
90% 36.63
100% 219.14
End of Forecasts
F-9
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumptions
Worksheet: [VarDp-Dep monte5.xls]Monte
Assumption: Activity selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C56
UTO BID 0-» 0X1 040 0» OM Ore OtO OBO 100
Assumption: Breath Selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C61
Dieafh Selector
Assumption: Clay load on arm (mg/cm2)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.04
0.35
3.00
Cell: C22
UWJ OJO OKI 0*0 I JO ' W l«0 2 <0 1*0 ."U
Assumption: Clay load on beverage (mg)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
Cell: C51
0.30
0.50
0.72
F-10
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Clay load on face (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
0.030
0.040
Cell: C40
Assumption: Clay load on feet (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
0.03
0.30
Cell: C34
Assumption: Clay load on food (mg)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.30
0.70
1.66
Cell: C49
d«y load on food (mq)
OM CM WO 100 1J> t«0 1*0
Assumption: Clay load on hand (mg/cm2)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
0.10
3.00
10.00
Cell:C17
Cta, lo.d „, h.nd
F-11
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Clay load on leg (mg/cm2)
Uniform distribution with parameters:
Minimum
Maximum
Cell: C28
Assumption: clothing selector
Uniform distribution with parameters:
Minimum
Maximum
0.10
0.70
0.00
1.00
Assumption: Dioxin cone in ball clay (pg/g)
Lognormal distribution with parameters:
Mean 808.00
Std. Dev. 318.00
Cell: C5
Assumption: Exposure Duration (hr/d)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
1.00
4.00
10.00
Cell: C7
Fxpcilif e Duration (fir/if)
F-12
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Fraction of ball clay in blend
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
Cell: C6
0.00
0.20
1.00
Assumption: M/F Selector
Uniform distribution with parameters:
Minimum
Maximum
0.00
1.00
Cell: C62
Assumption: MMAD (um)
Triangular distribution with parameters:
Minimum
Likeliest
Maximum
13.00
25.00
45.00
Cell: C60
xoo 1*00 xoo 1300 aoo *ooo
Assumption: Particle Concentration in Air(mg/m3)
Triangular distribution with parameters:
Minimum 0.08
Likeliest 0.44
Maximum 0.99
Cell: C59
010 930 030 9*9 9.9) Of) 0,70 090 OUO t TO
F-13
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Append ixF
Do Not Quote or Cite
Assumption: Total Body Surface Area (cm2)
Lognormal distribution with parameters:
Mean 18,000.00
Std. Dev. 37.40
Cell: C8
End of Assumptions
F-14
This document is a draft for review purposes only and does not constitute Agency policy .
-------
Appendix G
Evaluation of Clay Dust Inhalation
-------
1 APPENDIX G. EVALUATION OF CLAY DUST INHALATION
2
3 The methodology used to evaluate the dose of clay dust and associated dioxin received
4 via inhalation is discussed in this appendix. The appendix is divided into four sections: clay dust
5 size distribution, particle inhalability, respiratory deposition of clay dust, and delivered dose
6 estimates.
7
8 CLAY DUST SIZE DISTRIBUTION
9 As discussed in the main body of this report, the size distribution of clay dust was
10 measured using a Delron cascade impactor and a Climet during regular daily activities in the art
11 studio. The Climet optically determines particle concentration for six size bins with the
12 associated physical particle diameter (dp) of 0.3-0.5, 0.5-1, 1-2.5, 2.5-5, 5-10, and >10 |im.
13 Aerodynamic particle diameter (dae) can be estimated for the Climet's size bins by assuming that
14 the airborne clay dust has a density of 2.6 g/cm3, similar to that of bulk clay.1 Using this
15 approach, a clay particle with a dp of 10 jim has a dae of 16 jim. The Delron cascade impactor
16 fractionates particles directly, based on their dae, into the seven ranges of <0.5, 0.5-2, 2-4, 4-8,
17 8-16, 16-32, and >32|im.
18 During normal artisan activities (Subjects 1-8), 64 ± 9% (mean ± SD) of the aerosol is
19 associated with particles having a dae > 16 jam based on average Climet data. Based on average
20 impactor data, 63 ± 13% of the aerosol is associated with a dae > 16 jim (Subjects 1-8). The
21 particle size distributions to which the artisans were exposed was assumed to be log-normally
22 distributed.2 The cascade impactor data were selected for estimating particle size distributions
23 for the following reasons: (1) the impactor measures particle size based on the aerodynamic
24 behavior of particles, whereas the Climet uses light scattering to estimate a physical particle size;
25 (2) the impactor affords a better characterization of the large particles than does the Climet
26 because it contains an additional size bin of 16-32 jim; and (3) particle deposition in the
27 respiratory tract is a function of dae. Thus, uncertainty in estimates of respiratory deposition is
28 reduced by the direct measurement of dae by the impactor. The clay dust size distribution was
29 not estimated for runs where two or more of the impactor stages were below the nondetect level.
30 When engaged in normal artisan activities, the mass median aerodynamic diameter
31 (MMAD) of clay dust to which artisans were exposed ranged from 13 to 45 |im. Table G-l
Mae = dp {(clay density * Cc(dp) )/(H2O density * Cc(dae) )}05, where: Cc(dp) and Cc(dae) are the Cunningham slip
correction factor for the physical and aerodynamic particle size, respectively. For more information, the reader is
referred to ICRP (1994), page 239.
2For more information about particle sizing and the log-normal distribution, the reader is referred to Hinds (1999).
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-2 DRAFT-DO NOT CITE OR QUOTE
-------
1 provides a characterization of clay dust exposures for each subject. Figure G-l illustrates a log-
2 probability plot of a typical (i.e., near the average MMAD) clay dust particle size distribution
3 and a background sample from the studio. The prevalence of fewer large particles in the
4 background aerosol can be explained easily, based on particle-settling velocities. The settling
5 velocities for the dae of 1-, 10-, and 20-|im particles are 3.5 x 10"3, 0.3, and 1.2 cm/s,
6 respectively. Due to their rapidly settling velocities, large particles (dae > 10 jim) are maintained
7 in the air only by active generation or resuspension from surfaces. The substantive presence of
8 large particles (52% of mass associated with a dae > 10 jim) in the background sample is
9 suggestive of particle resuspension due to movement (e.g., walking and setting up sampling
10 equipment in the studio).
Table G-l. Clay dust size distribution and concentration during normal
activities
Subject
1
2
3
4
5
6
7
8
Mean ± SD
Size distribution3
MMAD (urn) og
26.9
44.6
18.5
n.a.
n.a.
20.2
13.0
26.7
25.0±11 3.*
3.9
4.8
4.3
n.a.
n.a.
3.0
3.6
3.3
I ±0.7
Total concentration
(mg/m3)
0.35
0.47
0.99
0.37
0.13
0.61
0.51
0.64
0.51 ±0.25
aThe aerosol size distribution is described in terms of the mass median aerodynamic diameter (MMAD) and
geometric standard deviation (og).
n.a. = not available
11 Data were also available for two subjects during specific activities (i.e., when sculpting
12 and using a pottery wheel) (see Table G-2). During pottery wheel operations, an average
13 MMAD of 33 jim with a geometric standard deviation (og) of 5.4 was observed. A dog was
14 present during two of the sculpting runs. The MMAD with the dog present was 21 |im versus
15 only 16 |im without the dog. The shift toward larger particles when the dog was present appears
16 to be consistent with particle resuspension due to the dog's movement around the studio.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-3 DRAFT-DO NOT CITE OR QUOTE
-------
i uu —
-
E
ZL
kT
1
Q
J>
o -in
•— 1 U
Is -
Q.
O
(0
c
T3
O
L.
n>
1
0
/
f
f
1
,
y
/
/
2
£
/
a'
/
/
-*1
jfl
/'/
/
/
/°
10
i 2
/
f
¥
/
P
O
/
/
30
0 4
/
f
X'
<
50
0 6
ji
/
f
/
i
f
/
/
70
0 8
/
' /
/
/
/
0 9°9
98
5 9
99.9
9
Cumulative Percent Less Than Indicated Size
Figure G-l. Clay dust particle size distribution during normal artisan activities
from analysis of cascade impactor data. Illustrated are the data for Subject 8 ( ) and
a background sample when work was not being done in the studio (o). The dashed and
solid lines illustrate the log-normal distribution for these respective data. The mass
median aerodynamic diameter (MMAD) of clay dust was 27 jim (og= 3.3) for Subject
8, whereas the background sample had an MMAD of 11 jim (og = 4.6).
1 PARTICLE INHALABILITY
2 For a given particle size, inhalability is the ratio of the particle concentration that enters
3 the respiratory tract through the nose or mouth to the concentration of these particles in the
4 ambient air. Inhalability depends mainly on particle size (i.e., dae), route of breathing, wind
5 speed, and a person's orientation with respect to wind direction. Wind speeds in the art studio
6 were assumed to be 0.3 m/s or less (Baldwin and Maynard, 1998). The artisans were presumed
7 to move about the studio such that their orientation was random with respect to wind direction.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-4 DRAFT-DO NOT CITE OR QUOTE
-------
Table G-2. Clay dust size distribution and concentration during specific activities
Subject
Subject 9
(Pottery wheel)
Run 1
Run 2
Run 3
Run 4
Run5
Mean ± SD
Subject 10b
(Sculpting work)
Run 1
Run 2
Run 3
Size
distribution3
MMAD urn og
33.7
n.a.
24.8
n.a.
39.3
32.6 ±7.3
21.2
20.4
16.0
6.2
n.a.
4.3
n.a.
5.6
5.4 ±0.9
3.9
3.2
3.5
Total
concentration
(mg/m3)
0.049
0.046
0.102
0.073
0.152
0.085 ± 0.044
0.48
0.24
0.24
aThe aerosol size distribution is described in terms of the mass median aerodynamic diameter (MMAD) and
geometric standard deviation (og).
bA dog was present during Runs 1 and 2 but not during Run 3. Therefore, these three runs were not averaged as
was done in the case of the pottery wheel work.
n.a. = not available
1 The clay dust aerosol present under normal activities in the art studio was observed to
2 have an average MMAD of 25 jim and og of 3.8. Hence, 50% (on average, by mass) of the
3 airborne clay dust is composed of particles having a dae of >25 jim, a size that is generally
4 considered to be unable to penetrate the thorax (ACGIH, 2004). These large particles
5 (dae >25 |im), if inhaled, will deposit almost completely and exclusively in the extrathoracic (ET)
6 airways. Thus, determining inhalability is key to estimating the delivered dose of these large
7 particles. For smaller particles, inhalability still describes the fraction of airborne particles that
8 may enter the respiratory tract and thereby the availability of these particles for deposition in the
9 lung.
10 Only limited data are available on the inhalability of particles from calm air (wind speeds
11 of 0.3 m/s and less). Inhalability from calm air depends on the route of breathing. Logistic
12 functions describing particle inhalability during nasal [P(!N)] and oral [P(Io)] breathing are given
13 by Menache et al. (1995) and Brown (2005):
14
1
exp(10.32-3.1141n(dae))
(G-l)
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-5 DRAFT-DO NOT CITE OR QUOTE
-------
1 44
P(I0) = : (G-2)
OJ l + 0.44exp(0.0195dae)
1
2 Note that these equations depend only on aerodynamic particle diameter, dae. Given by Eq G-l,
3 P(IN) begins a rapid decline from 0.95 at dae = 11 |im, to 0.5 at dae = 27.5 |im, and 0.1 at
4 dae = 56 |im. Equation G-2 predicts a slow decline in P(Io) from 0.95 at dae = 8 |im, to 0.5 at
5 dae = 74 jam, and 0.1 at dae =175 jim.
6 Figure G-2 illustrates particle inhalability predicted by Eqs G-l and G-2 (shown by solid
7 lines) along with relevant experimental data. Based on high wind speeds (1-8 m/s), the
8 American Conference of Governmental Industrial Hygienists (ACGIH) inhalability criterion is
9 also illustrated (shown by dashed lines) for comparative purposes. Equation G-l for P(!N)
10 describes the experimental nasal inhalability data well with an r2 of 0.86 (model sum of squares
11 divided by the total corrected sum of squares). A negative r2 is obtained for the fit of the
12 ACGIH (2004) criterion to these data.3 Equation G-2 describes the experimental oral
13 inhalability data with an r2 of 0.69, whereas the ACGIH criterion fit with an r2 of 0.32.
14
15 RESPIRATORY DEPOSITION OF CLAY DUST
16 Inhaled particles may be either exhaled or deposited in the ET, tracheobronchial (TB), or
17 pulmonary (PU) airways. The deposition of particles in the respiratory tract depends primarily
18 on inhaled particle size (i.e., dae), route of breathing (through the nose or mouth), tidal volume
19 (Vx), and breathing frequency (/"). Reference respiratory values for males and females were
20 adopted from the International Commission on Radiological Protection (ICRP, 1994). In
21 addition to breathing patterns (Table G-3) necessary for deposition calculations, males and
22 females were assumed to have a functional residual capacity of 3,300 mL and 2,680 mL,
23 respectively. The majority (70%) of the subjects were female; only Subjects 1, 2, and 5 were
24 male.
25 Particle deposition in the respiratory tract was predicted using the publicly available
26 Multiple Path Particle Dosimetry (MPPD) model.4 The MPPD model was developed by the
27 CUT
3An r2 is calculated as the model sum of squares (MSS) divided by the total corrected sum of squares (TSS). The
MSS equals the TSS minus the residual sum of squares (RSS). In typical linear regressions, when a model is fitted
to a data set, the resulting r2 must be non-negative because the least square fitting procedure assures RSS < TSS.
When r2 is computed on excluded data, i.e., data not used to fit the model, the RSS can exceed the TSS. In this
case, r2 (which is not the square of r) can be negative, indicating that the mean of the data is a better predictor than
the model.
4 The MPPD program is available on request from the CUT Centers for Health Research ().
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-6 DRAFT-DO NOT CITE OR QUOTE
-------
10
dae (Mm)
100
Figure G-2. Particle inhalability from calm air for nasal [P(!N)] and oral [P(Io)]
breathing as a function of aerodynamic particle diameter (dae) Left panel [
Equation G-l, • Breysse and Swift (1990), + Hinds et al. (1998), o Hsu and Swift
(1999), - - - ACGIH (2004)]. Right panel [ Equation G-2, o Aitken et al. (1999), •
Kennedy and Hinds (2002), - - - ACGIH (2004)].
Table G-3. Breathing patterns used in particle deposition calculations"
Activity
Sitting
Light exercise
VT (mL)
/(min1)
VT (mL)
/(min1)
Males
750
12
1,250
20
Females
464
14
992
21
1
2
3
4
5
Source: ICRP (1994), Table 8.
Centers for Health Research (CUT), United States, in collaboration with the National Institute of
Public Health and the Environment (RIVM), the Netherlands, and the Ministry of Housing,
Spatial Planning and the Environment, the Netherlands. The MPPD model may be used to
predict the deposition in the human respiratory tract for particles between 0.01 and 20 jim in
diameter. In the lung, the model considers deposition by the mechanisms of impaction,
sedimentation, and diffusion. Additional model details are available elsewhere (DeWinter-
Sorkina and Cassee, 2002). For the size of the clay dust, only impaction and sedimentation are
of concern.
Using the MPPD model, deposition was predicted for the ET, TB, and PU regions of the
respiratory tract. Particle deposition was estimated individually for oral and nasal breathing.
During oral breathing, deposition in the TB airways did not always reach zero by a dae of 20 jim
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-7 DRAFT-DO NOT CITE OR QUOTE
-------
1 (the upper limit for the MPPD model). For dae > 20 |im, deposition in the TB airways was
2 estimated by a best fit polynomial (3rd or 4th degree) determined using CurveExpert 1.3 (112B
3 Crossgate St., Starkville, MS 39759). This polynomial function was fitted to TB deposition
4 fractions for dae from 10 to 20 jim. The predicted ET deposition during oral breathing for a dae
5 >20 |im was taken as one minus the TB deposition fraction for oral breathing. For nasal
6 breathing, these additional steps were unnecessary because TB deposition was well under 1% at
7 a dae of 20 |im.
8 External to the MPPD model, all of the predicted deposition fractions were corrected for
9 particle inhalability using Eqs G-l and G-2. The current version of MPPD model offers an
10 inhalability correction for nasal breathing only. For a given dae, an inhalability corrected
11 deposition fraction is the product of the uncorrected deposition fraction and the predicted
12 inhalability for that dae. Unless otherwise specified, all mention of particle deposition fractions
13 in the main body of this report and subsequently in this appendix refer explicitly to inhalability
14 corrected deposition fractions.
15 The deposition fraction (DFr) of an aerosol in a region of the respiratory tract is the
16 integral of the deposition fractions across all particle sizes in the aerosol:
17
00
DFr (MMAD, ag) = J DFr (d, )p(d1 )5d, (G-3)
o
18
19 where:
20 DFr(di) = the deposition fraction in region, r, of particles having an aerodynamic
21 diameter of d;
22 p(di) = the mass fraction associated with the interval 5d;
23
24 The total deposition fraction for the respiratory tract is the sum of DFr for the ET, TB, and
25 PU regions. Equation G-3 can be approximated by summing the particle deposition fractions at
26 known intervals or percentiles of the particle size distribution. Here, the interval of 1% was used
27 and the approximation is:
28
i 0.99
DFr(MMAD,ag)* —^DF^) (G-4)
P=0.oi
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-8 DRAFT-DO NOT CITE OR QUOTE
-------
1 where:
2 DFr(di) = the deposition fraction in region, r, of particles having an aerodynamic diameter
3 d; (the particle size associated with a given percentile, P, of the size
4 distribution).
5
6 For a log-normal distribution, d; is given by:
7
(G-5)
8 where:
9 z(P) = the normal standard deviate for a given probability
10
1 1 Table G-4 provides the predicted regional deposition fractions for the clay dust in the
12 respiratory tract of each subject for oral and nasal breathing at two activity levels. These
13 deposition fraction estimates were based on each subject's measured aerosol exposure size
14 distribution (see Tables G-l and G-2). Subjects 4 and 5 lacked aerosol size distribution data and
15 were assumed exposed to an aerosol with an MMAD of 25 jim and og of 3.8, this being the
16 average for artisans during normal activities (see Table G-l). The deposition fraction estimates
17 for Subject 10 were based on Run 3, when the dog was not present in the studio.
18
19 DELIVERED DOSE ESTIMATES
20 The rate of particle deposition in a region of the respiratory tract may be expressed as:
21
Dr(t) = C(t) (t)VT(t)DFr(t) (G-6)
22
23 where:
24 br = the rate of deposition per unit time in region r
25 C = the exposure concentration
26 /= breathing frequency
27 VT = tidal volume
28 DFr = the deposition fraction in region r
29
30 Note that all of the variables in Eq G-6 may vary with time. The dose to a respiratory region is
3 1 determined by integrating Eq G-6 over the exposure duration.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-9 DRAFT-DO NOT CITE OR QUOTE
-------
1
2
Table G-4. Regional deposition fractions (corrected for inhalability) for clay
dust in the respiratory tract
Subject
1
2
3
4
5
6
7
8
9
10
Sitting
Nasal breathing
ET TB PU
0.441 0.015 0.022
0.336 0.011 0.016
0.472 0.028 0.033
0.447 0.021 0.022
0.458 0.016 0.023
0.526 0.023 0.022
0.549 0.035 0.041
0.451 0.018 0.017
0.368 0.020 0.023
0.533 0.030 0.033
Oral breathing
ET TB PU
0.473 0.082 0.058
0.412 0.059 0.042
0.431 0.104 0.067
0.471 0.091 0.050
0.479 0.086 0.061
0.521 0.108 0.053
0.432 0.128 0.085
0.507 0.087 0.041
0.396 0.077 0.047
0.462 0.118 0.072
Light exercise
Nasal breathing
ET TB PU
0.473 0.006 0.011
0.360 0.004 0.008
0.531 0.010 0.020
0.487 0.007 0.013
0.492 0.006 0.011
0.566 0.007 0.012
0.622 0.013 0.025
0.483 0.005 0.010
0.410 0.007 0.014
0.593 0.010 0.020
Oral breathing
ET TB PU
0.516 0.060 0.052
0.442 0.044 0.037
0.486 0.074 0.075
0.521 0.064 0.056
0.523 0.063 0.054
0.581 0.075 0.059
0.498 0.090 0.095
0.557 0.061 0.046
0.437 0.054 0.053
0.525 0.083 0.081
ET = extrathoracic; PU = pulmonary; TB = tracheobronchial
4
5
6
By assuming that aerosol characteristics and an individual's activity levels are fairly
constant over discrete periods of time, the dose to a respiratory region may be approximated by:
Dr =0.06X(VT/)J(CT)J[FmDFm^r = FNDFN>r]j
(G-7)
7 where:
8
9
10
11
12
13
14
15
16
Dr = the dose (jig) to region r of the respiratory tract
VT and/= tidal volume (mL) and breathing frequency (min"1) for a specified activity j
C and T = exposure concentration (mg/m3) and duration (hr) during activity j
Fm and FN = the fraction of a breath entering the respiratory tract through the mouth and
nose, respectively, during activity j
DFm>r and DFN>r = the deposition fraction for oral and nasal breathing, respectively, in
region r of the respiratory tract while performing activity]
Constant 0.06 = a unit conversion parameter
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-10 DRAFT-DO NOT CITE OR QUOTE
-------
1 As expressed, an "activity" in Eq G-7 could be associated with changes in exposure
2 concentration, the particle size distribution, and/or an individual's exertion level. For simplicity,
3 only two exertion levels (sitting and light exercise) and a single particle size distribution (see
4 Tables G-l and G-2) were considered for each subject.
5 The fraction of flow through the mouth (Fm in Eq G-7) increases with activity level and
6 varies between individuals. For the two activity levels considered here, most people (87%) will
7 breathe through their nose (Niinimaa et al., 1981). Hence, for these people, Fm= 0 and FN= 1 in
8 Eq G-7. However, 13% of people will be oronasal breathers even at rest, i.e., they will breathe
9 simultaneously through the nose and mouth (Niinimaa et al., 1981). This latter group is
10 commonly referred to in the literature as "mouth breathers" (e.g., ICRP, 1994). Derived from
11 Niinimaa et al. (1981), the fraction of air respired through the mouth (Fm) is well described by a
12 modified exponential function in the form of:
13
Fm=aexp
_J_
vVey
(G-8)
14 where:
15 Ve = minute ventilation
16 a = 0.748 and y=-7.09 (r2=0.997) in mouth breathers for 1QV£80 L/min and
17 35.3Ve80 L/min, a = 0.744, and y=-18.3 (r2=0.998) in normal augmenters
18
19 For Ve <35.3 L/min, normal augmenters breathe entirely through the nose, i.e., Fm = 0. FN is one
20 minus Fm regardless of the activity.
21 Table G-5 gives the estimated clay dust doses to regions of the respiratory tract for each
22 subject during nasal and oronasal breathing. Estimates are for a 4-hour exposure assuming that
23 the exposed individual spent 50% of his or her time sitting and 50% engaged in light exercise.
24 For oronasal breathing in Table G-5, there is a small positive bias in ET doses and a
25 corresponding negative bias in TB doses calculated by Eq G-7. In other words, this method of
26 calculating ET and TB doses shifts the pattern of deposition toward the head relative to the real-
27 life pattern of deposition. This shift occurs due to deposition being calculated at a higher airflow
28 rate through the nose and mouth than actually occurs during oronasal breathing. The deposition
29 calculations presumed that all inhaled airflow was through the nose or mouth. In reality, inhaled
30 air is partitioned between the nose and the mouth, and the actual flows (for sitting and light
31 exercise) are roughly half of that used in the deposition calculations. For breathing by a single
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-ll DRAFT-DO NOT CITE OR QUOTE
-------
1 route (nasal or oral), changing activity from sitting to light exercise approximately triples flow
2 rates but only slightly increases ET deposition and modestly decreases TB deposition (see Table
3 G-4). The effect of using Eq G-7 for calculating doses during oronasal breathing should
4 similarly affect the pattern of deposition. Ultimately, particles deposited in the ET and TB
5 regions will typically be cleared to the throat and swallowed within 24 to 48 hours
6 postdeposition (ICRP, 1994). Hence, the exact site of deposition (i.e., ET versus TB) is of little
7 significance because both regions effectively contribute to ingested doses.
8 Table G-6 provides estimates of the dioxin absorption in each subject for nasal and
9 oronasal breathing. Particles deposited in the ET and TB regions clear rapidly (within 1-2 days)
10 to the throat and are swallowed. The absorption of dioxin from particles deposited within the ET
11 and TB regions was treated as if the particles had been ingested. Dose estimates for oronasal
12 breathing are slightly more conservative from a safety or risk perspective than presuming nasal
13 breathing. However, nasal breathing may be considered as representative of the majority of the
14 population (87%). Oronasal breathing is thought to represent 13% of healthy individuals
15 (Niinimaa et al., 1981). In contrast to healthy subjects, Chadha et al. (1987) found that the
16 majority (11 of 12) of patients with asthma or allergic rhinitis breathe oronasally even at rest.
17 On average across all the subjects, dioxin doses are about 1.2 times greater for oronasal than for
18 nasal breathing.
19
Table G-5. Regional doses (jig) of clay dust in the respiratory tract3
Subject
1
2
3
4
5
6
7
8
9
10
Mean
SD
Nasal breathing
ET
664
678
1,677
580
256
1,114
1,011
997
110
455
754
460
TB
12
11
47
13
4.6
22
30
18
2.9
12
17
13
PU
20
19
75
19
7.7
29
49
24
4.5
18
27
21
Oronasal breathing
ET
693
757
1,612
598
264
1,126
917
1,067
114
431
758
445
TB
53
52
143
45
21
85
90
72
8.8
39
61
39
PU
48
47
154
41
19
70
100
57
9.2
39
58
42
a Doses calculated by Eq G-7 as described in the text.
ET = extrathoracic; PU = pulmonary; TB = tracheobronchial
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-12 DRAFT-DO NOT CITE OR QUOTE
-------
Table G-6. Estimates of dioxin absorption" (pg TEQ)
Subject
1
2
3
4
5
6
7
8
9
10
Mean
SD
ET and TB
0.033
0.034
0.084
0.029
0.013
0.055
0.051
0.049
0.005
0.023
0.038
0.023
Nasal breathing
b puc
0.003
0.003
0.010
0.002
0.001
0.004
0.006
0.003
0.001
0.002
0.004
0.003
Total
0.035
0.036
0.094
0.031
0.014
0.059
0.057
0.052
0.006
0.025
0.041
0.026
Oronasal breathing
ET and TBb PUC
0.036
0.039
0.085
0.031
0.014
0.059
0.049
0.055
0.006
0.023
0.040
0.023
0.006
0.006
0.020
0.005
0.002
0.009
0.013
0.007
0.001
0.005
0.007
0.006
Total
0.043
0.045
0.105
0.037
0.016
0.068
0.062
0.063
0.007
0.028
0.047
0.029
a Dioxin concentration was assumed to be 162 pg toxic equivalent (TEQ) per gram clay.
b Absorption fraction of 0.3 assumed, extrathoracic (ET) and tracheobronchial (TB) rapidly clear into the
gastrointestinal tract.
0 Absorption fraction of 0.8 assumed, due to slow clearance from pulmonary (PU) region.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-13 DRAFT-DO NOT CITE OR QUOTE
-------
1 REFERENCES
2
3 ACGIH (American Conference of Governmental Industrial Hygienists). (2004) TLVs and BEIs: based on the
4 documentation of the threshold limit values for chemical substances and physical agents and biological exposure
5 indices. Cincinnati, OH: ACGIH Worldwide.
6
7 Aitken, RJ; Baldwin, PEJ; Beaumont, GC; et al. (1999) Aerosol inhalability in low air movement environments. J
8 Aerosol Sci 30:613-626.
9
10 Baldwin, PEJ; Maynard, AD. (1998) A survey of wind speeds in indoor workplaces. AnnOccup Hyg 42:303-313.
11
12 Breysse, PN; Swift, DL. (1990) Inhalability of large particles into the human nasal passage: in vivo studies in still
13 air. Aerosol Sci Technol 13:459-464.
14
15 Brown, JS. (2005) Particle inhalability at low wind speeds. Inhal Toxicol 17:831-837.
16
17 Chadha, TS; Birch, S; Sacker, MA. (1987) Oronasal distribution of ventilation during exercise in normal subjects
18 and patients with asthma and rhinitis. Chest 92(6): 1037-1041.
19
20 De Winter-Sorkina, R; Cassee, FR. (2002) From concentration to dose: factors influencing airborne paniculate
21 matter deposition in humans and rats. Bilthoven, The Netherlands: National Institute of Public Health and the
22 Environment (RIVM); report no. 650010031/2002. Available online at
23 http://www.rivm.nl/bibliotheek/rapporten/650010031 .html.
24
25 Hinds, WC. (1999) Aerosol technology: properties, behavior, and measurement of airborne particles (2nd ed.). New
26 York, NY: Wiley-Interscience.
27
28 Hsu, DJ; Swift, DL. (1999) The measurement of human inhalability of ultralarge aerosols in calm air using
29 manikins. J Aerosol Sci 30:1331-1343.
30
31 ICRP (International Commission on Radiological Protection). (1994) Human respiratory tract model for
32 radiological protection: a report of a task group of the International Commission on Radiological Protection.
33 Oxford, United Kingdom: Elsevier Science Ltd. ICRP publication 66; Annals of the ICRP. Vol. 24, pp. 1-482.
34
3 5 Kennedy, NJ; Hinds, WC. (2002) Inhalability of large solid particles. J Aerosol Sci 33:237-255.
36
37 Menache, MG; Miller, FJ; Raabe, OG. (1995) Particle inhalability curves for humans and small laboratory animals.
38 AnnOccup Hyg 39:317-328.
39
40 Niinimaa, V; Cole, P; Mintz, S; et al. (1981) Oronasal distribution of respiratory airflow. Respir Physiol 43:69-75.
This document is a draft for review purposes only and does not constitute Agency policy.
9/25/07 G-14 DRAFT-DO NOT CITE OR QUOTE
-------
Appendix H
Skin Rinsing Data
-------
Table H-l. Weight of clay rinsed from skin of each subject during each
individual skin rinse (g)
Subject
1
2
3
4
5
6
7
8
Rinse 1
0.321
2.957
0.558
0.139
2.908
9.893
0.158
0.443
Rinse 2
NAa
2.804
0.427
0.126
1.919
12.522
0.149
1.018
Rinse 3
0.773
0.083
0.333
0.18
3.042
10.319
0.313
2.618
aSample lost during analysis.
Table H-2. Residual clay (nig)
Subject
Subject 9
Wheel
Subject 10
Sculpture
Right Hand
9,750
1,874
4,059
1,536
1,367
70
83
74
Left Hand
11,243
2,352
4,270
2,845
3,426
14
65
98
Arms
398.55
790.25
388.60
5,005.35
8,630.60
33.50
58.50
131.80
Legs
509.80
596.25
1,276.70
958.50
273.95
8.40
42.85
9.20
Feet
214.40
144.00
267.20
220.65
2,991.50
17.40
42.65
14.10
Face
16.70
0.00
4.35
9.60
524.60
0.00
9.80
25.70
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/2007 H-2 DRAFT—DO NOT CITE OR QUOTE
-------
Appendix I
Alternative Method for Estimating Dermal Absorption
-------
1 APPENDIX I. ALTERNATIVE METHOD FOR ESTIMATING DERMAL
2 ABSORPTION
O
4 This document uses the fraction absorbed approach to estimate dermal absorption, which
5 is the method recommended in current U.S. Environmental Protection Agency guidance (U.S.
6 EPA, 2004, 1992). This method does not accurately represent the mechanisms of dermal
7 absorption and presents difficulties in extrapolating experimental results to actual exposure
8 conditions. The discussion below presents an alternative approach using a more mechanistic
9 model. This method is based on work by Dr. Annette Bunge, as published in Bunge and Parks
10 (1998).
11
12 BASIC MODEL
13 Bunge and Parks (1998) present three approaches for estimating dermal dose from soil,
14 depending on whether absorption is small, large, or based on slow soil-release kinetics (i.e.,
15 desorption from soil is slow relative to dermal permeation). The slow-release kinetics approach
16 was selected as the best one to use because the high lipophilicity of dioxin, presence of organic
17 carbon in the clay, and small particles associated with clay all suggest that dioxin will be tightly
18 bound to the particles and slowly released. On this basis, the absorbed dermal dose (pg) is
19 estimated as follows:
AbsDose = CsmLoMsml [l-exp(-ksmlpsmlfareaAexptexp /Msml)\ (1-1)
2Q where:
21 Csoii,o = concentration of dioxin in soil at = 0 (pg/mg)
22 Msoii = mass of soil on exposed skin (mg)
23 ksoii = rate constant for first-order soil release kinetics (cm/s)
24 psoii = soil bulk density (mg/cm3)
25 farea = fraction of exposed area in contact with soil
26 Aexp = exposed skin area (cm2)
27 texp = exposure time (hr)
28
29 The rate constant and soil density terms can be combined into one term representing the
30 transfer rate from soil (k) with units of mg cm"2 hr"1. If the amount of dioxin absorbed is less
31 than about 10% of the original amount on the skin (i.e., Csoii,o x Mso;i), then Eq 1-1 simplifies to:
AbsDose = k /area 4xp t CSOI, 0 (1-2)
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 1-2 DRAFT-DO NOT CITE OR QUOTE
-------
1 ESTIMATING k
2 As discussed above, Eq 1-2 is based on the assumption of slow soil-release kinetics.
3 Assuming that desorption from soil is slow relative to dermal permeation, the rate of dermal
4 permeation can be used to estimate the rate of desorption from soil. This approach is used here.
5 As discussed in Section 5, this report derives the dermal absorption properties of dioxin
6 from Roy et al. (1990), who measured dermal absorption of tetrachlorodibenzo-/?-dioxin (TCDD)
7 in soil with an organic carbon content of 0.45% and applied at supermonolayer coverage
8 (monolayer estimated as 1.3 mg/cm2 and amount applied was 6 mg/cm2). The saturation limit
9 for TCDD in this soil was estimated as follows:
t,sat = Foe Koc Sw (I~3)
10 Csat = saturation limit for TCDD in soil (mg/kg)
11 Foc = fraction organic carbon in soil = 0.0045
12 Koc = organic carbon-to-water partition coefficient = 107 L/kg (U.S. EPA, 2003)
13 Sw = solubility of TCDD in water = 2 x 10"5 mg/L (U.S. EPA, 2003)
14
15 On this basis, the soil used by Roy et al. would have a saturation limit for TCDD of 0.8 mg/kg.
16 Roy et al. used soils with TCDD concentration of 1 mg/kg (1 ppm). Thus, the testing was
17 conducted at levels slightly above the saturation limit, which should yield maximum flux rates
18 through the skin.
19 The 24-hour average flux rate from Roy et al. was calculated as follows:
J = AbsDosel(Aexp texp) (I_4)where:
20 J = flux through the skin (ng cm"2 hr"1)
21 AbsDose = 0.048 ng (includes amount in skin)
22 Aexp=1.77cm2
23 texp = 24 hr
24
25 This yields a flux estimate of 0.0011 ng cm"2 hr"1. Now, an absorption rate constant (ka) can be
26 calculated as follows:
ka = JSM/Csat (I-5)where:
27 JSM = maximum flux for supermonolayer coverage = 0.0011 ng cm"2 hr"1
28 Csat = 0.8 mg/kg = 0.8 ng/mg
29
30 On this basis, ka is estimated to be 0.0014 mg cm"2 hr"1 and assumed equal to k.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 1-3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2 ESTIMATING THE ABSORBED DOSE
3 Finally, the absorbed dose can be calculated using Eq 1-2. As an example, the parameter
4 values for Subject 2 were used:
5
6 Csoii!0 = 162 pg/g = 0.162 pg/mg
7 Aexp = 970 cm2
8 texp = 4 hr
9 farea =1.0 (actual load exceeded monolayer)
10
11 This yields an absorbed dose of 0.88 pg. The absorbed dose calculation presented in Section 7
12 included an adjustment to reflect the observed difference between rat in vivo testing and rat in
13 vitro testing. These tests indicated that the absorbed dose in vivo was about twice as high as the
14 absorbed dose in vitro. Applying that factor to the dose estimate derived above yields an
15 absorbed dose of 1.8 pg. This is very similar to the value reported in Table 9 (1.65 pg) based on
16 the fraction absorbed approach. Note that the amount of dioxin in the monolayer can be
17 estimated as 97 pg (0.162 pg/mg x 0.62 mg/cm2 x 970 cm2). This means that the absorbed dose
18 is less than 10% of the applied dose and Eq 1-2 is valid to use.
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 1-4 DRAFT—DO NOT CITE OR QUOTE
-------
1 REFERENCES
2
3 Bunge, AL; Parks, JM. (1998) Soil contamination: theoretical descriptions. In: Roberts, MA; Walters, KA, eds.
4 Dermal absorption and toxicity assessment. New York, NY: Marcel Dekker; pp. 669-696.
5 Roy, TA; Yang, JJ; Krueger, AJ; et al. (1990) Percutaneous absorption of neat 2,3,7,8-tetrachlorodibenzo-p-dioxin
6 (TCDD) and TCDD sorbed on soils. Toxicology 10(1):308.
7 U.S. EPA (Environmental Protection Agency). (1992) Dermal exposure assessment: principles and applications.
8 Office of Science Policy, Office of Research and Development, Washington, DC; EPA/600/8-91/01 IB. Available
9 online at http://www.epa.gov/osa/spc.
10 U.S. EPA (Environmental Protection Agency). (2004) Risk assessment guidance for Superfund. Vol. I: human
11 health evaluation manual (part E, supplemental guidance for dermal risk assessment). Office of Superfund
12 Remediation and Technology Innovation, Washington, DC; EPA/540/R/99/005. Available online at
13 http://www.epa.gov/superfund/programs/risk/ragse/index.htm.
14
This document is a draft for review purposes only and does not constitute Agency policy.
9/24/07 1-5 DRAFT—DO NOT CITE OR QUOTE
-------
|