United States
  Environmental Protection
  Agency
EPA/6DO/R-09/135 November 2009 w\Anv.epa.gov/ord
                    A Scoping-Level
                    Field Monitoring Study of
                    Synthetic Turf Fields and
  Office of
  Research and Development

  National Exposure
  Research Laboratory
                                              .
                                  ;A \\y» 1, w*W*>K W&* ' "• t( W

-------
        EPA/600/R-09/135  November 2009 www.epa.gov/ord
A Scoping-Level Field  Monitoring Study of
   Synthetic Turf Fields and Playgrounds
                    Prepared by the
           National Exposure Research Laboratory
            Office of Research and Development
            U.S. Environmental Protection Agency

            with contributions from the Agency's
               Tire Crumb Science Workgroup

-------
                            Disclaimer

The information in this document has been funded by the U.S. Environmental
Protection Agency. It has been subjected to the Agency's peer and administrative
review and has been approved for publication as an EPA document. Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use.

-------
                               Acknowledgments

      This report was prepared for the U.S. Environmental Protection Agency (EPA) Tire
Crumb Committee, a cross-Agency workgroup led by Michael Firestone, Office of Children's
Health Protection and Environmental Education, and Ross Highsmith, National Exposure
Research Laboratory (NERL)/Office of Research and Development (ORD). A smaller Science
Workgroup was organized to consider options for collecting additional data from a small number
of U.S. synthetic turf fields and playgrounds. Initial Science Workgroup members included Cathy
Fehrenbacher, Office of Pollution Prevention and Toxics; Peter Grevatt and Tim Taylor, Office of
Solid Waste and Emergency Response; Linda Sheldon and Kent Thomas NERL/ORD; Urmila
Kodavanti, National Health and Environmental  Effects Research Laboratory/ORD; Souhail
AI-Abed, National Risk Management Research Laboratory/ORD; Jacqueline Moya, National
Center for Environmental Assessment/ORD; Jacqueline McQueen, Office of Science
Policy/ORD; Nora Conlon, Region 1; Mark Maddaloni, Region 2; Patti Tyler, Region 8; Arnold
Den, Region 9; and Dale Kemery, Office of Public Affairs. Some of the workgroup members
changed as the study progressed. NERL designed and implemented the limited scoping study
described in this report based on the Science Workgroup discussions and the availability of
resources.
      Ross Highsmith, Kent Thomas, Ron Wlliams, Don Whitaker, Sharon Harper, Karen
Bradham, Easter Coppedge, Fu-Lin Chen, Teri Conner, Bob Wllis, Carry Croghan, and Jeff
Morgan from NERL/ORD; Dennis Revel from Region 4; and Dan Boudreau and Dan Curran
from Region 1 contributed to the field sampling, sample analysis, and data compilation efforts.
Laboratory support was provided through the Senior Environmental Employment (SEE) and
Student Services Contracting Authority programs by Tom Gilmore, Charlie Bare, James Polk,
Doug Tilly, Lin Li, and Elena Arthur. Elizabeth Betz of NERL/ORD provided quality assurance
guidance and review. Numerous EPA program office, regional office, State, and local scientists
and staff provided valuable scientific input for use in developing the study design and technical
assistance in the implementation of the field study, especially gaining access to monitoring
sites. This study could not have been planned and implemented without their valuable
contributions.

-------

-------
                                      Abstract

       Recycled tire material, or "tire crumb," is used as a component in many recreational
fields, including synthetic turf fields and playgrounds. The use of tire crumbs in these
applications provides several benefits, including reduced sports injury. The public recently has
raised concerns regarding potential human health and environmental risks associated with the
presence of and potential exposures to tire crumb constituents in recreational fields, especially
with regard to children's exposures.
       In early 2008, U.S. Environment Protection Agency (EPA) Region 8 requested that the
Agency consider this issue. A cross-EPA workgroup inventoried and considered the limited
available scientific information: some laboratory studies of tire material content, off-gassing, and
leaching characteristics and a few European studies describing the extent and availability of tire
crumb constituents for potential human exposure. The workgroup recommended that research
be conducted to generate additional field monitoring data for potential U.S. environmental
conditions and potential exposures.
       A limited-scale study was conducted during the 2008 summer and fall seasons to
(1) gain experience conducting multiroute field monitoring of recreational surfaces that contain
   tire crumb by evaluating readily available methods for measuring environmental
   concentrations of tire crumb constituents; and
(2) generate limited field monitoring data that will be used by EPA to help the Agency determine
   possible next steps to address questions from the public regarding the safety of tire crumb
   infill in recreational fields.
       The field sites were selected based on availability and proximity to facilities of EPA's
National Exposure Research Laboratory; thus, the results reported here may not be
representative of environmental concentrations found at other sites. Because validated methods
for sampling synthetic turf fields or playgrounds did not exist, methods used for other
microenvironmental sampling were used. The full study protocol was implemented at two
synthetic turf fields and one playground. At each field and the playground, air sampling was
conducted to collect integrated particulate matter (PM10) and grab volatile organic chemical
(VOC) samples at two to three locations on each  turf field and playground and also at an upwind
background location. The air samples were collected at a height of 1 m  in close proximity to, but
without interfering with, planned recreational  activities. The VOC samples were collected around
2:00 p.m. Wipe samples were collected at the three turf field sampling locations, along with
readily available tire crumb infill and turf blade samples. Tire crumb material was collected from
the playground. The full protocol was implemented at one of the synthetic turf fields on a second
consecutive day providing repeat sampling data.  Selected samples were collected at a few
additional synthetic turf fields and one playground.
       Standard laboratory analysis methods were employed to analyze the environmental
samples for the targeted analytes. The PM10 samples were analyzed for PM mass, metals, and
particle morphology. The VOC samples were analyzed for 56 volatile organic analytes. The
wipe and material samples were analyzed for total extractable concentrations of several metals
and bioaccessible lead.
       Key findings are summarized  below.
(1) The study protocol and many of the methods  were found to be reliable and could be
   implemented in the field. Several  limitations are noted below.
   a.  Collecting integrated air samples provided  a high burden in terms of time and equipment.
   b.  Semivolatile  organic compounds were not  measured.
   c.  At any single site, there can be substantial variability in the materials used and the
      concentrations of contaminants measured. More work is needed  to determine where to
      collect samples and how many samples to collect to fully characterize a given site.

-------
   d. It was difficult to obtain access and permission to sample at playgrounds and synthetic
      turf fields. More work is needed to increase public and private owner participation if
      additional monitoring studies are conducted.
(2) Methods used to measure air concentrations of PM10 and metals were found to be reliable.
   a. Concentrations of PM10 and metals (including lead)  measured in air above the turf fields
      were similar to background concentrations.
   b. Concentrations of PM10 and metals at the playground site with high play activity were
      higher than background levels.
   c. All PM10 air concentrations were well below the National Ambient Air Quality Standards
      (NAAQS) for PM10 (150 ug/m3). All air concentrations for lead were well below the
      NAAQS for lead (150 ng/m3).
(3) Methods used to measure VOCs in air were found to be reliable.
   a. All VOCs were measured at extremely low concentrations that are typical of ambient air
      concentrations.
   b. One VOC associated with tire crumb materials (methyl isobutyl ketone) was detected in
      the samples collected on one synthetic turf field but was not detected in the
      corresponding background sample.
(4) Methods used to measure extractable metals from turf field blades, tire crumb materials, and
   turf field wipe samples were found to be reliable.  However,  the aggressive acid extraction
   procedure likely will  overestimate the concentration of metals that are readily available for
   human uptake. Since understanding uptake is a key component in understanding risk,
   methods to determine bioavailable metal concentrations still are needed.
   a. Total extractable  metal concentrations from the infill, turf blade samples and tire crumb
      material were variable in  the samples collected at a given site and between sites.
   b. The average extractable  lead concentrations for turf blade, tire crumb infill, and tire crumb
      rubber were low.  Although there are no standards for lead in recycled tire material or
      synthetic turf, average concentrations were well below the EPA standard for lead in soil
      (400 ppm).
   c. Likewise the average extractable lead concentrations for turf field wipe samples were
      low. Although there are no directly comparable standards, average concentrations were
      well below the EPA standard for lead in residential floor dust (40 ug/ft2).
(5) On average, concentrations of components monitored  in this study were below levels of
   concern; however, given the very limited nature of this  study (i.e., limited number of
   components monitored, samples sites, and  samples taken at each site) and the wide
   diversity of tire crumb material, it is not possible to reach any more comprehensive
   conclusions without  the consideration of additional data.
                                          VI

-------
                                Table of Contents

List of Tables	ix
List of Figures	ix

Executive Summary	xi

1. Introduction	1
       1.1 Background	1
       1.2 Exposure Science Questions	2
       1.3 Project Objectives	2
       1.4 Study Limitations	3

2. Conclusions and Recommendations	4
       2.1 Implementation of the Study Protocol	4
       2.2 Air Sampling and Analysis	4
       2.3 Surface Wipe, Tire Crumb, and Turf Blade Sampling and Analysis	5
       2.4 Conclusions with Regard to the Exposure Science Questions	5

3. Scoping Study Approach	7
       3.1 Scoping Study Goals	7
       3.2 Organizations	7
       3.3 Selection  of Target Analytes	7
       3.4 Proposed Sampling Sites and Sampling Locations	10
       3.5 Sampling Considerations	11
             3.5.1 Material Variability	11
             3.5.2 Activity Variability	11
             3.5.3 Sample Volume for  Detection Limit	12
             3.5.4 Meteorological Conditions	12
             3.5.5 Background Contribution	12

4. Methods	13
       4.1 Air VOC Samples	13
       4.2 Air PM10 Particle Samples for Mass and Metals Concentrations	13
       4.3 Air PM10 Particle Sample Collection for Scanning Electron Microscopy	13
       4.4 Surface Wipe Sample Collection—Synthetic Turf Fields	13
       4.5 Tire Crumb Infill Material Sample Collection—Synthetic Turf Fields	14
       4.6 Blade Material Sample Collection—Synthetic Turf Fields	14
       4.7 Tire Crumb Material Sample Collection—Playgrounds	14
       4.8 SEM Sample Preparation and Analysis	15
             4.8.1 Sample Preparation	15
             4.8.2 SEM Sample Analysis	15
       4.9 Surface Wipe, Tire Crumb, and Turf Blade Sample Metals Analysis	15
       4.10 Pb Bioaccessibility Analysis	17
             4.10.1 In Vitro Bioaccessibility Background Information	17
             4.10.2 In Vitro Pb Bioaccessibility Methodology	17
       4.11 Meteorological and Activity Information	18

5. Quality Control and Quality Assurance	19
       5.1 Air VOC Quality Control	19
                                         VII

-------
       5.2 Air PM10 Mass Quality Control	20
       5.3AirPM10 Metals Quality Control	20
       5.4AirPM10 SEM Quality Control	20
       5.5 Wipe, Tire Crumb, and Turf Blade Sample Quality Control and Quality
          Assurance	20
             5.5.1 Instrument Performance	20
             5.5.2 Recoveries	21
             5.5.3 Analysis of Blank Materials	21
             5.5.4 Measures of Precision	22
       5.6 In Vitro Bioaccessibility Analysis Quality Control and Quality Assurance	22
       5.7 Data Quality Assurance Review	22

6. Results	23
       6.1 Sampling Sites	23
       6.2 Site Characteristics	24
       6.3 Sample Collection	25
       6.4 Summary Measurement Results	30
       6.5 AirVOC Measurement Results	30
       6.6 Air PM10 Mass Measurement Results	34
       6.7 Air PM10 Metal Measurement Results	34
       6.8 Air PM10 SEM Measurement Results	35
       6.9 Total Extractable Metals in Synthetic Turf Field Surface Wipe, Tire Crumb
          Infill, and Turf Blade Samples and Playground Tire Crumb Rubber Samples	35
             6.9.1 Surface Wipes from Synthetic Turf Fields	36
             6.9.2 Tire Crumb Infill at Synthetic Turf Fields	36
             6.9.3 Turf Blades at Synthetic Turf Fields	37
             6.9.4 Tire Crumb Material from Playgrounds	37
             6.9.5 Lessons Learned with Regard to the Sampling and Analysis of Tire
                  Crumb Materials	38
       6.10 Pb Bioaccessibility Results	39
             6.10.1 Analysis of Turf Field Wipe, Tire Crumb, and Turf Blade Samples	39
             6.10.2 Lessons Learned with Regard to Bioaccessibility Data	39
       6.11 Methods Evaluation Summary	40

7. References	43

Appendix A: List of Sample Collection and Analysis Methods	44
Appendix B: Quality Control and Quality Assurance Results	45
Appendix C: Results from Analysis of Environmental Samples	74
Appendix D: NERL Report: SEM Analysis of Tire Crumb Samples	89
                                         VIM

-------
                                  List of Tables

Table 1. Summary of Sample Collection and Analysis Methods	8

Table 2. Sampling Site Information and Assigned Codes	23

Table 3. Overview of the Types of Samples Collected at Each Site	23

Table 4. Site Information	25

Table 5. Number of Samples Collected at Each Site	28

Table 6. Summary Results for Selected Analytes  in Air Samples Collected at Synthetic
Turf Fields and a Playground	31

Table 7. Summary Results for Total Extractable Pb, Cr, and Zn in Samples Collected at
Synthetic Turf Fields and Playgrounds	32

Table 8. Summary Results for Estimates of Pb Bioaccessibility in Samples Collected at
Synthetic Turf Fields and Playgrounds	33

Table 9. Overall Summary and Assessment of Methods Applied in This Scoping Study	41


                                 List of Figures

Figures 1 through 4. Site  F1 particle air samplers, surface wipe sample collection, green
turf blade with black granular tire crumb, and multiple turf blade colors	26

Figures 5 through 8. Tire  crumb  infill granules from site F2, shredded tire crumb at site P1,
and molded tire crumb material from site P2	27
                                          IX

-------

-------
                               Executive Summary

Background
       Recycled tire material, or "tire crumb", is used in many applications, including as a
component in synthetic turf fields and playground installations. The use of tire crumbs in these
applications provides several benefits, including but not limited to reduced impact injuries;
reduced or eliminated use of water, fertilizer, and pesticides needed to maintain grass fields;
reduced need for disposal of used tires in landfills; and increased availability of fields for
recreation. The public recently has raised concerns regarding potential human health and
environmental risks associated with the presence of and potential exposures to tire crumb
constituents in recreational fields, especially with regard to children's exposures.
       In early 2008, U.S. Environmental Protection Agency (EPA) Region 8 requested that the
Agency consider this issue, and a cross-EPA workgroup was formed. The workgroup
inventoried and considered the limited available scientific information: laboratory studies of tire
material content, off-gassing, and leaching characteristics. Also, a few European studies
reported data describing the extent and availability of tire crumb constituents for potential
human exposure through various routes and pathways (inhalation, ingestion, and dermal
contact).
       In the late spring  of 2008, a smaller EPA Tire Crumb Science Workgroup (science
workgroup) subsequently was formed and charged to consider the quality of the current science
and make recommendations regarding the need for future research. Because minimal
environmental or exposure data for U.S. populations were available, a limited scoping study was
proposed and designed to evaluate a protocol and methods for generating consistently collected
U.S. environmental data for select tire crumb constituents.
        This report provides the EPA scoping study results. The EPA scoping study results,
along with results from other studies conducted by Federal, State, and local organizations, such
as the Consumer Product Safety Commission (CPSC); the Agency for Toxic Substances and
Disease Registry; States including New Jersey, Connecticut, California, and New York; and
New York City, will be considered by EPA to identify possible next steps to address questions
from the public regarding the safety of tire crumb infill in ball fields and playgrounds.

Scoping Study Objectives
       The EPA science workgroup proposed a limited scoping-level study during 2008 that
included the following elements.
• Evaluate, through  real-world measurements, the application of readily available sampling and
  analysis methods for characterizing environmental concentrations of selected tire crumb
  contaminants in and around playgrounds  and synthetic turf fields.
• Evaluate the overall study protocol (monitoring, analytical,  and quality control [QC]
  procedures) for generating the quantity and quality of environmental measurement data
  needed to characterize the contribution of the tire crumb constituents to environmental
  concentrations.
• Collect a limited environmental dataset to help understand and assess methods for
  characterizing  potential route-  and pathway-specific exposures (inhalation, ingestion, and
  dermal) based on selected sentinel species.
• Generate a limited  set of consistently collected field measurement data from a very few
  playgrounds and synthetic turf fields that,  along with other study data, may be used to
  develop insights regarding the importance of the various exposure routes and pathways and
  to inform decisions regarding possible next steps to address questions from the public
  regarding the safety of tire crumb infill in ball fields and playgrounds.
                                           XI

-------
Study Approach
       The proposed final study design included the collection and analysis of selected air,
wipe, and material samples at one playground and one synthetic turf field site in the EPA
regions where the four National Exposure Research Laboratory (NERL) facilities are located.
This design (a total of eight sites) was based on the availability of NERL technical support.
During a single daytime period at each site, air samples were to be collected at up to three "on
field" or "on playground" sampling locations within the site boundaries in areas as close to
anticipated human activity as possible without interfering with routine activities. Air samples also
were to be collected at site background upwind sampling locations to characterize local ambient
background levels. A comparison of "on playground" or "on field" data with the background data
would be used to characterize the environmental availability of tire crumb constituents. Surface
wipe samples were to be collected at the  "on field" air sampling locations, but not at the
background sampling location. Tire crumb and synthetic turf blade samples were to be collected
at multiple sampling locations, but these were not always the same locations as  the air sampling
locations. The following samples were planned for collection and analysis.
• Grab air samples during  the hottest daytime period (-2:00 p.m.) to assess organic vapor
  concentrations (56 volatile organic compounds [VOCs])
• Integrated air particulate matter (PM10) samples to assess particle mass concentrations and
  concentrations of selected metals (including lead [Pb], chromium [Cr], zinc [Zn], and others)
• Integrated air PM10 samples to characterize ambient particles based on morphology (sizes
  and structure) using scanning electron  microscopy (SEM) and, if possible,  to estimate the
  relative contribution of tire crumb particles to the overall particle mass
• Wet surface wipe samples to assess environmental concentrations of metals (e.g., Pb, Cr, Zn,
  and others) associated with turf field materials (tire crumb rubber and turf blades)
• Turf field tire crumb infill granules, turf blades, and playground tire crumb material to assess
  concentrations of metals (e.g., Pb, Cr, Zn, and others) associated with these materials
• Field and laboratory QC samples to document the quality of the study data. Duplicate
  samples for each measure described above were collected where appropriate. Routine field
  and laboratory QC samples (e.g., blanks, spikes) also were analyzed.

Study Limitations
       This limited scoping-level study was designed to evaluate the methods for generating
quality environmental data  for selected tire crumb constituents and for understanding potential
exposure routes and pathways. The study was planned based on readily available resources
(personnel, equipment, media, etc.) and in consideration of the workgroup's desired study time
period (the summer and early fall of 2008). This  time period was  recommended,  as the
projected high ambient temperatures should result in conditions promoting the greatest potential
for the  environmental release of tire-related constituents.
       This study and the resulting data have many limitations. The study was not designed to
provide representative U.S. environmental measurement data for all tire crumb constituents or
applications. Nor was the study designed to inform conclusions regarding differences in U.S.
environmental concentrations or potential exposures to turf field and playground tire crumb
constituents based on geographical location, type, manufacturing materials, age, use, or
conditions. The study also was not designed to compare potential exposures to turf field and
playground tire crumb constituents with those at natural turf fields or playgrounds constructed
with other types of surfaces. The study collected limited environmental data to help understand
and assess methods for characterizing potential route- and pathway-specific exposures
(inhalation, ingestion, and dermal) based  on selected sentinel species. No personal exposure
data or related information  were collected. Validated sampling and analysis methods for
characterizing recreational  fields were not available, so existing methods used in similar studies
                                          XII

-------
were applied. The study did not evaluate methods for all the reported tire crumb constituents.
Semivolatile organic compounds (SVOCs [e.g., benzothiazole, aniline, polycyclic aromatic
hydrocarbons (PAHs)]) reported in some studies were not sampled or analyzed because of
resource limitations.

Sample Collection Results
Sampling Sites
       The full study protocol was implemented at only two synthetic turf field sites (F1 and F4)
and one playground site (P1), fewer than the planned four turf field and four playground sites.
Difficulties in identifying and arranging site access, logistical limitations, and personnel
requirements to operate the extensive array of equipment and sites were the key factors
impacting the number of sites monitored.
       Unplanned sampling also occurred and  is reported herein. The full protocol was
conducted at F1 on a second consecutive day providing repeat measures. A reduced set of
samples (without integrated air particle monitoring) was completed at a third synthetic turf field
site consisting of two collocated fields (F2 and F3). Some samples were collected for two
additional turf fields (F5 and F6) collocated with F4. Two F4 "on field" sampling locations were
very near a busy commuter road and parking deck.
       When a site consisted of multiple fields, one field was designated as the primary location
for implementing the protocol.  In total, samples were collected for six different synthetic turf
fields.
       Gaining access to playgrounds was very difficult and became even more difficult with
increased media  attention. The full sampling protocol was completed at only one playground
(P1) and at only two "on playground" sampling locations because of space limitations. Tire
crumb material molded to mimic wood bark was obtained from a second playground  site (P2).

Sample Collection and Analysis Methods
       AirVOCs. Grab air VOC samples (6-L Summa-polished stainless steel canisters) were
collected at each sampling  location at a 1-m inlet height during the hottest time of day
(-2:00  p.m.). The standard EPA Method TO-15 gas chromatography/mass spectrometry
(GC/MS) analytical method provided ambient-level concentration measurements for 56 VOC
analytes.
       Air Particulate Matter. Two integrated air PM10 samples (one for particle mass and
metals  analysis and another for scanning electron microscopy [SEM] analysis) were collected at
each sampling location at a 1-m inlet height over collection periods ranging from 5.8 to 7.8 h.
This resulted in individual sample air volumes ranging from approximately 7.0 to 9.2 m3 (3.5 to
5.0 m3 for SEM samples). PM10 mass was determined gravimetrically; metal concentrations by
X-ray fluorescence; and assessment of particle size and morphology and attempts to identify
the tire crumb component contribution by SEM.
       Synthetic Turf Field Surface Wipes. No known validated methods exist for
characterizing environmental concentrations of metals on synthetic turf surfaces comprised of
both turf blades and tire crumb rubber. A standard wet-wipe method (American Society for
Testing and Materials [ASTM]  E1728-03) used routinely to measure residential surface dust Pb
levels was used for this study. Advantages of this method were the availability of standard wipe
material and the existing, well-characterized, sampling and analytical methodologies. Samples
were collected at each "on field" turf field sampling location. Wipe samples were not collected at
the "on playground" or background sampling locations.  Each surface wipe, tire crumb, and turf
blade sample (described below) was extracted first using the EPA In Vitro Relative
Bioaccessibility Assessment Method 9200.1-86. (Note:  In vitro methods measure the
bioaccessibility [e.g., solubility] of metals during a simulated gastric extraction process to assess
the percentage of a metal in a material that may become available for absorption in the gastro-
                                          XIII

-------
intestinal [Gl] tract.) The same material from each sample then was extracted using EPA
Method 3050B. A total extractable concentration of Pb, Cr, Zn, arsenic (As), aluminum (Al),
barium (Ba), cadmium (Cd), copper (Cu), iron (Fe), manganese (Mn), and nickel (Ni) was
determined by an analysis of the combined bioaccessibility and Method 3050B extracts.
Extracts were analyzed by inductively coupled plasma (ICP)/MS using EPA Method 6020A. The
percent bioaccessible Pb was calculated from the relative amount in the bioaccessible extract
as compared with the total extractable amount.
       Synthetic Turf Field Tire Crumb Infill. Tire rubber infill was collected randomly at the
synthetic turf fields. In this study, it was decided to collect infill material that was readily
available at the surface rather than dislodging material trapped deep within the turf blades. This
decision was based partly on avoiding potential damage to field components but primarily
because the material  on the surface was more available for potential human contact. Infill
material was not available uniformly across the field surface.
       Synthetic Turf Field Blades. Blades were randomly collected at the synthetic turf fields.
Collecting blades of each color present at the field was attempted. Turf blade collection relied on
the availability of loose blades found on the field surface in lieu of a destructive (i.e., cutting)
method. Collection and analysis decisions were complicated by the limited availability of loose
blades and a later determination that a  minimum of 0.7 g of material was required for analysis.
       Playground Tire Crumb Rubber. Tire crumb samples were obtained from two
playground sites. It was not  clear how many pieces needed to be collected nor at what depth
(surface/subsurface)  for site characterization,  as the  crumb shifts with mechanical action. A
further challenge is that relatively small amounts (1 g or less) are required for analysis; large
amounts may overwhelm the digestion  and analytical systems. Intact tire crumb rubber pieces
were larger than  1 g.  A decision was made not to cut samples, as this would expose
unweathered surfaces and possibly impact the bioaccessible Pb estimate.

Conclusions
       The key study findings  are summarized below. The narrative and appendixes that follow
this Executive Summary provide additional details regarding the study, along with all of the
measurement and laboratory data. This descriptive report focuses on the study design and
methodologies; assessing the methodology for characterizing environmental concentrations of
tire crumb constituents in future studies; describing the quality of the scoping study data; and
providing recommendations for consideration  in the design of any future research, if needed.

In general, the study  protocol is expected to reliably yield data for assessing environmental
concentrations of selected tire crumb constituents and understanding potential exposure routes
and pathways. However, when considering future study designs and implementation, the
research needs to carefully consider issues associated with identifying and gaining site access,
the cost benefit of obtaining  the data versus the resource burden, and the implementation of
other methods for generating data to address specific research hypotheses. Future studies will
need a carefully developed and implemented communications  plan to promote the value of the
research and gain access to the required facilities.

(1) The study protocol and many of the methods were found to be reliable and could be
   implemented in the field. Several limitations are noted as follows.
   • Collecting integrated air samples  provided a high burden in terms of time and equipment.
   • SVOCs were not measured.
   • At any single site, there can be substantial variability in the materials used  and the
     concentrations  of contaminants measured.  More work  is  needed to determine where to
     collect samples and how many samples to collect to fully characterize a given site.
                                          XIV

-------
   • It was difficult to obtain access and permission to sample at playgrounds and recreational
     fields. More work is needed to increase public and private owner participation if these
     studies are to be implemented.
(2) Methods used to measure air concentrations of PM10 and metals were found to be reliable.
   • Concentrations of PM10 and metals (including Pb) measured in air above the turf fields
     were similar to background concentrations.
   • Concentrations of PM10 and metals at the playground site with high play activity were
     higher than background levels.
   • All PM10 air concentrations were well below the National Ambient Air Quality Standards
     (NAAQS) for PM10 (150 ug/m3). All air concentrations for Pb were well below the NAAQS
     forPb(150ng/m3).
(3) Methods used to measure VOCs in air were found to be reliable.
   • All VOCs were measured at extremely low concentrations that are typical of ambient air
     concentrations.
   • One VOC associated with tire crumb materials (methyl isobutyl ketone) was detected in
     the samples collected on one synthetic turf field but was not detected in the corresponding
     background sample.
(4) Methods used to measure extractable metals from turf field blades, tire crumb materials, and
   turf field wipe samples were found to be reliable. However,  the aggressive acid extraction
   procedure likely will overestimate the concentration of metals that  are readily  available for
   human uptake. Because understanding uptake is a  key component in understanding risk,
   methods to determine bioavailable metal concentrations are still needed.
   • Total extractable metal concentrations from the infill, turf blade samples, and tire crumb
     material were variable both between sites and at the same sites.
   • The average extractable lead concentrations for turf blade, tire crumb infill,  and tire crumb
     rubber were low. Although there are no standards for Pb in recycled tire material or
     synthetic turf, average concentrations were  well below the EPA  standard for lead in soil
     (400 ppm).
   • Likewise the average extractable Pb concentrations for turf field wipe samples were low.
     Although there are no directly comparable standards, average concentrations were well
     below the EPA standard for lead in residential floor dust (40 ug/ft2).
(5) On average, concentrations of components monitored in this study were below levels of
   concern; however, given the very limited nature of this study (i.e., limited number of
   components monitored, samples sites, and samples taken at each site) and the wide
   diversity of tire crumb material, it is not possible to reach any more comprehensive
   conclusions without the consideration of additional data.
                                          xv

-------

-------
                                  1. Introduction

1.1 Background
       Tire crumb or crumb rubber is produced from scrap tires or from the tire retreading
process. During the recycling process, steel and usually fiber are removed, and the remaining
rubber material is processed either by mechanical means or through freeze cracking into "chips"
or into various sizes of rubber mesh with a granular consistency. Tire crumb is used in several
commercial applications, including road construction, sidewalks, automobile parts, and in a
number of athletic and recreational applications. Recreational uses include ground cover (chips)
under playground equipment, landscaping mulch (chips), running track material (granular or
molded), and filler material used with many synthetic turf sports and playing fields (granular).
       The use of tire crumb materials for playground and turf fields provides numerous
benefits. First, it cushions falls, reducing sports injuries when compared with other playground
or athletic surfaces. Second, synthetic turf is a low-maintenance alternative to natural grass, as
there is no or reduced need for water, fertilizers, or pesticides. Because turf fields are installed
with below-ground drainage systems, there is reduced waiting time after storms, which
promotes their use. Third, reusing expended tires reduces their potential as disease vectors
(e.g., water hosting mosquitoes) and reduces the burden on landfills.
       There have been increased reports in the media of parents becoming alarmed when
their children returned home with tire crumb particles or fragments adhering to their socks and
clothing picked up while playing on tire-crumb-surfaced playgrounds and turf fields. The U.S.
Environmental Protection Agency (EPA) Region 8 asked several EPA program offices to help
understand the extent of crumb rubber recreational uses, fill critical data gaps, and assess the
available data to determine if there was any unreasonable exposure or risk, particularly to
children. In response to this request, an Agency-wide workgroup was formed to assess the
existing information and determine whether the Agency needed to collect additional information.
The workgroup included representatives from the various program, policy, scientific, and
communications staff, including the Office of Children's Health Protection and Environmental
Education (co-lead), the Office of Pollution Prevention and Toxic Substances, the Office of Solid
Waste and Emergency Response, the Office of Research and Development (ORD; co-lead),
and several EPA regional offices. The workgroup requested that a smaller science workgroup
familiar with planning and conducting environmental field studies be formed to consider the
quality of the current science and make recommendations regarding the need for future
research, if any  is needed.
        This scoping study was proposed, designed, and recommended by the science
workgroup as a  means for evaluating readily available methods and to generate consistently
collected U.S. data that could be used to help inform decisions regarding possible next steps to
address questions from the public regarding the safety of tire crumb infill used in  ball fields and
playgrounds. This study was not intended to address the very large number of variables that
might impact environmental concentrations or potential exposures (e.g., manufacturers,
materials, installation practices, spatial/temporal differences, age, use). The limited study data
were intended to complement data collected or planned for  collection by other State and
Federal agencies. Although this study included collection and analysis of environmental
samples that may be associated with several synthetic turf components, the focus of EPA's
work is developing  and evaluating methods for characterizing tire crumb constituents. Analysis
of the other components was included to better understand  the relative portion of any observed
tire crumb constituent environmental levels measured in the various samples. This study may
complement research performed by the Consumer Product  Safety Commission (CPSC); the
States of California, New Jersey, New York, and Connecticut; and New York City regarding
synthetic turf, but is distinct from the other studies in that the focus is on the tire crumb material.

-------
       Prior to preparing for a scoping study, a search of the scientific literature revealed limited
environmental or exposure measurement data associated with the use of tire crumb rubber for
U.S. recreational fields. Only a few peer reviewed laboratory or environmental studies were
reported, with many of these studies conducted in Europe.
       Although the results were limited, the search identified a number of compounds and
metals that may be found in tires, although not all of these compounds and metals are
contained in every tire nor are they contained in the same concentration in any tire at any given
time. These compounds and metals include those that follow.
Acetone                        Polycyclic aromatic              Manganese
Aniline                          hydrocarbons                 Mercury
Benzene                       Styrene-butadiene               Nickel
Benzothiazole                  Toluene                        Sulfur
Chloroethane                   Trichloroethylene                Zinc
Halogenated flame              Arsenic                         Pigments
  retardants                    Barium                         Nylon
Isoprene                       Cadmium                       Polyester
Methyl ethyl ketone              Chromium                      Rayon
Methyl isobutyl ketone           Cobalt                         Latex
Naphthalene                    Copper
Phenol                         Lead

1.2 Exposure Science Questions
       A series of general science questions was considered before the study protocol was
developed; they include the following ones.
• Can existing collection and analysis approaches and methods be used to assess
  environmental concentrations of tire crumb rubber constituents at synthetic turf fields and
  playgrounds?
• How well do such methods perform under real-world conditions?
• Do the methods produce data of sufficient quality to characterize potential exposure routes
  and pathways?
• Do the methods produce data of sufficient quality to characterize the contribution of
  constituents to various sources?
• Are the data and information produced through this research, when included with data from
  other studies, useful for developing hypotheses and informing the design of future research, if
  needed?
• What new methods are needed to fully characterize tire crumb environmental concentrations
  and to  understand potential exposure routes and pathways?

1.3 Project Objectives
       The science workgroup planned a very limited scoping-level field measurement study
during the 2008 summer/fall season to
• evaluate, through real-world measurements, the application of readily available sampling and
  analysis methods for characterizing environmental concentrations of selected tire crumb
  contaminants in and around playgrounds and synthetic turf fields;
• evaluate the overall study protocol (monitoring, analytical, and QC procedures) for generating
  the quantity and quality of environmental measurement data needed to characterize the
  contribution of the tire crumb constituents to environmental concentrations;
• generate a  limited set of consistently collected field measurement data from a few
  playgrounds and synthetic turf fields that, along with other study data, may be used  to

-------
  develop insights regarding the importance of the various exposure routes and pathways and
  inform the decision regarding future research, if any is needed; and
• understand the factors influencing the development and implementation of future study
  protocols.

1.4 Study Limitations
       This study was designed as a limited scoping-level methods evaluation study. It was
planned based on readily available resources (personnel, equipment, media, etc.) and in
consideration of the workgroup's desired study time period, the 2008 summer and early fall
months when high ambient temperatures should result in conditions promoting the greatest
potential for release of tire-related constituents. The study collected limited environmental data
to help understand and assess methods for characterizing potential route and pathway-specific
exposures (inhalation, ingestion, and dermal) based on selected sentinel species. This  study
and the resulting data have many limitations, which are described below.
• The study was not designed to provide representative U.S. environmental measurement data
  for all tire crumb constituents or applications, nor to make conclusions regarding differences
  in environmental concentrations or potential U.S. exposures to field and playground tire
  crumb constituents based on geographical location, type of recreational field, manufacturing
  materials, age, use, or conditions. Resource constraints prohibited the survey, coordination,
  and random selection of U.S. playgrounds and turf fields and the use of the study data in
  supporting statistical analysis or making statistical inferences. The study results can be used
  only to describe the playgrounds and turf fields monitored.
• The number of samples collected at each site was relatively small and will not necessarily
  support the spatial characterization of the species concentrations across the monitored area.
• Sampling was planned to be conducted only on one day. Therefore, temporal characterization
  of the targeted environmental contaminants will not be supported.
• No personal exposure data or related  information were collected.
• No scripted activities were planned or conducted. The study results were dependent on
  normal activity levels by the individuals using the playground or turf field. However, the limited
  data collected in this study likely will not be useful in  characterizing differences associated
  with these factors.
• The study did not evaluate methods for all the reported tire crumb constituents. Sampling and
  analysis of semivolatile organic compounds (SVOCs; e.g., benzothiazole, aniline, polycyclic
  aromatic hydrocarbons [PAHs]), reported in many studies, were not performed because of
  resource limitations.
• Validated sampling approaches and analysis methods were not available for real-world
  playground and synthetic turf field conditions. Currently accepted methods for measurement
  and analysis of the targeted species in indoor and outdoor microenvironments and in soils
  were used, with modifications required in some cases.
• QC/QA activities were implemented to document the quality of the sampling and analysis
  measurements; however, suitable QA/QC materials and standards were not available for
  some of the types of samples.

-------
                   2. Conclusions and  Recommendations

       The following narrative provides key highlights regarding the approach and methods that
were applied and assessed in this scoping study. The study results are being provided to the
workgroup for assessment and interpretation.

2.1 Implementation  of the Study Protocol
• A small scoping-level study protocol was fully implemented at two synthetic turf fields and one
  playground. The protocol was successfully implemented at one of these fields on a second
  day, providing a set of unplanned consecutive  day data. Additional data were collected from
  four turf fields and for tire crumb from a second playground.
• The study's success reflects the excellent collaborations and contributions of scientists and
  staff across many program offices, regions, States, and ORD.
• The protocol, and a majority of the corresponding methods employed in this study, generated
  quantitative data that can be used to characterize the contribution of tire crumb constituents to
  the environmental concentrations measured at the synthetic turf field, playground, and
  background sampling locations.
• Although none of the methods have been validated for this specific application, most methods
  were able to provide measurement data of known quality and at concentrations adequate for
  assessing potential tire crumb constituents.
• Air particle collection required considerable time, equipment, and expertise.
• Other collection procedures (air volatile organic compounds [VOCs]), wipe, and material
  collection) required much less time, equipment, and expertise.
• In general, the study protocol can be implemented and will yield data for assessing
  environmental concentrations and potential exposures for tire crumb constituents for various
  routes and pathways. However, when considering the design and implementation of future
  studies, the research needs to carefully consider
  - issues associated  with identifying and gaining site access,
  - the value of the data being generated versus the resource burden, and
  - the implementation of other methods for generating data to address specific research
    hypotheses.
• Any future study will  need a corresponding carefully developed  and implemented
  communications plan to help promote the value of the research and gain access to the
  required facilities.

2.2 Air Sampling and Analysis
• The air sample collection and analysis methods provided data suitable (both quality and
  concentration levels) for assessing environmental levels of particles, metals, and VOCs in air.
• Air particulate matter sampling employing relatively large (carry-on-size suitcase), battery-
  operated pumps and size selective inlets yielded sufficient particle mass for measuring
  selected metals at commonly reported ambient air levels. This sampling approach required
  significant resources (equipment and experienced field  staff) and long setup and sampling
  durations (8 to 10 h).
• Collecting air VOCs via grab sampling during the hottest daytime period (conditions when the
  greatest emissions from tire crumb material were anticipated) was simple  and required little
  time (~1 h).
• The air VOCs methods generated concentration data for many compounds. Slightly elevated
  MIBK levels were found at one turf field. The reproducibility in the data approximates what
  previously has been  reported in other field measurement studies. The use of an integrated

-------
  sampling pump method likely would increase the number of species with reportable data but
  would not necessarily generate substantially different data for characterizing tire crumb
  source contributions, environmental concentrations, or potential exposures.
• The air particulate matter (PM) mass and metals methods yielded reproducible results, with
  the turf field concentrations approximating the background levels. Concentrations on one
  playground site were somewhat higher than the background concentration.
• Tire crumb related fibers were not observed in the air samples analyzed by scanning electron
  microscopy (SEM). The SEM data were not sufficient for source apportionment or attribution
  of the data to any tire crumb constituent because of the variability of compositional or
  morphological characteristics of particles associated with the tire crumb material collected in
  this study.

2.3 Surface Wipe, Tire Crumb, and  Turf Blade Sampling  and Analysis
• No evaluated method was available for assessing dermal and indirect ingestion from tire
  crumb constituents in turf field or playground surfaces. A standard surface wet wipe sample
  collection method for residential lead (Pb) measurement was used at the synthetic turf fields.
  This method performed reasonably well for assessing extractable metals and required modest
  skills and time (~1 h).
• Collection of tire crumb infill and turf blade material at synthetic turf fields and tire crumb at
  playgrounds  was straightforward, requiring minimal skills and resources (~1 h). Convenience
  samples were collected in this study based on the materials being readily available on the
  surface. There is evidence that the material is not homogeneous with regard to some
  constituents  (Pb for example). Future site characterization studies should be considered to
  evaluate the  issue of sample heterogeneity and the  impact on data interpretation.
• Wipe, tire crumb, and turf blade samples were extracted using EPA Method 9200.1-86 for in
  vitro Pb bioaccessibility and EPA Method 3050B for total extractable Pb (and other metals).
  Both extraction techniques were combined with EPA inductively coupled plasma (ICP)/MS
  Method 6020A. These  methods require extensive skill and resources.  Multiple analyses of
  sample extracts with varying dilutions were required to capture  the range of elements and
  concentrations within appropriate calibration parameters.
• The in vitro Pb bioaccessibility method was judged not appropriate for the surface wipe
  samples. Because the  in vitro method  has been validated only for soil  samples, additional
  validation studies would be required to fully demonstrate the relevance of the method for tire
  crumb and turf blade materials.
• Although the methods appeared to perform reasonably well, a number of sample handling,
  size, and heterogeneity issues were discovered that may affect method performance and data
  interpretation.
• There is a lack of appropriate QC/QA materials and spiking methods. QA/QC materials and
  procedures need further development  for the methods as applied to these materials.
• The wipe, tire crumb, and turf blade data identified a potentially significant variability in source
  contribution based on turf field blade color and type, along with the tire crumb fraction being
  analyzed. Additional research is needed to understand the factors influencing the reported
  variability before future studies are designed and conducted. Understanding the variability is
  important in developing improved approaches for site characterization.

2.4 Conclusions with  Regard to the Exposure Science Questions
• Can existing collection and analysis approaches and methods be used to assess
  environmental concentrations of tire crumb rubber constituents  at synthetic turf fields and
  playgrounds? Yes,  existing air sampling and analysis methods can be used. Existing methods

-------
for analysis of metals in synthetic turf field and playground components can be successfully
applied, but may require additional validation assessments.
How well do such methods perform under real-world conditions? The air sampling and
analysis methods evaluated performed well. Methods for analysis of metals in synthetic turf
field and playground components showed good precision, but the assessment of recovery for
some metals was difficult because  of the  nonhomogeneity of the bulk materials.
Do the methods produce data of sufficient quality to characterize potential exposure routes
and pathways? In most cases, the  methods appeared to produce data of sufficient quality with
regard to sensitivity, precision, and accuracy. Additional validation efforts may be needed to
interpret measurement results, particularly with regard to bioaccessibility of metals in
synthetic turf field and playground components.
Do the methods produce data of sufficient quality to characterize the contribution of
constituents to various sources? Some of the methods generated data of sufficient quantity
and quality. Research is needed to better understand relative source contributions, in
particular for the wipe and air particle samples.
Are the data and information produced useful, when  included with data from other studies, for
developing hypotheses and informing the design of future research, if needed? The
assessment of approaches and methods  tested in this scoping study, in combination with
research recently completed and ongoing by other organizations, will be very useful for
developing hypotheses and informing the design of future research, if needed.
What new methods are needed to fully characterize tire  crumb environmental concentrations
and to understanding potential exposure routes and pathways? Testing and application of
personal sampling methods would  provide a more complete understanding of how
environmental concentrations translate into potential exposures. Methods for collection and
analysis of SVOCs were not tested in this scoping study but would be needed for a full
characterization.

-------
                          3. Scoping Study Approach

3.1 Scoping Study Goals
       The primary goal of this scoping study was to evaluate readily available methods and
approaches for characterizing environmentally available concentrations of selected
contaminants at synthetic turf fields and playgrounds that include tire crumb material. There are
currently no known validated sampling and analysis methods for these types of installations and
materials. Integrated and/or grab air, wipe, and material sample collection and analysis methods
(Table 1) were selected based on professional evaluation. Where available, standard methods
used routinely to characterize the targeted environmental contaminants in other
microenvironments were selected. However, because of time and resource constraints, none of
these methods were evaluated for the intended study application. A list of the sample collection
and analysis methods used or developed for this study is provided in Appendix A. The detailed
methods were included in the approved study QA project plan.
       A number of constraints influenced the decisions on proposed methodologies. These
included limited  resources (e.g., time, people); an anticipated lack of readily available electrical
power at the sites; uncertainty in sample collection times because of site availability or activity
issues; the  need for equipment that can be shipped to multiple sites across the country; and the
need for rugged, simple methods that could be implemented consistently by minimally trained
technical staff at several sites.

3.2 Organizations
     The scoping study approach was developed based on the cross-Agency collaborative
effort outlined below.
• National Exposure Research Laboratory (NERL): Prepare and ship sample collection
  equipment and media (less VOCs [see below]). Provide technical support for measurements.
  Analyze air filter media for mass, metals, and morphology. Analyze tire crumb material, turf
  blade material, and surface wipes for metals.
• EPA Regions 4, 5, and 9: Identify, assess, and coordinate access to the study sites.
  Communicate study to the public.
• EPA Region 1: Prepare VOC sampling media (canisters) and conduct TO-15 analyses.
• Workgroup: Assessment and interpretation of the study data provided in this report in context
  with other research data  and Agency compliance guidelines following receipt of this report.

3.3 Selection of Target Analytes
       Target analytes in this study were selected based on a combination of three factors:
(1) chemicals that have been associated with tire material (see Section 1.1);  (2) chemicals that
have been reported in other measurement studies at synthetic turf fields or playgrounds or are
of interest for these types of facilities; and (3) chemicals that could be analyzed using the
methods and resources that were readily available for this study. Methyl isobutyl ketone (MIBK)
was selected as a potential marker for emissions of volatile tire components into the air. PM10
was selected because it may occur from physical degradation of tire crumb material and its
potential for activity-related suspension into the air. PM10 particles are of interest because they
may be inhaled and also swallowed following trapping by mucus membranes. The metals Pb
and chromium (Cr) were of interest both because of their potential presence in tire material, and
also because they have been shown to be associated with pigments used in some types of
synthetic turf blades.
The metal zinc (Zn) was of interest as a potential marker for tire crumb material. Other metals
were of secondary interest  because of their potential association with tire material or because
they can provide additional particle source information. In some cases, additional VOC or metal

-------
Table 1. Summary of Sample Collection and Anal}
Sample Type
Air Particulate
Matter (PM10—
particles with
aerodynamic
size <10 |jm)







Air Metals in
PM10




Air Particles/
Fibers for
Scanning
Electron
Microscopy
(SEM)





Air Volatile
Organic
Compounds
(VOCs)










Sites
Synthetic
turf fields

Playgrounds








Synthetic
turf fields

Playgrounds


Synthetic
turf fields

Playgrounds







Synthetic
turf fields


Playgrounds









Sampling Method
ORD/NERL Research
Protocol

SKC pump, gel-cell
battery, Harvard
10-um impactor,
37-mm Teflon filter,
20-L/min flow rate
1-m sampling height,
three sites on/near
playground/field, one
site for background
Same sample as
collected for PM10
mass.



ORD/NERL Research
Protocol

SKC pump, gel-cell
battery, Harvard
10-um impactor,
polycarbonate filter,
10 L/min

Same sites and
height as PM10
Method TO-1 5

6-L Summa canisters

Grab sample
collected at approx.
2:00 pm, or hottest
time of day when
access to the field is
possible.
1-m sampling height,
three sites "on
playground/field", one
site for background
fs'\s Methods
Analytical Method
ORD/NERL Research
Protocol

Gravimetric analysis








ORD/NERL Research
Protocol

XRF (X-ray fluorescence)


ORD/NERL Research
Protocol

SEM (scanning electron
microscopy)






TO-15GC/MS(gas
chromatography/mass
spectrometry)











Target Analytes
PM-io mass











Primary:
Pb, Cr, Zn

Secondary:
Ca, Cl, Cu, Fe,
K, Mn, S, Si, Ti
Particle
morphology

Particle size
distribution

Attempt to
characterize tire
crumb
composition
signature
Primary:
Methyl-isobutyl-
ketone

Secondary:
55 other alkane,
aromatic,
oxygenated, and
halogenated
compounds





-------
Table 1. Summary of Sample Collection and Analysis Methods (cont'd.)
Sample Type
Surface Wipe





















Tire Crumb
Material
(crumbs from
playgrounds
and infill
material from
synthetic turf
fields)









Sites
Synthetic
turf fields




















Synthetic
turf fields

Playgrounds













Sampling Method
ASTME1 792 wipes
(Ghost Wipes) and
ASTME1 728 dust
collection method.
This is a standard wet
wipe method.

Collect wipes from
three 1 -ft2 sites on turf
fields.

Wipes placed in
precleaned 50-mL
polyethylene
container.
Collect a second
sample wipe next to
each original
sampling location for
archival or possible
metals bioavailability
analysis
ORD/NERL Research
Protocol

Collect samples of
crumb material from
three sampling
locations on each
playground or field.

Add material to clean
HOPE bottle.
Collect a second set
of samples for
archival for possible
metals bioavailability
analysis or SVOC
analysis.
Analytical Method
EPA Method 3050B, acid
digestion with
determination by EPA
Method 6020A (ICP/MS
inductively coupled
plasma mass
spectrometry)

RBALP in vitro extraction
(EPA Method 9200. 1-86)
for bioaccessible lead,
and determination using
ICP/MS by EPA Method
6020A








EPA Method 3050B, acid
digestion, and
determination by EPA
Method 6020A (ICP/MS)

RBALP in vitro extraction
(EPA Method 9200. 1-86)
for bioaccessible lead,
and determination using
ICP/MS by EPA Method
6020A






Target Analytes
Primary:
Pb, Cr, Zn

Secondary:
Al, As, Ba, Cd,
Cu, Fe, Mn, Ni

Pb














Primary:
Pb, Cr, Zn

Secondary:
Al, As, Ba, Cd,
Cu, Fe, Mn, Ni

Pb










-------
Table 1. Summary of Sample Collection and Analysis Methods (cont'd.)
Sample Type
Turf Blades


















Sites
Synthetic
turf fields

















Sampling Method
ORD/NERL Research
Protocol

Collect loose blades
from field surface in
several sampling
locations on the field,
collect samples of
different colors where
possible, place blades
in a clean
50-mL polyethylene
container.
Where there is
sufficient material,
archive material for
possible metals
bioavailability
analysis.
Analytical Method
EPA Method 3050B, acid
digestion, and
determination by EPA
Method 6020A (ICP/MS)

RBALP in vitro extraction
(EPA Method 9200. 1-86)
for bioaccessible Pb, and
determination using
ICP/MS by EPA Method
6020A








Target Analytes
Primary:
Pb, Cr, Zn

Secondary:
Al, As, Ba, Cd,
Cu, Fe, Mn, Ni

Pb











analytes were included because they could be measured as part of the routine analysis of
analytes of higher interest. As noted earlier, semivolatile chemicals, such as benzothiazole and
PAHs, were of interest but were not measured in this scoping study because of the lack of
readily available resources.

3.4 Proposed Sampling Sites and Sampling Locations
      A study goal was to collect real-world environmental samples in four geographical areas
across the United States (located near the four NERL laboratory locations) in late summer and
fall of 2008:
• Athens, GA (EPA Region 4),
• Research Triangle Park, NC (EPA Region 4),
• Cincinnati, OH  (EPA Region 5), and
• Las Vegas, NV (EPA Region 9).
      The recommended approach relied on available NERL technical staff to implement the
sampling protocol and the use of the laboratories' facilities as staging areas. In each
geographical region, two sampling sites were to be defined: (1) a playground with crumb rubber
material and (2) a synthetic turf field with tire crumb rubber infill. The proposed design would
result in sampling at four playground sites and four synthetic turf field sites. Based on availability
and access, alternate approaches were to be considered regarding the number of sites to be
monitored  in an area.
      At a given sampling site (turf field or playground), four sampling locations were to be
selected: three "on field" sampling locations and a background sampling location. The proposed
"on field" sampling location configuration was an isosceles triangle, with one sampling location
near the center of the playground or field and the other two at approximately equally distanced
downwind  positions. Actual deployment configuration was dependent on the site layout, planned
activities, and wind direction. The background sampling location was intended to be within
100 m of the field or playground, over a natural grass surface when possible, and not in close
proximity to likely pollutant sources. At each of the four sampling locations, all the following
environmental samples and measures were to be collected (nominal, except where noted):
                                          10

-------
• integrated air PM10 sample for particle mass and metals composition,
• integrated air PM10 sample for particle morphology (size and shape),
• grab air sample for VOCs,
• wet wipe for metals composition (not at the background sampling location), and
• materials (tire crumb and turf fibers).
       All air sampling was to be conducted with the inlet at a height of 1 m. Additional
duplicate samples and measures were collected at one sampling location at each site. Standard
meteorological measures (temperature, relative humidity [RH], and wind speed and direction)
were taken periodically at each site sampling location. Activity levels (e.g., number of activities,
number of individuals, type of activity) also were recorded. A 1-m2 square plastic child barrier
was set up around each sampling location to prevent the participants from running into or falling
on the sampling equipment.

3.5 Sampling Considerations
     There is no standard approach for determining the number of sample collection locations
or the timing of sample collection at any one playground or synthetic turf field site. Key factors
that were considered included
• potential variability of materials and chemical concentrations within a site;
• potential variability of activities at a site over time;
• meteorological conditions, particularly moisture, temperature, and wind speed and direction;
  and
• contribution of ambient background levels or nearby source contribution of the targeted
  chemicals to onsite measurements.
Each key factor is briefly described below, along with the proposed approach taken to minimize
or characterize the impact on the resulting data.

3.5.1 Material Variability
       Factor: Materials and chemical concentrations could  vary within a site resulting in
variation of targeted species across space and time at the playground or turf field. However, few
data were available regarding the variability in contaminant concentrations within a playground
or synthetic turf field site.  Also, there was little information available to guide optimum locations
for sampling at a site.
       Proposed Approach: Three sampling locations were selected within the boundaries of
the playground or turf field in areas close to the anticipated activity. These sampling locations
were positioned such as to not interfere with normal activity or use. An additional background
sampling location was selected near (-20 to 100 m) and upwind from the playground or turf field
to characterize ambient background levels. This approach was implemented successfully for air
samples with the exception that only two "on playground"  sampling locations were set up at the
playground because of the small size of the area. In addition, the tire crumb infill material and
synthetic turf blades were collected where available, rather than at predetermined locations.

3.5.2 Activity Variability
       Factor: Activities could vary over time at a site. Activity levels for the sites and sample
collection locations could be highly variable within and between sites. This study was a scoping
environmental measurement and methods development study. Therefore, no scripted activities
and no personal measurements were implemented. Activity levels may affect air particle
measurements. However, activity levels are unlikely to affect air VOC measurements, surface
wipes, and tire crumb and turf blade material grab sampling. The normal use or activities at one
site might require one or more of the samplers to be deployed near but not directly on the
                                           11

-------
playground or turf field. In this case, the samplers were placed as close to the activity as
permitted.

3.5.3 Sample Volume for Detection Limit
       Adequate sample volumes are needed to obtain reasonable detection limits for airborne
particles and metals on particles. Therefore, equipment and methods for particle air sampling
were selected to provide flow rates of 20 L/min over a nominal 8-h sample collection period. If
the particle air sampling duration was applied to match an ongoing  site activity of less than 8 h,
then detection limits for those samples would be higher. A minimum sample collection time of
2 h for air particle  samples was selected.
       Proposed Approach: Where practical and permitted, air particle samples were collected
in close proximity to where activities were ongoing,  but without interfering with the normal
activities  or use of the playground or turf field. Information about extant activities at each site
was collected. This approach was implemented successfully.

3.5.4 Meteorological Conditions
       Factor: Meteorological conditions, particularly moisture, temperature, and wind speed
and direction, might impact sample collection decisions and potential  emissions. Meteorological
conditions may influence air particle and air VOC measurements at playground and turf field
sites. Wind will transport airborne pollutant species  away from the site and will transport ambient
pollutant species onto the site. Suspension and resuspension of particles likely will be affected
by meteorological  conditions. Temperature likely will influence the VOCs that might be emitted
from tire crumb or other synthetic turf materials. With higher temperatures, higher levels of
VOCs emissions would be anticipated.
       Proposed Approach: This study was designed to collect air samples during those
meteorological conditions that likely would result in  the highest emissions (i.e., hot, dry, and
calm days). Sampling was scheduled in August and September on  days when no rain had
occurred  on the previous day and when no rain was anticipated, with  anticipated wind speeds
<10 mph. Air VOC samples were collected during the hottest time of day (~2 p.m.) at each "on
field" or "on playground" sampling location at the site.  Air sampling  locations were selected,
where possible, to offset potential changes in wind direction. Portable meteorological
measurement stations were not deployed. Basic information about  meteorological conditions
(temperature, wind speed, and  approximate wind direction) was collected at each site using a
handheld measurement device. In general, this approach was implemented successfully.

3.5.5 Background Contribution
       Factor: Ambient background pollutant levels could contribute to onsite measurements,
particularly for air samples. Ambient contaminants also may contribute to the total burden at the
playground or turf field as a result of aerosol or dust deposition.
       Proposed Approach: At  each playground or  synthetic turf field site, one background
sampling location was collected upwind from the playground or turf field. A set of air samples
identical to the other sample collection locations was collected at the  background sample
collection location. The resulting data were intended to be used to characterize the potential
contribution of ambient or background air contaminants to the playground or turf field. This
approach was implemented successfully.
                                          12

-------
                                    4. Methods

4.1 Air VOC Samples
       Grab air samples for VOCs were collected using evacuated 6-L Summa-polished
stainless steel canisters during the hottest time of day (~2 p.m.). This collection time was
selected for likely highest air and surface temperature conditions promoting VOC emissions. Air
temperature and surface temperature were measured and recorded at the time the samples
were collected. Each sample was collected by opening the canister valve and allowing the
evacuated canister to fill with air over an interval of approximately 20 s. The canister valve then
was closed, and the canisters stored at ambient temperature until analysis. The canister
samples were analyzed at the EPA Region 1 Office of Environmental Measurement and
Evaluation for 56 VOCs by gas chromatography/mass spectrometry (GC/MS) following EPA
Method TO-15.

4.2 Air PMio Particle Samples for Mass and Metals Concentrations
       PM10 is defined as airborne particulate matter with an aerodynamic size less than 10
(im. Ambient air was sampled at a nominal flow rate of 20 L/min (LPM) using metered, direct-
current-supplied active samplers (SKC-HV-30 air pumps) and Harvard Impactor inlets (Air
Diagnostic and Engineering), enabling PM10 mass loading on 47-mm Teflo filter media (Williams
et al., 2008). Air monitoring was  initiated for all  monitors in quick order on their setup and
calibration and continued  without interruption through the monitoring event (day). At the
conclusion of the sampling event, filter samples were recovered, stored in sealed transportation
containers, and returned to the laboratory under ambient temperatures. The sampler ending
flow rate was checked.
       Filters were returned to the NERL Research Triangle Park, NC, gravimetric weighing
facility, which operates under Federal Reference guidelines  for  temperature and relative
humidity specifications (22 ± 0.5 °C, 35 ± 1% RH). The filters underwent a 24-h equilibration
period prior to mass loading determination (Chen et al., 2007). Filter mass loadings were
determined as the difference between presampling (tare) weights and those obtained following
postsampling using a Sartorius MC 5 microbalance. The differential mass loading and data
pertaining to the total volume of air sampled through each individual filter then was used to
calculate the air mass PM10 concentration in units of micrograms per cubic meter for each
sampling location.  Immediately following gravimetric analysis, the PM10 mass concentration
filters were released to the NERL X-ray fluorescence (XRF)  laboratory for metals analysis.
Metals analysis was performed for 44 selected metals using the NERL's unique Lawrence
Berkeley National Laboratory-designed spectrometer (Williams  et al., 2008).

4.3 Air PMio Particle Sample Collection for Scanning Electron  Microscopy
       Sample collection  for assessing air particle morphology  and selected particle metals
composition  using SEM was conducted similarly to the primary  PM10 mass and metals sample
collection method. Identical SKC HV-30 pumps and similar Harvard Impactor samplers were
used, the only differences being  the operation of the units at a lower flow rate (10 LPM) to
overcome observed filter  pressure drop issues  affecting pump battery life and run time.
Specialized 37-mm polycarbonate filter material (Nuclepore) needed for SEM analyses was
used.

4.4 Surface Wipe Sample Collection—Synthetic Turf Fields
       Surface wipe samples were collected at synthetic turf field sites using a wet (water) wipe
(Environmental Express, Ghost Wpe No. 4210) conforming  to American Society for Testing and
Materials (ASTM) E1792-03 requirements. Samples were collected at times when it was safe to
                                         13

-------
do so with regard to any activities occurring on the field. Sample collection time was not critical
for these samples. Two co-located samples were scheduled for collection at the three "on field"
sampling locations. No background sampling location wipe sample was collected. Samples
were collected following the ASTM E1728-03 method, a standard wet-wipe method for collecting
dust from indoor floor surfaces that used water as the wetting agent. Specifically, a 1-ft2
template was placed on the surface of the field. Using clean, powderless plastic gloves, the field
sampling technician removed the wet wipe from the foil packet.  Using one side of the wipe, the
turf surface was wiped in a U-shaped pattern within the template area. After folding the wipe in
half to get a fresh wipe surface, the area was wiped again in a U-shaped pattern perpendicular
to the first wipe pattern. The wipe was then folded in half again and the edges near the interior
portion of the template were wiped. Finally, the wipe was folded again and placed in a
precleaned 50-mL polyethylene tube (Environmental Express, Disposable Digestion Cup No.
SC475) for storage. The tube was tightly capped and transported at ambient temperature to the
laboratory, where the samples were placed in a freezer at -20 °C.

4.5 Tire Crumb Infill Material Sample Collection—Synthetic Turf Fields
       Tire crumb infill material was collected at the synthetic turf field sites. Samples were
collected from one or more areas primarily based on availability of infill material (small  tire
crumb granules) at the surface of the field. These sampling locations did not necessarily
correspond to the air particle sampling sites. Samples were collected at times when it was safe
to do so with regard to any activities occurring on the field. Sample collection time was not
critical for these samples.  No background  sample was collected. Infill material was scooped into
a precleaned 50-mL polyethylene tube (Environmental Express, Disposable Digestion  Cup No.
SC475) for storage. The tube was tightly capped and transported at ambient temperature to the
laboratory, where the samples were placed in a freezer at -20 °C.

4.6 Blade Material Sample Collection—Synthetic Turf Fields
       An attempt was made to collect samples of the loose "grass blades" at synthetic turf field
sites. No destructive sample collection was allowed, so blades were not cut or harvested from
the turf fields. Where possible, samples  were to be taken for each color of turf blades on the
field. Sampling locations did not necessarily  correspond to the air particle sampling sites.
Samples were collected at times when it was safe to do so with  regard to any activities
occurring on the field. Sample collection time was not critical for these samples. No background
sample was collected. Blades were collected using cleaned plastic forceps and were placed into
a precleaned 50-mL polyethylene tube (Environmental Express, Disposable Digestion  Cup No.
SC475) for storage. The tube was tightly capped and transported at ambient temperature to the
laboratory, where the samples were placed in a freezer at -20 °C.

4.7 Tire Crumb Material Sample Collection—Playgrounds
       Two different approaches were used for sample collection at playgrounds. For the first
approach, sample collection locations were approximately adjacent to the "on playground"
sampling locations. Sample collection time was not critical for these samples. No background
sample was collected. Tire crumb material was intended for collection from an approximate
4" x 4" square, with material collected from the surface down to  ground level at each site.
Material was collected using forceps or another appropriate tool, and crumbs were placed into a
250- or precleaned 500-mL, high-density polyethylene wide-mouth bottle (SciSpec Scientific
Specialties Service, Inc, No. 353008 or No. 353016) for storage. The bottle was tightly capped
and transported at ambient temperature to the laboratory, where the samples were placed in a
freezer at  -20 °C. At the second playground, a simple collection of tire crumb rubber material
was performed. Samples were placed in polyethylene bags and were mailed to the laboratory.
                                          14

-------
Once received there, the samples were placed into high-density polyethylene bottles and stored
in a freezer at -20 °C.

4.8 SEM Sample Preparation and Analysis

4.8.1 Sample Preparation
      Ambient samples: 5 mm x 5 mm sections were cut from each polycarbonate filter using a
stainless steel scalpel. Each section was affixed to a standard 12-mm aluminum specimen stub
using a double-sided, sticky C tab. The samples were then coated with -200 A of C to minimize
sample charging by the electron beam during SEM analysis.
      Source material samples: Individual crumbs from the bulk  material sample, typically 2 to
3 mm in size, were deposited "as is" on a sticky C tab. Source particles closer in size to the
ambient sample were generated by shaving pieces from larger crumbs using a stainless steel
razor blade. Source samples were coated with -200-A film of conductive C to minimize charge
buildup on the sample during SEM analysis.

4.8.2 SEM Sample Analysis
      Samples were analyzed by SEM and Energy-Dispersive X-ray Spectrometry (EDX)
using the Personal SEM (R.J.  Lee Instruments Ltd.) in the NERL Electron Microscopy
Laboratory. Manual SEM/EDX analysis was first conducted on the bulk tire crumb source
samples. Chemistry and morphological features characteristic of the tire crumb material were
identified to help identify tire crumb particles in the ambient samples. Ambient samples were
analyzed by computer-controlled SEM (CCSEM/EDX). Instrument parameters for the CCSEM
analyses included 20-kV accelerating voltage, backscattered electron (BSE) imaging mode,
16-mm working distance, and  zero tilt. The BSE mode yields a more uniform background than
the secondary electron (SE) mode,  necessary for computer-controlled SEM, but at the expense
of some loss in sensitivity for small carbonaceous particles; carbonaceous tire crumb particles
about 1 (im or smaller can be difficult to distinguish from the polycarbonate filter substrate in
CCSEM analyses. Thus, small carbonaceous particles may be underreported in  these analyses.
      The CCSEM analysis was set up to analyze particles with  average diameters between
1 and 20 urn. Few particles >10 urn, however, were observed in any sample. All  particles within
this size range were sized automatically and analyzed by EDX for chemistry. Based on the
analyses of the tire crumb source samples sulfur (S), Zn, and C were identified as possible
indicators of tire crumb material. Rules were developed to optimize the search for tire-crumb-like
particles by extending the X-ray analysis time (10 s) and saving low-resolution images for all
particles containing S, Zn, or C. Images and spectra for these particle types were reviewed
manually offline, and particles were judged subjectively to be either tire-crumb-like, or not tire
crumb material based on the particle morphology and chemistry.
      Only a small fraction of the 6.7 cm2 deposit area of each ambient filter was analyzed by
CCSEM, typically about  1 mm2, to complete each analysis in a reasonable time.  Following
CCSEM analyses, the EDX spectra and images of the particles of interest were reviewed
manually, particles were relocated in the SEM for further examination, and suspected tire crumb
particles were flagged.

4.9 Surface Wipe, Tire Crumb,  and Turf Blade Sample Metals Analysis
      Surface wipe samples, tire infill, tire crumb, and turf blade  samples were received and
prepared for analysis. Detailed sample descriptions were recorded because it was observed
that the blade, infill, and  crumb samples were not homogeneous. The playground tire crumb
sample pieces were quite large and heavier than the normal sample size used for Pb in vitro
                                         15

-------
bioaccessibility extractions (typically 1 g). Five turf blade samples were not processed because
they did not meet the minimum sample size of 0.7 g.
       Sample Selection and Processing. For wipe samples, the entire wipe was used after
removing any obvious turf blades or pieces of the infill material. For the crumb rubber infill
samples from synthetic turf fields,  1.0 g aliquots were weighed out after rotating the field
collection tubes in the x, y, and z axes for 1 min.  For synthetic turf blade samples, the samples
were processed with consideration of blade color. If more than one color of blade was present in
the sample, representative blades of each color were selected for extraction. For the crumb
rubber samples from playgrounds, pieces that appeared representative of the entire sample
were chosen for extraction. In the situation where all pieces were very heavy, the piece closest
to 1 g was used. A decision was made not to cut the samples as that would introduce fresh
unaged surfaces, which potentially could impact the Pb bioaccessibility of the sample. Duplicate
sample aliquots were chosen for extraction and spiking where there were adequate quantities.
       Sample Extraction. The samples were treated as "soil" for the scoping study. Existing
extraction procedures already in place were used. A consecutive extraction approach was taken
because of the small number and amount of samples collected in the field.
       First, the Pb in vitro extraction procedure  EPA 9200.1-86 May 2008 "Standard Operating
Procedure for an In Vitro Bioaccessibility Assay for Lead in Soil" was used. Two  10-mL aliquots
from the 100-mL extract total were removed for analysis and storage. The extracts were
adjusted to 2% nitric acid (HNO3; v/v) prior to analysis.
       To obtain total leachable metals, SW-846 Method EPA 3050B "Acid Digestion of
Sediments, Sludges and Soils" was used next. The method was used as written  with the
following minor modifications.
• After quantitative transfer of the 80 ml_ of Pb in vitro extract and solids from the in vitro
  extraction bottles to 250-mL glass beakers, 5 ml_ of concentrated HNO3 was added to the in
  vitro extracts, and the extracts were reduced in volume to 5 ml_ on  a hot plate.
• A maximum of 2 h of acid refluxing was performed (similar to the hot block option extraction
  time). Neither the tire crumb infill nor tire crumb samples from playgrounds completely
  dissolved.
• Samples were filtered through a Whatman 25-mm GD/X 0.45-um cellulose acetate membrane
  syringe filter, as centrifugation was not adequate to separate the particulates from the solution
  for analysis.
• Final samples extracts for EPA Method 3050B were in 5% HNO3 (v/v).
       Analysis by ICP/MS. A new X-Series II quadrapole ICP/MS was designated as the
preferred instrument despite it still being in "start-up" mode. Therefore, instrument,  software,
and data processing routines were developed and evaluated concurrent with the samples'
analysis. After dilution, the in vitro extracts were 2% HNO3 (v/v). The EPA 3050B extracts were
received as 5% HNO3 (v/v), and all subsequent dilutions were made by weight with 5% HNO3
(v/v).
       Quantitative analysis for total extractable  mass concentration was performed for the
primary metals of interest: Cr, Pb,  and Zn. In addition, the metals aluminum (Al), arsenic (As),
barium (Ba), cadmium (Cd), copper (Cu), iron (Fe), manganese (Mn), and nickel (Ni) were
reported; however, QC assessment was not as extensive for these metals.
       The instrumental details for the Thermo X-Series ICP/MS are shown in Appendix B as
follows: Table B-4 lists the operating parameters, Table B-5 lists masses used and interference
correction information, Table B-6 lists the calibration standards  used, and Table  B-7 lists
method detection limits.
       The sample extract analysis followed procedures outlined in EPA SW-846 Method
6020A (http://www.epa.gov/epawaste/hazard/testmethods/sw846/online/index.htm). Instrument
performance indicators are the QC solutions listed in Appendix B, Table B-8. For samples that
                                          16

-------
needed multiple dilutions to insure that all metals of interest were in the calibration range, the
lowest dilution was used for a metal, where it did not exceed the top standard. Total extractable
concentrations are calculated from the prorated combination of the in vitro extract
concentrations and the EPA 3050B extract concentrations.

4.10 Pb Bioaccessibility Analysis

4.10.1 In Vitro Bioaccessibility Background Information
       Methods for assessing Pb bioavailability in soil include in vivo animal studies, in vitro
(referred to as bioaccessibility) studies, and mineralogical/speciation studies. In vivo studies
quantify the metal present in various tissues and excreta of animals after an animal feeding
bioassay is conducted. In vitro methodologies are physiologically based extraction tests
designed to mimic the human gastrointestinal system. In vitro methods measure the
bioaccessibility (e.g., solubility) of metals during a simulated gastric extraction process.
       Bioavailability for this study is defined in the Guidance for Evaluating the Oral
Bioavailability of Metals in Soils for Use in Human Health Risk Assessment (OSWER 9285.
7-80) as "The fraction of an ingested dose that crosses the gastrointestinal epithelium and
becomes available for distribution to internal target tissues and organs
(http://www.epa.gov/superfund/bioavailability/guidance.htm)." A related term pertaining to
bioavailability assessment is bioaccessibility. Bioaccessibility refers to a measure of the
physiological solubility of the metal at the portal of entry into the body (NRC, 2003). The U.S.
EPA guidance document describes the methodologies for predicting lead bioavailability in soil
using either an in vivo swine bioavailability bioassay or an in vitro bioaccessibility assay (IVBA).
These methods have undergone extensive testing and evaluation, and they "are scientifically
sound and feasible methodologies for predicting bioavailability of lead in soil" (OSWER 9285.
7-77). EPA recently published a standard operating procedure (SOP) for an in vitro
bioaccessibility extraction for Pb that has been validated against the juvenile swine model (EPA
Method 9200.1-86). The in vivo and in vitro methods described are specific to Pb-contaminated
soils and Pb bioavailability. Currently, these  methods have not been validated for testing other
contaminants or media (e.g., tire crumb materials), and these have only been validated by EPA
for Pb in soil.

4.10.2 In Vitro Pb Bioaccessibility Methodology
       As noted above, validated in vitro methods did not exist for tire crumb samples when this
study was conducted. The samples were extracted according to EPA Method 9200.1-86 May
2008 "Standard Operating Procedure for an  In Vitro Bioaccessibility Assay for Lead in Soil." This
SOP defines the proper analytical procedure for the validated in vitro bioaccessibility assay for
Pb in soil (soil which has been homogenized and processed for optimal reproducibility) to
describe the typical working range and limits of the assay and to indicate potential interferences.
Users of this SOP are cautioned that deviations in the assay method may impact the results
(and the validity of the method). Two 10-mL aliquots were removed  from the 100-mL extract for
analysis and storage. Samples were analyzed by ICP/MS following  procedures  outlined in EPA
SW-846 Method 6020A.
       Calculations. The amount of Pb in the in vitro bioaccessibility extraction is calculated by
multiplying the extract concentration by the total volume of the bioaccessible extract, which was
100 ml_. The in vitro percent bioaccessibility values were determined by dividing the amount of
Pb extracted in the in vitro extraction by the total extractable amount of Pb in  the sample and
multiplying by 100.
                                           17

-------
4.11 Meteorological and Activity Information
       Meteorological information, including air and surface temperatures, wind speed, and
wind direction information was collected periodically during each monitoring period when
integrated air sampling was conducted. Air and surface temperatures were always measured at
the time of air VOC sample collection. Meteorological measurements were made using a
handheld Kestral 3000 device. This portable device was used so that multiple measurements
could be made at various sampling locations on and around a site. Surface temperatures were
made by laying the device on the field or playground surface and waiting for the temperature
reading to stabilize. Activities occurring at the synthetic turf field and playground locations, if
any, were noted periodically during each monitoring daytime period.
                                          18

-------
                     5. Quality Control and Quality Assurance

       Appendix B includes the data resulting from the QC/QA procedures implemented in this
study and as outlined in the approved QA project plan developed for this study (U.S. EPA
2008).  The following narrative summarizes these results.

5.1 Air VOC Quality Control
       Air VOC QC results are shown in Tables B-1, B-2, and B-3. Potential background
contamination in the sampling and analysis procedure was assessed using field and laboratory
blanks  (Table B-1). Field blanks were 6-L canisters filled with clean air at the laboratory and
transported and stored with the samples collected at three sampling locations. Potential
background in the laboratory was assessed using laboratory blanks that were  clean air
delivered to the analytical system. A laboratory blank was analyzed with each  of the five sets of
samples. Except for methyl ethyl ketone (MEK), no analytes were detected in any laboratory
blank. MEK was measured at levels below the method detection limit in two of the five
laboratory blanks. MEK was the only analyte detected in two of the three field  blanks, at levels
that were about 5 to 10 times lower than the concentrations measured in air samples. The third
field blank contained numerous target analytes, often at concentrations exceeding those
measured in the samples. For example, the concentration of benzene was 1.1 ppbV, and the
concentration of toluene was 29 ppbV. It is possible that this canister had a leak and was
contaminated during the storage and transport process. The pressure of each evacuated
canister used for sample collection was measured prior to collecting the sample to ensure that
the canister did not leak. Two canisters intended for sampling were found to be at ambient
pressure prior to use and were not used for sample collection. It is possible that this field blank
canister also had a leak, but, because it was filled to ambient pressure at the laboratory, it was
not possible to directly assess this prior to field deployment.
       Recovery of target analytes (Table B-2) was assessed through the analysis of field and
laboratory controls. Field controls were 6-L canisters filled with air fortified with a subset of 30
target analytes. These field controls were transported and stored with the samples collected at
three sampling locations. The same mixture of analytes also was analyzed as  three laboratory
controls to assess recoveries at the analytical step. Except for 1,1,1-trichloroethane and
1,2,4-trichlorobenzene, the mean recoveries of target analytes in field and laboratory controls
were within the range of 84% to 114%. The mean recoveries for 1,1,1-trichloroethane and
1,2,4-trichlorobenzene in field controls were 125% and 62%, respectively.
       Sampling and analysis precision was assessed using duplicate sample collection and
repeat  analysis of samples in the laboratory (Table B-3). Duplicate samples were collected at
the same site and within a minute of the collection time of the primary sample  at three sampling
locations to assess the precision of the sampling and analysis procedures. Air collected in
sample canisters at three sampling locations was analyzed a second time to assess laboratory
analysis precision. Laboratory precision was assessed for 11 analytes with sufficient
measureable results.  Mean relative percent differences (RPDs) ranged from 2.9% to 15.6% for
repeat  analyses. Field sampling and analysis precision was assessed for  12 analytes with
measurable results. For eight of those analytes, mean RPD values ranged from  1.8 to 21.1%.
The mean RPD values were higher for four analytes, including hexane (59.7%),  meta- and
para-(m&p)-xylenes (38.4%), toluene (85.8%), and methylene chloride (32.4%).  Two of the
duplicate samples had concentrations of hexane, toluene, and m&p-xylenes that were much
lower than their corresponding samples. As discussed below, these two field samples also had
higher  concentrations of these three analytes than the other samples collected at different sites
at the same sampling locations. It is possible that these two field samples were contaminated,
resulting in the poor measurement precision.
                                          19

-------
5.2 Air PM10 Mass Quality Control
       A single certified flow calibration device (Bios) was used to document the set point and
final flow rate of each sampler deployed in this effort. Flow rates, start and stop times, and
sample identification notations were logged into predesignated data collection sheets employed
for the field effort. A two-party check system was used during the field sampling effort to ensure
that proper recovery of this data was performed, and that sample identification was correct.
Two-party checks involved the primary researcher making the initial notation with the reviewer
(second party) checking the notation independently for correctness. In addition to the data
review conducted in the field, duplicate (collocated) samples were collected during each
monitoring episode. This involved setting up a second monitoring station beside the original unit
at one of the designated site sampling locations (A,B,C,D). The duplicate sampler was run
under the same conditions as the primary unit, and its data were used to determine the
precision of the employed methodology. Likewise, field and  laboratory blank samples were
utilized. Laboratory blanks (filter samples from the same single lot of filters used in the study)
were maintained in the gravimetric laboratory to determine the amount of filter mass  changes
under control conditions. Similarly, field blank samples were deployed in the field at all sites.
These samples were transported to the field, placed in the sampling apparatus but did not have
any volume of air pulled through them. Filters treated in such a manner would represent the
"artifact" mass associated with the sampling effort itself. Following the review of the data
validation component of this study, it was determined that both the laboratory  blanks and the
field blanks had consistently insignificant quantities of mass loadings (<2 ug/m3), and, therefore,
no blank correction was performed on the sample data. Results from comparisons of the field
duplicate samples indicate that precision errors <10% or mass concentration differences  of ~1
to 2 ug/m3 were observed between replicates across all sites. All of these are  highly acceptable
values relative to normal data quality indicators for field monitoring efforts.

5.3 Air PM10 Metals Quality Control
       The NERL XRF laboratory employs a sophisticated QA/QC review of instrumentation
during all analyses. For example, the unit associated with this analysis underwent audit trials
during the sample analysis runs.  Such audits, using National Bureau of Standards or other
reference materials, provide the means to determine the accuracy of the current instrument
calibration, as well as other parameters. The instrument has to have both precision (±5%) and
accuracy (±10%) values from such trials to be considered operational. All such parameters were
achieved for the reported analysis results.

5.4 Air PMio SEM Quality Control
       As previously described for the PM10 gravimetric sample collections, field checks were
conducted pre- and postsampling relative to flow rates.  Duplicate samples were collected at
every regional site to assess precision. The two-party review system again was used to ensure
proper documentation of field data collections. Laboratory and field blanks were obtained and
used to assess data quality (reported in Appendix D. These samples indicate that no sample
handling or storage issues impacted SEM data quality.

5.5 Wipe, Tire Crumb, and Turf Blade Sample Quality Control and Quality
Assurance

5.5.1 Instrument Performance
       Data summarizing the ICP/MS operating parameters, calibration, and method detection
limits are provided in Tables B-4 through B-7. Table B-8 provides the ICP/MS  criteria for
acceptable data. All ICP/MS instrument QCs met the criteria shown in Table B-8 with the
                                          20

-------
exception of one serial dilution result for Zn (15%) and one serial dilution result for Al (26%) that
exceeded the 10% target value. Additional data regarding the results of instrument duplicate
aliquot analysis, postdigestion spike recoveries, and agreement between serial dilutions are
provided in the sections below.

5.5.2 Recoveries
       Extraction performance indicators consisted of recoveries from solution spikes and
spikes of sample media (Tables B-9, B-10, and B-11 and National Institute of Standards and
Technology (NIST) Standard Reference Material (SRM) 2710 spikes (Tables B-12 and B-13).
For the eight spikes prepared in the extraction reagent blank (Table B-10), recoveries for Cr, Pb,
and Zn were all above 92%. Results for analyte recovery from solutions spiked onto the media
of interest varied by media (Table B-11). The Ghost wipes were a uniform media. Recoveries for
the three Ghost wipe spiked samples for Cr, Pb, and Zn were greater than 93%.  For the tire
crumb rubber infill and tire crumb material from playgrounds, there was considerable
heterogeneity of the material.  There was a visual difference between the sample aliquot used
for spiking and the aliquot designated as "sample," which was used to correct for the sample's
contribution to the total spiked sample concentration. Recoveries for the four spiked tire crumb
rubber infill samples for the three metals of interest varied from 16% to 553%. For the tire crumb
samples from playground samples, recoveries range from 17% to 255%. It is probable that
these recovery ranges reflect the variability in existing metal content across samples and the
inability to correctly subtract the existing content uniformly without additional sample
homogenization.
       Extraction reagent blanks and Ghost wipe samples were spiked with NIST SRM 2710,
where the values for Pb and Zn are certified, and the Cr concentration is provided as
"information only" in Tables B-12 and B-13, respectively. For the six extraction reagent blanks
spiked with NIST SRM 2710 (Table B-12), the average recoveries for Pb, Zn, and Cr were 87%,
76%, and 37%, respectively. For the three Ghost wipe samples spiked with NIST SRM 2710
(Table B-13),  average recoveries for Pb, Zn, and Cr, were 87%, 79%, and 32%,  respectively.
According to the NIST SRM 2710 certificate addendum and EPA 3050B,the median recoveries
for Pb, Zn, and Cr in 2710 using EPA 3050B are 92%, 85%, and 49%, respectively. Note that
EPA 3050B is a not total digestion technique but designed to dissolve almost all  metals that
could  become "environmentally available." By design, metals bound in silicate structures
normally are not dissolved by the EPA 3050B procedure, as they are not usually mobile in the
environment.

5.5.3 Analysis of Blank Materials
       Nine extraction reagent blanks (identified as "bottle blanks"), one with each batch of
samples, were processed with the samples (Table B-14). The sample data are reagent blank
corrected using the bottle blank for the specific batch. Contamination in bottle blanks used for
correction may cause overcorrection for some samples on some metals. However, the field
samples also may be subject to this apparently random contamination. Table B-14 shows some
situations where some metals, such as Cr, Fe, Mn, and Ni, are quite high in the 3050B bottle
blank data compared to other bottle blank concentrations. The in vitro bottle blanks overall had
very low concentrations but also showed a few high values.
       Five 3050B-only reagent blanks were prepared with processing beginning at the hotplate
step. Data from these samples were not used for sample correction. Results were similar to the
3050B bottle blanks. However, one sample did show very elevated concentrations of Cd, Cr, Fe,
Mn, and Ni.
       Three laboratory and three field blank ghost wipes were analyzed as samples
(Table B-15).  The two sets of wipes show similar concentrations. The 3050B extracts also
                                          21

-------
showed a few very high concentrations. The data from these blanks were not used to correct
the ghost wipe sample data.

5.5.4 Measures of Precision
       Analytical precision was assessed through the repeat analysis of sample extracts. The
RPD for repeat analyses was less than the <20% target for the eleven metals (Table B-9).
       Sampling precision also often is assessed by the collection of duplicate (collocated)
samples. For the surface wipe and tire crumb material collected in this scoping study, this type
of precision assessment also may assess the homogeneity of the sample media. Results for
analysis of duplicate aliquots of tire infill material from a  sample container or from different
pieces of tire crumb collected from a playground are reported  in Table B-16. For some media,
there was a large RPD. Given the good analytical precision, the large RPD for duplicate sample
aliquots or pieces strongly suggests that the materials, as collected and analyzed in this study,
were not homogeneous with regard to the total extractable amount of some metals.

5.6 In Vitro Bioaccessibility Analysis Quality Control  and Quality Assurance
       A series of procedures and limits were followed to ensure the quality of the in vitro
extractions and analyses. These QA/QC protocols are reviewed below, and any occurrence of
nonconformance is noted and addressed. QA/QC for the extraction procedure consisted of a
series of QC samples (controls, control limits, and corrective actions), as listed in Table  B-17.
       All bioaccessibility QC results are summarized in Tables B-18 through B-24) and meet
the criteria shown in Table B-18, with the exception of the RPD values for the tire crumb
duplicates (Tables B-23 and B-24) that is believed to result from sample heterogeneity issues.
This extraction method was designed specifically for soils that have been processed in a
manner used to create homogeneous samples. The RPD between duplicate extractions of
these samples ranged from 2.7% to 124% for the infill samples and 4% to  183% for the  crumb
samples. Duplicate extractions of the wipes and blades were not possible either because there
was a unique sample (wipes) or insufficient sample quantity (blades).
       For the eight Pb spikes prepared in the in vitro extraction solutions (Table B-19),
recoveries for Pb were all above 90%. The six NIST SRM 2710 extractions (Table B-20)
resulted in an average RPD of 4.5% (range 0.4% to 8.9%).  The RPDs for the three Ghost wipes
spiked with NIST SRM 2710 (Table B-21) resulted in an average of 4.7% (range 3.5% to 6.2%).
The NIST SRM RPD values  are based on the mean in vitro bioaccessibility values of 75% for
this SRM (EPA Method 9200.1-86).
       Recoveries for blank Ghost wipes and tire crumb samples spiked with Pb solutions are
listed in Table B-22. The media were extracted without spikes and used to correct for the
sample's contribution to the total spiked samples. Recoveries  for the four spiked infill samples
for Pb ranged from 89% to 104%.  Recoveries for the spiked crumb samples, ranged from 87%
to 103% for Pb, whereas the recoveries for the spiked wipe samples ranged from 87% to 99%.
Analysis of duplicate tire crumb infill and playground tire crumb aliquots (Tables B-23 and B-24,
respectively) likely reflects the significant difference in heterogeneity of these samples.

5.7 Data Quality Assurance Review
       Data generated in NERL/ORD laboratories in this scoping study were reviewed
independently by a trained QA officer. The review included  data generation, calculations, and
transcriptions for a subset of the data. The Region 1 laboratory followed established laboratory
QA/QC procedures in their analysis and review of air VOC results.
                                         22

-------
                                      6. Results
6.1 Sampling Sites
       The full study protocol (collection of all planned air, wipe, and material samples) was
completed at two synthetic turf field sites (including repeated sampling on a second consecutive
day at one site, F1). A reduced set of samples that did not include the integrated air particle
sampling systems was completed at a third synthetic turf field site (F2). In addition, there were
multiple synthetic turf fields at two sites  (F2 and F3 at one  site and F4, F5, and F6 at the other),
and selected wipe and material samples were collected across different fields at these sites.
The full study protocol, including collection of air samples,  was completed at one playground
(P1). However,  because of the size of this playground, only two "on playground" sampling
locations were operated instead of the three planned. Tire  crumb material was obtained from a
second playground, but no other sampling or site characterization was performed at this
playground site. Information about each sampling site is provided in Table 2. The samples
collected at each site are shown in Table 3.

Table 2. Sampling Site Information and Assigned Codes
Site
EPA Region 5, Field 1, Day 1a
EPA Region 5, Field 1 , Day 2
EPA Region 4, Field 2D
EPA Region 4, Field 3
EPA Region 4, Field 4C
EPA Region 4, Field 5
EPA Region 4, Field 6
EPA Region 3
EPA Region 4
Type
Synthetic turf field
Synthetic turf field
Synthetic turf field
Synthetic turf field
Synthetic turf field
Synthetic turf field
Synthetic turf field
Playground
Playground
Assigned Code
F1D1
F1D2
F2
F3
F4
F5
F6
P1
P2
 Samples were collected at one synthetic turf field on 2 consecutive days.
bTwo synthetic turf fields (F2 and F3) were part of the complex at this site.
cThree synthetic turf fields (F4, F5, and F6) were part of the complex at this site.

Table 3. Overview of the Types of  Samples Collected at Each Site
Site
F1, Day 1
F1,Day2
F2
F3
F4
F5
F6
P1
P2
Air
VOC
•
•
•

•


•

Air
PM10
Mass
•
•


•


•

Air
PM10
Metals
•
•


•


•

Air
PM10
SEM
•
•


•


•

Surface
Wipes
•
•
•

•
•
•


Tire
Infill
•

•

•
•
•


Turf
Blades
•

•
•
•
•
•


Tire
Crumb







•
•
                                           23

-------
       Fewer synthetic turf fields and playgrounds were monitored than originally planned, and
at sites near only three of the four NERL facilities. This was primarily because of difficulties in
identifying and arranging access to sites in combination with the logistical difficulties posed by
the extensive array of equipment and skill required to operate the active particle sampling
equipment at multiple sampling locations per site. Sampling at fewer sites had minimal impact
on accomplishing the study goals, as the original design was very limited. Implementing the full
protocol at only two synthetic turf field sites resulted in little impact in the methods evaluation
study. Samples were collected at these two sites during relatively hot, dry days and with high
activity levels (conditions favorable for the evaluation). The consecutive day measures at one
site yielded additional data for assessing the methodology and understanding potential changes
in site conditions from day to day. Sampling at only one playground  site, although one with
relatively high levels of activity, did provide insights regarding the  practical issues regarding
implementing the protocol at playground sites but resulted in limited data for the workgroup.

6.2 Site Characteristics
       Descriptive information about each sampling site, including the age and type of material,
maximum temperatures, wind speed, and activity information is provided in Table 4. Some
anecdotal information may be relevant with regard to  interpreting the study results. Sampling
was conducted at field F1 on 2 consecutive days (F1D1 for day 1  and F1D2 for day 2). New tire
crumb infill material recently had been applied to field F1. During the first day at field F1 (F1D1),
heat thermals were observed coming off of the field during the hottest times of day, and there
was a smell that generally is associated with tires or tire crumbs. There was considerable
activity on this field throughout the day, including multiple physical education classes, as well  as
football and soccer practices. On the second day at field F1 (F1D2), the temperature was
somewhat cooler,  no thermals were observed, a similar smell was noted, and there were lower
activity levels. Air PM sampling equipment was either on or immediately adjacent to field F1
during the activities. Fields F2 and F3 were adjacent fields at the same regional site. There was
no activity at these fields during the  monitoring period; therefore, no air PM samples were
collected. Fields F4, F5, and F6 were collocated at another regional site. Air PM samples were
collected on the sidelines of field F4. There was sporadic moderate  activity on field  F4 and on
the immediately adjacent field (F5) during the monitoring day, including physical education
classes, flag football, and a soccer game. One air PM sampling location was placed between
fields F4 and F5. Two air PM sampling locations were placed on the opposite field F4 sideline
based on wind direction at the beginning of the monitoring. The wind shifted later in the day and
may have transported VOCs and particles to these two sampling locations (particularly to
sampling location A) from the adjacent parking deck and nearby road with moderate commuter
traffic. At playground P1, the playground was used by up to 60 preschool and early elementary
students twice during the school day, and sampling continued for approximately 90 min into
after-school use by approximately 12 to  20 students. The tire crumb material at playground P1,
with embedded and visual fibers, was prepared and provided by a local supplier.
       The air monitoring equipment setup at one sampling location at site F1 is shown in
Figure 1, wipe sampling is shown in Figure 2, the turf blade and tire crumb infill at the surface is
shown in Figure 3, and the multiple colors of turf blade are shown in Figure 4. A laboratory
close-up of a sample vial containing tire crumb infill granules collected at site F2 is shown in
Figure 5. Tire crumb material at site P1 is shown in Figure 6, and a laboratory close-up of this
tire crumb material, with exposed fibers, is shown in Figure 7. A laboratory close-up of the
"wood bark" tire crumb material collected at site P2 is shown in Figure 8.
                                          24

-------
Table 4. Site Information
Site
F1
Day 1
(F1D1)
F1
Day 2
(F1D2)
F2
F3
F4
F5
F6
P1
P2
Surface
Age
2 years0
2 years0
4 years
5 years
4 years
3 years
2 years
4 years
Not
known
Type of Surface Material
Polyethylene turf blades0; green
field, red end zones, black and
white lines; with granular tire
crumb rubber infill
Polyethylene turf blades0; green
field, red end zones, black and
white lines; with granular tire
crumb rubber infill
Polyethylene turf blades0; green
field, red center circle, white
lines; with tire crumb rubber
infill
Polyethylene turf blades0; green
field, red center circle, white
lines; with tire crumb rubber
infill
Polyethylene turf blades0; green
field; yellow and white lines;
with tire crumb rubber infill
Polyethylene turf blades0; green
field, yellow and white lines;
patched area appeared to be
green nylon turf blades0; with
tire crumb rubber infill
Polyethylene turf blades0; green
field, yellow and white lines;
with tire crumb rubber infill
Shredded tire material; black
color; much of the tire crumb
had fiber material still included.
Tire crumb material processed
and formed to simulated bark
appearance; multiple green and
brown colors.
Temperature at
Time of AirVOC
Sample Collection3
32 °C Air above field
50 °C Field surface
Wind 2-11 mph
35 °C Air above field
46 °C Field surface
Wind 1-6 mph
28 °C Air above field
44 °C Field surface
Wind 1-11 mph
a
30 °C Air above field
44 °C Field surface
Wind calm (2 mph)


30 °C Air above
playground
36 °C Playground
surface
Wind calm (1 mph)

General Activity
Levels on
Monitoring Day
Est. number: 20-70 at
a time
Est. duration:
45-120 min at a time
Est. number: 20-70 at
a time
Est. duration:
30-45 min at a time
Number: 0
Duration: 0

Est. number: 10-35 at
a time
Est. duration:
45-120 min at a time


Est. number: 12-60 at
a time
Est. duration:
30-90 min at a time

aVOC air samples collected at hottest time of day (~2 p.m.).
bAdditional new tire infill applied during summer prior to sampling.
cType of turf blade material based on visual assessment (not confirmed through material analysis).
dNot measured or not monitored.
6.3 Sample Collection
The numbers and types of samples collected at each site are shown in Table 5. Air samples for
VOCs were collected in evacuated 6-L Summa-polished stainless steel canisters at a 1-m
sampling height. Four air VOC samples were collected at each of three synthetic turf field sites
                                            25

-------
•
  •
                                                                          ••?
                                                                                .  --
Figures 1 through 4. Site F1 particle air samplers (top), surface wipe sample collection (left), green
turf blade with black granular tire crumb (middle right), and multiple turf blade colors (lower right).

(F1, F3, and F4) with tire rubber infill material. Three air samples were collected at one
playground site (P1) with tire crumb rubber material. Samples were collected at three different
sampling locations (designated as A, B, and C, respectively) directly above each of the synthetic
turf fields and at two different sites (A and B) directly above the playground. A background
sample (designated as D) also was collected at each site a short distance upwind from the field
                                            26

-------
Figures 5 through 8. Tire crumb infill granules from site F2 (top left), shredded tire crumb at site
P1 (top and bottom  right), and tire crumb material from site P2 (bottom left).

or playground. With one exception, all air samples were collected at each site on 1 day and at
approximately the same time of day. The exception was the collection of a set of four air
samples on 2 consecutive days at one synthetic turf field. All grab air VOC samples were
collected during the hottest time of day (~2 p.m.). In general, these grab air VOC samples were
simple to collect, required little onsite collection time, and modest technical  expertise. We  found
it was very important to verify the canister pressure prior to sampling to ensure that the
evacuated canisters had not leaked prior to sample collection.
       Integrated air PM10 samples for mass and metals  concentration measurement, as  well
as separate integrated air samples for SEM analysis, were collected at a 1-m sampling height at
the four VOC sampling locations at synthetic turf field F1 on 2 consecutive days, and on 1 day
at field F4. Air PM10 samples were collected at the three VOC sampling locations on 1 day at
playground P1. The limited sampling performed at playground P1 was based on the small space
that did not allow the full complement of the normal sampling routine to be performed. In all
events, active sampling locations were established quickly on  the site, as well as from a
background sampling location. Air monitoring was initiated for all monitors in quick order on their
setup and calibration and continued without interruption through the monitoring event (daytime).
This resulted  in sample collections ranging from approximately 5.8 to 7.8 h  and corresponding
individual air volumes ranging from approximately 7.0 to 9.2 m3 over the course of a sampling
day. Collection of up to 10 air particle samples at each site required considerable equipment
(enough to fill a van), considerable setup and retrieval time (approximately  1 h each), extensive
staff time onsite (typically 8 to 10 h for two  people), and moderate technical monitoring
expertise. The monitoring approach, equipment, and sampling durations were selected to obtain
                                          27

-------
Table 5. Number of Samples Collected at Each Site
Site3
Onsite
Samples
Background
Samples
Duplicate
Samples
Archival
Samples
Field
Blanks
Field
Controls
AirVOC
F1D1
F1D2
F2
F4
P1
3
3
3
3
2
1
1
1
1
1
1
—
—
1
1
—
—
—
—
—
1
—
—
1
1
1
—
—
1
1
Air PM10 for Mass and Metals
F1D1
F1D2
F4
P1
3
3
3
2
1
1
1
1
1
1
1
1
—
—
—
—
1
1
1
1
—
—
—
—
AirPM10forSEM
F1D1
F1D2
F4
P1
3
3
3
2
1
1
1
1
1
1
1
1
—
—
—
—
1
1
1
1
—
—
—
—
Surface Wipes
F1D1
F1D2
F2
F4
F5
F6
3
3
3
1
1
1
—
—
—
—
—
—
3
3
3
1
—
—
3
3
—
1
1
1
1
—
1
1
—
—
—
—
—
—
—
—
Tire Infill
F1D1
F2
F4
F5
F6
3
3
1
1
1
—
—
—
—
—
—
3
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Turf Blades
F1D1
F1D2
F2
F3
F4
F5
F6
3
1
3
4
1
1
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Tire Crumb
P1
P2
2
1
—
—
1
—
2
—
—
—
—
—
 F = synthetic turf field, D = day 1 or day 2, P = playground.

adequate limits of detection for PM10 mass and a range of metal analytes to ensure that levels
typically found in ambient air would be measurable.
       Surface wipe samples were collected at three "on field" sampling locations at synthetic
turf fields F1 (on 2 consecutive days) and F2. At a third site, a single surface wipe sample was
                                          28

-------
collected from each of three fields (F4, F5, and F6), as the three fields making up this complex
were of different ages. A standard wet-wipe method (ASTM  E1728-03) that is used to measure
residential surface dust Pb levels was employed for sample  collection in this study. In the
absence of a validated wipe method for synthetic turf surfaces, this method was selected for
evaluation in this study. In general, the advantages of this method were the availability of
standard wipe material and existing analytical methodology. In practice, samples could be
collected by those with moderate technical expertise at multiple sampling locations in a
relatively short period of time. However, the sampling procedure was not as simple as for
smooth floor materials, with the synthetic turf blades requiring additional patience and control.
The relationship between the dust collected on the wipe sample  that comes from turf blades
versus the dust from the infill and is available for human contact is not clear and may need
further investigation. Most wipes also picked up a  few pieces of the rubber infill material and turf
blades. A decision was made to remove these relatively large discrete materials in the
laboratory prior to extraction as these larger materials would be characterized as part of the
additional material samples. It is not expected that removal of these large materials would
impact the measurement of the small dust particles that the  surface wipe sample is designed to
collect.
       Tire crumb rubber infill used in these synthetic turf field installations was collected from
three sampling locations at fields F1 and F2  and at single sampling locations from each of three
fields (F4, F5, and F6) at the third site. Sample collection could be completed in a short time by
persons with minimal technical expertise. In this study, it was decided to collect infill material
that was already available at the surface rather than by dislodging material  trapped deep within
the turf blades. This decision was based partly on avoiding potential damage to field
components but primarily because the material on the surface was more available for potential
human contact. However, infill material was not available uniformly across the field surface.
       Synthetic turf blade samples were collected at all field sites in this study. The synthetic
turf blades were not a primary interest in this study, as the characterization  of this type of
material is being performed by other organizations. However, samples of turf blades were
collected to enable an improved understanding of the surface wipe measurements with regard
to the potential differential contributions from the infill and blades. A decision was made not to
perform destructive collection (i.e., there was no cutting of material from the fields). Instead, the
collection relied on the availability of loose blades  found at the surface of the fields. Collection
and analysis decisions were complicated by the limited availability of loose  blades and a later
decision that a minimum of 0.7 g of material was required for analysis. Collection of blades of
each color type was attempted. For fields F1, F2, and F3, the colors were collected separately
and kept separate for analysis. None of the green blade samples from the site complex
comprised of fields F2 and F3 achieved the "postsampling" requisite 0.7-g sample size, and
they, therefore,  were not analyzed. In retrospect, a decision to combine green blades from
several different sampling locations would have enabled the analysis of a composite site
sample. For fields F4,  F5, and F6 the different colors  for each field were mixed together in the
sample to best achieve adequate sample sizes, while, at the same time, obtaining samples from
fields of different ages.
       Tire crumb rubber samples were obtained  from two playground sites. The material used
at the P1 playground  consisted of shredded tire particles containing fibers, whereas the material
from playground P2 was processed and colored to simulate tree bark. Collection of tire crumb
material is a simple process.  However, the material is relatively heterogeneous and it is not
clear how many pieces or which pieces need to be collected for site characterization. It is also
not clear whether  it is most appropriate to collect tire crumb  pieces at the surface that may  have
experienced different weathering and contact than underlying  pieces. A further challenge is that
relatively small amounts (1 g or less) are required for analysis because larger amounts may
overwhelm digestion and analytical systems. Intact tire crumb rubber pieces are generally larger
                                           29

-------
than 1 g. A decision was made not to cut the tire crumb pieces because exposing unweathered
surfaces might impact the Pb bioaccessibility measurement.

6.4 Summary Measurement Results
      Tables 6 through 8 provide key data summaries for the synthetic turf field and
playground sites. All of the individual site sample data are provided in Appendix C.

6.5 Air  VOC Measurement Results (Appendix C, Table C-1)
      Key highlights from the grab air VOC measurements are summarized below.
• Twelve of the 56 target VOC analytes were present in multiple samples. Thirty-seven of the
  56 analytes were not detected in any sample. Seven of the remaining 19 analytes had
  detectable levels in only one or two  samples.
• Measured VOC concentrations were generally low. No analyte concentration exceeded
  1 ppbV in any sample. Detection limits ranged from 0.070 to 0.079 ppbV for most analytes,
  and 0.14 to 0.16 ppbV for 1,3-butadiene and m&p-xylenes.
• MIBK  previously has been identified as a tire-related VOC. MIBK was present in three "on
  field" samples collected at one synthetic turf field (F1) on the first monitoring day when heat
  thermals were observed and in two  "on field" samples on the second consecutive day of
  monitoring. MIBK concentrations were low (<0.2 ppbV)  in all five samples. MIBK was not
  detected in the background samples collected near the field on either day. MIBK was not
  detected at any other site in this study.
• Concentrations of the other VOCs routinely measured over the field or playground sampling
  locations were similar to the concentrations measured in corresponding  upwind background
  samples collected nearby, or they likely could be explained by documented local sources near
  the site.
• MEK was measured in all the study samples, with the "on field" MEK levels being similar to
  levels  in the upwind background samples at each site.
• Hexane was present in most of the "on field" and background  samples. At the F4 synthetic
  turf field, the hexane concentrations at all three "on field" sites were higher than the
  background site, but all concentrations were  low (<0.2 ppbV).
• The aromatic analytes benzene, toluene, and m&p-xylenes are ubiquitous atmospheric
  pollutants and were present at measureable levels in most samples collected in this study.
• Benzene concentrations were similar for the background and "on field" samples at all sites.
  The highest benzene concentration  (0.32 ppbV) was measured at location A on site F4. This
  concentration was  higher than at sites B or C or the background  site D. Other aromatic VOC
  concentrations also were elevated somewhat at this sampling location and site. Sampling
  location A was closest to a parking garage exit and may have been impacted by traffic and
  vehicle exhaust.
• Toluene, m&p-xylenes, and hexane concentrations at site P1 location B, and at site F1
  location B, were higher than concentrations at the other sampling locations for these sites.
  The reason for this is not clear; however, elevated levels (>4 ppbV) of these compounds were
  measured in one field blank, and contamination of the sampling canisters cannot be ruled out.
  Both of these samples had a corresponding duplicate sample collected at location B.
  Concentrations of these  three analytes were present in the duplicate samples at ratios
  ranging from 0.15 to 0.63 of the concentrations in the samples. In fact, the concentrations
  measured in the duplicate samples were similar to the concentrations measured in the other
  samples at each site, further suggesting that these two  samples  with slightly higher
  concentrations may have been contaminated.
• Several halogenated VOCs were measurable in all or most of the samples. These included
  carbon tetrachloride, methylchloride, methylene  chloride, dichlorodifluoromethane (Freon 12),
                                         30

-------
Table 6. Summary Results for Selected Analytes in Air Samples Collected at Synthetic Turf Fields and a Playground



Air Samples
Air PM Mass
Air PM Metals


AirVOCs














Analyte
Particle Mass
Pb
Cr
Zn
Methyl Isobutyl Ketone
Benzene
Toluene
m&p-Xylenes
Hexane
Methyl Ethyl Ketone
Carbon Tetrachloride
Dichlorodifluoromethane
Methylchloride
Trichloro-fluoromethane
Trichloro-trifluoroethane
Methylene Chloride



Unit
pg/m3
ng/m3
ng/m3
ng/m3
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
ppbV
Synthetic Turf Fields
F1D1
On Back-
Field3 grnd.b
27.8 29.5
NDC ND
2.9 2.0
10.8 23.8
0.13 ND
0.09 0.07
0.42 0.15
0.17 0.08
0.21 0.06
0.47 0.44
0.09 0.10
0.52 0.55
0.47 0.48
0.26 0.28
0.08 0.08
0.07 0.06
F1D2
On Back-
Field3 grnd.b
29.8 29.5
7.7d 6.3
3.6 3.3
11.8 11.6
0.12 ND
0.08 0.09
0.11 0.12
0.10 ND
0.08 0.08
0.38 0.36
0.10 0.10
0.50 0.56
0.47 0.46
0.26 0.27
0.08 0.08
ND ND
F2
On Back-
Field3 grnd.b
— —
— —
— —
— —
ND ND
0.11 0.12
0.18 0.19
0.07 0.08
0.08 0.05
0.41 0.37
0.09 0.08
0.56 0.51
0.45 0.45
0.27 0.25
0.08 0.07
0.06 0.06
F4
On Back-
Field3 grnd.b
31.8 28.6
ND ND
ND ND
31.4 21.7
ND ND
0.20 0.12
0.28 0.19
0.14 ND
0.14 0.05
0.43 0.44
0.09 0.10
0.48 0.54
0.48 0.52
0.24 0.30
0.07 0.15
0.06 0.06
Playground
P1
On Back-
Play.3 grnd.b
26.7 14.2
5.1d ND
3.4 ND
104 10.5
ND ND
0.09 0.09
0.29 0.16
0.13 0.05
0.30 ND
0.41 0.38
0.08 0.09
0.53 0.54
0.46 0.45
0.26 0.26
0.07 0.07
0.09 0.07
 Average of two or three "on field" or "on playground" measurements (any nondetect values were not included in the average).
bSingle measurement from upwind background location.
cNot detected.
dPb was measured in only one of three "on field" samples at F1D2 and one of two "on playground" samples at PL

-------
Table 7. Summary Results for Total Extractable Pb, Cr, and Zn in Samples Collected at Synthetic Turf Fields and
Playgrounds


Sample
Turf Field Surface Wipe Samples


Turf Field Infill Crumb Rubber


Turf Field Blades


Playground Tire Crumb




Analyte
Pb
Cr
Zn
Pb
Cr
Zn
Pb
Cr
Zn
Pb
Cr
Zn


Unit
pg/ft2
pg/ft2
pg/ft2
pg/g
pg/g
pg/g
pg/g
pg/g
pg/g
pg/g
pg/g
pg/g
Synthetic Turf Fields
F1D1
Range
0.3-1.9
0.1-0.5
21.3-43.3
13.1-34.7
0.3-1.0
5,050-19,200
2.8-389c
1.0-73.1
316-730
NC
NC
NC
F1D2
Range
0.4-1.4
0.1-0.3
26.4-40.6
NC
NC
NC
NC
NC
NC
NC
NC
NC
F2, F3
Range
0.3-0.5
ND.b
9.2-19.3
20.6-61.2
0.4-0.9
3,120-12,300
2.4-2. 8d
1.2-1.9
199-255
NC
NC
NC
F4, F5, F6
Range
0.1-0.2
ND-0.3
4.3-13.6
10.7-47.7
0.3-1.0
2,660-11,400
2.1-7018
3.7-177
131-206
NC
NC
NC
Playgrounds
P1
Range
NCa
NC
NC
NC
NC
NC
NC
NC
NC
1.0-443f
0.3-1.7
4,300-17,500
P2
Range
NC
NC
NC
NC
NC
NC
NC
NC
NC
3.4-7.8g
1.6-3.0
12,100-18,000
aNot collected.
bNot detected.
cDifferences noted for different blade colors (red = 389 pg/g; white, green, and black all <4.3 pg/g).
dAnalysis of red and white blades.
eHighest level (701  pg/g) found in a field with a repaired area; levels in blades from two adjacent fields ranged from 2.0 to 77 pg/g.
'Analysis of seven pieces of tire crumb; six of those pieces had Pb levels <50 pg/g.
gAnalysis of two pieces of tire crumb.

-------
Table 8. Summary Results for Estimates of Pb Bioaccessibility in Samples Collected at Synthetic Turf Fields and
Playgrounds3
Sample
Turf Field Infill Crumb Rubber
Turf Field Blades
Playground Tire Crumb
Analyte
Pb
Pb
Pb
Synthetic Turf Fields
F1D1
Range
1 .6%-9.6%
2.3%-86.8%c
NC
F1D2
Range
NCD
NC
NC
F2, F3
Range
1 .7%-7.6%
38.7%-40.3%d
NC
F4, F5, F6
Range
1.7%-10.1%
0.2%-54.4%e
NC
Playgrounds
P1
Range
NC
NC
0.3%-10.7%f
P2
Range
NC
NC
1 .8%-7.4%
 The in vitro bioaccessibility method has not been validated for these materials.
bNot collected.
Differences noted for different blade colors (red = 2.3%, white and green = 40.9% to 43.0%, and black = 86.8%); also, the lowest bioaccessibility value (2.3%)
corresponds to the sample with the highest total extractable Pb (389 ug/g).
dAnalysis of red and white blades.
el_owest bioaccessibility value (0.2%) corresponds to the sample with the highest total extractable Pb (701 ug/g).
'Lowest bioaccessibility value (0.3%) corresponds to the sample with the highest total extractable Pb (443 ug/g)

-------
  trichlorofluoromethane (Freon 11), and trichlorotrifluoroethane (Freon 113). In all cases, the
  levels in the "on field" samples were similar to levels in the upwind background sample at
  each site.

6.6 Air PMio Mass Measurement Results (Appendix C, Table C-2)
       PM10 concentrations observed across all of the sites ranged from approximately 24 to
33 ug/m3. PM10 mass concentrations collected on or adjacent to synthetic turf fields were
generally equivalent to concentrations in ambient air measured at the upwind background sites.
Mass concentrations across a given synthetic turf field or playground were often consistent
within themselves. That is, PM10 concentrations from field sampling locations A, B, and C at a
given site often varied by only 2 to 3 ug/m3, which was within the precision error typically
observed for the duplicates.  This mass consistency was generally true regardless of the range
of activities taking place on the field and the proximity of such activities to a given monitor. Such
a statement, however, cannot be made for the one playground site monitored. Comparison of
data from the P1  site indicates an approximately 15 ug/m3 higher PM10 mass concentration was
obtained from the monitor located near the highest density of playground activity (location B).

6.7 Air PMio Metal Measurement  Results (Appendix C, Table C-3)
Air PM10 sample filters were analyzed  for 44 metals. As part of the analysis, the statistical
uncertainty of the measurement was determined; the measured metal concentrations must be at
least three times the uncertainty concentration to be considered a measured result. Based on
this assessment,  the full list of 44 metals was reduced to 12 that are reported in Appendix C-3,
including the primary analytes Pb, Cr, and Zn (see Table 6), as well as 8 other elements (Ca
[calcium], Cl [chlorine], Cu, Fe, K [potassium], Mn,  S, Si [silicon], and Ti [titanium]) with sufficient
number of measurable results for assessment within and across sampling sites.
       Measurement results from synthetic turf field F1 show that upwind background levels of
Si, S, Cl, K, Ca, Ti, Cr, Mn, Fe, Cu,  Zn, and Pb were similar to the "on field" concentration
measurements at thee sampling locations (A, B, and C). This would indicate  that the infill tire
crumb rubber and other materials on the turf field examined in F1 (on both day  1 and day 2)
provided little contribution to the measured metal concentrations associated with airborne
aerosols. Slightly higher concentrations of Ti (-36%), Cr (-90%), Mn (-100%),  Fe (-74%), Cu,
(-57%), and Zn (-65%) were measured at the "on field" monitors relative to the upwind
background sampling location at the F4 synthetic turf field. This site was bordered by both a
busy urban commuter road,  as well as a parking deck. Because of this, the additional
contribution of some of the metals to the collected "on field" air samples, notably those that often
are observed in near-roadway air samples (Fe,  Zn, and Cu)  might not be singularly reflective of
contributions from a single source (i.e., this might be indicative of near-road influence, as well as
of any contribution from any resuspended tire crumb rubber infill aerosol).
       It would appear that the "on playground" samples associated with the P1 site had
consistently higher levels of the 12  selected metals discussed above, as compared with the
background site.  For example, "on playground" aerosol metal concentrations for Si, Cl, K, and
Ca were sometimes 50% to  700% higher, as compared with the background. These metals
often are associated with crustal (soil)  related sources. "On playground" samples had much
higher levels of metals that might be of relevance to tire crumb rubber components. For
example, Ti concentrations were more than four times higher for one "on playground" sampling
location, with nonspeciated Cr concentrations (-3 ng/m3) substantially higher than the
background monitor.
Mn  concentrations were -7 times higher (15 ng/m3), with Fe as much as 4 times more
concentrated (-1,000 ng/m3). Of particular interest are Zn levels, which were 8  to 11 times
higher than background levels for the two "on playground" monitors (82 to 117 ng/m3). On the
                                          34

-------
other hand, these metals also may be found in soils, so the relative contribution of tire crumb
particles to the increased levels could not be determined readily from these results.
       Pb concentrations were at measurable levels in only 3 of the 18 air PM samples
collected in this study. These three samples had concentrations (<7.7 ng/m3) that were near the
method detection limit. The maximum concentration of 7.7 ng/m3 was measured in a sample
collected at a synthetic turf field and,  given the precision of the method, is considered
indistinguishable from the corresponding background level (6.3 ng/m3). A Pb level of 5.1 ng/m3
was measured at one "on playground" sampling location; again, a level near the method
detection limit. The samples for another "on playground" sample, its duplicate sample, and the
background sampling location had Pb values below the method detection limit.

6.8 Air PMio SEM Measurement  Results (Appendix C, Table C-4)
       Air PM10 samples were collected on filters at three sites with tire crumb material for SEM
analysis.  Air samples collected at two "on field" or "on  playground" sampling locations and at
one upwind background sampling location were analyzed for synthetic turf fields F1 and F4 and
playground P1. In preparation for analyzing the air samples, samples of the tire infill material
from a  field and the crumb material from a playground were analyzed to determine whether
metal or morphological "source profiles" could be identified that would assist in identifying tire
crumb-related particles in the air samples. A detailed report describing the  SEM analysis
procedure and measurement results is provided in Appendix D. Key findings from the SEM
analysis are described below.
• Prior to analyzing the air filter samples,  particles were generated from the tire crumb materials
  collected at the turf fields and playground to try to identify a signature morphology and metals
  composition for tire crumb particles. These particles did not show a  unique, easily identifiable
  X-ray spectrum for metals composition or supporting a definitive source attribution analysis.
  C and S were consistently present  in the tire crumb  particles. Zn usually but not always was
  observed in tire crumb particles, often at trace levels. Tire crumb particles from the source
  material varied considerably in morphology, making it difficult to identify a typical or
  characteristic tire crumb morphology.
• Very few fibers were observed in any of these air samples, and none could be attributed to
  tire crumb. This was true even for the playground site (P1) air samples, which had tire crumbs
  with  exposed and embedded fibrous material.
• The ability to quantify the tire crumb concentration in these samples hinges on the tire crumb
  particles having a unique composition or morphology that would enable the analyst to identify
  tire crumb particles with a high degree of confidence. This does not seem to be the case for
  tire crumb particles collected in this study, as seen in the variety of morphologies and
  compositions on air filters.
• At the two synthetic turf fields, mass concentrations  for postulated tire crumb particles were
  estimated to be only a very small fraction of the total PM10 mass concentrations measured at
  these sites. At  the playground site,  estimated mass concentrations for postulated tire crumb
  particles were a relatively small fraction of the total PM10 mass concentrations, but a higher
  fraction than was measured at the synthetic turf fields. However, the variability in tire crumb
  particle composition and morphologies introduces large uncertainties in these results.

6.9 Total Extractable Metals in Synthetic Turf Field Surface Wipe, Tire Crumb
Infill, and Turf  Blade Samples and Playground Tire Crumb Rubber Samples
(Appendix C, Tables C-5 through C-8)
       The total extractable measurement results for 11 metals in wipe, tire crumb, and
synthetic turf blades collected in the scoping study are shown in Appendix  C. Primary target
metals were Pb, Cr, and Zn (see Table C-5), and metals of secondary interest included Al, As,
                                          35

-------
Ba, Cd, Cu, Fe, Mn, and Ni. Results from surface wipe samples collected at synthetic turf fields
are shown in Table C-5, those for tire crumb infill from synthetic turf fields are shown in
Table C-6, and those for synthetic field turf blades are shown in Table C-7. Results for tire
crumb material at playgrounds are shown in Table C-8.

6.9.1 Surface Wipes from Synthetic Turf Fields (Table C-5)
       Surface wipe samples were collected at different sampling locations from several
synthetic turf fields. The sampling locations were in close proximity to the air monitoring "on
field" locations. In some cases, duplicate samples were collected side-by-side. Some wipe
samples were collected from different color turf blade areas at one site (F1) where there were
large areas of different color turf blades. A wipe sample also was collected at the field (F5) with
visually different turf materials. Total extractable metal measurement result highlights include
those that follow.
• Total extractable Pb was less than 2.0 ug/ft2 in all surface wipe samples collected at the
  synthetic turf fields in this study. Most results were less than  1.0 ug/ft2.
• Many of the sample analysis results were similar to levels measured in field blanks, which had
  Pb values ranging from 0.14 to 0.54 ug/ft2.
• The  highest total extractable  Pb value (1.9 ug/ft2) was measured on a surface wipe collected
  at site F1 on red synthetic turf blades.
• Surface wipe samples collected side-by-side, in some cases, had similar total extractable Pb
  levels; in other cases, the differences in side-by-side measurements ranged up to
  approximately twofold. Some of the side-by-side measures taken at one turf field site (F1)
  were from an area of mixed turf blade colors.
• Surface wipe samples collected at different locations on synthetic turf fields had up to sixfold
  differences in total extractable Pb concentrations. The greatest differences  appeared to be
  associated with wipes taken from areas with different blade colors.
• Total extractable Cr was <0.6 ug/ft2 in all surface wipe samples.
• Total extractable Zn in surface wipe samples ranged from 4.0 to 43 ug/ft2.
• Measurements of As were very low (<0.1 ug/ft2) and similar to amounts found on field blanks.
• Most Cd measurements were less than the method detection limit; the remainder were
  <0.025 ug/ft2.

6.9.2 Tire Crumb Infill at Synthetic Turf Fields (Table C-6)
       Samples of tire crumb infill granules were collected at different sampling locations from
several synthetic turf fields.  The sampling locations did not necessarily correspond to the
sampling locations where the other samples were collected. In  some cases, duplicate samples
were collected side-by-side. In addition, second aliquots of material collected  in each sample
container were analyzed, so that there were two analysis results for each sample or duplicate
sample.
       It is important to remember that the methods for collection and analysis have not been
validated. Total extractable metal measurement result highlights include the following.
• Total extractable Pb concentrations in tire crumb infill from synthetic turf fields ranged from 11
  to 61 ug/g.
• There was considerable variability (up to an  approximately fourfold difference)  in total
  extractable Pb concentrations for different aliquots randomly taken from the same sample
  container.
• The  variability among locations at a site and among different sites was similar to the variability
  in Pb measurement results for aliquots taken from the same container (up to approximately
  fourfold).
• Total extractable Cr concentrations ranged from not detected to 1.0 ug/g.
                                           36

-------
• Total extractable Zn concentrations ranged from 2,600 to 19,000 |jg/g.
• The variability in Zn concentrations for aliquots taken from the same container was less than
  the variability observed in Pb concentrations.
• Concentrations  of As ranged from not detected to 0.55 |jg/g.
• Cd concentrations ranged from not detected to approximately 1.5 |jg/g.

6.9.3 Turf Blades at  Synthetic Turf Fields (Table C-7)
   Characterization of the turf blade material was not a primary goal of this scoping study; the
focus of this study was on the tire crumb components. However, these samples were collected
and analyzed to help  improve interpretation of the surface wipe measurements. It is important to
remember that the methods for collection and analysis have not been validated for this material.
Samples of synthetic  turf blades were collected at different sampling locations from several
synthetic turf fields. Where possible, samples of different colors were collected but, because of
the small sample sizes, could not always be analyzed separately. The sampling locations did
not necessarily correspond to the locations where the other samples were collected. Total
extractable metal  measurement result highlights are as follows.
• Total extractable Pb concentrations from synthetic turf  blades ranged from 2.4 to 700 ug/g.
• The highest total extractable Pb concentration (700 ug/g) was measured from  blades in a
  sample collected at a turf field (F5), which included an area that had apparently been repaired
  with a section of turf material visually different from the rest of the field. Mixed-color turf blade
  samples taken from two adjacent fields (F4 and F6) at the same site had Pb levels ranging
  from 2.0 to 77 ug/g.
• The second highest total extractable Pb concentration (389 ug/g) was measured in red blades
  at  another turf field  site (F1). Pb concentrations for green, white, and black blades collected
  from this same site ranged from 2.8 to 4.3  ug/g.
• Total extractable Cr concentrations ranged from 0.1 to  180 ug/g. The level of Cr generally
  appeared to be  lower than but correlated with the corresponding Pb concentrations.
• Total extractable Zn concentrations ranged from 130 to 730 ug/g. Zn levels in white and black
  blades collected at  one site (F1) were about twofold higher than in green and red blades at
  the same site and about three to five times higher than levels measured in blades from the
  other two synthetic turf field sites.
• Concentrations  of As and Cd were less than 0.6 ug/g in all samples.

6.9.4 Tire Crumb Material from Playgrounds (Table C-8)
      Samples of tire crumb pieces were collected at different sampling locations at two
playgrounds. Seven pieces of crumb rubber from one playground (P1; shredded tires with
exposed fibers) and two pieces from  the second playground (P2; with simulated bark tire crumb
material) were analyzed. It is important to remember that the methods for collection and
analysis have not  been assessed or validated. Total extractable metal measurement result
highlights include  those described below.
• Total extractable Pb concentrations in five  pieces of shredded tire crumb from  P1, the
  playground with fibrous materials, ranged from 1.0 to 6.3 ug/g. A Pb concentration of 46 ug/g
  was measured in a sixth piece, and 440 ug/g Pb was measured in a seventh piece,
  documenting the heterogeneity of Pb in these site samples.
• Total extractable Pb concentrations from two simulated bark tire crumb samples collected at a
  second playground (P2) were 3.4 and 7.8 ug/g.
• Total extractable Cr concentrations ranged from 0.3 to  3.0 ug/g.
• Total extractable Zn concentrations ranged from 4,300 to 18,000 ug/g.
• The variability in Zn concentrations was less than the variability observed in Pb
  concentrations.
                                          37

-------
• Concentrations of As ranged from 0.04 to 0.96 |jg/g, except for a value of 15 ug/g that was
  measured in the same P1 crumb piece with the highest Pb level (440 ug/g Pb).
• Cd concentrations ranged from 0.09 to 10.5 ug/g, with the highest concentrations
  corresponding to samples with the highest Pb concentrations.

6.9.5 Lessons Learned with Regard to the Sampling and Analysis of Tire Crumb Materials
       Again, there are no validated sample collection and analytical procedures for
characterizing tire crumb materials for assessing potential environmental concentrations or
potential exposures by various routes and pathways at synthetic turf fields and playgrounds. An
objective of this scoping study was to apply an existing wipe sample collection method and
analytical procedures developed for soil media and to assess their performance. A few of the
lessons learned are described below.
• There was no evaluated protocol available for measuring tire crumb rubber constituents at turf
  fields and  playgrounds. As such, this scoping study was conducted to evaluate the methods
  and identify key factors (e.g., resources, accessibility, practicality, activity levels) that would
  need to be considered in designing future studies. Some factors that may need to be
  considered for future research include the number and  placement of sampling locations (i.e.,
  how many samples at how many locations need to be collected and analyzed to adequately
  characterize a site), representative and duplicate wipe samples for turf fields with mixed
  colors, retaining or not retaining turf infill and fibers on wipe samples, representative material
  samples, the relationship between the wipe sample and material  sample results.
• In some cases, the amount of material available for analysis was not optimal, generally a 1-g
  sample is specified in methods. For tire crumb material from playgrounds, most of the  crumb
  pieces were much larger than 1 g. It was decided that the material would not be cut because
  that would open fresh surfaces  that potentially would result in different extractable amounts.
  Decisions  not to cut up samples prevented sample size matching to extraction procedure
  requirements and prevented homogenization procedures.  In addition, the larger samples
  created some extraction and analysis problems with regard to the extraction vessel  and need
  for multiple dilutions of some sample extracts. On the other hand, some samples of synthetic
  turf blades collected in this study were not adequate for analysis. Only nondestructive
  collection of loose blades was performed in this study. In future work, sample sizes and
  decisions regarding tire crumb subsampling should  be considered.
• Information from this and other studies regarding the metals of most interest would improve
  analytical optimization, reporting, and the selection of appropriate QC materials.
• The heterogeneity of these samples create analysis and data interpretation challenges. For
  example, multiple dilutions and reanalyses of many samples were required to obtain
  measurements in the instrument calibration range. Based on excellent results from analytical
  QC analyses (serial dilution and postdigestion spikes), the new ICP/MS instrument used for
  this study appeared to produce quantitative total extractable results for the wipe samples, turf
  blades, tire infill, and tire crumb media for multiple metals with low detection limits.
• Spiking levels appropriate to the tire material and blade material concentrations need to be
  considered in future study designs to improve  results. However, given the heterogeneity of
  the materials, it is not clear whether spiking samples to assess recoveries will be feasible.
  Spike recoveries need to be reevaluated with truly homogeneous samples to determine
  whether the extraction procedure or the sample heterogeneity was  the source of variable
  recoveries during the scoping study.
• Sporadic contamination of laboratory bottle blanks was found, especially for metals that might
  be associated with steel. In this study, the blank result for each analysis batch was subtracted
  from measured results, following the  standard procedure. In the future, blank correction could
  be based on the average of the bottle blanks, with the option of removing outliers.
                                          38

-------
  Investigation is needed of possible sources of contamination and, perhaps, changing to a less
  "open" extraction process.

6.10 Pb Bioaccessibility Results (Appendix C, Tables C-9, C-10, and C-11)

6.10.1 Analysis of Turf Field Wipe, Tire Crumb, and Turf Blade Samples
       The bioaccessible values for Pb in the field samples are provided in Appendix C.
Because of the variability observed for different sample aliquots, duplicate samples, and
analyses of multiple sample pieces, we have reported all analyses individually rather than
averaging the results across the two measurements for each sample. It is important to
recognize that different bioaccessibility procedures may yield different Pb bioaccessibility
results. The EPA method employed  in this study has been validated for Pb in soil, but no
methods have been validated for tire crumb or synthetic turf blade materials. Highlights from the
bioaccessability analyses include those below.
• Synthetic turf field tire crumb infill Pb bioaccessibility ranged from 1.6% to 10.7% (mean 4.7%;
  Table C-9).
• Synthetic turf field blade Pb  bioaccessibility (Table C-10) ranged from 0.2% to 86.8% (mean
  34.2%). The three samples with the highest total extractable  Pb (77 to 700 ug/g) had the
  lowest Pb bioaccessibility values (0.2% to 2.3%). Gaining a clear understanding of this
  observation requires additional research.
• Playground tire crumb Pb bioaccessibility (Table C-11) ranged from 0.3% to 7.4% (mean
  4.3%). The two samples with the highest total extractable Pb (46 and 440 ug/g) had the
  lowest Pb bioaccessibility values (both 0.3%). Gaining a clear understanding of this
  observation requires additional research.
• Up to a fourfold difference in Pb bioaccessibility was found between two aliquots of tire crumb
  infill analyzed from the same sample vial. Up to a 36-fold difference was found between the
  analyses of seven pieces of tire crumb material from the same playground. These results
  suggest substantial heterogeneity in Pb bioaccessibility from tire crumb rubber samples.
  Gaining a clear understanding of this observation requires additional research.
• The in vitro Pb bioaccessibility method was judged to be inappropriate  for the surface wipe
  samples. The blank media bioaccessible Pb values were similar to the values observed in the
  field samples. Given the relatively small amount of dust collected on the wipe, as compared
  with the large mass of wipe material, and the relatively low amounts of Pb measured, it is
  likely that any calculated bioaccessibility attributed to the dust likely is to be impacted
  significantly by the background levels in the sampling and analysis procedures. Gaining a
  clear understanding of this observation requires additional research.

6.10.2 Lessons Learned with Regard to Bioaccessibility Data
       An objective of this scoping study was to apply existing wipe collection and material
analytical procedures developed for soil media and to assess their performance. A few of the
lessons learned are described below.
• Sufficient quantities of samples are needed to meet extraction requirements for EPA SOP
  9200.1-86; this should be  considered in sampling designs.
• Limitations on cutting samples prevented further homogenization of samples.
• Better method detection limits (MDLs) for the in vitro extractions were achieved relative  to the
  EPA SOP 9200.1-86 requirements and values previously published.
• Additional methods development and validation research is recommended before this wipe
  method is applied in future studies.
                                          39

-------
6.11 Methods Evaluation Summary
      A summary evaluation of the sample collection and analysis methods for the several
types of samples collected in this scoping study is provided in Table 9.
                                        40

-------
Table 9. Overall Summary and Assessment of Methods Applied in This Scoping Study
 Sample Type
    Sample Collection
      Sample Analysis
    Other Comments
 AirVOCs
• Grab samples simple to
  collect.
• Short time needed.
• Can be collected directly
  over field or playground.
• Standard Method TO-15
  (GC/MS) provided data for
  56 analytes with good
  precision and accuracy.
• Sensitivity was sufficient to
  measure analytes in ambient
  air.
• Overall, this method
  was simple to
  implement and provided
  adequate data.
 AirPM-,0
 (mass, metals,
 SEM)
• Large amount of
  equipment needed.
• Technical expertise
  required.
• Required 8-10 h of time
  to collect.
• Could not always be
  placed directly on field
  because of activity.
• Standard research methods
  provided data for mass
  concentration and for
  multiple metals.
  Sensitivity sufficient to
  measure analytes in ambient
  air.
  Precision and accuracy were
  good.
• Overall, this method
  was somewhat complex
  and time consuming to
  implement and provided
  adequate data.
• Identification of tire
  crumb particles was
  difficult because  of lack
  of standard morphology
  or composition.
 Surface Wipes
 (turf fields)
  Moderately simple to
  collect.
• Short time needed.
• Method not validated for
  synthetic turf surfaces.
  Wipes can collect infill
  particles and turf blades
  (that were removed for
  these analyses).
• Standard EPA Methods
  3050B and 6020 applied with
  good recovery and analytical
  precision.
• Sensitivity was very good
  and sufficient to measure low
  levels of multiple metals.
• EPA Pb in vitro
  bioaccessibility method
  9200.1-86 found not to be
  appropriate for wipe
  samples.
• Overall, this method
  was relatively simple to
  implement.
  The method provided
  quantitative
  measurement results,
  but the method has not
  been validated for use
  on synthetic turf field
  surfaces.
• The in vitro Pb
  bioaccessibility method
  was judged to be
  inappropriate for the
  surface wipe samples.
 Tire Crumb
 Infill (turf fields)
  Simple to collect.
• Short time needed.
• Decisions needed on
  area and depth of
  material collection.
• Standard EPA Methods
  3050B and 6020 applied with
  good analytical precision.
• EPA Pb in vitro
  bioaccessibility method
  9200.1-86 applied with good
  analytical precision.
• Nonhomogeneous material
  made assessment of analyte
  recoveries difficult.
• Sensitivity was very good
  and sufficient to measure low
  levels of multiple metals.
• Overall, this method
  was relatively simple to
  implement.
  The method provided
  quantitative
  measurement results,
  but the method has not
  been validated for use
  on tire crumb particles.
  Improved quality control
  methods and materials
  required to assess
  metal recoveries.
  Nonhomogeneity of Pb
  has implications for site
  sampling, analysis, and
  data interpretation.
                                             41

-------
Table 9. Overall Summary and Assessment of Methods Applied in This Scoping Study
(cont'd.)	
 Sample Type
     Sample Collection
     Sample Analysis
    Other Comments
 Blades
 (turf fields)
  Simple to collect.
  Relatively short time
  needed.
• Limited in this study to
  collecting loose blades
  where available.
  Larger sample size
  needed for analysis (>1 g
  each) than were collected
  for some samples in this
  study.
• Collection of different
  colors revealed different
  Pb levels at some fields;
  this should be considered
  in site sampling plans.
• Standard EPA Methods
  3050B and 6020 applied
  with good analytical
  precision.
• EPA Pb in vitro
  bioaccessibility method
  9200.1-86 applied with
  good analytical
  precision.
  Nonhomogeneous
  material made
  assessment of analyte
  recoveries difficult.
\» Sensitivity was very
  good and sufficient to
  measure low levels of
  multiple metals of
  interest.
• Overall, this method
  was relatively simple to
  implement.
• The method provided
  quantitative
  measurement results,
  but the method has not
  been validated for use
  on synthetic turf blades.
• Improved quality control
  methods and materials
  required to assess
  metal recoveries.
• Differences for Pb
  depending on blade
  color have implications
  for site sampling and
  analysis.
 Tire Crumb
 Rubber
 (playgrounds)
• Simple to collect.
• Short time needed.
• Decisions required on site
  sampling plan with regard
  to location and number of
  areas and depth of
  material  for collection.
• Standard EPA Methods
  3050B and 6020 applied
  with good analytical
  precision.
• EPA Pb in vitro
  bioaccessibility method
  9200.1-86 applied with
  good analytical
  precision.
\» Sensitivity was very
  good and sufficient to
  measure low levels of
  multiple metals of
  interest.
  Nonhomogeneous
  material made
  assessment of analyte
  recoveries difficult.
  Tire crumb pieces were
  larger than the 1 g
  desired for analysis;
  homogenization and
  subsampling may be
  needed in future work.
  Overall, sample
  collection was simple to
  implement, but analysis
  was more difficult
  because of sample size
  issues.
  The method provided
  quantitative
  measurement results,
  but the method has not
  been validated for use
  on tire crumb particles.
  Improved quality control
  methods and materials
  required to assess
  metal recoveries.
  Nonhomogeneity of Pb
  has implications for site
  sampling, analysis, and
  data interpretation.
                                             42

-------
                                     7. References

Chen, F., Williams, R., Svendsen, E., Yeatts, K., Creason, J., Scott, J., Terrell, D., Case. M. (2007).
       Coarse participate matter concentrations from residential outdoor sites associated with the North
       Carolina Asthma and Childrens Environment Studies (NC-ACES). Atmospheric Environment,
       41:1200-1208.
National Research Council (2003). Bioavailability of Contaminants in Soils and Sediments: Processes,
       Tools, and Applications. National Academies Press: Washington, DC.
       http://www.nap.edu/openbook/0309086256/html/.
U.S. EPA (2008). Quality Assurance Project Plan. Scoping Study: Evaluation of Exposure Methods and
       Approaches for Characterizing Environmental Concentrations Resulting from the Use of Tire
       Crumb Components in Playgrounds and Artificial Turf Fields. National Exposure Research
       Laboratory.
Williams, R., Suggs, J., Rea., A., Leovic, K., Vette, A., Croghan, C., Sheldon, L, Rodes, C.,
       Thornburg, J., Ejire, A., Herbst, M.,  Sanders, W. Jr. (2003). The Research Triangle Park
       particulate matter panel study: PM mass concentration relationships. Atmospheric Environment,
       37:5349-5363.
Williams, R., Rea, A., Vette, A., Croghan C., Whitaker, D., Wilson, H., Stevens, C., McDow, S., Burke, J.,
       Fortmann, R., Sheldon, L., Thornburg, J., Phillips, M., Lawless, P., Rodes, C., Daughtrey, H.
       (2008). The design and field implementation of the Detroit Exposure and Aerosol Research Study
       (DEARS). Journal of Exposure Science and Environmental Epidemiology, doi:101038.
                                             43

-------
                                    Appendix A

                 List of Sample Collection and Analysis Methods

  NOTE: The following methods have not been evaluated for collecting and analyzing samples
            from synthetic turf fields or from playgrounds with tire crumb rubber.

Research Operating Procedure for the Collection of Participate Matter (PM) Air Samples at
Playgrounds and Synthetic Turf Fields

Research Operating Procedure for the Collection of Tire Crumb Material at Playgrounds

Research Operating Procedure for the Collection of "Grass Blade" Fibers from Synthetic Turf
Fields

Research Operating Procedure for the Collection of Infill Material from Synthetic Turf Fields

Research Operating Procedure for the Collection of Volatile Organic Chemicals in Air Using
Canisters

ASTM E1792-03: Standard Specification for Wipe Sampling Materials for Lead in Surface Dust

ASTM E1728-03: Standard Practice for Collection of Settled Dust Samples Using Wipe
Sampling Methods for Subsequent Lead Determination

Recommended Operating Procedure for Elemental Analysis of Particulate Matter on Membrane
Filters by the LBL XRF Spectrometer

Standard Operating Procedure for the Gravimetric Determination of Particle Mass on Teflon Air
Sampling Filters

Standard Operating Procedures for the USEPA-NERL Scanning Electron Microscopy/
Energy-Dispersive X-Ray Analysis (SEM/EDX) Laboratory

Compendium Method TO-15: Determination of Volatile Organic Compounds (VOCs) in Air
Collected in Specially-Prepared Canisters and Analyzed by Gas Chromatography/Mass
Spectrometry (GC/MS); (Region 1 EIASOP-AIRCAN9: "Standard Operating Procedure for the
Analysis of Volatile Organic Compounds in Air by Gas Chromatography/lon Trap Detector)"

EPA SW-846 Method 3050B: Acid Digestion of Sediments, Sludges, and Soils. Revision 2.
December 1996

EPA SW-846 Method 6020A: Inductively Coupled Plasma-Mass Spectrometry. Revision 1.
February 2007

EPA 9200.1-86. Standard Operating Procedure for an In Vitro Bioaccessibility Assay for Lead in
Soil. May 2008
                                         44

-------
                                    Appendix B

                  Quality Control and Quality Assurance Results

Table B-1. Field and Laboratory Blank Measurement Results for Volatile Organic Compounds in
Air

Table B-2. Percent Recovery of Air Volatile Organic Compounds in Field and Laboratory
Controls

Table B-3. Relative  Percent Difference in Measurement Results for Air Volatile Organic
Compounds in Field Duplicate Canister Samples and in Repeat Analysis of Canister Samples in
the Laboratory

Table B-4. ICP/MS Operating Parameter Settings

Table B-5. ICP/MS Isotopes and Interference Corrections

Table B-6. Concentrations of Individual Metals in Working Calibration Standards

Table B-7. ICP/MS Method Detection  Limits

Table B-8. Summary of ICP/MS QC Criteria

Table B-9. Summary of ICP/MS Instrumental QC Results

Table B-10. Recovery of Metals Spiked in Extraction Reagent Blank

Table B-11. Recovery of Metals-Spiked Solution on Matrix of Interest

Table B-12. Total Extractable Recoveries for NIST SRM  2710 Spiked into Extraction Solution

Table B-13. Total Extractable Recoveries for NIST SRM  2710 Spiked onto Ghost Wipe Media

Table B-14. Bottle Blank Data Used for Sample Correction and Reagent Blank Data Method
3050B

Table B-15. Laboratory and Field Ghost Wpe Blank Samples

Table B-16. Relative Percent Differences for the Analysis of Duplicate Aliquots of Tire Crumb
Infill from Synthetic  Fields and Playground  Tire Crumb Samples

Table B-17. Recommended Control Limits  for In Vitro Soil Quality Control Samples According to
EPA Method 9200.1-86

Table B-18. Summary Control Limit Results for In Vitro Pb Bioaccessibility Quality Control
Samples

Table B-19. Percent Recovery Results for In Vitro Blank  Spikes
                                         45

-------
Table B-20. Summary of Pb In Vitro Extractable Values for NIST SRM 2710

Table B-21. Summary of Pb In Vitro Extractable Values for NIST SRM 2710 Spiked onto Ghost
Wipe Media

Table B-22. Results for In Vitro Pb Solution Spikes onto Blank Ghost Wipes and Tire Crumb
Samples

Table B-23. In Vitro Pb Bioaccessibility Extraction Duplicates for Synthetic Turf Field Tire Crumb
Infill

Table B-24. In Vitro Pb Bioaccessibility Extraction Duplicates for Playground Tire Crumb
                                          46

-------
Table B-1. Field and Laboratory Blank Measurement Results (ppbV) for Volatile Organic Compounds (VOCs) in Air
voc
1,1,1 -Trichloroethane
1 , 1 ,2,2-Tetrachloroethane
1 , 1 ,2-Trichloroethane
1,1-Dichloroethane
1,1-Dichloroethylene
1 ,2,4-Trichlorobenzene
1 , 2, 4-Trimethyl benzene
1,2-Dibromoethane
1,2-Dichlorobenzene
1,2-Dichloroethane
1 ,2-Dichloropropane
1, 3, 5-Trimethyl benzene
1,3-Butadiene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
2-Hexanone
4-Ethyltoluene
Acrylonitrile
Allyl Chloride
Benzene
Benzylchloride
Bromodichloromethane
Bromoform
c-1 ,2-Dichloroethylene
c-1,3-Dichloropropylene
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Cyclohexane
Dibromochloromethane
Dichlorodifluoromethane
Dichlorotetrafluoroethane
Ethyl benzene
Heptane
Hexachloro-1,3-butadiene
Hexane
m&p-Xylenes
Methyl Ethyl Ketone
Field Blank
MDL
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.10
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.10
0.05
Sample
MDL
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.1 5C
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.1 5C
0.076a
Field Blanks
P1
NDD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.100
F4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.059
F1D1
ND
ND
ND
ND
ND
ND
0.360
ND
ND
0.073
ND
0.120
0.250
ND
ND
ND
0.380
ND
ND
1.11
ND
ND
ND
ND
ND
0.040
0.510
ND
0.049
12.0
ND
0.270
ND
1.63
ND
ND
6.20
4.27
3.83
Laboratory Blanks
P1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.040
F2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.044
F4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F1D1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F1D2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND

-------
Table B-1. Field and Laboratory Blank Measurement Results (ppbV) for Volatile Organic Compounds (VOCs) in Air (cont'd.)
voc
Methyl Isobutyl Ketone
Methylbromide
Methylchloride
Methylene Chloride
Methyl-t-butyl ether
o-Xylene
Styrene
t-1 ,2-Dichloroethylene
t-1 ,3-Dichloropropylene
Tetrachloroethylene
Tetrahydrofuran
Toluene
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
Vinyl Bromide
Vinylchloride
Field Blank
MDL
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
Sample
MDL
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
0.076a
Field Blanks
P1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F1D1
0.190
0.042
0.440
0.400
ND
1.35
1.10
ND
ND
0.160
ND
29.0
0.180
ND
ND
ND
ND
Laboratory Blanks
P1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F1D1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
F1D2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
a Range from 0.070 to 0.079 ppbV.
bND = not detected.
0 Range from 0.14 to 0.16 ppbV.

-------
Table B-2. Percent Recovery'
Laboratory Controls
            of Air Volatile Organic Compounds (VOCs) in Field and
voc
1,1,1-Trichloroethane
1 ,1 ,2,2-Tetrachloroethane
1,1,2-Trichloroethane
1,1-Dichloroethane
1,1-Dichloroethylene
1 ,2,4-Trichlorobenzene
1 ,2,4-Trimethylbenzene
1,2-Dichlorobenzene
1,2-Dichloropropane
1 ,3,5-Trimethylbenzene
1,3-Dichlorobenzene
Benzene
c-1 ,2-Dichloroethylene
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Dichlorodifluoromethane
Dichlorotetrafluoroethane
Ethylbenzene
Hexachloro-1 ,3-butadiene
m&p-Xylenes
Methylchloride
Methylene Chloride
o-Xylene
Tetrachloroethylene
Toluene
Trichloroethylene
Trichlorofluoromethane
Vinylchloride
Spiking
Level
(ppbV)
4.83
4.99
4.82
4.90
4.78
4.09
5.04
4.97
4.85
4.87
5.00
4.83
4.80
4.84
4.86
4.81
4.90
4.05
3.84
4.88
4.41
4.97
4.38
4.91
4.60
4.86
4.83
4.92
4.76
3.09
Field Controls
P1
124
98
112
112
103
72
101
91
112
111
92
116
107
101
103
96
110
113
103
79
92
106
90
111
108
109
107
113
117
104
F4
114
86
104
99
97
44
84
77
105
97
77
105
97
26
97
66
99
95
96
94
72
95
103
101
93
100
101
103
106
97
F1D1
136
99
119
108
122
70
102
91
119
112
92
120
113
132
105
112
106
107
109
109
89
111
98
114
111
108
115
112
120
116
Mean
125
94
112
106
108
62
96
86
112
107
87
114
106
87
101
91
105
105
103
94
84
104
97
108
104
106
108
109
114
106
Std.
Dev.
10.8
7.5
7.8
6.8
12.8
15.7
10.1
8.4
6.7
8.8
8.5
7.7
8.1
55.0
4.4
23.2
5.8
9.0
6.6
15.0
10.7
8.3
6.5
6.7
9.7
5.0
7.1
5.3
7.3
9.2
Laboratory Controls
P1
120
93
108
106
106
69
96
88
102
111
89
114
106
76
102
90
103
97
100
84
90
106
89
105
102
101
107
98
114
103
F4
125
93
110
99
101
61
98
85
114
108
83
118
99
107
102
101
104
95
104
104
81
108
87
105
105
107
108
107
116
105
F1D2
128
93
109
93
109
70
96
86
109
109
86
111
99
120
104
91
96
101
99
105
103
106
82
101
104
105
108
109
110
102
Mean
124
93
109
99
105
66
97
86
108
109
86
114
101
101
102
94
101
98
101
97
91
107
86
104
104
104
108
105
113
103
Std.
Dev.
4.3
0.1
1.1
6.2
3.9
5.0
1.2
1.7
5.8
1.3
2.9
3.5
3.9
22.7
1.3
6.1
4.6
3.0
2.5
12.0
11.1
1.0
3.8
2.4
1.2
3.0
0.3
5.5
3.1
1.4
      Found
     Expected
'100
                                        49

-------
Table B-3. Relative Percent Difference (RPD)a in Measurement Results for Air Volatile
Organic Compounds (VOCs) in Field Duplicate Canister Samples and in Repeat Analysis
of Canister Samples in the Laboratory
voc
Benzene
Carbon Tetrachloride
Dichlorodifluoromethane
Hexane
m&p-Xylenes
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methylchloride
Methylene Chloride
Toluene
Trichlorofluoromethane
Trichlorotrifluoroethane
Field Duplicates
RPD
P1
4.8
9.3
3.8
44.9
	
11.8
	
8.5
44.4
107.1
0.0
2.7
RPD
F4
13.3
3.4
0.0
28.6
2.6
4.8
	
2.1
20.3
6.5
0.0
1.4
RPD
F1D1
9.5
8.3
7.7
105.5
74.2
28.6
21.1
4.1
—
143.9
11.3
1.2

Mean
RPD
9.2
7.0
3.8
59.7
38.4
15.0
21.1
4.9
32.4
85.8
3.8
1.8
Std.
Dev
RPD
4.3
3.1
3.8
40.5
50.6
12.2

3.3
17.0
71.1
6.5
0.8
Laboratory Repeat Analysis
RPD
P1
12.8
12.4
1.9
	
	
0.0
	
0.0
4.3
6.9
0.0
2.7
RPD
F4
0.0
1.1
4.1
0.0
8.7
2.0
	
0.0
8.7
5.4
0.0
2.8
RPD
F1D1
7.9
1.0
5.6
31.2
3.7
14.6
	
8.7
12.0
30.8
11.3
7.7

Mean
RPD
6.9
4.9
3.9
15.6
6.2
5.6

2.9
8.3
14.4
3.8
4.4
Std.
Dev.
RPD
6.4
6.6
1.9
22.1
3.5
7.9

5.0
3.8
14.2
6.5
2.8
a100 * ABS(M1 - M2) / [(M1+M2) / 2}
                                       50

-------
Table B-4. ICP/MS Operating Parameter Settings
Instrument Settings
RF Power
Ar Gas Flow Rates:
Cool
Auxiliary
CCT Gas Flow (He 93%/ H 7%)
Sampler Cone (Ni/Cu)
Skimmer Cone (Ni/Cu)
Nebulizer
Spray Chamber
Detector Dead Time
Internal Standard Solution

1,200-1, 260 W

13 Lpm
0.9-1.0 Lpm
-10 uL/min
1 . 1 -mm diameter orifice sample cone
0.75-mm diameter skimmer cone
Concentric nebulizer, 35 PSI, 1 mL/min
Air-cooled cyclone
55 ns
40-200 ppb solution of 6 Li 45 Sc 89 Y In115and 159Tb

Acquisition Parameters (Normal Mode)
Major
Extraction
Lens 1
Lens 2
Focus
D1
D2-140
Pole Bias 0
Hexapole Bias -4.0
Minor
Lens
Forward power 1,400
Horizontal
Vertical
DA
Cool
Auxiliary
Sampling depth 130-140
Global
Standard resolution
High resolution
Analogue detector
PC detector




Add. Gases
CCT-0







Acquisition Parameters (Collision Cell Technology [CCT] Mode)
Major
Extraction
Lens 1
Lens 2
Focus
D1
D2-140
Pole Bias -17.0
Hexapole Bias -20.0
Minor
Lens
Forward power 1,400
Horizontal
Vertical
DA
Cool
Auxiliary
Sampling depth 130-140
Global
Standard resolution
High resolution
Analogue detector
PC detector




Add. Gases
CCT- -10 uL/min








Operating Parameters
Standard Resolution
High Resolution
Integration Type
Calibration Type
Number of scans per replicate
Number of replicates (runs)
0.75± 0.1 amu
0.30± 0.1 amu
Average
Linear
1
3-7












                                      51

-------
Table B-5. ICP/MS Isotopes and Interference Corrections
Analyte
Pb
Pb
Cr
Zn
Al
As
Ba
Cd
Mass
207
208
52
66
27
75
137
111
Interference








Analyte
Cd
Cu
Fe
Fe
Mn
Ni
Ni
Mass
114
65
54
56
55
60
62
Interference
-0.027*1 18 Sn

-0.028* 52 Cr
-0.15*43 Ca

-0.002* 34 Ca

Table B-6. Concentrations of Individual Metals in Working Calibration Standards
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
LoCaM
(ppb)
250
250
250
250
250
250
250
250
250
250
250
LoCal2
(ppb)
500
500
500
500
500
500
500
500
500
500
500
HiCaM
(ppm)








50


HiCal2
(ppm)








100


LLQC CRI
(ppb)
1
2
2
30
1
10
1
2

1
1
LoCalS
(ppb)
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
LoCal2
1:100 (ppb)
5
5
5
5
5
5
5
5
5
5
5
Internal
Standards
Li
Sc
Y
In
Tb
200
200
40
40
40






























Table B-7. ICP/MS Method Detection Limits (ppb)a
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Isotope
208
52
66
27
75
137
111
65
56
55
60
In Vitro Method
0.082
0.201
0.409
1.99
0.366
0.668
0.219
0.335
6.4
0.178
0.268
EPA Method 3050B
0.092
0.175
0.388
2.47
0.246
0.634
0.224
0.350
4.3
0.144
0.332
 Calculated by formula in 40 CRF Part 136. MDLs reported here are for sample extracts delivered to the instrument.
For In vitro matrix: First CRI chosen per day over 6 analysis days plus 1 extra. (Note: One metal (Fe) not spiked into
CRI so concentration at reagent blank level.
For EPA 3050B matrix: One CRI chosen per day over 7 analysis days.
                                            52

-------
Table B-8. Summary of ICP/MS QC Criteria3
Name
Initial Calibration Verification
Initial Calibration Blank
Continuing Calibration Verification
Continuing Calibration Blank
Interference Check Solution A
Interference Check Solution AB
Contract Required Quantitation Limit Check
Duplicate Samples Relative Percent Difference
Serial Dilution Samples Percent Difference
Post Digestion Spike Samples

Sample Extract QC
Spike
SRM
Acceptance Criteria
90%-110%

-------
Table B-9. Summary of ICP/MS Instrumental QC Results
In EPA 3050B 5% HNO3 Matrix, n=4
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
DUPa Criteria
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
SER° Criteria
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
PDS6 Criteira
75%- 125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
DUP RPDb Found
0.117%-7.83%
0.145%-3.54%
0.081%- 1.26%
0.023%-5.26%
0.929%-12.8%
0.563%-1.41%
1 .048%-1 .26%
0.011%-1.40%
0.110%-2.37%
1.09%-3.61%
0.478%-2.50%
SER Difference
2.54%-7.24%
0.167%
2.41%-7.74%, 14.8%
1.50%-4.31%
*d
0.938%-4.99%
*
2.34%-5.72%
*
*
3.43%
PDS Recovery
100%-104%
86.7%-89.0%
99.6%-109%
*
98.7%-105%
77.2%-100%
94.6%-103%
96.5%-101%
*
*
92.8%-99.2%
Status (pass/fail)
P
P
P
P
P
P
P
P
P
P
P
Status (pass/fail)
P
P
F
P
N/A
P
N/A
P
N/A
N/A
P
Status (pass/fail)
P
P
P
N/A
P
P
P
P
N/A
N/A
P
                                     54

-------
Table B-9. Summary of ICP/MS Instrumental QC Results (cont'd.)
In In Vitro 2% HNO3 Matrix, n=5
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Metal
Pb
Cr
Zn
Al
As
Ba
Cd
Cu
Fe
Mn
Ni
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
Mass
(amu)
208
52
66
27
75
137
111
65
56
55
60
DUP Criteria
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
<20%
SER Criteria
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
<10%
PDS Criteria
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
75%-125%
DUP RPD Found
0.684%-4.70%
*
0.104%-1.98%
1 .90%-1 1 .0%
*
0.142%-2.38%
*
0.142%-9.38%
*
*
*
SER Difference
0.062%
*
0.169%-9.40%
8.49%, 26.0%
*
0.045%-2.51%
*
*
*
*
*
PDS Recovery
94.7%-106%
91.2%-102%
95.4%-110%
*
90.0%-103%
88.0%-108%
97.3%-107%
91.2%-111%
*
*
97.2%-108%
Status (pass/fail)
P
N/A
P
P
N/A
P
N/A
P
N/A
N/A
N/A
Status (pass/fail)
P
N/A
P
F
N/A
P
N/A
N/A
N/A
N/A
N/A
Status (pass/fail)
P
P
P
N/A
P
P
P
P
N/A
N/A
P
 DUP = duplicate aliquot of extract analyzed independently.
bRPD = relative percent difference.
CSER = serial dilution of extract.
dConcentrations too low to meet QC criteria.
ePDS = postdigestion spike.
                                               55

-------
Table B-10. Recovery of Metals Spiked in Extraction Reagent Blank
Sample ID
Description
Spiked Amount
(total ug spiked)
Net Total Extractable
(total ug found)
Percent
Recovery
Primary Metals of Interest
Cr
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=8)
Pb
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=8)
Zn
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=8)

Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


640
640
2,500
2,500
2,500
2,500
2,500
2,500


1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000


800
800
2,500
2,500
2,500
2,500
2,500
2,500


600
616
2,340
2,350
2,320
2,390
2,370
2,380


949
1,070
987
985
980
974
999
1,000


774
747
2,390
2,390
2,370
2,420
2,400
2,400


93.7
96.3
93.8
94.1
92.7
95.8
94.8
95.0
94.5

94.9
107
98.7
98.5
98.0
97.4
99.9
100
99.3

96.8
93.4
95.7
95.5
94.8
96.7
95.9
96.1
95.6
Secondary Metals of Interest
Al
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X
As
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=6)

Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


0
0
0.2
0.2
0.2
0.2
0.2
0.2


0
0
500
500
500
500
500
500


5.40
31.4
44.2
38.3
36.7
29.3
32.9
20.6


23.1
22.9
449
459
449
423
453
463


Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked

Not spiked
Not spiked
89.8
91.8
89.8
84.5
90.6
92.5
89.8
                                       56

-------
Table B-10. Recovery of Metals Spiked in Extraction Reagent Blank (cont'd.)
Sample ID
Ba
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=6)
Cd
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=8)
Cu
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=8)
Fe
TC1-2
TC2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X
Mn
TC1-2
TC-2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=2)
Description

Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike


Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike

Spiked Amount
(total ug spiked)

0
0
2,510
2,510
2,510
2,510
2,510
2,510


160
160
497
497
497
497
497
497


800
800
2,540
2,540
2,540
2,540
2,540
2,540


0
0
0
0
0
0
0
0


400
400
0.02
0.02
0.02
0.02
0.02
0.02

Net Total Extractable
(total ug found)

2.97
4.58
2,420
2,460
2,420
2,480
2,490
2,490


157
150
475
480
465
483
480
480


793
800
2,410
2,410
2,400
2,450
2,420
2,430


0
27.7
3.00
0.222
2.12
0.213
0.668
0.219


377
386
0.311
0.004
0.028
0
0.010
0.121

Percent
Recovery

Not spiked
Not spiked
96.6
98.1
96.3
98.6
99.2
99.1
98.0

98.0
93.6
95.6
96.6
93.5
97.3
96.6
96.5
96.0

99.1
100
95.0
95.0
94.6
96.3
95.4
95.5
96.4

Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked

94.2
96.4
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
Not spiked
95.3
                                        57

-------
Table B-10. Recovery of Metals Spiked in Extraction Reagent Blank (cont'd.)
Sample ID
Ni
TC1-2
TC-2-2
TC3-2
TC4-2
TC5-2
TC6-2
TC7-2
TC8-2
X(n=6)
Description

Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike

Spiked Amount
(total ug spiked)

0
0
1,250
1,250
1,250
1,250
1,250
1,250

Net Total Extractable
(total ug found)

0.869
0.312
1,220
1,220
1,200
1,240
1,220
1,230

Percent
Recovery

Not spiked
Not spiked
97.2
97.7
95.8
99.2
97.5
98.3
97.6
                                        58

-------
Table B-11. Recovery of Metals-Spiked Solution on Matrix of Interest
Media and Sample
Spiked
Amount (ug)
Net Spiked
Total
Extractable (ug)
Net Media
Total
Extractable (ug)
Media
Corrected
Spike (ug)
Percent
Recovery
Pb on Wipes
Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike
Mean


1,000


991


991



921


994


995


0.227


0.694


1.08




920


993


994



92.0


100


100
99.4

Pb on Infill
F4-L1-S1-A1
F4-L1-S1-A1

F4-L1-D1-A1
F4-L1-D1-A1

F1-L3-S1-A1
F1-L3-S1-A1

F2-L2-D1-A1
F2-L2-D1-A1


1,000


1,000


991


991


974


214


854


889

41.1


24.8


20.6


36.4



933


190


834


853


93.3


19.0


84.1


86.0

Pbon Crumb
P1-LA-TC1
P1-LA-TC1

P1-LA-TC2
P1-LA-TC2

P1-LB-TC1
P1-LB-TC1


1,000


991


991


317


233


276

0.989


6.31


443



316


227


-167


31.6


22.9


N/A

Cr on Wipes
Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike
Mean


640


2,490


2,490



580


2,400


2,370


0.091


6.24


13.0




580


2,390


2,360



90.7


96.1


94.8
93.9

Cron Infill
F4-L1-S1-A1
F4-L1-S1-A1

F4-L1-D1-A1
F4-L1-D1-A1

F1-L3-S1-A1
F1-L3-S1-A1

640


640


2,490

596


106


2,160
0.602


0.352


0.544


595


106


2,160

93.0


16.6


86.7
                                         59

-------
Table B-11. Recovery of Metals-Spiked Solution on Matrix of Interest (cont'd.)
Media and Sample
F2-L2-D1-A1
F2-L2-D1-A1
Mean

Spiked
Amount (ug)

2,490


Net Spiked
Total
Extractable (ug)

2,080


Net Media
Total
Extractable (ug)
0.018



Media
Corrected
Spike (ug)

2,080


Percent
Recovery

83.6
69.9

Cron Crumb
P1-LA-TC1
P1-LA-TC1

P1-LA-TC2
P1-LA-TC2

P1-LB-TC1
P1-LB-TC1
Mean


640


2,490


2,490



115


598


788


0.281


0.721


0.761




115


598


787



17.9


24.0


31.6
24.4

Zn on Wipes
Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike

Blank Ghost Wipe
Wipe with Spike
Mean


800


2,490


2,490



763


2,420


2,430


4.06


13.0


23.6




759


2,400


2,410



94.8


96.5


96.7
96.0

Zn on Infill
F4-L1-S1-A1
F4-L1-S1-A1

F4-L1-D1-A1
F4-L1-D1-A1

F1-L3-S1-A1
F1-L3-S1-A1

F2-L2-D1-A1
F2-L2-D1-A1
Mean


800


800


2,490


2,490



8,960


6,040


21,700


11,700


9,940


4,880


7,930


10,300




-989


1,159


13,800


1,420



N/A


1,459


555


56.8
251

Zn on Crumb
P1-LA-TC1
P1-LA-TC1

P1-LA-TC2
P1-LA-TC2

P1-LB-TC1
P1-LB-TC1
Mean

800


2,490


2,490


5,270


13,100


14,000

4,330


6,730


17,500



940


6,380


-3,430


117.5


256


N/A
186
                                         60

-------
Table B-12. Total Extractable Recoveries for NIST SRM 2710 Spiked into Extraction
Solution
Metal
As





Mean
Cra





Mean
Pb





Mean
Zn





Mean
Al





Mean
Ba





Mean
Cd





Mean
Batch
1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

SRM Certified
Concentration
(ug/g)
626
626
626
626
626
626

39
39
39
39
39
39

5,530
5,530
5,530
5,530
5,530
5,530

6,950
6,950
6,950
6,950
6,950
6,950

64,400
64,400
64,400
64,400
64,400
64,400

707
707
707
707
707
707

21.8
21.8
21.8
21.8
21.8
21.8

Found
Concentration
(ug/g)
535
502
514
492
502
498

14.2
16.3
15.1
10.1
14.8
13.8

5,100
4,740
4,830
4,830
4,740
4,630

5,640
5,040
5,580
5,300
5,140
5,200

18,300
18,900
18,400
16,900
17,600
17,200

329
311
267
298
304
322

17.0
17.8
20.2
18.8
17.9
18.7

Percent
Recovery
85.5
80.3
82.0
78.5
80.2
79.5
81.0
36.3
41.8
38.6
26.0
38.0
35.4
37.5
92.2
85.7
87.4
87.3
85.6
83.7
87.0
81.2
72.5
80.3
76.2
73.9
74.8
76.5
28.3
29.4
28.5
26.3
27.3
26.8
27.8
46.6
44.0
37.8
42.1
43.0
45.5
43.2
78.1
81.4
92.7
86.3
82.2
85.7
84.4
SRM
Percent
Leachable
Recovery
94
94
94
94
94
94

49
49
49
49
49
49

92
92
92
92
92
92

85
85
85
85
85
85

28
28
28
28
28
28

51
51
51
51
51
51

92
92
92
92
92
92

                                        61

-------
Table B-12. Total Extractable Recoveries for NIST SRM 2710 Spiked into Extraction
Solution (cont'd.)
Metal
Cu





Mean
Fe





Mean
Mn





Mean
Ni





Mean
Batch
1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

1
2
3
4
6
8

SRM Certified
Concentration
(ug/g)
2,950
2,950
2,950
2,950
2,950
2,950

33,800
33,800
33,800
33,800
33,800
33,800

10,100
10,100
10,100
10,100
10,100
10,100

14.3
14.3
14.3
14.3
14.3
14.3

Found
Concentration
(ug/g)
2,630
2,490
2,570
2,500
2,490
2,510

28,500
24,100
27,500
25,400
23,400
23,000

7,250
6,890
7,080
6,610
6,860
6,880

7.89
9.89
8.70
5.50
8.39
7.96

Percent
Recovery
89.0
84.3
87.2
84.6
84.4
85.0
85.8
84.4
71.2
81.4
75.0
69.4
67.9
74.9
71.8
68.2
70.1
65.4
67.9
68.2
68.6
55.2
69.2
60.8
38.5
58.7
55.6
56.3
SRM
Percent
Leachable
Recovery
92
92
92
92
92
92

80
80
80
80
80
80

76
76
76
76
76
76

71
71
71
71
71
71

aNot certified for SRM.
                                         62

-------
TableB-13
Media
Total Extractable Recoveries for NIST SRM 2710 Spiked onto Ghost Wipe
Metal
As


As


As

Mean
Cra


Cr


Cr

Mean (n=3)
Mean (n=2)
Pb








Mean
Zn








Mean
Al








Mean
Type
SRM
Blank

SRM
Blank

SRM
Blank

SRM
Blank

SRM
Blank

SRM
Blank



SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank

Batch
1
1

8
8

9
9

1
1

8
8

9
9



1
1

8
8

9
9


1
1

8
8

9
9


1
1

8
8

9
9

SRM Certified
Total Cone.
(ug/g)
626


626


626


39


39


39




5,530


5,530


5,530



6,950


6,950


6,950



64,400


64,400


64,400


Found
Cone.
(ug/sample)
507
0.15

543
0.13

496
0.12

14.1
0.09

17.7
6.23

13.8
13.0



4,950
0.23

5,040
0.69

4,500
1.08


5,570
5.11

5,800
12.9

5,260
23.7


18,100
0.37

20,000
2.14

18,500
7.71

Blank Corrected
Cone.
(ug/g)
507


543


496


14.0


11.4


0.76




4,950


5,040


4,500



5,560


5,780


5,240



18,100


20,000


18,500


Percent
Recovery
80.90


86.80


79.2

82.3
36.0


29.3


1.96

22.4
32.6

89.5


91.1


81.4

87.3

80.0


83.2


75.4

79.5

28.2


31.1


28.8

29.3
SRM Percent
Leachable
Recovery
94


94


94


49


49


49




92


92


92



85


85


85



28


28


28


                                      63

-------
Table B-13. Total Extractable Recoveries for NIST SRM 2710 Spiked onto Ghost Wipe
Media (cont'd.)
Metal
Ba








Mean
Cd








Mean
Cu








Mean
Fe








Mean
Mn








Mean
Type

SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank


SRM
Blank

SRM
Blank

SRM
Blank

Batch

1
1

8
8

9
9


1
1

8
8

9
9


1
1

8
8

9
9


1
1

8
8

9
9


1
1

8
8

9
9

SRM Certified
Total Cone.
(ug/g)

707


707


707



21.8


21.8


21.8



2,950


2,950


2,950



33,800


33,800


33,800



10,100


10,100


10,100


Found
Cone.
(ug/sample)

316
6.84

353
22.7

324
25.5


16.4
0.03

18.2
0.06

17.9
0.02


2,540
0.02

2,630
0.25

2,450
0.44


27,500
0.20

29,100
26.7

23,800
61.2


7,050
0.03

7,310
10.3

6,980
23.3

Blank Corrected
Cone.
(ug/g)

309


330


298



16.4


18.1


17.9



2,540


2,630


2,450



27,500


29,100


23,700



7,050


7,300


6,960


Percent
Recovery

43.8


46.7


42.2

44.2

75.0


83.2


81.9

80.1

86.2


89.3


83.1

86.2

81.4


86.0


70.2

79.2

69.8


72.3


68.9

70.3
SRM Percent
Leachable
Recovery

51


51


51



92


92


92



92


92


92



80


80


80



76


76


76


                                       64

-------
Table B-13. Total Extractable Recoveries for NIST SRM 2710 Spiked onto Ghost Wipe
Media (cont'd.)
Metal
Ni








Mean (n=3)
Mean (n=2)
Type

SRM
Blank

SRM
Blank

SRM
Blank


Batch

1
1

8
8

9
9


SRM Certified
Total Cone.
(ug/g)

14.3


14.3


14.3



Found
Cone.
(ug/sample)

7.86
0.00

9.96
3.42

7.98
7.08


Blank Corrected
Cone.
(ug/g)

7.86


6.55


0.91



Percent
Recovery

55.0


45.8


6.34

35.7
50.4
SRM Percent
Leachable
Recovery

71


71


71



aNot certified for SRM.
                                        65

-------
Table B-14. Bottle Blank Data Used for Sample Correction and Reagent Blank Data
Method 3050B (ng/mL)

As
Cr
Pb
Zn
Al
Ba
Cd
Cu
Fe
Mn
Ni
For In Vitro 2% HNO3
Bottle Blank ID
TC1-1
TC2-1
TC3-1
TC4-1
TC5-1
TC6-1
TC7-1
TC8-1
TC9-1
Mean (n=9)
Std. dev.
%RSD
MDL
0.13
-0.016
0.126
0.023
0.002
0.005
-0.022
-0.015
-0.031
0.057
0.065
114
0.367
0.252
0.059
0.466
0.035
0.109
0.132
0.027
0.043
0.041
0.129
0.145
112
0.201
0.292
0.114
0.293
0.05
0.106
0.138
0.087
0.07
0.071
0.136
0.093
68.3
0.082
68.0
10.4
97.3
7.72
54.4
58.5
9.60
8.04
8.63
35.8
34.1
95.2
0.409
23.0
3.46
31.1
3.29
16.2
17.8
2.81
3.22
1.70
11.4
10.9
95.6
1.99
125
11
174
8.18
119
126
11.4
9.04
7.31
65.6
68.4
104.3
0.668
0.007
-0.001
0.05
0.001
0.001
0.001
0
0
0.002
0.008
0.017
222
0.219
1.36
0.926
1.51
0.23
0.716
1.27
0.372
0.42
0.261
0.784
0.499
63.6
0.335
0.006
0.0
0.007
0.000
0.003
0.003
0.001
0.001
0.000
0.003
0.002
98.0
0.006
0.132
0.08
0.132
0.017
0.043
0.061
0.023
0.06
0.012
0.062
0.046
73.1
0.178
0.121
0.038
0.208
0.015
0.013
0.071
0.026
0.009
0.074
0.064
0.066
102
0.268
For EPA 3050B 5% HNO3
Bottle Blank ID
TC1-1
TC2-1
TC3-1
TC4-1
TC5-1
TC6-1
TC7-1
TC8-1
TC9-1
Mean (n=9)
Std. dev.
%RSD
MDL
1.04
1.28
0.755
1.40
0.761
0.814
0.383
0.426
1.47
0.925
0.398
43.0
0.246
1.45
1.10
1.29
48.6
5.49
3.38
40.5
5.67
6.97
12.7
18.3
144
0.175
6.91
2.69
1.84
4.407
1.65
1.54
0.791
2.54
3.22
2.84
1.86
65.3
0.092
192
148
157
229
288
185
208
182
170
195
42.4
21.7
0.388
128
85.6
91.6
100
102
88.5
90.7
103
82.0
96.9
13.9
14.4
2.47
179
153
197
193
192
178
183
191
169
182
14.0
7.7
0.634
0.088
0.061
0.034
0.811
0.413
0.19
1.40
0.18
1.06
0.471
0.500
106
0.224
13.4
7.63
3.61
7.88
14.8
5.53
2.90
5.31
8.52
7.72
4.08
52.8
0.350
0.113
0.047
0.042
0.191
0.034
0.041
0.194
0.056
0.029
0.083
0.067
80.5
0.004
9.40
2.48
1.47
52.1
2.70
4.15
48.8
9.25
4.13
14.9
20.4
136
0.144
3.71
2.91
2.75
27.8
2.90
2.50
25.6
3.24
7.73
8.80
10.3
117
0.332
For EPA 3050B Only Reagent Blanks 5% HNO3 Not Stored in Bottles
Reagent Blank ID
TC-4-13
TC5-13
TC6-13
TC7-13
TC9-13
Mean (n=4)
Std. dev.
%RSD
Lost
0.936
0.971
0.074
0.382
0.591
0.438
74.1
Lost
5.20
1.02
8.15
271
71.2
133
187
Lost
1.56
7.04
1.39
2.76
3.19
2.64
82.8
Lost
301
166
156
285
227
76.3
33.6
Lost
92.6
95.9
78.6
60.8
82.0
16.0
19.5
Lost
173
168
187
167
174
9.15
5.3
Lost
0.472
0.035
0.314
2.74
0.890
1.24
140
Lost
11.3
6.31
1.37
3.44
5.60
4.28
76.6
Lost
0.044
0.109
0.038
1.043
0.308
0.491
159
Lost
2.69
11.2
9.01
460
121
226
187
Lost
12.3
2.93
5.63
152
43.2
72.6
168
                                      66

-------
Table B-15. Laboratory and Field Ghost Wipe Blank Samples (not blank corrected;
ng/mL)
Blank
Wipe ID
As
Cr
Pb
Zn
Al
Ba
Cd
Cu
Fe
Mn
Ni
 Laboratory Blank Wipes
ForlnVitro2%HNO3
TC1-11
TC8-12
TC9-2
Mean
std dev
%RSD
MDL
0.144
0.065
0.038
0.082
0.055
66.9
0.367
0.183
0.13
0.113
0.142
0.037
25.7
0.201
0.282
0.695
0.227
0.401
0.256
63.7
0.082
67.7
55.5
57.6
60.3
6.53
10.8
0.409
13.4
12.9
15.3
13.8
1.29
9.3
1.99
95.6
118
115
110
12.4
11.3
0.668
0.083
0.012
-0.001
0.048
0.050
106
0.219
1.45
0.944
0.902
1.099
0.306
27.8
0.335
0.006
0.005
0.003
0.005
0.002
32.7
0.006
0.224
0.167
0.086
0.159
0.069
43.6
0.178
0.077
0.034
0.064
0.058
0.022
37.8
0.268
For EPA 3050B 5% HNO3
TC1-11
TC8-12
TC9-2
Mean
std dev
%RSD
MDL
2.45
1.61
2.47
2.17
0.49
22.6
0.25
2.34
67.9
136
68.9
67.0
97.3
0.17
9.20
8.10
13.7
10.3
2.95
28.5
0.09
232
207
297
245
46.4
18.9
0.39
133
97.5
124
118
18.0
15.3
2.47
261
177
179
205
47.8
23.3
0.63
0.252
0.777
1.24
0.75
0.49
65.2
0.22
9.58
6.61
11.6
9.27
2.53
27.2
0.35
0.114
0.313
0.63
0.35
0.26
73.9
0.00
9.41
112
246
123
119
96.8
0.14
2.16
37.4
78.2
39.3
38.0
96.9
0.33
 Field Blank Wipes
For In Vitro 2% HNO3
TC8-4
TC8-6
TC8-7
Mean
std dev
%RSD
0.043
0.037
0.037
0.04
0.00
8.9
0.045
0.134
0.144
0.11
0.05
50.6
0.242
9.13
0.2
3.19
5.14
161
12.9
54.8
60.4
42.7
26.0
60.8
4.16
16.0
20.3
13.5
8.37
62.0
9.60
115
126
83.7
64.4
77.0
0.084
0.009
0.006
0.03
0.04
134
0.402
0.926
0.857
0.73
0.28
39.1
0.003
0.005
0.005
0.004
0.001
26.6
0.157
0.129
0.133
0.1400
0.015
10.8
0.055
0.049
0.058
0.054
0.005
8.5
Repeat Extract Analysis
TC8-6
0.034
0.146
9.27
55.0
15.8
115
0.009
0.926
0.005
0.153
0.048
 For EPA 3050B 5% HNO3
TC8-4
TC8-6
TC8-7
Mean
std dev
%RSD
1.72
1.17
1.35
1.42
0.28
19.8
9.68
7.37
665
227
379
167
9.75
4.72
7.84
7.44
2.54
34.1
246
209
371
276
85.1
30.9
109
127
138
125
14.9
12.0
189
186
173
183
8.79
4.8
1.40
0.18
3.01
1.53
1.42
92.6
7.75
6.30
7.12
7.06
0.73
10.3
0.15
0.14
2.63
0.97
1.43
148
19.6
14.2
1,170
401
666
166
5.960
5.04
358
123
203
166
                                         67

-------
Table B-16. Relative Percent Differences (RPD) for the Analysis of Duplicate Aliquots of
Tire Crumb Infill from Synthetic Fields and Playground Tire Crumb Samples (ug/g)
Sample Cr Pb Zn As
Synthetic Field Infill Samples
F4-L1-S1-A1
F4-LI-S1-A2
RPD

F4-L1-D1-A1
F4-L1-D1-A2
RPD

F4-L2-S1-A1
F4-L2-S1-A2
RPD

F4-L3-S1-A1
F4-L3-S1-A2
RPD

F1-L1-S1-A1
F1-L1-S1-A2
RPD

F2-L1-S1-A1
F2-L1-S1-A2
RPD

F2-L1-D1-A1
F2-L1-D1-A2
RPD

F1-L3-S1-A1
F1-L3-S1-A2
RPD

F1-L2-S1-A1
F1-L2-S1-A2
RPD

F2-L3-S1-A1
F2-L3-S1-A2
RPD

F2-L2-D1-A1
F2-L2-D1-A2
RPD

F2-L2-S1-A1
F2-L2-S1-A2
RPD

F2-L3-D1-A1
F2-L3-D1-A2
RPD
0.60
0.25
84.0

0.35
0.33
6.0

0.14
0.37
88.9

1.03
0.98
5.0

1.01
0.95
6.2

0.35
0.92
90.2

0.36
0.24
41.1

0.54
0.24
76.0

0.33
0.54
47.2

NR
NR


NR
NR


NR
NR


NR
NR

41.1
10.7
118

24.8
47.7
63.2

13.6
19.3
34.4

23.7
20.0
16.7

29.2
18.5
44.9

20.6
26.5
25.3

61.2
36.0
51.9

20.6
14.4
35.2

13.1
34.7
90.2

21.6
29.1
29.6

36.4
27.5
27.9

33.0
30.6
7.7

43.7
22.4
64.7
9,940
5,320
60.6

4,880
4,070
18.0

2,660
4,310
47.6

11,400
8,190
33.1

17,200
19,200
10.9

5,690
9,930
54.4

5,890
3,120
61.4

7,930
5,047
44.4

9,050
8,541
5.8

10,700
10,300
3.5

10,300
10,700
3.19

10,500
10,100
3.6

10,200
12,300
18.7
0.11
0.08
33.3

NRa
NR


NR
NR


0.40
0.28
35.9

0.55
0.18
102

0.23
0.24
3.9

0.18
0.22
17.7

0.29
0.20
38.1

0.21
0.25
21.1

0.22
0.25
11.4

0.44
0.24
58.2

0.28
0.13
74.7

0.20
0.26
29.1
                                       68

-------
Table B-16. Relative Percent Differences (RPD) for the Analysis of Duplicate Aliquots of
Tire Crumb Infill from Synthetic Fields and Playground Tire Crumb Samples (ug/g)
(cont'd.)

Cr
Pb
Zn
As
Playground Tire Crumb Samples
P1-LA-TC1
P1-LA-TC2
RPD

P2-TC1
P2-TC2
RPD

P1-LA-TC4
P1-LA-TC5
RPD

P1-LB-TC1
P1-LB-TC2
RPD
0.52
1.66
104

1.61
2.97
59.5

0.72
0.76
5.8

0.76
0.26
98.2
2.43
46.3
180

7.75
3.42
77.5

6.31
4.64
30.4

443
0.99
199
9,720
11,100
13.7

18,000
12,100
39.4

6,730
8,250
20.3

17,500
6630
89.9
0.15
0.96
146

0.25
0.28
10.1

0.28
0.59
72.1

15.0
0.08
198
 aNR = Not reported.
Table B-17. Recommended Control Limits for In Vitro Soil Quality Control Samples
According to EPA Method 9200.1-86
In Vitro QCs
Reagent Blank
Bottle Blank
Blank Spike (10 mg/L)
Matrix Spike (1 0 mg/L)
Duplicate Sample
Control Soil (NIST 2710)
Frequency
Once per batch
5%a
5%a
10%a
1 0%a
5%a
Control Limits
<25 |jg/L Pb
<50 |jg/L Pb
85%- 11 5% recovery
75%-125% recovery
±20% RPDD
± 1 0% RPDC
Corrective Actions
Make new fluid and rerun all
analyses.
Check calibration and reanalyze
as necessary.
Check calibration and/or source
of contamination and reanalyze.
Flag
Flag
Flag
 Minimum 1 in 20.
bRPD = relative percent difference.
cThe National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) RPD is based on
certified values and mean RBA-Pb values of 75% for SRM 2710.
                                          69

-------
Table B-18. Summary Control Limit Results for In Vitro Pb Bioaccessibility Quality
Control Samples (Note: This method has not been validated for the types of samples
collected in this scoping study.)
In Vitro QC Parameters
Reagent Blank
Bottle Blank
Blank Spike (10 mg/L)
Blank Ghost Wipe with Spike
(10 mg/L)
Blank Ghost Wipe with NIST SRM
271 Ob
NIST SRM 27 10b
Infill Spike (10 mg/L)
Crumb Spike (10 mg/L)
Duplicate Samples
Blank Ghost Wipe
Analysis Frequency
9 Reagent blank
(batches 1-9)
9 Blank runs
(batches 1-9)
8 Blank spike runs
(batches 1-8)
3 Blank wipe with
spike runs
(batches 1,8,9)
3 Wipes with
SRM runs
(batches 1,8,9)
6 SRM runs
(batches 1-4, 6, 8)
4 Infill spike runs
(batches 1-4)
3 Crumb spike runs
(batches 1,5,9)
13 Pairs of infill
samples and 4 pairs
of crumb samples
3 Blank ghost wipe
runs
(batches 1,8,9)
Control Limit
Results
<5 ug/L Pb
<5 ug/L Pb
90-105%
recovery
87%-99%
recovery
3%-6% RPDC
0%-9% RPDC
89%-104%
recovery
87%- 103%
recovery
2.7%-124%for
infill samples and
4%-183%for
crumb samples
<10 ug/wipe Pb
Corrective Actions
None
None
None
None
None
None
None
None
Noted in report
None
 Minimum 1 in 20; matrix spikes for crumbs and infill only performed if enough material was available.
bThe National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) RPD is based on
certified values and mean RBA-Pb values of 75% for SRM 2710.
CRPD = relative percent difference.

Table B-19. Percent Recovery Results for In Vitro Blank Spikes
Batch Number
1
2
3
4
5
6
7
8
Mean
Sample ID
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike
Blank Spike

Spiked
Amount
(mg/L)
10
10
10
10
10
10
10
10

Recovered In Vitro
Extractable Pb (mg/L)
9.0
9.9
10.2
10.2
10.2
10.5
10.0
10.1

Percent Spike
Recovery
89.7
99.5
102
102
102
105
100
101
100
                                            70

-------
Table B-20. Summary of Pb In Vitro Extractable Values for MIST SRM 2710
Batch Number
1
2
3
4
6
8
Mean
SRM Certified
Concentration (ug/g)
75.0
75.0
75.0
75.0
75.0
75.0
75.0
Found
Concentration (ug/g)
81.7
73.6
74.1
79.5
81.3
75.3
77.6
RPD
8.9
1.9
1.2
6.0
8.3
0.4
4.5
Table B-21. Summary of Pb In Vitro Extractable Values for SRM 2710 Spiked onto Ghost
Wipe Media
Type
SRM

SRM

SRM

Mean
Batch
1

8

9


SRM %IVBAa
(based on EPA SOP)
75

75

75


Found
(%IVBA)
79.7

78.3

77.6

78.5
RPD
6.22

4.39

3.45

4.69
 In vitro bioaccessibility.
                                        71

-------
Table B-22. Results for In Vitro Pb Solution Spikes onto Blank Ghost Wipes and Tire
Crumb Samples
Media and Spike
Pb on Wipes
Ghost Wipe with Spike
Blank Ghost Wipe
Ghost Wipe with Spike
Blank Ghost Wpe
Ghost Wipe with Spike
Blank Ghost Wpe
Pb on Tire Crumb Infill
F4-L1-S1-A1 Spiked
F4-L1-S1-A1 Unspiked
F4-L1-DS1-A1 Spiked
F4-L1-DS1-A1
Unspiked
F1-L3-S1-A1 Spiked
F1-L3-S1-A1 Unspiked
F2-L2-DS1-A1 Spiked
F2-L2-DS1-A1
Unspiked
Pb on Tire Crumb
P1-LA-TC1 Spiked
P1-LA-TC1 Unspiked
P1-LA-TC4 Spiked
P1-LA-TC4 Unspiked
P1-LB-TC1 Spiked
P1-LB-TC1 Unspiked
Spiked
Amount
(mg/L)

10

10

10


10

10

10

10


10

10

10

Net Spiked
In Vitro
Extra ctable
(mg/L)

8.65

9.93

9.66


8.86

10.0

9.91

10.5


8.88

10.3

10.2

Net Media
In Vitro
Extractable
(mg/L)


0.00

0.01

0.00


0.01

0.01

0.01

0.02


0.00

0.00

0.02
Media
Corrected
Spike

8.65

9.93

9.66


8.86

10.0

9.90

10.5


8.87

10.3

10.2

Percent
Recovery

86.5

99.3

96.6


88.6

100

99.0

105


88.7

103

102

                                      72

-------
Table B-23. In Vitro Pb Bioaccessibility Extraction Duplicates for Synthetic Turf Field Tire
Crumb Infill
Sample
F4-L1-S1-A1
F4-L1-S1-A2
RPD
F4-L1-DS1-A1
F4-L1-DS1-A2
RPD
F4-L2-DS1-A1
F4-L2-DS1-A2
RPD
F4-L3-S1-A1
F4-L3-S1-A2
RPD
F1-L1-S1-A1
F1-L1-S1-A2
RPD
F2-L1-S1-A1
F2-L1-S1-A2
RPD
F2-L1-DS1-A1
F2-L1-DS1-A2
RPD
Pb (%IVBA)a
1.7
7.3
124
3.9
2.4
46.3
3.8
2.7
34.7
8.5
10.1
17.4
9.6
5.3
57.2
3.7
3.8
2.7
1.7
2.9
52.9
Sample
F1-L3-S1-A1
F1-L3-S1-A2
RPD
F1-L2-S1-A1
F1-L2-S1-A2
RPD
F2-L3-S1-A1
F2-L3-S1-A2
RPD
F2-L2-DS1-A1
F2-L2-DS1-A2
RPD
F2-L2-S1-A1
F2-L2-S1-A2
RPD
F2-L3-DS1-A1
F2-L3-DS1-A2
RPD
Maximum RPD
Minimum RPD
Number
Pb (%IVBA)
4.2
4.4
4.6
5.0
1.6
105
5.0
4.2
17.9
5.0
4.5
11.3
7.6
3.6
72.3
3.1
5.5
54.7
124
2.7
13
 %IVBA = percent in vitro bioaccessibility.

Table B-24. In Vitro Pb Bioaccessibility Extraction Duplicates for Playground Tire Crumb
Sample
P1-LA-TC2
P1-LA-TC3
RPD
P2-TC1
P2-TC2
RPD
P1-LA-TC4
P1-LA-TC5
Pb (%IVBA)a
4.6
0.3
176
1.8
7.4
122
2.4
5.2
 a%IVBA = percent in vitro bioaccessibility.
Sample
P1-LB-TC1
P1-LB-TC2
RPD
Maximum RPD
Minimum RPD
Number
Pb (%IVBA)
0.3
6.4
183
183
73.2
4
                                           73

-------
                                     Appendix C

              Compilation of Environmental Sample Analysis Results

Table C-1. Results of Analysis for Grab VOC Air Samples

Table C-2. Results of Analysis for Particle Mass on Integrated Air Samples

Table C-3. Results of Analysis for Metals in Integrated Air Samples

Table C-4. Results of SEM Analysis for Postulated Tire Crumb Particles on Integrated Air
Samples

Table C-5. Results of Analysis of Wet Wipe Samples for Total Extractable Metals

Table C-6. Results of Analysis of Synthetic Turf Field Tire Crumb Infill for Total Extractable
Metals

Table C-7. Results of Analysis of Synthetic Turf Field Blades for Total Extractable Metals

Table C-8. Results of Analysis of Playground Tire Crumb for Total Extractable Metals

Table C-9. Results of Analysis of Synthetic Turf Field Infill Sample Analysis for Bioaccessible Pb

Table C-10. Results of Analysis of Synthetic Turf Field Blade Sample Analysis for Bioaccessible
Pb

Table C-11. Results of Analysis of Playground Tire Crumb Sample Analysis for Bioaccessible
Pb
                                          74

-------
Table C-1. Results of Analysis for Grab VOC Air Samples
VOC
Benzene
Toluene
m&p-Xylenes
o-Xylene
Ethylbenzene
1 ,2,4-Trimethylbenzene
4-Ethyltoluene
Hexane
Site"
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
0.090
0.081
0.098
0.320
0.088
0.140
0.120
0.180
0.510
0.140
NDd
0.075
0.047
0.290
ND
ND
ND
ND
0.120
ND
ND
ND
ND
0.110
ND
ND
ND
ND
0.082
ND
ND
ND
ND
0.082
ND
ND
0.098
ND
0.150
ND
B
0.110
0.081
0.120
0.120
0.086
0.98
0.110
0.180
0.190
0.430
0.170
0.130
0.074
0.055
0.130
ND
ND
ND
ND
0.045
0.073
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.320
0.062
0.049
0.110
0.300
C
0.074
0.088
0.120
0.160
C
0.140
0.110
0.190
0.150
—
ND
ND
0.091
0.077
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
0.092
ND
0.110
0.160
—
D
0.073
0.092
0.120
0.120
0.087
0.150
0.120
0.190
0.190
0.160
0.083
ND
0.077
ND
0.050
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.063
0.078
0.045
0.049
ND
M9/mJ
Sampling Location at Siteb
A
0.287
0.255
0.319
1.021
0.287
0.527
0.452
0.677
1.919
0.527
ND
0.303
0.217
1.257
ND
ND
ND
ND
0.520
ND
ND
ND
ND
0.477
ND
ND
ND
ND
0.393
ND
ND
ND
ND
0.393
ND
ND
0.352
ND
0.528
ND
B
0.351
0.255
0.383
0.383
0.287
3.688
0.414
0.677
0.715
1.618
0.737
0.564
0.303
0.217
0.564
ND
ND
ND
ND
0.173
0.303
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.126
0.211
0.176
0.387
1.055
C
0.223
0.287
0.383
0.510
—
0.527
0.414
0.715
0.564
—
ND
ND
0.390
0.347
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
0.317
ND
0.387
0.563
—
D
0.223
0.287
0.383
0.383
0.287
0.564
0.452
0.715
0.715
0.602
0.347
ND
0.347
ND
0.217
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.211
0.281
0.141
0.176
ND
                                      75

-------
Table C-1. Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
Cyclohexane
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
2-Hexanone
Carbon Tetrachloride
Dichlorodifluoromethane
Methylchloride
Trichlorofluoromethane
Sitea
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
0.400
0.490
0.400
0.370
0.340
0.088
0.120
ND
ND
ND
ND
ND
ND
ND
ND
0.090
0.097
0.088
0.088
0.083
0.519
0.490
0.500
0.470
0.520
0.430
0.460
0.420
0.470
0.470
0.250
0.260
0.250
0.240
0.260
B
0.330
ND
ND
ND
0.250
0.560
0.340
0.440
0.500
0.480
0.110
0.110
ND
ND
ND
ND
ND
0.046
ND
ND
0.092
0.100
0.100
0.089
0.082
0.540
0.530
0.640
0.480
0.540
0.500
0.470
0.460
0.500
0.450
0.280
0.260
0.290
0.240
0.260
C
ND
ND
ND
ND
—
0.440
0.320
0.380
0.410
—
0.180
ND
ND
ND
—
ND
ND
ND
ND
—
0.084
0.093
0.092
0.086
—
0.500
0.470
0.540
0.490
—
0.470
0.470
0.470
0.470
—
0.260
0.250
0.270
0.240
—
D
ND
ND
ND
ND
ND
0.440
0.360
0.370
0.440
0.380
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.097
0.098
0.084
0.098
0.094
0.550
0.560
0.510
0.540
0.540
0.480
0.460
0.450
0.520
0.450
0.280
0.270
0.250
0.300
0.260
M9/mJ
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
1.180
1.445
1.180
1.091
1.003
0.368
0.491
ND
ND
ND
ND
ND
ND
ND
ND
0.566
0.610
0.554
0.554
0.522
2.566
2.423
2.472
2.324
2.571
0.888
0.95
0.867
0.97
0.97
1.405
1.461
1.405
1.349
1.461
B
1.134
ND
ND
ND
0.859
1.651
1.003
1.298
1.474
1.415
0.450
0.450
ND
ND
ND
ND
ND
0.204
ND
ND
0.579
0.629
0.629
0.560
0.516
2.670
2.621
3.165
2.373
2.670
1.033
0.97
0.95
1.033
0.93
1.573
1.461
1.630
1.349
1.461
C
ND
ND
ND
ND
—
1.298
0.94
1.121
1.209
—
0.736
ND
ND
ND
—
ND
ND
ND
ND
—
0.528
0.585
0.579
0.541
—
2.472
2.324
2.670
2.423
—
0.97
0.97
0.971
0.97
—
1.461
1.405
1.517
1.349
—
D
ND
ND
ND
ND
ND
1.298
1.062
1.091
1.298
1.121
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.610
0.616
0.528
0.616
0.0.591
2.720
2.769
2.522
2.670
2.670
0.99
0.95
0.93
1.074
0.93
1.573
1.517
1.405
1.686
1.461
                                      76

-------
Table C-1. Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
Trichlorotrifluoroethane
Methylene Chloride
Chloroform
1,1,1-Trichloroethane
1,1,2,2-
Tetrachloroethane
1,1,2-Trichloroethane
1,1-Dichloroethane
1,1-Dichloroethylene
1 ,2,4-Trichlorobenzene
Sitea
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
0.080
0.077
0.073
0.069
0.074
ND
ND
0.058
0.055
0.071
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
0.081
0.079
0.085
0.072
0.074
0.074
ND
0.069
0.055
0.110
ND
ND
0.062
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
0.075
0.076
0.075
0.072
—
0.062
ND
0.062
0.065
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
0.081
0.084
0.072
0.150
0.074
0.062
ND
0.059
0.061
0.066
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
M9/mJ
Sampling Location at Siteb
A
0.613
0.613
0.537
0.537
0.537
ND
ND
0.208
0.208
0.243
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
0.613
0.613
0.613
0.537
0.537
0.243
ND
0.243
0.174
0.382
ND
ND
0.293
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
0.537
0.613
0.537
0.537
—
0.208
ND
0.208
0.208
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
0.613
0.613
0.537
1.150
0.537
0.208
ND
0.208
0.208
0.243
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
                                      77

-------
Table C-1. Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
1 ,2-Dibromoethane
1,2-Dichlorobenzene
1,2-Dichloroethane
1,2-Dichloropropane
1 ,3,5-Trimethylbenzene
1,3-Butadiene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
Acrylonitrile
Sitea
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
M9/mJ
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
                                      78

-------
Table C-1. Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
Allyl Chloride
Benzylchloride
Bromodichloromethane
Bromoform
cis-1 ,2-Dichloroethylene
cis-1 ,3-
Dichloropropylene
Chlorobenzene
Chloroethane
Dibromochloromethane
Sitea
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
M9/mJ
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
                                      79

-------
Table C-1. Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
Dichlorotetrafluoroethane
Heptane
Hexachloro-1 ,3-
butadiene
Methylbromide
Methyl-t-Butyl Ether
Styrene
trans-1 ,2-
Dichloroethylene
trans-1 ,3-
Dichloropropylene
Tetrachloroethylene
Sitea
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
M9/mJ
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
                                      80

-------
 Table C-1.  Results of Analysis for Grab VOC Air Samples (cont'd.)
VOC
Tetrahydrofuran
Trichloroethylene
Vinyl Bromide
Vinylchloride
Site"
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
F1D1
F1D2
F2
F4
P1
ppbV
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
M9/mJ
Sampling Location at Siteb
A
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
B
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
C
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
ND
ND
ND
ND
—
D
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
aSite identification: P = playground; F = synthetic turf field; D = day 1 or day 2.
""Sampling location: A, B, and C are "on field" or "on playground"; D is "upwind background."
°Only two "on playground" sampling locations.
dND = not detected.
                                                   81

-------
Table C-2. Results of Analysis for Particle Mass on Integrated Air Samples (|jg/m3)
Site3
F1D1
F1D2
F4
P1
Sampling Location0
A
27.5
29.6
33.4
23.9
A Dup
c
—
31.0
24.3
B
28.6
30.0
30.9
29.5
B Dup
29.0
—
—
—
C
27.4
—
31.0
—
D
29.5
29.5
28.6
14.2
Field Blank
0.6
0.4
0.7
0.2
aF = turf field, P = playground, D = day, Dup = duplicate sample.
bA, B, and C are "on field" or "on playground," D is "upwind background.
cNot collected.
Table C-3. Results of Analysis for Metals in Integrated Air Samples (ng/m3)3
Site and Sampling
Location at the Siteb
F1D1 location A
F1D1 location B
F1D1 location B duplicate
F1D1 location C
F1D1 location D
F1D1 field blank
F1D2 location A
F1D2 location B
F1D2 location C
F1D2 location D
F1D2 field blank
F4 location A
F4 location A duplicate
F4 location B
F4 location C
F4 location D
F4 field blank
P1 location A
P1 location A duplicate
P1 location B
P1 location D
P1 field blank
Pb
c
—
—
—
—
—
7.7
—
NRd
6.3
—
—
—
—
—
—
—
5.1
—
—
—
—
Cr
2.3
—
2.8
3.6
2.0
—
4.3
2.8
NR
3.3
	
3.4
5.0
—
—
2.7
—
3.6
2.5
3.1
—
—
Zn
6.6
15.3
13.6
10.6
23.8
—
17.0
6.6
NR
11.6
9.5
36.0
31.9
33.0
25.2
21.7
—
90.7
82.5
117
10.5
—
Ca
449
230
370
349
396
34.5
834
812
NR
881
	
487
303
338
319
466
12.9
317
365
414
238
—
Cl
14.1
—
—
—
—
12.2
21.0
—
NR
20.9
	
25.5
—
—
—
21.1
—
118
177
189
95.5
—
Cu
8.9
—
—
—
—
—
10.3
—
NR
—
—
21.3
20.0
13.4
12.2
12.3
—
11.8
—
—
10.8
—
Fe
230
201
199
226
240
—
364
364
NR
356
	
803
559
663
573
456
—
672
748
987
294
—
K
134
91.7
99.2
126
128
8.6
164
158
NR
178
	
302
191
228
238
227
—
196
240
312
97.2
—
Mn
6.1
—
6.3
—
7.5
—
9.3
10.6
NR
8.4
4.3
18.5
13.0
13.4
10.9
10.9
—
8.6
13.6
15.0
—
—
S
4,004
3,688
3,018
3,946
3,667
—
3,680
3,784
NR
3,933
	
3,044
1,516
2,845
2,952
2,976
—
664
882
903
751
—
Si
1,291
555
1,497
796
1,405
—
1,248
858
NR
1,548
	
2,646
2,116
1,912
1,942
1,890
—
2,784
2,435
3,455
516
—
Ti
20.4
30.8
22.3
24.7
25.2
—
29.4
27.9
NR
31.9
	
58.7
44.0
58.2
51.4
41.2
—
57.6
63.7
92.3
21.4
—
aValues represent those less than three times the measurement uncertainty limit following field blank corrections.
bF = turf field, P = playground, D = day, Dup = duplicate sample. bDesignations A, B, and C refer to "on field" or "on playground"
locations; D locations represent background sites.
°— Represents values less than three times the measurement uncertainty limit.
dNR = sample filter inverted during collection; analysis results not reported.
                                                    82

-------
Table C-4. Results of SEM Analysis for Postulated Tire Crumb Particles on Integrated Air
Samples3
Sample IDb
F1D1 location A
F1 D1 location A duplicate
F1D1 location A dup. Repeat
F1D1 location B
F1D1 location D
F4 location A
F4 location A duplicate
F4 location B
F4 location D
P1 location A
P1 location B
P1 location D
Number of
Postulated
Tire Crumb
Particles0
5
15
5
18
7
0
4
6
6
63e
58f
5
Mass of
Postulated
Tire Crumb
Particles0
(pg)
6
80
65
90
16
0
19
47
15
3,330
2,530
40
Scaled
Mass of
Postulated
Tire Crumb
Particlesd (ug)
0.01
0.02
0.03
0.08
0.01
0.00
0.01
0.03
0.01
3.9
2.1
0.03
Estimated
Concentration of
Postulated Tire
Crumb Particles
(ug/m3)
0.001
0.004
0.007
0.019
0.003
0.000
0.003
0.007
0.002
0.77
0.42
0.005
aGiven the lack of unique composition and morphology for tire crumb particles from the collected materials, the
estimates in this table have considerable uncertainty.
bF= synthetic turf field; P = playground; A and B are "on field" or "on playground;" D = "upwind background."
cRaw numbers, not normalized to the same analyzed area. (CCSEM areas analyzed ranged from 0.6 mm2 to 3.6
mm2).
dEstimated tire crumb mass scaled to total exposed filter area of 6.7 cm2.
eMass median diameter -2.6 urn.
fMass median diameter -2.2 urn.
                                                83

-------
Table C-5. Results of Analysis of Wet Wipe Samples for Total Extractable Metals (ug/ft2;
based on total extractable amounts from consecutive in vitro and Method 3050B
extractions)
Site and Sample
Location3
F1D1
F1D1
F1D1
F1D1
F1D1
F1D1
F1D2
F1D2
F1D2
F1D2
F1D2
F1D2
F2
F2
F2
F4
F4
F5
F6
F1D1
F4
F4
Loc. A, S
Loc. A, DS
Loc. B, S°
Loc. B, DS°
Loc. C, S
Loc. C, DS
Loc. A, S
Loc. A, DS
Loc. B, S°
Loc. B, DS°
Loc. C, S
Loc. C, DS
Loc. A, S
Loc. B, S
Loc. C, S
Loc. A, S
Loc. A, DS
Loc. B, S
Loc. C, S
Field Blank
Field Blank
Field Blank
Pb
0.500
0.489
1.46
1.91
0.347
0.323
0.370
0.407
0.731
1.39
0.688
0.346
0.456
0.280
0.289
0.177
0.139
0.129
0.184
0.181
0.541
0.135
Cr
0.112
0.178
0.413
0.517
0.214
0.077
0.096
0.197
0.278
NRd
NR
NR
NR
NR
NR
NR
NR
NR
0.245
0.096
0.046
16.0
Zn
35.9
43.3
34.4
38.4
30.9
21.3
26.4
34.9
37.4
29.1
40.6
25.7
19.3
13.0
9.24
5.28
4.25
6.29
13.6
1.79
3.17
7.36
Al
19.5
26.3
27.0
22.5
18.6
17.3
17.6
21.1
30.4
28.9
30.0
18.2
52.1
28.0
33.3
42.7
36.1
19.6
16.4
0.17
1.25
1.76
As
0.051
0.055
0.062
0.102
0.033
0.028
0.045
0.039
0.050
0.058
0.049
0.045
0.049
0.026
0.024
0.022
0.018
0.020
0.027
0.034
0.021
0.025
Ba
NR
0.351
0.083
NR
5.05
NR
NR
NR
0.387
0.564
0.280
NR
0.125
7.31
1.16
0.085
0.384
6.10
5.94
NR
5.60
5.84
Cd

-------
Table C-6. Results of Analysis of Synthetic Turf Field Tire Crumb Infill for Total
Extractable Metals (ug/g; based on total extractable amounts from consecutive in vitro
and Method 3050B extractions)
Site, Sampling
Location, Sample,
and Aliquot3
F1
F1
F1
F1
F1
F1
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F4
F4
F4
F4
F5
F5
F6
F6
L1.S1, A1
L1.S1, A2
L2, S1, A1
L2, S1, A2
L3, S1, A1
L3, S1, A2
L1.S1, A1
L1.S1, A2
L1, DS1, A1
L1, DS1, A2
L2, S1, A1
L2, S1, A2
L2, DS1, A1
L2, DS1, A2
L3, S1, A1
L3, S1, A2
L3, DS1, A1
L3, DS1, A2
L1.S1, A1
L1.S1, A2
L1, DS1, A1
L1, DS1, A2
L1, S1, A1
L1.S1, A2
L1.S1, A1
L1.S1, A2
Pb
29.2
18.5
13.1
34.7
20.6
14.4
20.6
26.5
61.2
36.0
33.0
30.6
36.4
27.5
21.6
29.1
43.7
22.4
41.1
10.7
24.8
47.7
13.6
19.3
23.7
20.0
Cr
1.01
0.95
0.33
0.54
0.54
0.24
0.35
0.92
0.36
0.24
NRb
NR
NR
NR
NR
NR
NR
NR
0.60
0.25
0.35
0.33

-------
Table C-7. Results of Analysis of Synthetic Turf Blades for Total Extractable Metals (ug/g;
based on total extractable amounts from consecutive in vitro and Method 3050B
extractions)
Site and
Blade Color3
F1
F1
F1
F1
F3
F3
F4
F5
F6
Red
Green
White
Black
Red
White
Green, white,
yellow mix
Green, white,
yellow mixc
Green, white,
yellow mix
Pb
389
3.84
4.28
2.76
2.40
1.97
2.08
701
77.1
Cr
73.1
9.71
0.99
1.91
1.20
0.08
3.72
177
18.9
Zn
351
316
688
729
199
255
206
131
175
Al
1,090
2,090
1,320
1,290
947
336
2,120
1,620
1,150
As
0.40
0.47
0.60
0.29
0.22
NR
0.25
0.12
0.05
Ba
141
88
114
111
1,950
38
50
40
303
Cd
0.16
0.07
NRb
NR
NR
NR
NR

-------
Table C-9. Results of Analysis of Synthetic Turf Field Infill Sample Analysis for
Bioaccessible Pb
Site, Sampling Location, Sample,
and Aliquot3
F1
F1
F1
F1
F1
F1
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F2
F4
F4
F4
F4
F5
F5
F6
F6
Mean
Minimum
Maximum
L1.S1.A1
L1,S1, A2
L2, S1.A1
L2, S1, A2
L3, S1.A1
L3, S1, A2
L1.S1.A1
L1.S1.A2
L1, DS1, A1
L1.DS1.A2
L2, S1, A1
L2, S1.A2
L2, DS1, A1
L2, DS1.A2
L3, S1, A1
L3, S1.A2
L3, DS1, A1
L3, DS1.A2
L1.S1, A1
L1,S1, A2
L1.DS1.A1
L1, DS1, A2
L1.S1.A1
L1,S1, A2
L1.S1.A1
L1,S1, A2



Total Extractable Pb
(M9/9)
29.2
18.5
13.1
34.7
20.6
14.4
20.6
26.5
61.2
36.0
33.0
30.6
36.4
27.5
21.6
29.1
43.7
22.4
41.1
10.7
24.8
47.7
13.6
19.3
23.7
20.0



Pb In Vitro
Bioaccessibility Value (%)b
9.6
5.3
5.0
1.6
4.2
4.4
3.7
3.8
1.7
2.9
7.6
3.6
5.0
4.5
5.0
4.2
3.1
5.5
1.7
7.3
3.9
2.4
3.8
2.7
8.5
10.1
4. 7 ±2.3
1.6
10.1
al_ = sampling location at site, S = sample collected at the location, DS = duplicate sample collected at the location,
A = aliquot of tire crumb infill from sample (~1 g each).
bThe in vitro bioaccessibility values were determined by dividing the amount of Pb extracted in the in vitro extraction
by the total extractable amount of Pb.
                                                 87

-------
Table C-10. Results of Analysis of Synthetic Turf Field Blade Sample Analysis for
Bioaccessible Pb
Site3 and Blade Color
F1
F1
F1
F1
F3
F3
F4
F5
F6
Mean
Minimum
Maximum
Red
Green
White
Black
Red
White
Green, white, yellow mix
Green, white, yellow mix
Green, white, yellow mix



Total Extractable Pb
(ug/g)
389
3.84
4.28
2.76
2.40
1.97
2.08
701
77.1



Pb In Vitro
Bioaccessibility Value (%)a
2.3
40.9
43.0
86.8
40.3
38.7
54.4
0.2
1.0
34. 2 ±28. 8
0.2
86.8
3 F = field site.
bThe in vitro bioaccessibility values were determined by dividing the amount of Pb extracted in the in vitro extraction
by the total extractable amount of Pb.
Table C-11. Results of Analysis of Playground Tire Crumb Sample Analysis for
Bioaccessible Pb
Site, Sampling Location, and
Tire Crumb Piece3
P1
P1
P1
P1
P1
P1
P1
P2
P2
Mean
Minimum
Maximum
Loc. A, TC1
Loc. A, TC2
Loc. A, TC3
Loc. A, TC4
Loc. A, TC5
Loc. B, TC1
Loc. B, TC2
TC1
TC2



Total Extractable Pb
(ug/g)
0.99
2.43
46.3
6.31
4.64
443
0.99
7.75
3.42



Pb In Vitro
Bioaccessibility Value (%)b
10.7
4.6
0.3
2.4
5.2
0.3
6.4
1.8
7.4
4.3 ±3.5
0.3
10.7
 L = sampling location at site, TC = tire crumb piece analyzed from the location.
bThe in vitro bioaccessibility values were determined by dividing the amount of Pb extracted in the in vitro extraction
by the total extractable amount of Pb.
                                               88

-------
                                    Appendix D

                           Air PM10 SEM Analysis Report

           UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                 NATIONAL EXPOSURE RESEARCH LABORATORY
                   Human Exposure and Atmospheric Sciences Division
                           Research Triangle Park, NC 27711


                      SEM Analysis of Tire Crumb Samples

                                      Bob Willis
                                  February 23, 2009
Sample Collection
PMio samples were collected by the EPA on playgrounds and turf fields that used tire crumb
(TC) for a base. Samples were collected at two synthetic turf fields in EPA Regions 4 and 5 and
at one playground in EPA Region 3. Samples were collected on 37mm polycarbonate filters
(0.4 jim pore) using a Harvard Impactor employing a 10 |im inlet. Samples were collected for
approximately 400 min at a flow rate of 10 1pm, giving a total sampled volume of about 4 m3.
Duplicate samples were collected to assess precision and background samples were collected for
comparison from nearby playgrounds/turf fields that did not use tire crumbs.

In addition to ambient samples collected on the playgrounds and turf fields, TC particles
collected from the field's crumb base were provided from each of the three sites to provide
"source profiles"  to assist in identifying TC-related particles in the ambient samples.

Sample Preparation
Source samples:  Individual "crumbs" from the bulk sample, typically a couple of mm in size,
were deposited "as is" on a sticky carbon tab. Source particles closer in size to the ambient
sample were generated by shaving pieces from larger crumbs using a  stainless steel razor blade.
Source samples were coated with -200  A film of conductive carbon to minimize charge build-up
on the sample during SEM analysis.

Air PM10 samples: 5 mm x 5mm sections were cut from each polycarbonate filter using a
stainless steel scalpel. Each section was affixed to a standard 12-mm aluminum specimen stub
using a double-sided sticky carbon tab.  The samples were then coated with -200 A of carbon to
minimize sample charging by the electron beam during SEM analysis.

Sample Analysis
Samples were analyzed by Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray
Spectrometry (EDS) using the Personal SEM (RJ. Lee Instruments Ltd.) in the NERL Electron
Microscopy Laboratory. Manual SEM/EDX analysis was first conducted on the bulk TC source
samples provided. Chemistry and morphological features characteristic of the TC material were
                                         89

-------
identified to help identify TC particles in the ambient samples. Ambient samples were analyzed
by computer-controlled SEM (CCSEM/EDX). Instrument parameters for the CCSEM analyses
included: 20 kV accelerating voltage, Backscattered electron (BSE) imaging mode, 16 mm
working distance, zero tilt. The BSE mode yields a more uniform background than the secondary
electron (SE) mode, necessary for computer-controlled SEM, but at the expense of some loss in
sensitivity for small carbonaceous particles: carbonaceous TC particles about  1 micron or smaller
can be difficult to distinguish from the polycarbonate filter substrate in CCSEM analyses. Thus,
small carbonaceous particles may be under-reported in these analyses.

The CCSEM analysis was set up to analyze particles with average diameters between 1 and
20 |im. Few particles >10 |im, however, were observed in any sample. All particles within this
size range were automatically sized and analyzed by EDS for chemistry. Based on the analyses
of the tire crumb source samples, Sulfur, Zinc, and Carbon were identified as possible indicators
of TC material. Rules were developed to optimize the search for TC-like particles by extending
the  X-ray analysis time (10 s) and saving low-resolution images for all particles containing S,
Zn, or C. Images and spectra for these particle types were manually reviewed  off-line and
particles were subjectively judged o be either TC-like, or not TC material based on the particle
morphology and chemistry.

Only a small fraction of the 6.7 cm2 deposit area of each ambient filter was analyzed by CCSEM,
typically about 1 mm2, to complete  each analysis in a reasonable time.
Following CCSEM analyses, the EDX spectra and images of the particles of interest were
manually reviewed, particles were relocated in the SEM for further examination and suspected
TC  particles were flagged.

Results
Source samples: Figures  1-10 are SEM photos of tire crumb material used at the three sampling
sites. The TC particles examined did not show a single, unique, easily identifiable x-ray
spectrum or particle morphology. Many of the tire crumbs displayed an "exterior"  and an
"interior" surface  (Figs. 1, 2, 3a, 4a). The exterior surface often had a rough, fractured surface
decorated with super-micron crustal-like aluminosilicate particles, quartz, and Fe-rich particles.
(Figs. 3b, 4a). Freshly exposed interior surfaces tend to have a smooth surface embedded with
many sub-micron Zn-rich inclusions (Figs. 4b, 5-7). It is postulated that some of the sub-micron
Zn-rich particles observed in the ambient samples (e.g., Figs. 15, 21, 24, 37, 39, 43, 45, 46) may
have been liberated from the TC matrix as part of the mechanical wear process. Carbon and
sulfur were consistently present in TC particles; zinc was usually, but not always observed in TC
particles (Fig. 8), and often was found at a trace level (Fig. 9). TC particles varied considerably
in morphology such that it is difficult to identify a typical or characteristic TC morphology.
Infrequently, TC source sample particles had the appearance of a bundle of fibers (Fig. 10).

It is questionable whether the morphologies observed in these very large source particles are of
any relevance in identifying PMio TC particles collected on the air filters.

Air PMio samples: Particle loadings on ambient samples were excellent for CCSEM, with
relatively few instances of particles touching or overlapping. Representative field images are
shown in Figs. 11 and 12  (samples from Site Fl Location B and Site Fl Location D,
                                           90

-------
respectively). Figures 13-46 show images and X-ray spectra of a subset of postulated TC
particles found in samples from the three sites.

Table 1 below summarizes the results of the CCSEM analyses of 11 samples. The third column
of Table 1 (#TC) shows the number of TC-like particles identified in each sample after manual
review of the CCSEM data. The area analyzed by CCSEM differed somewhat for each sample,
and the number of TC-like particles reported is not normalized to the area, so these numbers
cannot be strictly compared between samples. The TC mass associated with each particle is
estimated by assuming the particle to be a prolate sphere, calculating the volume from the
projected area of the particle, and assuming a density for each particle. (For particles which are
primarily carbonaceous, a density of 1.5 was assumed. For particles which are mostly
noncarbonaceous, the density is the  average weighted density calculated from the EDS spectrum
where the density of each metal detected in the particle (excluding carbon) is weighted by its
fraction of X-ray counts in the EDS  spectrum). The Scaled TC Mass is the estimated TC mass on
the entire filter, assuming that the area analyzed is representative of the filter as a whole, and that
the exposed filter area is 6.7 cm2. The estimated tire crumb concentration in each sample

   Table 1. Concentration of TC-like particles in ambient samples.
Sample ID
F1D1 Location B
F1D1 Location A
F1 D1 Location A Duplicate
F1D1 Location A Dup
Repeat
F1D1 Location D
(background)
Run#
10C
21D
9D
21B
21C
#TCa
18
5
15
5
7
Mass
TCapg
90
6
80
65
16
Scaled TC
Massb |jg
0.08
0.01
0.02
0.03
0.01
TC
Hg/m3
0.019
0.001
0.004
0.007
0.003

F4 Location B
F4 Location A
F4 Location A Duplicate
F4 Location D
(background)
P1 Location A
P1 Location B
P1 Location D
(background)
28C
28D
28B
28A
22A
16E
16D
6
0
4
6
63C
58d
5
47
0
19
15
3330
2530
40
0.03
0.00
0.01
0.01
3.9
2.1
0.03
0.007
0.000
0.003
0.002
0.77
0.42
0.005
aThese are raw numbers, not normalized to the same analyzed area. (CCSEM areas
analyzed ranged from 0.6 mm2 to 3.6 mm2).
Estimated TC mass scaled to total exposed filter area of 6.7 cm2.
cMass median diameter ~2.6 urn.
dMass median diameter ~2.2 urn.
                                         91

-------
assumes a sample volume of 4 m3. Samples highlighted in yellow (D sites) were collected at
playgrounds or turf fields which did not use TC material, but which were located near the A and
B sites.

There is no independent means or SEM calibration standard to check the accuracy of the SEM
analyses reported in Table 1. These are crude estimates whose uncertainty could be at least a
factor of ±5. (See caveats in Discussion section).

Discussion
Even though the numbers in Table 1 have large uncertainties in absolute terms, the two PI
samples (Locations A and B) collected at a playground using TC material, clearly stand out from
all other samples. These two samples showed much higher number and mass concentrations of
TC-like particles than samples collected at the background sampling location (Location D) and at
all Fl and F4 sampling locations. The TC-like mass concentrations at the two PI sampling
locations were estimated to be 90x and 160x higher than the background sample, with estimated
concentrations of about 0.53 and 0.96 |ig/m3, respectively in the PMi0 size fraction. In eight of
the remaining ten samples, there were < 7 TC-like particles, too few to support any conclusions
regarding differences between the "on field" and background sampling locations. Although the
data suggest possible enrichment of TC-like particles in Fl samples at Locations B and A
(duplicate), an earlier analysis  of the collocated sample from Location A did not show a
significant enrichment in TC mass concentration compared to the background sample. The repeat
analyses of the Location A duplicate sample provides a rough assessment of the overall analysis
precision including the CCSEM analysis and the subjective, manual data interpretation. This
precision is about 50% and is attributed to the small number of candidate particles coupled with
the subjective nature of the particle classification. The estimated mass median diameters of the
TC-like particles in samples in the PI Location A and Location B samples were 2.6 jim (63
particles) and 2.2 jim (58 particles), respectively.

Concern has been expressed about the potential for inhalation exposure to fibers from TC
material.  Very few fibers were observed in any of these samples and none that could be
attributed to TC.

A number of factors may contribute to the observed enrichment in TC-like particles at the PI  site
and the lack of enrichment at the Fl and F4 sites:  (1) The PI site was a playground (as opposed
to a turf field); (2) the PI site may have had more and/or more vigorous activity during the
sampling period than the other sites; (3) the PI sampler may have been located in an area of
unusually high ambient TC concentrations; (4) TC materials vary in composition and mechanical
wear properties. In particular, Zn concentrations in tires can vary nearly a factor of 10: the TC
material used in the PI site may have had elevated Zn compared to the TC material used at the
Fl and F4 sites.  Since the range of Zn concentrations in tires approximates the EDS detection
limit for Zn, any increase in the Zn concentration  of the TC material would have a major effect
on the number of TC-like particles identified by CCSEM. (ICP analyses of the bulk TC samples
should answer this question).

Caveats: The ability to quantify the TC concentration in these samples by SEM/EDS hinges on
the TC particles having distinct and unique composition and/or morphology which would enable
                                          92

-------
the analyst to identify TC particles with a high degree of confidence. This does not seem to be
the case for TC particles, as seen in the variety of morphologies and compositions in Figs.  1-10.
The identification of the TC particles in Table 1 was a subjective judgment and large
uncertainties should be assumed in the estimated TC contributions in Table 1. Reasonable
arguments could be made that the TC concentration estimates in Table 1 are either
underestimates or overestimates. For example, the chemical markers used to identify TC
particles (C, S, and Zn) are not unique to tire crumbs and can be found in particles from common
non-TC sources, potentially resulting in an overestimate. On the other hand, Zn, being the most
critical tracer for TC, is poorly detected by EDS. It is possible that many Zn-bearing TC particles
are undetected in the CCSEM analysis because the Zn concentration is just below the minimum
detectable level, thus resulting in an underestimate of the TC concentration.

As seen in Fig. 11, most particles 1  jim and smaller are sulfate particles whose X-ray spectra
show both C and S (C is contributed by the polycarbonate substrate). Distinguishing TC particles
from these sulfate particles based on chemistry alone is difficult, if not impossible. And, as
discussed above, identifying TC particles by morphology for particles this small is probably not
feasible. However, many, if not most, sulfate particles are damaged by the focused electron beam
which leaves a visible hole in the particle. TC particles are not expected to exhibit similar beam
sensitivity, so this may provide a means of distinguishing TC particles from most sulfate
particles. Isolated Zn-S particles (e.g., Figs. 19 and 20) were observed in most samples and
attributed to TC, even though there  may be industrial sources of ZnS particles other than TC. It is
also possible that TC particles may  be generated from traffic tire wear with airborne
transportation to and across the field and playground sites.
                                           93

-------
          Selected SEM Images and EDS Spectra of Tire Crumb Source Particles

Figure 1. TC particle showing smooth
interior and rough exterior surfaces. Image
was acquired in BSE mode in which particle
brightness increases with atomic number.
(SiteFl)
Figure 2. Rough fractured exterior surface
of TC particle. (Site Fl). The carbonaceous
matrix (darker) is sprinkled with bright
crustal-like particles of aluminosilicates,
 uartz, and Fe-rich.
Figure 3a. Rough exterior surface and
freshly cleaved interior surface. The bright
features on the rough exterior surface are
mostly super-micron crustal particles.
(Site F4).
Figure 3b. The magnified image (upper
right) is the area shown in the square in the
low-mag image at left. The EDS spectrum
was acquired at the spot located by the small
box in the right-hand image (bright,
rectangular particle just below center of
image).  The  particle appears to be an
Fe-rich aluminosilicate crustal particle.
                                           94

-------
Figure 4a. Two TC particles showing
smooth interior surface (upper particle) and
rough exterior surface (lower particle) (Site
PI). The EDS spectrum acquired in the
exterior surface shows a matrix rich in S, Si,
and Al with a trace of Fe and Zn.
 rsonal SEM V4.0£i Sep  5, £088  E.P.fl.
          '      ' 188 urn  £8.8 kV  17 mm  35.EX sp
Figure 4b. Interior surface is much
smoother and is not decorated with super-
micron crustal particles. Spectrum shows
S and trace of Zn.
Figure 5. EDS spectrum was acquired from
the bright, micron-sized Zn inclusion
(zoomed image on right) in the interior
surface of the tire crumb. (Site Fl)
Figure 6. EDS spectrum collected from the
bright, micron-sized Zn inclusion (zoomed
image on right) shows C, S, and Zn.
                                           95

-------
Figure 7. EDS spectra of TC particles
typically show C, S, and Zn. The spectrum
was acquired from the bright inclusion at
center of right-hand image. (Site Fl
Figure 8. EDS spectrum from this TC
particle showed no Zn in the TC matrix.
(Site PI)
Figure 9. EDS spectrum shows C, S and
trace Zn in the interior surface of this TC
particle. (Site PI)
Figure 10. TC particle from shredded tire
sample (Site PI) showing fibrous
morphology. EDS spectrum shows only C.
                                         96

-------
Selected SEM Images and EDS Spectra of Air PM10 Samples.
Figure 11. Representative field image for
Site Fl Location B sample. Magnification =
600x. Tiny sub-micron black dots are pores
in the polycarbonate filter. Sub-micron gray
dots are sulfate particles.
Figure 12. Representative field image for
sample Site Fl Location A Duplicate
sample. Magnification = 600x.
Site Fl Air PMio Samples: Postulated TC Particles
Figure 13. C-S-rich particle. Site Fl
Location B sample.
Figure 14. C-S-Si-rich particle. Site Fl
Location B sample. Note the dull-gray,
micron-sized sulfate particles.
                                          97

-------
Site Fl Air PM10 Samples: Postulated TC Particles
Figure 15. C-Zn-S-rich particle. Site Fl
Location B sample, #224
Figure 16. C-Zn-S-Si-rich. Site Fl Location
B sample, #306
Figure 17. C-S-Si-Zn-rich. Site Fl Location
B sample, #906
Figure 18. C-S-Zn-Si-rich SiteFl Location
B sample, #1060
                                          98

-------
Site Fl Air PM10 Samples: Postulated TC Particles
Figure 19. C-Zn-S-rich particle. Site Fl
Location A Duplicate sample, #179
Figure 20. C-Zn-S-rich particle. Site Fl
Location A Duplicate sample, #794
Figure 21. C-Zn-S-rich particle. Site Fl
Location A Duplicate sample, #1148
Figure 22. C-Fe-Zn-S-rich particle. Site Fl
Location A Duplicate sample, #1241
                                          99

-------
Site Fl Air PM10 Samples: Postulated TC Particles
Figure 23. Site Fl Location A Duplicate
sample, #1471
Figure 24. Site Fl Location A Duplicate
sample, #136
Figure 25. Site F1 Location A Duplicate
sample, #519
Figure 26. Site Fl Location A Duplicate
sample, #153
                                         100

-------
Site PI Air PM10 Samples: Postulated TC Particles
Figure 27. Site PI Location A sample,
#515
Figure 28. Site PI Location A sample,
#1546
Figure 29. Site PI Location A sample,
#1712
Figure 30. Site PI Location A sample, #162
                                        101

-------
Site PI Air PM10 Samples: Postulated TC Particles
Figure 31. Site PI Location A sample,
#1500
DT= 57. CPS= £15 FD=
YFS= 183 (manual)
T = 60  720BX
    X=250  V=
Figure 33. Site PI Location A sample,
#1115
                           Figure 32. Site PI Location A sample, #49
                           Figure 34. Site PI Location A sample,
                           #1537
                                          102

-------
Site PI Air PM10 Samples: Postulated TC Particles
Figure 35. Site PI Location D sample, #834  Figure 36. Site PI Location D sample,
                                        #1065
Figure 37. Site PI Location D sample,
#1328
Figure 38. Site PI Location D sample,
#1370
                                        103

-------
Site F4 Air PM10 Samples: Postulated TC Particles
Figure 39. Fe-Zn-S-rich. Site F4 Location B
sample, #146
Figure 40. Si-Al-Zn-Fe. Site F4 Location B
sample, #761
Figure 41. S-K-Zn-Ca. Site F4 Location B
sample, #1038
Figure 42. Si-Al-Zn-S-Fe. Site F4 Location
B sample, #1561
                                        104

-------
Figure 43. Fe-Zn-S-Mn. Site F4 Location D
sample, #2011
Figure 44. Zn-Fe-S-Si. Site F4 Location D
    le, #2118
Figure 45. C-S. Site F4 Location D sample,
#1480
Figure 46. S-Ti-Fe-Zn. Site F4 Location D
sample, #2262
                                         105

-------




               w
                 •;;.

SCIENCE
 x>EPA
  United States
  Environmental Protection
  Agency
  Office of Research and Development (8101R)
  Washington, DC 20460

  Official Business
  Penalty for Private Use
  $300
PRESORTED STANDARD
POSTAGE & FEES PAID
  EPA
PERMIT NO. G-35
       Recycled/Recyclable Primed on paper ihm contains a minimum of
       50% poslconsurner liber concert! processed chlorine free

-------