United States
Environmental Protection
Agency
Prevention, Pesticides,
and Toxic Substances
EPA747-R-93-005
March 1993
&ERA Evaluation of Asbestos Fiber
Release During Maintenance
of Asbestos-Containing
Floor
-------
EVALUATION OF ASBESTOS FIBER RELEASE
DURING MAINTENANCE OF
ASBESTOS-CONTAINING FLOOR TILE
FINAL REPORT
Chemical Management Division
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, DC 20460
March, 1993
-------
OPPT DISCLAIMER
This report was prepared under contract to an agency of the
United States Government. Neither the United States Government
nor any of its employees, contractors, subcontractors, or their
employees makes any warranty, expressed or implied, or assumes
any legal liability or responsibility for any third party's use
of or the results of such use of any information', apparatus,
product, or process disclosed in this report, or represents that
its use by such third party would not infringe on privately owned
rights.
Publication of the data in this document does not signify
that the contents necessarily reflect.the joint or separate views
and policies of each sponsoring agency. Mention of trade names
or commercial projects does not constitute endorsement or
recommendation for use.
11
-------
AUTHORS AND CONTRIBUTORS
This document was prepared under contracts 68-DO-0061 and
68-DO-0099 for the U.S. Environmental Protection Agency. The.
document was written by Arnold Greenland and David C. Cox with
the assistance of Carolyn Foster and David M. Lawrence, all of
David C. Cox & Associates. The design and conduct of the study
was developed by a working group. The group consisted of:
Margaret Conomos, Dr. Robert Jordan, Betsy Dutrow, and Kin Wong
of the EPA's Office of Pollution Prevention and Toxics; Roger
Wilmoth and Bruce A. Hollett of EPA's Office of Research and
Development; and, Dr. Jean Chesson of Chesson Consulting, Inc.
The field work was performed under EPA contract 68-CO-0006 by PEI
Associates, Inc., under the direction of Mr. John R. Kominsky.
Mr. Pat Clark and Dr. Kim Brackett performed transmission
electron microscopy (TEM) analysis for the field samples.
Material contained in Chapters 3 and 4 was supplied by John
R. Kominsky. Bruce A. Hollett prepared Chapter 5. Substantial
material for this report was also updated from the Quality
Assurance Project Plan prepared by Chesson Consulting, Inc., and
PEI Associates, Inc.
ACKNOWLEDGMENTS
Valuable assistance was also provided by the following
organizations and persons:
Dr. Joseph Breen, EPA
Chemical Specialty Manufacturing Association
Dr. Jay Glatz, EPA
International Sanitary Supply Association
Prince Georges County School System
Resilient Floor Covering Institute
Mr. Joseph Schechter, EPA
Mr. Brad Schultz, EPA .
Ms. Cindy Stroup, EPA
111
-------
CONTENTS
OPPT DISCLAIMER ii
AUTHORS AND CONTRIBUTORS iii
ACKNOWLEDGMENTS iii
ACRONYMS ' viii
EXECUTIVE SUMMARY . . . . .".--.- ix
1.0 INTRODUCTION 1
1.1 BACKGROUND 1
1.2 OBJECTIVES 1
1.3 APPROACH 2
2.0 CONCLUSIONS 5
3.0 STUDY DESIGN 7
3.1 EXPERIMENTAL DESIGN 7
3.2 SAMPLE COLLECTION 8
4.0 FIELD METHODS 11
4.1 SAMPLING AIRBORNE ASBESTOS 11
4.1.1 Air Sampling Strategy 11
4.1.2 Sampling Methodology 14
4.1.3 Sampling Conditions 15
4.2 FIELD DOCUMENTATION AND CONTROL 15
4.2.1 Sample Documentation 15
4.2.2 Traceability Procedures 15
4.3 SPECIALIZED FIELD PROCEDURES 19
4.3.1 Filter Handling Procedures 19
4.3.2 Air Flow Calibration Procedures 19
5.0 LABORATORY METHODS 21
5.1 SAMPLE CONTROL 21
5.1.1 Sample Receipt and Log-In 21
5.1.2 Sample Storage and Disposal 21
5.1.3 Recordkeeping and Filing 21
5.2 LABORATORY PROCEDURES 21
5.2.1 Sample Preparation 21
5.2.2 Analytical Methods 22
5.2.3 Analysis Documentation 22
6.0 QUALITY ASSURANCE 23
6.1 FIELD QUALITY CONTROL CHECKS 23
6.2 LABORATORY QUALITY CONTROL CHECKS 23
6.3 EXTERNAL QUALITY ASSURANCE ANALYSES 24
6.4 QUALITY CONTROL CHECKS FOR DATA PROCESSING .... 25
-------
7.0 STATISTICAL ANALYSIS 27
7.1 INTRODUCTION 27
7.2 DATA ANALYSIS 29
7.2.1 Analysis of ANOVA model assumptions .... 30
7.2.2 Analysis of the basic ANOVA model 32
7.2.3 Analysis of the multiple comparisons .... 35
7.2.4 Other models investigated 36
7.3 FINDINGS 37
REFERENCES * 39
APPENDICES
APPENDIX A. WAX STRIPPING PROCEDURES 41
APPENDIX B. PLOTS OF AIRBORNE ASBESTOS MEASUREMENTS FOR
AREA SAMPLES 45
APPENDIX C. PCM PERSONAL SAMPLE STUDY DATA 53
APPENDIX D. TEM AREA SAMPLE STUDY DATA . 57
APPENDIX E. PERSONAL TEM SAMPLE STUDY DATA 65
TABLES
Table 7-1. Arithmetic mean airborne asbestos
concentrations (s/cc by TEM) for area
samples 27
Table 7-2. Arithmetic mean airborne asbestos
concentrations (f/cc by PCM) for personal
samples 28
Table 7-3. Arithmetic mean airborne concentrations of
asbestos fibers longer than 5 /xm (f/cc by TEM)
for area samples 29
Table 7-4. Analysis of variance table of log airborne
asbestos concentrations (s/cc by TEM) from
area samples 33
Table 7-5. Summary of main effects for log airborne !
asbestos concentrations (s/cc by TEM) from
area samples 34
Table 7-6. Summary of multiple comparisons
(Bonferroni) . . . ; 35
VI
-------
FIGURES
Figure 3-1. Layout of Test Area in Bowie High
School Annex 9
Figure 4-1. Sampling Data Form ' 16
Figure 4-2. PEI Request for Analysis Form 17
Figure 4-3. PEI Chain-of-Custody Form 18
Figure 7-1. Residuals from the ANOVA model for
airborne asbestos concentrations (TEM)
plotted against fitted values 31
Figure 7-2. Probability plot of residuals from ANOVA
model of airborne concentrations
(area samples by TEM) 32
Figure B-l. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the first pair of
rooms 46
Figure B-2. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the second pair of
rooms 47
Figure B-3. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the third pair of
rooms 48
Figure B-4. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the fourth pair of
rooms 49
Figure B-5. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the fifth pair of
rooms 50
Figure B-6. Maximum, minimum, and arithmetic average
airborne asbestos concentrations (TEM) by
method and period for the sixth pair of
rooms 51
VII
-------
ACRONYMS
AHERA Asbestos Hazard Emergency Response Act
ANOVA Analysis of Variance
EDS Energy Dispersive Spectrometry
BED Exposure Evaluation Division
EPA (U.S.) Environmental Protection Agency
f/cc fibers per cubic centimeter
HEPA High Efficiency Particulate Air
HVAC Heating, Ventilation and Air-Conditioning
NIOSH National Institute of Occupational Safety and
Health
NBC National Broadcasting Company.
OSHA Occupational Safety and Health Administration
PEI International Technology Corporation (formerly PEI
Associates, Inc.)
PCM Phase Contrast Microscopy
P.G. Prince Georges (County Public Schools)
QA Quality Assurance
QAPjP Quality Assurance Project Plan
QC Quality Control
RREL Risk Reduction Engineering Laboratory
SAED Selected Area Electron Diffraction
s/cc (asbestos) structures per cubic centimeter
TEM Transmission Electron Microscopy
TWA Time Weighted Average
VAT Vinyl Asbestos Tile
Vlll
-------
EXECUTIVE SUMMARY
In 1990, the U.S.. Environmental Protection Agency (EPA),
with the assistance of the resilient flooring manufacturing and
floor care industries, developed interim guidance (the "EPA
method") on appropriate procedures for maintenance of asbestos-
containing floor coverings (U.S. EPA 1990). The guidance was
provided in response to numerous questions raised by school
districts and building managers regarding the feasibility of
stripping wax from asbestos-containing floor tile. Some of these
questions had been prompted by a limited study conducted by a
local television station (WRC-TV, the Washington, B.C., NBC
affiliate station). Subsequently, a designed experiment was
conducted to evaluate the interim guidance, and to develop
reliable information on airborne asbestos levels during ^stripping
of wax from asbestos-containing floor tile. The experiment
compared the EPA method to a (presumed) more aggressive method in
use by the Prince Georges County, Maryland, public schools (the
"P.G. method"). The study was conducted at the Bowie High School
Annex, in Upper Marlboro, Maryland.
The objectives of the study were to:
1. Compare airborne asbestos levels during vinyl asbestos
tile floor wax stripping by the EPA and P.G. methods to
levels in the room before wax stripping began;
2. Compare the effects of the two wax stripping methods on
airborne asbestos levels during floor stripping and
rebuilding of the wax layer;
3. Compare fiber concentrations measured by Phase Contrast
Microscopy (PCM) during wax stripping with the OSHA
action level of 0.1 f/cc;
4. Compare the effect of the two wax stripping methods on.
residual airborne asbestos levels after the work had
been completed.
TO meet these objectives, a study was designed which took
measurements from both area and personal monitors in six matched
pairs of rooms in the school. For the P.G. method, a 3M black
pad was used while a 3M green pad was used in the EPA method.
The study design called for the same procedures for each type of
pad to allow for comparisons during identical stripping and
rebuilding phases. The design specified a total of 240 area
measurements to be taken and analyzed using transmission electron
microscopy (TEM) techniques. Of the total, 120 measurements were
to be taken for each method, with. 30 each in four periods (before
stripping began, during stripping of wax, during the rebuilding
of the wax layer, and after the process was complete). Of the
240 planned measurements, 237 were actually obtained. A total of
72 personal measurements were taken during the stripping and
rebuilding periods, 36 for each method of wax stripping (3 for
ix
-------
each room). The personal samples, in accordance with OSHA
practice, were analyzed using the NIOSH 7400 Phase Contrast
Microscopy (PCM) protocol. An additional 9 personal samples were
taken during stripping and analyzed by TEM.
The study found no significant difference between airborne
asbestos concentrations, measured by TEM, from area samples for
the P.O. and EPA methods of stripping wax from vinyl asbestos
tile. More specifically, although there was a difference between
the overall mean airborne asbestos concentrations for the two
methods during stripping (0.53 s/cc for the EPA method and 1.18
s/cc for the P.G. method), the difference was not statistically
significant, because of the variability between rooms in the
experiment.
Differences between airborne asbestos concentrations over
the time period of interest '(before stripping began, during
stripping, during rebuilding of the wax layer, and after the
process was complete) were a statistically significant factor in
the study. Specifically, for both the EPA method and the P.G.
method airborne asbestos concentrations during stripping were
significantly elevated as compared to all other times in the
process. There were, however, no significant differences in
concentrations between the other three time periods. This can be
interpreted to mean that there are no short term residual effects
of the stripping process (since airborne levels are not
significantly different before or after wax stripping). Also,
there were no differences in this regard between the two methods
of wax stripping.
The largest of all of the 72 airborne asbestos
concentrations obtained from personal monitors and analyzed by
PCM in the study was 0.056 fibers/cc. These measurements are not
strictly comparable with the OSHA action level of 0.1 f/cc 8 hour
Time Weighted Average (TWA), since the maximum sampling time was
52 minutes. The total time elapsed during the maintenance
process (stripping and rebuilding combined) averaged 1 hour and
35 minutes. The most conservative approach to comparing the
measurements in this study to the OSHA level is to assume the
worker would be exposed at the same levels of asbestos for a full
8-hour shift. This would result in TWA's exactly equal to the
figures reported in this study. Clearly such a situation is the
worst case. Therefore, it was concluded that neither floor
stripping protocol exposed maintenance workers in this study to
TWA airborne concentrations of asbestos exceeding OSHA's action
level.
Although the personal PCM samples were all below the OSHA
action level of 0.1 f/cc, considerably higher exposures are
indicated by the TEM area samples.- This is confirmed by the 9
personal TEM samples taken during stripping, which ranged in
concentration from 0.26 s/cc to 1.49 s/cc. There are two reasons
for this. First, PCM cannot detect fibers thinner than 0.25 /im.
Second, the PCM protocol used in this study does not count fibers
shorter than 5 /xm. The TEM analysis results show that PCM did
-------
not account for all the fibers present in the workplace. For
example, the concentration of fibers longer than 5 /im in the TEM
area samples is 2 orders of magnitude lower than the total
structure concentration for these samples. Thus, caution should
be exercised in interpreting the PCM measurements taken in this
study.
XI
-------
1.0 INTRODUCTION
1.1 BACKGROUND
In 1990, the U.S. Environmental Protection Agency (EPA),
with the assistance of resilient flooring manufacturing and floor
care industries, developed interim guidance (the "EPA method") on
appropriate procedures for maintenance of asbestos-containing
floor coverings (U.S. EPA 1990). The guidance was provided in
response to numerous questions raised by school districts and
building managersregarding the feasibility of stripping wax from
asbestos-containing floor tile. Some of these questions had been
prompted by a limited study conducted by a local television
station (WRC-TV, the Washington, B.C., NBC affiliate station).
Following the airing of the WRC-TV program, a number of
school districts conducted studies of asbestos exposure during
floor-tile maintenance. At least one State placed a ban on
mechanical maintenance of floor tile. In order to evaluate its
own interim guidance, as well as to develop reliable information
on airborne asbestos levels during stripping of wax from
asbestos-containing floor tile, EPA decided to conduct a designed
experiment. The experiment compared the EPA method to a
(presumed) more aggressive method in use by the Prince Georges
County, Maryland, public schools (the "P.O. method"). The study
was conducted at the Bowie High School Annex, in Upper Marlboro,
Maryland.
This report describes the study design, field and laboratory
methods employed, statistical analysis of the data, and
conclusions derived from the study.
1.2 OBJECTIVES
The objectives of the study were to:
1. Compare airborne asbestos levels during vinyl asbestos
tile floor wax stripping by the EPA and P.G. methods to
levels in the room before wax stripping began;
'2. Compare the effects of the two wax stripping methods on
.airborne asbestos levels during stripping and
rebuilding of the wax layer;
i
3. Compare fiber concentrations measured by Phase Contrast
Microscopy (PCM) during wax stripping with the OSHA
action level of 0.1 f/cc;
4. Compare the effect of the two wax stripping methods on
residual airborne asbestos levels after the work had
been completed..
-------
1.3 APPROACH
The study was conducted in a school building with asbestos-
containing floor tile. Site preparation included cleaning,
construction of containment areas, and pre- and post-study
sampling to ensure that the school was returned to the school
district in as good or better condition than at the beginning of
the study.
Two methods of wax stripping were compared in the study:
A method based on EPA's recommended wax stripping
procedure; and
Prince George's County Public Schools' (presumed) more
aggressive .wax stripping procedure.
The methods are described in Appendix A. Since it was not
possible to explore all possible tile maintenance procedures in a
single study, the stripping methods selected were those of
greatest interest to the study participants. EPA's primary
objective was to document the performance of a procedure that is
consistent with the EPA Interim Guidelines, developed with the
assistance of industry. Prince George's County Public School
System was interested in documenting the performance of its own
procedures that are somewhat more aggressive than those
recommended by EPA. Even though the study did not provide
information on every maintenance procedure that might be applied
to floor tile, it may provide a basis for decision-making by
individual building owners, as well as defining areas of future
research.
The experimental design was a randomized block with six
groups (blocks) of two rooms each. The two rooms within any
given group were chosen to be as similar as possible to each
other in terms of floor tile (type, color, wear, etc.), physical
layout (size, fittings, air circulation, etc.), and any other
factor that could affect the outcome of the experiment. The two
treatments were assigned at random to the rooms within a group.
Ideally, both treatments in a given group should be applied
simultaneously. Since this was impossible, the order of
treatments was randomized and both treatments in a given group
were carried out as close in time to each other as possible.
Airborne asbestos concentrations were measured before wax
stripping for each of the two methods under consideration and
while the floor wax stripping was in progress using both area and
personal samples. Passive sampling techniques were employed.
During the wax stripping process, there were two sampling phases:
(1) while wax was being stripped from the floor, and (2) during
rebuilding of a new wax surface. Area samples were analyzed by
transmission electron microscopy (TEM) using a direct transfer
technique (AHERA, 40 CFR Part 763, Appendix A to Subpart E).
-------
While wax stripping was being performed, personal samples
were collected according to OSHA sampling procedures. These
samples were analyzed by Phase Contrast Microscopy (PCM) using
the NIOSH 7400 protocol (Revision 3, June 5, 1989, National
Institute of Occupational Safety and Health Manual of Analytical
Methods). However, an additional 9 personal samples were
analyzed by TEM.
After each treatment was completed, area samples were
collected using a modified aggressive sampling technique to
obtain a measure of residual asbestos levels.
-------
2.0 CONCLUSIONS
The specific conclusions reached, by objective, follow as
"1" through "4". Conclusion 5 addresses the differences between
PCM and TEM measurements taken in the study. Conclusion 6
discusses qualitative differences between the EPA and P.G.
samples observed by the laboratory analysts.
1. Compare airborne asbestos levels during vinyl asbestos tile
floor wax stripping (using either of the two methods for
stripping which are described in Appendix A) with levels in
the room before wax stripping began.
Differences between airborne asbestos concentrations over
the time period of interest (before wax stripping began,
during stripping, during rebuilding of the wax layer, and
after the process was complete) were a statistically
significant factor in the study. Specifically, there was a
statistically significant increase in airborne asbestos
concentrations (for both methods) from before stripping to
during stripping.
2. Compare the effect of each wax stripping method on airborne
asbestos levels during application.
The study found that there was not a significant difference
between airborne asbestos concentrations measured from area
samples for the P.G. method of stripping wax from vinyl
asbestos tile as opposed to the floor stripping method
consistent with EPA interim guidelines.
There was an apparent difference between the overall mean
airborne asbestos concentration for the two methods during
stripping (0.53 s/cc for the EPA method and 1.18 s/cc for
the P.G. method). However, the variability between blocks
of rooms in the experiment forced the conclusion that the
difference in levels was not statistically significant. A
larger study, involving more rooms and, possibly, more
schools, would be needed to determine whether the EPA method
truly results in lower exposures than the P.G. method. Such
a study, if conducted, should consider fewer measurements
per room, as the variability in levels within rooms was
small in the present study. It might; also be useful to
evaluate a less abrasive pad (e.g., a> red pad) than the
green pad used in the EPA method in the present study.
3. Compare fiber concentrations from personal monitors measured
by Phase Contrast Microscopy (PCM) during wax stripping with
the OSHA action level of 0.1 f/cc.
The largest of all of the 72 airborne asbestos
concentrations obtained from personal monitors and analyzed
using Phase Contrast Microscopy (PCM) techniques in the
study was 0.056 fibers/cc which is smaller than the OSHA
-------
action level of 0.1 f/cc. There were therefore no
exceedances of the OSHA action level during this study.
4. Compare the effect of each of the two wax stripping methods
on residual asbestos levels after the work has been
completed.
There was no significant difference, for either of the two
floor stripping methods, between measurements taken before
stripping began and after the process was complete. Thus,
there is no evidence of a short-term residual effect of the
stripping process on airborne asbestos levels.
5. Although the personal PCM samples were all below the OSHA
action level of 0.1 f/cc, considerably higher exposures are
indicated by the TEM area samples.
There are two reasons for this. First, PCM cannot detect
fibers thinner than 0.25 /im. Second, the PCM protocol used
in this study does not count fibers shorter than 5 /xm. The
TEM analysis results show that PCM did not account for all
the fibers present in the workplace. In particular, the
concentration of fibers longer than 5 /zm in the TEM area
samples is generally over 2 orders of magnitude lower than
the total structure concentration. Thus, caution should be
exercised in interpreting the PCM measurements taken in this
study.
6. In most cases, more complex matrix structures, with larger
attached particles and longer fibers and bundles protruding
from the matrix material, were found on the filters from
samples taken during stripping by the P.G. method.
Based on the professional opinion of the laboratory
analysts, chlorine and titanium peaks identified by EDS
indicate a VAT origin for the matrix material from the P.G.
stripping samples. The matrix material often had ill-
defined edges and probably.represented vinyl material
softened and swollen during the preparation of the samples.
While these matrices are not countable under AHERA rules,
they may have originally represented smaller matrices with
exposed asbestos fibers. This deserves further study. In
future experiments, it would also be of value to examine the
surface of VAT floor samples under a scanning electron
microscope before and after the stripping operation to
observe the overall condition of the surface on a
microscopic scale, and to assess the prevalence of exposed
asbestos structures (if such sampling would not create
unacceptable damage to the floor under study).
-------
3.0 STUDY DESIGN
3.1 EXPERIMENTAL DESIGN
The experimental design employed for this study was a
randomized block with six groups (blocks) of two rooms each. A
(complete) randomized block design is one in which each
experimental treatment is applied, in random order, to each
specimen in the test. Because each set of treatments is compared
on the same specimen, the influence of variability between the
specimens on the results is minimized. In this study, a true
randomized block would have involved applying the EPA and PG
methods to the same room. This was not possible for many
reasons, the principal one being that contamination introduced
during the first test in a room could influence the results of
the second test. Therefore, an approximation to a randomized
block was devised, in which, instead of applying the two
treatments to the same room, a matched pair of rooms (called a
block) was used. The assignment of method (PG or EPA) to rooms
within a pair was at random.
The two rooms within any given block were chosen to be as
similar as possible to each other in terms of floor tile (type,
color, wear, etc.), and physical layout (size, fittings, air
circulation patterns, etc.). However, the condition of the wax
was not considered as a test variable, nor was it tested for
asbestos content.
Airborne asbestos concentrations were measured before
application of each method and while the wax stripping was in
progress using area samples and a passive sampling protocol. For
purposes of comparison, the same stripping and rebuilding
protocols were followed for each method (except for the pad
used), regardless of the outcome. For example, the length of
time during stripping was the same for both methods. There were
two sampling phases during wax stripping: (1) while wax was
being stripped from the floor, and (2) during rebuilding of a new
wax surface. Samples were analyzed by Transmission Electron
Microscopy (TEM) using a direct transfer technique (the "AHERA"
protocol).
While stripping and rebuilding were being carried out,
personal samples were collected according to OSHA sampling
protocols. These samples were analyzed by Phase Contrast
Microscopy (PCM) using the NIOSH 7400 protocol. '
After a treatment was completed, area samples were collected
using a modified aggressive sampling technique to obtain a
measure of residual asbestos levels.
-------
3.2 SAMPLE COLLECTION
The study was conducted at the Bowie High School Annex of
the Prince George's County Public School System in Upper
Marlboro, Maryland. The test area consisted of individual
classrooms all located on the ground floor in an unused area of
the building (Figure 3-1). Each classroom contained
approximately 900 square feet of 9" x 9" asbestos-containing
floor tile. Based on samples collected by Prince George's County
Public Schools in December 1988, it was thought that tiles and
mastic from one classroom contained approximately 10 percent and
20 percent chrysotile asbestos, respectively. The samples had
been analyzed using polarized light and dispersion staining
microscopy (40 CFR Part 763, Vol. 52, No. 210). The asbestos
content of the floor tile was investigated further by EPA RREL
prior to the present study. RREL analyzed each type of floor
tile from each classroom and determined that the floor tiles
contained greater than 1 percent asbestos. A confirmatory
analysis of ten tile samples collected by EPA-RREL on May 24,
1990, showed that the floor tile contained 14 to 26 percent
chrysotile asbestos. The EPA-RREL samples were analyzed by TEM
using Chatfield's Method (SOP-1988-02, Revision No. 1: Analysis
of Resilient Floor Tile). Other sources of asbestos in the
building include thermal system insulation on the hot water tank
and boiler, and on pipes, elbows, and valves, all located only in
the boiler room.
The test site was partitioned into three zones to facilitate
air flow and negative-air pressure conditions. These zones were
further partitioned by air-locks created from triple curtained 6-
mil polyethylene doorways. The locations of the air-locks and
high efficiency particulate air (HEPA) filtration systems are
shown in Figure 3-1. The HEPA units were used to maintain a
negative-air pressure (-0.02 inches water gauge) in the study
area during the experiments. The units also served to ventilate
the classrooms after each experiment; the test area was also
ventilated with outside air.
Prior to conducting the experiments, the classrooms were
wet-cleaned and double-flap 6-mil polyethylene curtains were
installed at the entrance to each room. The perimeter wall-
mounted heating, ventilation, and air-conditioning (HVAC) units
were sealed with 6-mil polyethylene. The student lockers in each
corridor and the ceiling vents were also covered with 6-mil
polyethylene.
Details of each method of floor wax stripping are given in
Appendix A. The specifications (e.g., rotational pad speed) of
the floor care machines are also included. The floor care
machine selected for the EPA procedure was randomly selected from
five different machines which all had rotational pad speeds
between 170 to 300 RPM. The wax-stripping chemicals used in the
EPA procedure were selected based on consultation with resilient
floor tile manufacturers.
-------
SITE
ACCESS
ORIGINAL LOCATION
(PRIOR TO FIRST CLEANING)
NOTE:
DOUBLE-CURTAINED
DOORWAYS TO "
CLASSROOMS
LOCKERS IN
CORRIDORS
COVERED WITH
6-mil POLYETHYLENE
CLASSROOM HVAC
UNITS SEALED WITH
6-mil POLYETHYLENE
TRIPLE-CURTAINED
AIRLOCK
119
120
121
EPA LAB
AREA
BOILER
ROOM
HEPA
FILTRATION UNIT
(EXHAUST)
118 L .
A h
117
A
ZONE
US
8
J
113\
D
lie
122
123
114
0
112
B
OPEN
COURT
101
0
102
C
20NE1
CRAWL
SPACE
CRAWL
SPACE
GROUND FLOOR
EXHAUST DUCT
FINAL LOCATION
(AFTER SECOND
CLEANING)
TRIPLE-CURTAINED .
AIRLOCK
HEPA
FILTRATION UNIT
(EXHAUST)
Figure 3-1. Layout of Test Area in Bowie High School Annex,
-------
4.0 FIELD METHODS
4.1 SAMPLING AIRBORNE ASBESTOS
4.1.1 Air Sampling Strategy
Pre-study site evaluation -- Pre-study site evaluation
samples were collected to document the background asbestos
contamination in each classroom. The primary purpose of these
samples was for comparison with the post-study evaluation samples
if airborne asbestos levels at the end of the study were elevated
with respect to outdoor , levels. The sampling was conducted using
modified aggressive sampling techniques, i.e., floors and walls
(up to a height of five feet) of each classroom were swept with
the exhaust of a one-horsepower leaf blower. One stationary fan
(18-inch diameter axial flow) per 10,000. ft2 of floor area was
positioned with the air directed toward the ceiling to maintain
air movement during sampling. One air sample was collected in
each of 15 classrooms and in each of the three corridors. Five
air samples were collected outdoors. Two quality assurance
samples (one closed and one open field blank) were also collected
in each of the three zones, and one open field blank was
collected outdoors. The field blanks were collected in
accordance with the procedures specified in the AHERA final rule.
This yielded a total of 30 prestudy air samples.
Four of the five outdoor samples showed no asbestos
structures; the fifth had a concentration of 0.01 s/cc. However,
the prestudy samples collected in the classrooms and corridors
showed elevated levels of background contamination inside the
test area. The average airborne asbestos concentrations in Zones
1, 2, and 3 (see Figure 3-1) were 0.044, 0.145, and 0.041 s/cc,
respectively. Therefore, the test area was recleaned before
beginning the study. The classrooms and corridors were dry-
vacuumed with a HEPA-filtered vacuum cleaner, and then the floors
were wet-mopped. Three of the six classrooms in Zone 2 were
eliminated' from the study design (rooms 105, 106 and 107) because
of the high airborne asbestos concentrations in these rooms and
in the corridor outside these three rooms (average concentration,
0.231 s/cc). The test area was therefore reduced to two zones by
reconfiguring Zone 3 to include the other three classrooms from
Zone 2 (rooms 109, 110 and 111). Additional samples were then
collected to document background asbestos concentrations in the
test area after the classrooms were cleaned. The sampling was
again conducted using modified aggressive sampling techniques.
Stationary fans were used to maintain air movement during
sampling. One air sample was collected in each of 12 classrooms
and in each of the two corridors. Five air samples were
collected outdoors. One closed and two open field blanks were
also collected in each of the two zones, and one open field blank
was collected outdoors. .. This yielded a total of 26 air samples.
All five outdoor samples showed zero asbestos structures. The
average airborne concentrations in Zones 1 and 3 after the
classrooms were cleaned were 0.011 s/cc and 0.042 s/cc,
respectively. The overall average concentration in the 12 rooms
11
-------
in which the experiment was performed was 0.03 s/cc. However, we
note that the pre-study cleaning did not have a lasting effect:
airborne asbestos concentrations in the 12 rooms, measured by
passive sampling just prior to the start of the experiment,
averaged 0.3. s/cc (see Table 7-1) .
Experiment baseline -- Before each experiment, air samples
were collected to establish a baseline airborne asbestos
concentration in the classroom for comparison with the
concentration measured during the maintenance treatment. These
samples were collected under passive sampling conditions. Five
baseline air samples were collected in each classroom. One open
field blank and one closed field blank were also collected. This
provided 7 baseline samples for each experiment for a total of 84
baseline samples.
Five area air samples were collected to document the
background concentrations of airborne asbestos in the perimeter
of the test area before the study was conducted. These samples
were collected under passive sampling conditions. One open and
one closed field blank were also collected. This provided a
total of 7 baseline perimeter samples.
During experiment -- During each experiment, area air
samples were collected in the classroom, under passive sampling
conditions, for comparison with the experimental baseline
samples. Five area air samples were collected in each classroom
during stripping of the floor wax. One open and one closed field
blank were also collected. This provided 7 area samples during
stripping for each room, a total of 84 samples. The same
sampling plan was followed during rebuilding of the wax layer in
each room, resulting in 84 area samples for the rebuilding period
also. The average sampling time for the area samples was 50
minutes during stripping and 45 minutes during rebuilding.
Additionally, two personal air samples were collected on
each worker performing the floor maintenance treatments. Two
workers (an equipment operator and a helper) performed the
stripping treatments, and one worker performed the wax
refinishing. These samples were collected under passive sampling
conditions. One open and one closed field blank were also
collected. This provided 6 personal samples during each
stripping treatment, and 4 personal samples during each rebuild
treatment, for a total of 120 personal air samples to be analyzed
by Phase Contrast Microscopy using the NIOSH 7400 protocol. An
additional 9 personal samples were collected during stripping and
analyzed by TEM. The sampling times for the personal samples
were slightly shorter than for the area samples, averaging 44
minutes during stripping and 35 minutes' 'during rebuilding.
To document the background concentrations of airborne
asbestos in the perimeter of the test area while the experiments
were being conducted, area air samples were collected during each
of the four days of sampling. The samples were collected under
passive sampling conditions. Five perimeter samples were
12
-------
collected on each of the first two days of the study, whereas two
perimeter samples were collected on each of the third and fourth
day of sampling. Fewer perimeter samples were collected on the
third and fourth day of sampling because three additional pumps
were needed to collect experimental samples during the
maintenance treatments on these days. One open and one closed
field blank were also collected in the perimeter area on each day
of sampling. This provided a total of 22 perimeter samples
during the study.
Post Experiment -- After each experiment, area air samples
were collected to determine if the .maintenance treatments left
the classrooms with an elevated level of airborne asbestos, i.e.,
to establish the residual level of contamination in the
classroom. The sampling was conducted using modified aggressive
sampling techniques, i.e., floors and walls (up to a height of
five feet) of each classroom were swept with the exhaust of a
one-horsepower leaf blower for 20 minutes. Five area air samples
were collected in the classroom after each experiment. One open -
and one closed field blank were also collected after each
experiment. This provided 7 area samples after each experiment.
A closed blank was not collected for one experiment; therefore, a
total of 83 post experiment samples were collected.
Poststudy site evaluation -- Post-study site evaluation
samples were collected to determine if the site (the classrooms
and corridors) was acceptable for reoccupancy. Similar to the
post experiment step, the sampling was conducted using modified
aggressive sampling techniques. One air sample was collected in
each of the 15 classrooms and in each of the corridors of the
original three test zones. Five air samples were also collected
outdoors (all showed zero asbestos structures). One open and one
closed field blank were also collected in each of the three
zones, and one open field blank was collected outdoors. This
yielded a total of 30 poststudy air samples. The average
airborne asbestos concentration for the 18 worksite samples was
0.029 s/cc, which is comparable to the levels obtained in the
pre-study evaluation after re-cleaning of the site.
13
-------
4.1.2 Sampling Methodology
Area air samples -- The area air samples were collected on
open-face, 25-mm diameter, 0.45-/im pore-size, mixed cellulose
ester membrane filters with a 5-/im pore-size mixed cellulose
ester backup diffusing filter and cellulose support pad contained
in a three piece cassette. The filter cassettes were positioned
approximately five feet above the floor with the filter face at
approximately a 45-degree angle toward the floor. The filter
assembly was attached to an electric-powered (110 VAC) 1/6-
horsepower- vacuum pump operating at a flowrate of approximately 9
liters per minute. The sampling pumps were calibrated both
before and after sampling with a calibrated precision rotameter
(Manostat Model 36-54b-215). The precision rotameter is a
secondary air flow standard in that it has mechanical moving
parts; therefore, it was calibrated with a primary air flow
standard. The primary standard used was an electronic bubble
flowmeter (see Section 4.3.2).
The range of air volumes for each type of area sample is
shown below:
Type of area sample
Prestudy site evaluation
Experiment baseline
During experiment
Post experiment
Poststudy site evaluation
Air volume rancre (liters)
1248 to 1610
1225 to 1794
217 to 685
858 to 1898
1758 to 2781
Some of the air volumes in the "During" phase are below the 560
liter minimum required under the EPA interim TEM analytical
method. In such cases, the laboratory analysts compensated by
reading additional grid openings to achieve the desired
analytical sensitivity.
Blanks --At the sampling site, the open blanks were opened
for not more than 30 seconds at the time of sampling. The blanks
accompanied the regular samples through the field operations and
transport to the laboratory and were handled in an identical
fashion. Closed blanks were not opened. Apart from this, they
were treated in the same way as open blanks.
Personal air samples -- Personal samples were collected on
the floor maintenance personnel, i.e., each worker wore a
personal sampling pump with the filter assembly positioned in his
breathing zone area. The samples for analysis by PCM were
collected on open-face, 25-mm diameter, 0.8-/im pore-size, mixed
cellulose ester membrane filters with cellulose support pad
contained in a three piece cassette with a 50-mm conductive cowl.
The filter assembly was attached to a constant-flow, battery-
powered vacuum pump operating at a flowrate of approximately 2
liters per minute. The sampling assembly was worn by the worker
during the entire duration of the maintenance treatment. The
sampling pumps were calibrated before and after sampling by using
a calibrated electronic mass flow meter (Kurz Model 580). The
14
-------
mass flow meter is a secondary air flow standard, therefore, it
was calibrated with a primary air flow standard (refer to Section
4.3.2) .
4.1.3 Sampling Conditions
The environmental conditions (dry bulb temperature and
relative humidity) were measured inside the test area and in the
perimeter area each day during the study. The average dry-bulb
temperature during the week the study was conducted was 86
degrees. The average relative humidity was 75 percent.
4.2 FIELD DOCUMENTATION AMD CONTROL
4.2.1 Sample Documentation
An important part of any field program is the observations
and accurate records of the field team. This information was
recorded (in ink) on bound notebooks and/or data forms. At a
minimum, the following information was recorded on a Sampling
Data Form (Figure 4-1):
Name of person collecting the sample
Experiment number
Date of record
Location of sample
Type of sample (e.g., personal, area, etc.)
Sample number
Rotameter number and reading (start/stop)
Sample time (start/stop).
Relevant notes describing sampling and environmental
conditions, technical problems/solutions, equipment performance,
operator work practices, etc., were recorded in the bound
notebook.
4.2.2 Traceabilitv Procedures
Standard PEI sample documentation procedures were used to
ensure sample traceability. Chain-of-custody procedures
documented the identity of each sample and its handling from its
first existence as a virgin filter until analysis and data
reduction were completed. Chain-of-custody records traced each
sample from its collection until it was transferred to the
analytical laboratory. Internal laboratory records then
documented the custody of the sample through its final
disposition.
Each sample was issued a unique project identification
number. This identification number was recorded on a PEI
Sampling Data Form (Figure 4-1) along with the other information
specified on the form. ..After the labelled sample cassettes were
recovered from the sampling trains, the on-site industrial
hygienist filled out (in ink) a Request for Analysis Form (Figure
4-2) and a Chain-of-Custody Record (Figure 4-3) . This form
accompanied the samples, and each person having custody of the
15
-------
H
CT>
CO
JU
I
a
PJ
ft
{U
^
o
PEI Associates, Inc.
11499 Chester Road
Cincinnati, Ohio 45246
Sampling Oala Form
InveslQalor _
Eipemnenl
Dale: / /
Page ol._. ...
Rolametef No.
Miss Flow Ueuu Mo
Lin. Reg. Equation
Con. Cocll.
Sample descijptkxvtocalion
Sample
numbet
Pump Mo J
T«x>
Rolamelw teading
Start
Slop
Flow tale. \pm
Start
Slop
1*ne
Start
Stop
Ouralon.
mm.
An volume.
liters
Piep»ed by PC I As$otiai«i. Inc . Cncmnah. Ohto
-------
HI AMMUMI. KM.
mount rot ANUVSIS
UT1LWH.ni
Ilt1 T_
MOJCTUMMII.
u« annum
tt»Ll«««»a
-------
PEI Associates, Inc.
CHAMOF-CUSTOOY RECORD
PROJECT MAM E/NUMI
SAMPLE TEAM UEM8
SAU>LE
tuueta
f«s rjjxaeiuwivnii i wr>
IOCA1CH AND KSCAFTIDN
OMEANOTtC
CCLLECTH)
SAUPUE
TYW
OONTAM8I
COWrTONONRECCFT
««*, .-rrBv-nr.*.
POSSIBLE SAMPLE HAZARDS.
SGMATURES: (KAME.COUPAMY.DAT&ANDniE) .
BKOVEDBY:
2.RSJNOUISHEDBY:
RECEDED BY:
«MTE-T«
IRBJCXBSHEOBr: ,
RECSVEDBT:__
RECEINCDBV:
Figure 4-3. PEI Chain-of-Custody Form.
18
-------
samples noted receipt of same and completed an appropriate
section of the form. Samples were hand delivered to the TEM
laboratory. The laboratory's sample clerk examined the shipping
container and each filter cassette for any evidence of damage or
tampering, noted any damage or indication of tampering on the
accompanying chain-of-custody form, and then forwarded the form
to the PEI Project Manager.
4.3 SPECIALIZED FIELD PROCEDURES
4.3.1 Filter Handling Procedures
The following procedures were followed to ensure consistent
handling of all samples collected in the field:
Use of pre-loaded filter cassettes from a lot
prescreened by TEM to minimize the possibility of
contamination.
After sampling, the filter cassettes were sealed and
placed in boxes so that the cassettes would remain in
an upright position. No material other than closed
cassettes was shipped in the boxes.
Hand delivery of all samples to the laboratory for
analysis.
4.3.2 Air Flow Calibration Procedures
The following precautions were followed in the handling of
the air flow equipment employed in the field.
Area samples -- A limiting orifice was used to regulate the
sampling flow rate through the sampling train during sampling.
The air flowrate was determined both before and after sampling by
a calibrated precision rotameter (Manostat Model 36-54b-215).
Personal samples -- The personal samples were collected
using a constant flow (±5%) sampling pump, i.e., the pump has an
in-line feedback control to monitor air flow through the pump
mechanism and to adjust pump speed to maintain constant air flow.
This control system ensures, within limits, constant flow
regardless of load variations or other factors that would
normally change the flow rate. The air flowrate was determined
both before and after sampling by a calibrated electronic mass
flow meter (Kurz Model 580) .
The precision rotameter and electronic mass flow meter were
calibrated in the field using a primary standard airflow
calibrator (Gilibrator brand electronic bubble flowmeter).
19
-------
A detailed record was maintained. It included a written
record (in ink) of all relevant calibration data, including the
following elements:
Rotameter model and serial number.
Sampling device (0.45-/im or 0.8-/xm pore-size, 25-mm
mixed cellulose filter contained in a three piece
cassette) in line during calibration.
X- and Y-coordinate calibration data.
Intercept, slope, and correlation coefficient from a
linear regression analysis of the calibration data.
Linear regression equation to be used to determine a
flowrate.
Dry bulb temperature.
Barometric pressure.
Relevant calculations.
Name of person performing the calibration.
20
-------
5.0 LABORATORY METHODS
5.1 SAMPLE CONTROL
5.1.1 Sample Receipt and Log-In
Samples were received in the laboratory by the Sample
Custodian and checked against the sample receiving form to verify
information. The chain of custody form was checked for
completeness and signed and dated to document receipt. . In the
laboratory, a unique sample identification number -was assigned to
each sample batch and to each sample using a standard protocol.
The sample number was recorded on the sample container
label, and samples were placed in a bag or box labelled with the
sample batch number. The sample numbers were recorded on the
sample receiving form and on the chain of custody form.
Sample numbers were recorded chronologically along with the
date of receipt, project number, the name of the sample project
leader, type of sample media, analysis to be performed, and the
initials of the .sample recipient. The information was
computerized on a WordPerfect chain-of-custody file.
5.1.2 Sample Storage and Disposal
After log-in, the samples, chain of custody form, and sample
receiving form were placed in the prep laboratory. Unused
portions of samples were archived. The disposition of the
samples was noted on the chain-of-custody form.
5.1.3 Recordkeeping and Filing
The completed sample receiving and chain of custody forms
were maintained in a file folder along with all analysis
documents. The folders were stored chronologically by project
number in a file cabinet in the laboratory. Project completion
was documented on the computerized WordPerfect file. A permanent
record of this data will be maintained.
5.2 LABORATORY PROCEDURES
5.2.1 Sample Preparation
The outside surfaces of the cassettes containing samples to
be analyzed were carefully wiped with a lint-free cloth that had
been dampened with particle-free water to remove contamination.
The sample containers were then placed on the clean bench in the
HEPA laminar flow hood.
Portions of the filters were placed on slides and collapsed
with the Chatfield DMF mixture. The slides were placed in sealed
petri dishes and taken to the main laboratory for low temperature
plasma ashing and carbon coating. The coated filters were again
placed in sealed petri dishes and returned to the prep laboratory
21
-------
for placement on grids and dissolution of the filter material on
Jaffe Wicks. Until they were analyzed, the prepared grids were
stored in numbered grid boxes in the main laboratory.
5.2.2 Analytical Methods
The analytical methodology used for TEM examination of the
area samples followed the procedures detailed in 40 CFR Part 763,
AHERA Non-Mandatory Method. The .length and width of all
asbestos-containing structures was also recorded. Clusters and
matrices were sized by measuring the length and diameter of the
longest asbestos fiber within the structure.
r
5.2.3 Analysis Documentation
Identification of Structure by Energy Dispersive
Spectrometrv (EDS) and Selected Area Electron Diffraction (SAED):
EDS and SAED were used to confirm the identification of asbestos
and non-asbestos structures which was noted on the TEM Asbestos
Analysis Data form. The identifications were periodically
documented by printing out EDS spectra. EDS print-outs were
obtained for each type of asbestos present in the sample.
Micrograph and spectrum numbers were recorded on the TEM Asbestos
Analysis Data form.
Data Recording: All analytical data were carefully recorded
so that the details of the analysis were complete and clear.
Counting and sizing data were recorded on "TEM Asbestos Analysis
Data" forms. The original forms were retained in the project
files along with identification validation data such as EDS
spectra, SAED patterns, and photomicrographs. Analysis data was
initialled and dated by the analyst. Data was transferred to
Lotus 1-2-3 files, stored on disk, and entered into the EPA
mainframe computer database.
Results Calculation: Results were calculated according to
the standard AHERA formulae. Calculations were performed by
computer. Hand calculations, computer entries, and data
transfers were double checked by a second analyst. Calculations
were hand checked for at least one out of every hundred samples.
Reporting: Analytical results were reported only if quality
control results were acceptable (see Section 6.0). Reports were
signed and dated by the analyst. The reports were reviewed by
the Chief of the Toxics Control Branch and the RREL QA Manager
before the data was released.
22
-------
6.0 QUALITY ASSURANCE
The Quality Assurance Project Plan (QAPjP) for this study
mandated quality control (QC) checks to be performed for each of
the three key functions within the study: field data collection,
laboratory analysis, and data analysis. This chapter contains a
brief discussion of the QC checks implemented. For details of
the field, laboratory or data analysis methods, the reader is
referred to Chapters 4, 5 and 7 respectively.
6.1 FIELD QUALITY CONTROL PROCEDURES
QC procedures for the field sampling aspects of this study
included the following:
The use of standardized forms (e.g., Figures 4-1 and
4-2), checklists, and field notebooks to ensure
completeness, traceability and comparability of data
and samples collected.
Strict adherence to the sample chain-of-custody
procedures outlined in the.QAPjP for this project.
Selection of sample locations in an unbiased manner as
outlined in the QAPjP.
Field cross checking of data forms to ensure accuracy
and completeness by the field coordinator/supervisor
from the organization responsible for field data
collection.
6.2 LABORATORY QUALITY CONTROL CHECKS
A total of 48 field (open) blanks, one per sampling period
in each room, were analyzed by RREL. In addition, 47 sealed
blanks, one per period per room with the exception of the "after"
period for the"EPA method in the 6th group of rooms, were
analyzed. The blanks were not distinguishable from regular
samples to the analysts. All but two of the 95 blanks showed
zero structures. One field blank and one sealed blank showed a
single asbestos structure in the 10 grid openings examined. A
laboratory blank was added to each group of samples prepared.
One of the 38 lab blanks analyzed contained 2 asbestos structures
in 10 openings, 5 contained 1 structure, and the remaining 32
were free of any structures.
To measure laboratory precision, 3 samples per experimental
group were selected at random and analyzed a second time by RREL.
Two of the three were replicates, that is, repeat counts of the
same preparation. The third sample in each group was a
duplicate, i.e., a new preparation from a different filter
quadrant. Thus, a total of 12 replicates and 6 duplicates were
analyzed. The repeat analyses were carried out blind by the
analysts.
23
-------
For the replicates, 11 of the 12 coefficients of variation
(CV's) for the two analyses were less than 25%. The 12th CV was
130%. All 6 CV's for the duplicates were 20% or less. These
results are well within acceptable ranges for repeat direct TEM
analyses, although it is curious that the CV's for the duplicates
are smaller than those for the replicates.
The total number of TEM analyses performed in the study was
397: 237 regular samples, 48 open blanks, 47. sealed blanks, 18
repeat analyses by the same laboratory, 9 personal samples
analyzed by direct TEM, and 38 laboratory blanks prepared with
the study samples. The pre- and post-study, site evaluations were
additional. '
6.3 EXTERNAL QUALITY ASSURANCE ANALYSES
As a check on the performance of the RREL laboratory,
certain filters were selected for reanalysis by an external
laboratory (RJ Lee Group). Each reanalysis consisted of
preparation and direct TEM analysis of a different filter
quadrant from the one originally analyzed by RREL. Pairwise
comparison of the measured airborne asbestos concentrations was
then used to assess any differences in the analytical performance
of the two laboratories.
The QAPjP called for the selection of 3 samples at random
from each of the 6 pairs of rooms for external QA analysis, for a
total of 18 external QA samples. The actual selection of samples
deviated slightly from the QAPjP. For each combination of 2
methods (EPA, P.G.) and 3 periods (Before, During/Rebuild,
After), 3 external QA samples were selected at random, for a
total of 18. In addition, 4 more samples were selected entirely
at random from the study samples, giving a total of 22 external
QA samples analyzed by RJ Lee Group. Thus, although the sample
selection procedure specified in the QAPjP was not followed,
there is no reason to believe that the external QA samples
actually selected are not adequate to represent the laboratories'
performance.
The airborne asbestos concentrations reported by RJ Lee
Group and RREL are highly correlated (correlation coefficient of
0.96), but the RJ Lee results are generally higher than those of
RREL. Of the 22 samples, 16 have higher concentrations reported
by RJ Lee, a statistically significant result (p = 0.03). A
paired t-test of the difference between the natural logarithms of
the RREL and RJ Lee measurements also indicates a statistically
significant elevation of the RJ Lee measurements (p = 0.003) .
Among the external QA samples, orders of magnitude differences in
airborne asbestos concentrations are present, so that an analysis
on the log scale is most appropriate. This gives an estimate of
the geometric mean ratio, between RJ Lee and RREL measurements of
1.32, with a 95% confidence interval of 1.11 to 1.57. Thus, on
average, the RJ Lee measurement is 32% higher than the RREL
measurement, with 95% confidence bounds of 11% to 57%.
24
-------
The above analysis indicates that there is a measurable, but
small, positive bias of the RL Lee laboratory versus the RREL
laboratory. Relative interlaboratory biases of the magnitude
found are common in asbestos analysis, and can be due to
extremely subtle differences in procedures. Thus, there is no
cause for concern in the results found, and the external QA
analyses are satisfactory.
6.4 QUALITY CONTROL CHECKS FOR DATA PROCESSING
PEI provided information linking sample ID'S with sampling
locations and any other field data relevant to the interpretation
of the results. Laboratory results for each sample ID were also
provided, in computer readable form.
. The field and laboratory data were entered into a computer
data base. Each entry was verified against the information
provided by RREL and PEI. Any discrepancies were documented and
their solution and subsequent correction recorded. A traceable
link was retained between the original data and all data sets
that were created and used for statistical analysis.
25
-------
7.0 STATISTICAL ANALYSIS
7.1 INTRODUCTION
Tables 7-1 and 7-2 contain arithmetic mean airborne asbestos
levels for area (TEM) and personal (PCM) samples respectively.
The concentration values and the number of samples on which each
is based are shown in Table 7-1, broken down by wax stripping
method. The two methods appearing as columns in the table
correspond to the two methods for stripping wax from the vinyl
Table 7-1. Arithmetic mean airborne asbestos concentrations
(s/cc by TEM) for area samples.
EPA Method P.O. Method
Arithmetic
Mean
Before 0.11
During stripping 0.53
During rebuild 0.23
After 0.19
Number
of samples
29
30
29
29
Arithmetic Number
Mean of samples
0.50 30
1.18 30
0.32 30
0.21 30
asbestos tile floors: the method used by Prince Georges County
Public Schools (P.G. method) and a method consistent with EPA
Interim Guidelines (EPA method). The four row headings in the
tables refer to the four stages of the wax stripping process at
which measurements where taken:
before wax stripping
during the actual floor stripping process
during the process of rebuilding the wax layer
after the process was complete.
The mean values shown in Table 7-1 are arithmetic means for the
30 measurements that were taken for each combination of method
and time period (only 29 measurements were available for EPA
"Before", "Rebuild", and "After"). The units of measurement are
asbestos structures per cubic centimeter (s/cc). These
measurements !were obtained using direct Transmission Electron
Microscopy (TEM). The 30 measurements are comprised of five
measurements per room in each of six rooms (there are three
missing values: one in the third room for EPA "Before"; one in
the sixth room for EPA "After"; one in the sixth room for EPA
"Rebuild"). A full description of"the design of the experiment
is given of Chapter 4 of., this report.
Table 7-2; contains arithmetic means, the range of values
measured (minimum to maximum), and the number of samples broken
down by wax stripping method and time period of the airborne
27
-------
asbestos PCM concentrations. These measurements were obtained
from personal monitors worn by the maintenance staff who stripped
and waxed the floors (as discussed in Chapter 5). As data was
collected only during the stripping and rebuilding steps of the
process, only those two time periods are shown in the table.
During stripping each of the two workers wore two personal
monitors each making four measurements in each of six rooms (for
a total of 24) . During rebuilding, there was only one worker,
therefore, the two monitors per room for six rooms makes a total
of 12 measurements per method. The concentrations shown in Table
7-2 are in units of fibers per cubic centimeter (f/cc). These
measurements were obtained using Phase Contrast Microscopy (PCM)
in accordance with the NIOSH protocol 7400.
Table 7-2. Arithmetic mean airborne asbestos concentrations
(f/cc by PCM) of personal samples.
EPA Method P.G. Method
During stripping
Arithmetic mean 0.0070 0.0092
Range 0-0.030 0-0.026
Number of samples 24 24
During rebuild
Arithmetic mean 0.010 0.0070
Range 0-0.056 0-0.025
Number of samples 12 12
In this study, the PCM measurements should be carefully
interpreted. While the personal samples show the largest
concentration to be less than the OSHA action level of 0.1 f/cc,
PCM, as a tool of sample analysis, is likely to miss fibers
released from the tile. The data produced by TEM analysis of
area samples collected during stripping provide evidence that PCM
did not measure all fibers present in the workplace air. On
average, the concentrations reported by TEM analysis during
stripping reached 1.18 s/cc, whereas the highest concentration
reported by PCM analysis was 0.056 f/cc. In addition, the 9
personal samples taken during stripping and analyzed by TEM had
an average concentration of 0.75 s/cc, with a range from 0.26
s/cc to 1.49 s/cc.
An examination of the limitations of PCM may support the
findings. Practically, PCM is capable of resolving fibers of.
width greater than 0.25 /xm. Further, the protocol employed for
PCM analysis in this study counts only fibers greater than 5 /zm
in length. The great majority of the fibers released from
stripping asbestos-containing floor tile were shorter than 5 /xm.
Table 7-3 shows arithmetic mean airborne asbestos concentrations
of fibers longer than 5 jan, broken down by method and period, for
28
-------
the same area samples reported in Table 7-1 (a small number of
amosite fibers longer than 5 p.m have been excluded from these
counts because of their non-VAT origin). The concentration of
fibers longer than 5 /im is typically over two orders of magnitude
lower than the total structure concentration.
Table 7-3. Arithmetic mean airborne concentrations of non-
amosite asbestos fibers > 5 /xm (f/cc by TEM) for area samples.
EPA Method
Arithmetic
Mean
Number
of samples
P.O. Method
Arithmetic
Mean
Number
of samples
Before 0.0000
During stripping 0.0054
During rebuild 0.0031
After 0.0012
29
30
29
29
0.0008
0.0076
0.0010
0.0008
30
30
30
30
7.2 DATA ANALYSIS
The data analysis was designed to satisfy the four study
objectives stated in Section 1.2 of this report. Three of the
research objectives are met by examining comparisons of mean
values obtained from different breakdowns of the data using an
analysis of variance (ANOVA) model. The other goal, relating to
how airborne concentrations obtained from personal monitors
compared to OSHA action levels, has already been discussed. In
order to systematically present an analysis of the remaining
three objectives, this data analysis chapter is structured as
follows:
discussion of the underlying assumptions of the model.
analysis of the high level ANOVA model.
discussion of the multiple comparisons inherent in the
model (among which are the basic study objectives
mentioned in Chapter 2).
discussion of alternative methods of analysis that were
considered.
29
-------
7.2.1 Analysis of ANOVA model assumptions
ANOVA is based upon a linear model of the effects of the
components of the experiment which were varied. In this case,
the experiment consisted of two different types of floor wax
removal, four different times of measurement, and six different
pairs of rooms in which measurements were taken. Each of these
three variables is called a factor in the model. Further, the
model assumes that varying the different factors results in a
linear change in the overall mean of the measurement, in this
case, airborne asbestos concentration. The final aspect of this
model is that the measurements are not used in the scale f/cc or
s/cc directly. As is common with environmental measurements, a
transformation was applied to the data to conform to the
assumptions underlying the ANOVA model.
To use ANOVA models confidently, one must check that certain
basic underlying assumptions of those models are satisfied by the
data under consideration. Consider first the assumptions of
normality of the analysis of variance (ANOVA) model and equality
of the variance of the error terms. As mentioned above, it is
common with environmental measurements that a transformation of
the raw measurements be made so that adherence to the normality
assumption is ensured. Therefore, this data set was transformed
using a natural logarithm transformation.
To check the normality of the resulting residuals, two
diagnostic graphical displays were obtained. Those are shown in
Figures 7-1 and 7-2. Figure 7-1 is a residual plot. It shows
the residuals, the differences between the actual measurements
and the fitted values predicted by the ANOVA model, plotted
against the fitted values. When the model conforms to the
standard assumptions of normality of residuals and equality of
the variance of the error terms, the plot will appear to have no
pattern. The residual values will be clustered about the value
zero with the density of residuals diminishing as the distance
from zero increases (both of which are present in Figure 7-1). A
sign that the model does not conform to the assumption of
equality of error variance is a funnel-shaped appearance of the
residuals, where the funnel tends to increase as the size of the
fitted value increases. There is no evidence in Figure 7-1 of
such behavior. The only aspect of the graphic shown in Figure 7-
1 that requires some discussion is the single value shown at the
bottom of the graph, with an associated residual value of
approximately -1.0. This is the only value that appears to be
out of the cluster of values. Since there is only one such value
and since it corresponds to a error of less than one standard
deviation, we conclude that the assumption of equality of the
error variance, required by ANOVA, is acceptable.
The normality plot shown in Figure 7-2 is a graphical device
to determine whether the assumption of normality of residuals is
satisfied. The closer this graph is to a straight line, the
stronger is the assumption of normality. The normal probability
plot shown in Figure 7-2 is visually linear in appearance. The
30
-------
E '
0.2
0. 1
0.0
-0.1
0.2
0.3
0.4
0.6
-0.8
s .
*
F i tied Value
Figure 7-1. Residuals from the ANOVA model for airborne
asbestos concentrations (TEM) plotted against fitted values.
same residual value mentioned with regard to Figure 7-1 is also
present in Figure 7-2, and it represents the only outlier in the
linear appearance of the normal probability plot. We conclude
that adherence to both of the assumptions of normality and
equality of error terms were sufficiently satisfied so as to
warrant application of the ANOVA model.
31
-------
0.6
0.5
0.4 '
0.3
0.2
0.1-
o.o
-0. 1
-0.2
0.3
0.1
-0.5 -
-0.6
-0.7 -
0.8
-0.9
1.0
I
I
0
NormaIized Rank
Figure 7-2. Probability plot of residuals from ANOVA model of
airborne asbestos concentrations (area samples by TEM).
7.2.2 Analysis of the basic ANOVA model
The study was designed as a randomized block with rooms
nested within the wax stripping method; therefore, the analysis
proceeded by first obtaining the analysis of variance components.
Table 7-4 contains the basic ANOVA table. This table shows the
main effects and all interactions relating to the effect of the
wax stripping method, period of measurement (before wax stripping
began, during stripping, during rebuilding, and after the process
was complete), and the block effect of matched rooms. All
statistical tests relating to the main and interaction effects
inherent in this model can be derived from this table. Following
accepted practice with ANOVA models, one looks first at the main
effects of the model. Table 7-5 contains a summary of both the
32
-------
main and interaction effects. The most important effect to be
considered is that of the wax stripping method, EPA versus P.O.
The results of Table 7-5 reveal that the overall effect was not
significant (with an F-ratio of 1.07 and p-value of 0.35) . The
p-value is interpreted as the probability that, under the
hypothesis of equality of population means, the difference
between means which was encountered (or a .larger difference)
would occur. It is customary to infer that the assumption is
false if the probability of such an occurrence is 0.05 or less.
Since the p-value encountered in this case is 0.35, it is not
possible to make the desired inference; and it is concluded that
there is not a significant quantitative difference between the
two wax stripping methods.
Table 7-4. Analysis of variance table of log airborne asbestos
concentrations (s/cc by TEM) from area samples.
Sum of Mean Square
Source DF Squares Error
Rooms 5 31.757 6.351
(Stripping) Method 1 6.541 6.541
Rooms*Method 5 30.553 6.111
Period 3 92.064 30.688
Rooms*Period 15 83.418 5.561
Method*Period 3 5.794 1.931
Rooms*Method*Period 15 22.678 1.512
This result appears to contradict the large difference shown
in Table 7-1 between the arithmetic means of area measurements of
airborne asbestos obtained during the stripping process for the
EPA versus the P.G method. The explanation is as follows. It
was the goal of this study to infer the existence of differences
between these two methods when applied to any randomly selected
set of rooms in which the wax stripping methods could be applied.
The study used a specific set of rooms. To make an inference to
all rooms, one must use what is called the Random Effects Model
to obtain tests of significance. This model assumes that the set
of rooms used is a sample of rooms'as opposed to the complete set
of all rooms about which., the inference is to be made. Because
the study revealed a large variability between room pairs as
compared to the difference between methods within room pairs, the
tests of significance resulted in F-values which were not
significant. This means that, although the magnitude of the
33
-------
differences revealed in Table 7-1 is substantial, it cannot be
concluded that the result would be significant for a different
selection of rooms.
This study was designed to detect a tenfold difference in the
geometric means of concentrations. The ratio of the geometric
mean of the P.G. measurement during stripping (0.84 s/cc) to the
geometric mean of the EPA measurement during stripping (0.48
s/cc) is less than 2. Therefore, the study was not designed to
find such a difference as significant.
Table 7-5. Summary of main effects for log of airborne asbestos
concentrations (s/cc by TEM) from area samples.
Source
DF
F-Ratio
P-value
Rooms
Method *
Rooms *Met hod
Period *
Rooms*Period
Method*Period *
5
1
5
3
15
3
115
1
111
5
101
1
.60
.07
.22
.52
.21
.28
< 0
0
< 0
0
< 0
0
.0001
.3483
.0001
.0093
.0001
.3180
indicates the use of a fixed effects model to determine
the F-ratio; all other effects are assumed random.
Two comments relating to other values in Table 7-5 are
required. One comment relates to the test whether there is a
difference between mean airborne asbestos concentrations over the
time period of measurement (before wax stripping, during
stripping, during rebuild, and after). As shown in Table 7-5,
this difference was significant at the 0.0093 level even using
the random effects model. This indicates a very clear effect
deriving from when the measurement was taken during the treatment
process. This effect is discussed further in the next
subsection.
A second comment concerns the impact of the difference
between pairs of rooms in the experiment. As Table 7-5 shows,
there was a very strong difference between the room pairs (F =
115.6 and p < 0.0001). This ties together with the inability to
infer a difference between wax stripping methods. The room pairs
were sufficiently different that any difference between the EPA
and P.G. methods which may be present could not be detected.
34
-------
7.2.3 Analysis of the multiple comparisons
Three of the four objectives given in Section 1.2 of this
report can be derived directly by looking at comparisons between
means of subgroups within the context of the study design. Since
it is planned to make numerous comparisons of such means, it is
appropriate to employ standard multiple comparison tests (Miller
1981). The Bonferroni method of multiple comparisons was
selected. This method is applied by allocating a fraction of the
total significance level that is to be used in the totality of
all tests to be made, usually 0.05, to each of the separate
tests. For example, if ten comparisons are to be made, each will
be considered "significant" if a p-value of 0.05/10 or 0.005 or
less is achieved. This method was used to analyze all pairs of
differences between mean asbestos concentrations obtained for wax
stripping methods and time period combinations. Since there were
2 x 4 or 8 wax stripping method by time period combinations,
there are (7 x 8)/2 = 28 different comparisons of means. Using
the Bonferroni test, we counted as significant, any difference
that produced a p-value of less than 0.05/28 = 0.0018.
Table 7-6. Summary of multiple comparisons (Bonferroni).
Hypothesis Test
Significance
EPA strip (0.48) versus EPA before (0.10)
P.G. strip (0.84) versus P.G. before (0.19)
P.G. strip (0.84) versus EPA strip (0.48)
P.G. after (0.16) versus EPA after (0.18)
P.G. after (0.16) versus P.G. before (0.19)
EPA after (0.18) versus EPA before (0.10)
S
S
NS
NS
NS
NS
Note: Geometric mean TEM concentrations shown as (s/cc).
Table 7-6 contains a summary of the outcome of the key
multiple comparison tests that were examined in this analysis.
The table shows the comparison made and whether it was
significant or not. A significant difference is denoted with the
letter, S, while a finding of no significant difference is
labeled, NS. The geometric means in s/cc are also supplied and
are shown in parentheses after each of the comparisons is
described. The key results are the following:
There was a significant difference for both the P.G. and
EPA wax stripping methods between airborne asbestos
. levels before and during the wax stripping. This
35
-------
reflects the existence of the overall significant effect
of time period of treatment in the experiment.
It could not be concluded that there is a significant
difference between the P.G. and EPA methods of treatment
during stripping. The inability to reach this conclusion
for the two comparison tests mirrors the overall
situation for the variable, "wax stripping method". The
variability between rooms overshadowed the difference in
airborne asbestos levels between methods.
For both the P.G and EPA methods, we could not conclude
that there was a difference between airborne asbestos
levels from before the experiment to after the process
was completed. There is no evidence from" this experiment
of a re.sidual effect on airborne asbestos levels from
either the P.G. or the EPA method.
7.2.4 Other models investigated
Although the above discussion concentrated on the main
analysis of the data, a great many different scenarios were
examined during the data analysis phase of the project. The
multiplicity of analyses derived from the need to investigate an
anomaly in the data which occurred in the sixth pair of rooms.
This anomaly can best be understood by examining plots of the
range of airborne asbestos concentrations which are contained in
Appendix B. This appendix contains Figures B-l through B-6 which
graph the arithmetic mean and range of values measured for the
six pairs of rooms in the study.
In the first five pairs of rooms, the pattern over time is
similar. The range of "before" values is generally below the
range of values "during" stripping. The range of values for the
"rebuild" period usually is lower than the range of values
"during" stripping; and, finally, the range of values for
"after" is close to (sometimes higher and sometimes lower than)
the "rebuild" values. The picture in the sixth pair of rooms is
starkly different, as is apparent from Figure B-6. The "before"
values for the rooms in which the P.G. method of wax stripping
was applied are the highest set of measurements made over the
four time periods. Careful review of the sample traceability
documentation maintained during the data collection indicated
that there was no mixup of the samples; however it was decided
that it was worth performing some exploratory analyses to see
whether the sixth pair of rooms needed to be dropped from the
study or whether a reclassification of the data would bring a
different conclusion to the analysis. To deal with this issue,
exploratory analyses were done which looked at the following
configuration:
excluding the sixth pair of rooms from the analysis.
reclassifying the measurements for the sixth pair of
rooms for the P.G. method which were labeled "before" as
"during" and vice-versa.
36
-------
We emphasize there was no intention of reporting reclassified
data. Rather, there was a desire to determine whether the
conclusions of the study were robust, given the possibility of a
problem with the P.G. data from the sixth pair of rooms.
There was a second concern about building an ANOVA model
including the "before" measurements. Since the rooms were
matched and assigned at random to the EPA and P.G. methods, there
is no reason to classify the "before" measurements as being
associated with either the EPA or P.G. methods. Therefore, three
approaches were suggested:
leaving out the before data
analyzing only the "strip" data
using the "before" data as a covariate in the analysis.
The third approach implements an analysis of covariance model
rather than the ANOVA model described above. The idea is that
the concentrations during later periods may be modeled as a
linear function of the "before" measurements. The factors:
method of wax stripping, time period (now excluding the "before"
measurement), and, pairs of rooms are still part of the model.
Such models can be considered as a combination of a linear
regression model which relates the concentration to the "before"
measurement and an ANOVA model which relates the concentration to
the levels of the various factors in the analysis.
All of the above scenarios were investigated. Although the
specific F-ratios and p-values of the various tests varied
somewhat, the general findings were the same and the same general
conclusions were reached. These consistent results were that no
significant difference could be found between levels of airborne
asbestos measured during the EPA and the P.G. methods of wax
stripping, that there was a significant difference between levels
measured during the wax stripping as opposed to other time
periods in the wax process, and that the variability of airborne
asbestos levels measured between pairs of rooms was large in
comparison to the variability between the two methods of wax
stripping.
The consistency of results for the various scenarios
described above lends credibility to the findings of the study.
7.3 FINDINGS
This section summarizes the results discussed above.. The
key findings are: .
(1) The study data revealed no significant difference
between the EPA and P.G.-methods of floor wax
stripping. This finding is driven by the large
variability in airborne asbestos levels between pairs
of rooms in the study. It suggests that, if a future
study is contemplated, resources be concentrated on
increasing the number of rooms included in the study at
37
-------
the expense of a decrease in the number of measurements
per room.
(2) There was a significant difference between measurements
observed during the wax stripping process and
measurements taken at the three other times in the
process. This finding was true for both the P.G. and
EPA methods of wax stripping.
(3) Excluding the "during stripping" period, there were no
other significant differences between period of
measurement within a wax stripping method. In
particular, before and after measurements were not
significantly different, nor were there significant
differences between "during rebuilding" and the
"before" and "after" periods. Therefore, the study
found no evidence of short term residual effects of the
two wax stripping methods.
(4) All 72 of the personal monitoring samples obtained
during stripping and rebuilding and analyzed by PCM in
the study displayed a concentration lower than the OSHA
action level of 0.1 f/cc.
(5) The exposures indicated by the TEM area measurements,
as well as by the limited number of TEM personal
samples, are considerably higher than those shown by
the PCM personal samples. This is because PCM did not
measure the smaller fibers which predominated in this
experiment.
38
-------
REFERENCES
Abler, Roger, 3M Building Service and Cleaning Products Division,
letter to Mr. Bob Jordan of EPA concerning an experiment
measuring asbestos fiber in the air during a floor stripping
procedure.
"Asbestos Floor Tile: The Polish is Off," The Marcor News Vol.
3, No. 1. Article talks about WRC report, Jan-Feb-March, 1990.
Brzezowski, Edward H., "Vinyl Asbestos Floor Tiles: Operations
and Maintenance Field Test Results," ECON: Environmental
Contractor, February, 1990.
Burdett, Garry J., Ainslie, Vicki H., Papanicolopolous, C. D.,
and Smith, Joseph I., "Residue Levels of Asbestos in Buildings
After Asbestos Floor Tile Removal," paper presented at the
National Asbestos Council Conference, San Antonio, TX, February
20, 1990.
Crandlemere, R. Wayne, Kevin P. McCarthy and Amy B. Ginsberg,
Certified Engineering and Testing Company, "Asbestos Floor Tile
Removal Techniques," paper presented at the AAC World Congress
in Las Vegas, NV and appeared in Asbestos Abatement,
September/October, 1988.
ENVIRON Corporation. Report containing data collected by Fowler
Associates with CTC analyses (Fowler-Chatfield Study), letter
sent to Betsy Dutrow, January 12, 1990.
ENVIRON Corporation. "Evaluation of Exposures to Airborne Fibers
During Maintenance of Asbestos Containing Resilient Floor Tiles
Using Recommended Work Practices," report prepared for Resilient
Floor Covering Institute and Armstrong World Industries, Inc.,
September 10, 1990.
ENVIRON Corporation. "Supporting Data for Evaluation of Worker
Exposure to Airborne Fibers During the Removal of Resilient Sheet
Vinyl Floor Covering Using Recommended Work Practices," report
prepared for Resilient Floor Covering Institute and Armstrong
World Industries, Inc., March 30, 1990. (Contains analysis
results from studies conducted in PA, TN, OS, VA, MO, and ID .
ENVIRON Corporation. "Evaluation of Worker Exposure to Airborne
Fibers During the Removal of Asphaltic Cutback Adhesive. Using
Recommended Work Practices," report prepared for Resilient Floor
Covering Institute and Armstrong World Industries, Inc., March
30, 1990.
Fine, David N. S. C. Johnson & Son, Inc., (Johnson Wax), letter
to Mr. Robert Jordan of EPA concerning information about routine
floor maintenance.
39
-------
MacDonald, Hector S. and John Bedore, Jr. of the Brookfield
Academy and the MacDonald Research Group, Inc., "A Comparison of
Methods for Determining the Asbestos Content of Floor Tiles,"
ECON: Environmental Contractor, August, 1989.
MacDonald, H. S., V. M. Kushnaryov and E. Kolinski, "The Release
of Asbestos Fibers from Asbestos Containing Floor Tiles," ECON:
Environmental Contractor, June, 1988.
MacDonald, Hector S., Daniel J.. Grosse, Vladimir M. Kushnaryov
and Scott Kolinski, "Maintenance of Vinyl Asbestos Tile Floors:
The Effect of Floor Finishes on the Release of Asbestos from VAT
Floors," Report andy later, an article in ECON: Environmental
Contractor, July, 1990.
Miller, R. G. Simultaneous statistical inference. Springer-
Verlag, New York, 1981.
New Jersey VAT Study, prepared for Mr. Ed Brzezowski of NJ DOH by
Atlantic Environmental Inc., May 30, 1989.
NIOSH. Method 7400. National Inst. for Occupational Safety and
Health. NIOSH Manual of Analytical Methods. Third Ed., Vol 2.
Cincinnati, OH. U.S. Dept. of Health and Human Services. DHHS
(NIOSH) Publication No. 84-100.
Public Law 99-519. Asbestos Hazard Emergency Response Act of
1986, Sec 2. Amendment to Toxic Substances Control Act, Title II
- Asbestos Hazard Emergency Response. Signed October 22, 1986.
Redden, William D. of the County of Riverside, CA Department of
Health, "For Unified School District: Air Monitoring Study
Conducted During the Cleaning of Asbestos-Containing Floor Tile
at (Blank) High School California," in a letter sent to Bob
Jordan of EPA describing the Riverside County, CA Floor Tile
Study, February 27, 1990.
Shagott, David, "Defining and Selecting Chemical Mastic
Removers," ECON:. Environmental Contractor, February, 1990.
U.S. EPA (1990) . "Interim Guidelines for Maintenance of
Asbestos-Containing Floor Coverings". Office of Toxic
Substances.
Wallace, B. and K. McCollum, "Removal of Asbestos-Containing
Floor Tile Using Infrared Technology," paper given at some
conference (Session 16), not dated, report references NY
Department of Health, Office of Mental Health, Adirondack
Environmental Services, Inc., Calibrations, Inc., Middletown
Psychiatric Center, Middletown, NY..
Ya.mate, G. , S. Agarwal, and R. Gibbons, "Methodology for the
Measurement of Airborne Asbestos by Electron Microscopy", Draft,
U.S. EPA, July 1984.
40
-------
APPENDIX A. WAX STRIPPING PROCEDURES
4i
-------
A.I PRINCE GEORGE'S COUNTY PUBLIC SCHOOLS PROCEDURE FOR
REFINISHING (POLISHING) OF ASBESTOS-CONTAINING RESILIENT FLOORS
FOLLOWING STRIPPING
1. Equipment and Materials
Applicator: 24-ounce cotton string mops. All new mops will
be washed and dried twice prior to use.
2. Floor Wax
Metal Cross Linked Floor Polish containing 18% solids.
3. Procedure
Dry sweep the stripped floor to remove any settled
debris. Do not use a treated mop or sweeping compound.
Apply the polish from a mop bucket. Wring the mop out
until slow dripping wet.
Using a cotton string mop apply the wax as evenly as
possibly covering about 100 square feet per 24-ounce mop
head.
Let each coat dry thoroughly before the next application,
typically 20 to 30 minutes or longer depending on
temperature and humidity. The polish should be dry to
the touch.
Apply two uniform coats of polish.
42
-------
A.2 PRINCE GEORGE'S COUNTY PUBLIC SCHOOLS PROCEDURE FOR
STRIPPING OF ASBESTOS-CONTAINING RESILIENT FLOORS
1. Equipment and Materials
Machine: 300 rpm rotary
Pad: 3M Black
2. Stripping Solution
Ammoniated liquid wax stripper. Dilution - 1 part cleaner :
8 parts water. .
3. Procedure
Sweep floor thoroughly with sweeping compound (for this
study completed during room preparation).
Mix stripping solution in a bucket of water (dilute
1 part cleaner with 8 parts water).
Apply generously with a clean mop and allow to stand on
the floor for at least 5 minutes. Check to be sure the
solution is still standing on the floor. If not, rewet.
Double scrub the floor with a black pad, first in one
direction and then in the opposite direction. Do not
allow the solution to dry.
Use a second bucket and mop with clean water to pickup
the spent stripper solution and polish.
Use a third bucket and mop with clean water to rinse the
floor and remove all residue.
Use a fourth bucket and mop with clean water for a second
rinse.
Repeat the above procedure as.: necessary on spots where
the removal was not adequate.
Allow the floor to dry.
Check floor. If free from residue, proceed to wax. If
not, rerinse, dry, and check again.
43
-------
A.3 U.S. EPA'S PROCEDURE FOR STRIPPING OF ASBESTOS-CONTAINING
RESILIENT FLOORS AT BOWIE HIGH SCHOOL ANNEX IN BOWIE, MARYLAND
1. Equipment and Materials
Machine: 175 rpm
Pad: 3M green
2. Stripping Solution
Ammoniated liquid wax stripper. Dilution - 1 part cleaner :
8 parts water.
3. Procedure
Sweep floor thoroughly with sweeping compound (for this
study completed during room preparation).
Mix stripping solution in a bucket of water (dilute
1 part cleaner with 8 parts water).
Apply generously with a clean mop and allow to stand on
the floor for at least 5 minutes. Check to be sure the
solution is still standing on the floor. If not, rewet.
Double scrub the floor with a green pad, first in one
direction and then in the opposite direction. Do not
allow the solution to dry.
Use a second bucket and mop with clean water to pickup
the spent stripper solution and polish.
Use a third bucket and mop with clean water to rinse the
floor and remove all residue.
Use a fourth bucket and mop with clean water for a second
rinse.
Repeat the above procedure as necessary on spots where
the removal was not adequate.
Allow the floor to dry.
Check floor. If free from residue, proceed to wax. '' If
not, rerinse, dry, and check again.
44
-------
APPENDIX B.
PLOTS OF AIRBORNE ASBESTOS MEASUREMENTS FOR AREA SAMPLES
45
-------
Concentration (Direct TEM) s/cc
1.1-
0.6-
o.o-
g, '
Before
Method:
Strip
Rebui
I
Id After
Per i od
y P.G. i EPA
Figure B-l. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the first
pair of rooms.
46
-------
Concentration (Direct TEM) s/cc
1.421
0,321
0,221
1.121
0.0411
Before
Strip
Rebuild
After
Per i od
Method:
P.G.
EPA
Figure B-2. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the second
pair of rooms.
47
-------
Concentration (Direct TEM] s/cc
1.7
0.9-
0.01
T
Before
Strip
Rebuild
After
Period
Method:
P-G'
EPA
Figure B-3. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the third
pair of rooms.
48
-------
Concentration (Direct TEM)
2.7-
1.8-
0.9-
o-
B ^
s/cc
Before Strip
Method:
r
B
Rebuild After
Per i od
P P.G. J EPA
Figure B-4. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the fourth
pair of rooms.
49
-------
Concentration (Direct TEM) s/cc
3-
2-
1 -
o-
T
Before Strip
Method:
Rebuild
P.G. J
:
After
Per i od
EPA
Figure B-5. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the fifth
pair of rooms.
50
-------
Concentration (Direct TEM) s/cc
2.8-
1.8-
0.8-
o-
-
'
I
5
Before Strip
Method:
f.
B
Rebui Id
P.G. J EPA
r 1
±
After
Per i od
Figure B-6. Maximum, minimum, and arithmetic average airborne
asbestos concentrations (TEM) by method and period for the sixth
pair of rooms.
51
-------
APPENDIX C. PCM PERSONAL SAMPLE STUDY DATA
53
-------
OBS REPNUM
1
2
3
4
5
6
7
8
9
'10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
PERIOD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
METHOD
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G. '
P.G.
EPA
EPA
P.G.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
EPA
EPA
P.G.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
EPA
EPA
P.G.
P.G.
EPA
EPA
EPA .
EPA
P.G.
P.G.
P.G.
P.G.
EPA
EPA
LOCATION
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
111-
111
111
111
102
102
102
102
111
111
102
102
113
113
113
113
101
101
101
101
113
113
101
101
112
112
112
112
109
109
109
109
112
112
109
109
115
115
115
115
114
114
114
114
115
115
SAMPLE
ID
01E-001D2
01E-002D2
01E-003D2
01E-004D2
01P-001D2
01P-002D2
01P-003D2
01P-004D2
01E-001R2
01E-002R2
01P-001R2
01P-002R2
02E-001D2
02E-002D2
02E-003D2
02E-004D2
02P-001D2
02P-002D2
02P-003D2
02P-004D2
02E-001R2
02E-002R2
02P-001R2
02P-002R2
03E-001D2
03E-002D2
03E-003D2
03E-004D2
03P-001D2
03P-002D2
03P-003D2
03P-004D2
03E-001R2
03E-002R2
03P-001R2
03P-002R2
04E-001D2
04E-002D2
04E-003D2
04E-004D2
04P-001D2
04P-002D2
04P-003D2
04P-004D2
04E-001R2
04E-002R2
CONG
(F/CC)
0.0280
0.0136
0.0000
0.0033
0.0000
0.0089
0.0059
0.0000
0.0365
0 .0000
0.0000
0.0245
0.0298
0.0032
0.0065
0.0000
0.0000
0.0213
0.0000
0.0109
0.0000
0.0082
0.0000
0.0000
0.0043
0.0000
0.0000
0.0000
0.0024
0.0097
0.0162
0.0023
0.0555
0.0093
0.0249
0.0249
0.0078
0.0077
0.0000
0.0052
0.0068
0.0045
0.0000
0.0000
0.0056
0.0000
54
-------
OBS REPNUM
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
4
4
5
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
PERIOD
REBUILD
REBUILD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
METHOD
P.O.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
EPA
EPA
P.G.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
EPA
EPA
P.G.
P.G.
LOCATION
ROOM
ROOM
. ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
114
114
117
117
117
117
118
118
118
118
117
117
118
118
110
110
110
110
103
103
103
103
110
110
103
103
SAMPLE
ID
04P-001R2
04P-002R2
05E-001D2
05E-002D2
05E-003D2
05E-004D2
05P-001D2
05P-002D2
05P-003D2
05P-004D2
05E-001R2
05E-002R2
05P-001R2
05P-002R2
06E-001D2
06E-002D2
06E-003D2
06E-004D2
06P-001D2
06P-002D2
06P-003D2
06P-004D2
06E-001R2
06E-002R2
06P-001R2
06P-002R2
CONG
(F/CC)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.0000
.0000
.0193
.0000
.0055
.0000
.0103
.0207
.0207
.0183 [
.0000 .
.0048
.0052
.0051
.0128
.0000
.0124
.0075
.0058
.0058
.0257
.0242
.0000
.0000
.0000
.0000
55
-------
APPENDIX D. TEM AREA SAMPLE STUDY DATA
57
-------
OBS REPNUM PERIOD
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35 "
36
37
38
39
40
41
42
43
44
45
46
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
' 'i
i
2
2
2
2
2
2
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
METHOD
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
LOCATION
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM'
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
111
111
111
111
111
102
102
102
102
102
111
111
111
111
111
102
102
102
102
102
111
111
111
111
111
102
102
102
102
102
111
111
111
111
111
102
102
102
102
102
113
113
113
113
113
101
SAMPLE
ID
01E-001B
01E-002B
01E-003B
01E-004B
01E-005B
01P-001B
01P-002BR
01P-003B
01P-004B
01P-005B
01E-001D1
01E-002D1
01E-003D1
01E-004D1D
01E-005D1
01P-001D1
01P-002D1
01P-003D1
01P-004D1
01P-005D1
01E-001R1
01E-002R1
01E-003R1
01E-004R1
01E-005R1
01P-001R1
01P-002R1
01P-003R1
01P-004R1
01P-005R1
01E-001A
01E-002AR
01E-003A
01E-004A
01E-005A
01P-001A
01P-002A
01P-003A
01P-004A
01P-005A
02E-001B
02E-002B
02E-003B
02E-004B
02E-005BR
02P-001B
SAMPLE
TYPE
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
TOTAL
STRUCTURE
CONC
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(S/CC)
.09440
.10580
.10160
.11330
.09740
.06310
.10730
.11230
.05800
.05360
.71730
.08450
.97100
.80580
.81580
.52160
.49140
.39590
.42200
.30730
.47370
.38130
.34960
.32000
.29750
.20490
.26320
.33140
.39870
.20240
.18870
.14860
.17230
.11940
.14460
.09720
.10970
.07630
.09960
.06760
.12250
.11690
.06950
.08060
.09265
.08690
CONC
FIBERS
> 5 /im
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
(F/CC)
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
007102
000000
000000
011439
007921
000000
000000
014846
000000
005038
000000
009778
020265
015000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
004571
58
-------
OBS REPNUM PERIOD
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
. 2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
BEFORE
BEFORE
' BEFORE
BEFORE
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
STRIP
STRIP
STRIP
METHOD
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
'P.G.
P.G.
.P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
LOCATION
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 113
ROOM 113
ROOM 113
ROOM 113
ROOM 113
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 113
ROOM 113
ROOM 113
ROOM 113
ROOM 113
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 113
ROOM 113
ROOM 113
ROOM 113
. ROOM 113
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 101
ROOM 112
ROOM 112
ROOM 112
ROOM 112
ROOM 109
ROOM 109
ROOM 109
ROOM 109
ROOM 109
ROOM 112
ROOM 112
ROOM 112
SAMPLE
ID
02P-002B
02P-003B
02P-004B
02P-005B
02E-001D1
02E-002D1D
02E-003D1
02E-004D1
02E-005D1
02P-001D1
02P-002D1
02P-003D1
02P-004D1
02P-005D1
02E-001R1
02E-002R1
02E-003R1
02E-004R1
02E-005R1
02P-001R1
02P-002R1
02P-003R1
02P-004R1
02P-005R1
02E-001A
02E-002A
02E-003A
02E-004A
02E-005A
02P-001AR
02P-002A
02P-003A
02P-004A
02P-005A
03E-001B
03E-002B
03E-004B
03E-005B
03P-001B
03P-002B
03P-003B
03P-004B
03P-005B
03E-001D1
03E-002D1
03E-003D1
SAMPLE
TYPE
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
TOTAL CONG
STRUCTURE FIBERS
CONG > 5 Mm
(S/CC)
0.07630
0.07400
0.05330
0.09930
0.26410
0.24945
0.24140
0.27040
0.40900
0.19950
0.19150
0.17680
0.24930
0.15590
0.16190
0.17280
0.16680
0.11010
0.18380
0.07540
0.04060
0.06030
0.06400
0.05580
0.15310
0.14250
0.11120
0.14700
0.11420
0.13000
0.09700
0.13260
0.07090
0.16760
0.17300
0.11730
0.11990
0.14170
0.10400
0.22840
0.12520
0.11930
0.14050
0.44490
0.41360
0.38940
(F/CC)
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.009468
0.004745
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.010110
0.004866
0.000000
59
-------
OBS REPNUM PERIOD
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134 .
135
136
137
138
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE"
BEFORE
BEFORE
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
METHOD
EPA
EPA
P.G.
P.O.
P.G.
P.G.
P.G.
EPA
EPA
EPA
' EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
_EPA
"EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
LOCATION
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
112
112
109
109
109
109
109
112
112
112
112
112
109
109
109
109
109
112
112
112
112
112
109
109
109
109
109
115
115
115
115
115
114
114
114
114
114
115
115
115
115
115
114
114
114
114
SAMPLE
ID
03E-004D1
03E-005D1
03P-001D1
03P-002D1
03P-003D1R
03P-004D1
03P-005D1
03E-001R1D
03E-002R1
03E-003R1
03E-004R1
03E-005R1
03P-001R1
03P-002R1
03P-003R1
03P-004R1
03P-005R1
03E-001AR
03E-002A
03E-003A
03E-004A
03E-005A '
03P-001A
03P-002A
03P-003A
03P-004A
03P-005A
04E-001B
04E-002B
04E-003B
04E-004B
04E-005B
04P-001BR
04P-002B
04P-003B
04P-004B
04P-005B
04E-001D1
04E-002D1
04E-003D1
04E-004D1
04E-005D1
04P-001D1
04P-002D1
04P-003D1
04P-004D1
SAMPLE
TYPE
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
TOTAL
STRUCTURE
CONG
CONG
FIBERS
> 5 fj.m
(S/CC)
0.
0.
1.
1.
1.
1.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
2.
2.
2.
41020
55320
47410
16540
33385
66830
41060
23820
18660
17240
19300
16240
78790
81190
78350
77410
60950
18755
19180
21250
20980
18170
13490
15930
13350
09810
12880
16770
14290
16070
16180
11750
09875
10460
15930
10680
17960
40940
29120
44930
45660
68110
85500
19740
28530
21790
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
(F/CC)
010005
000000
026091
000000
005876
032083
027390
002481
000000
000000
000000
000000
000000
000000
015514
000000
000000
000000
000000
004620
000000
000000
004353
000000
000000
003924
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
009870
000000
010034
007095
000000
043513
000000
020924
'60
-------
OBS REPNUM PERIOD METHOD LOCATION
TOTAL CONG
STRUCTURE FIBERS
SAMPLE SAMPLE CONC > 5 /xrn
ID TYPE (S/CC) (F/CC)
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
4
4
4
4
4
4
4
4
4
4.
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5 .
5
5
5
5
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
114
115
115
115
115
115
114
114
114
114
114
115
115
115
115
115
114
114
114
114
114
117
117
117
117
117
118
118
118
118
118
117
117
117
117.
117
118
118
118
118
118
117
117
117
117
117
04P-005D1
04E-001R1
04E-002R1
04E-003R1
04E-004R1
04E-005R1
04P-001R1
04P-002R1
04P-003R1
04P-004R1
04P-005R1
04E-001A
04E-002A
04E-003AR
04E-004A
04E-005A '
04P-001A
04P-002A
04P-003AD
04P-004A
04P-005A
05E-001B
05E-002B
05E-003B
05E-004B
05E-005B
05P-001B
05P-002B
05P-003B
05P-004B
05P-005B
05E-001D1
05E-002D1
05E-003D1
05E-004D1R
05E-005D1
05P-001D1
05P-002D1
05P-003D1
05P-004D1R
05P-005D1
05E-001R1
05E-002R1
05E-003R1
05E-004R1
05E-005R1
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
. AREA
AREA
AREA
AREA
AREA
AREA
AREA
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
1
2
2
2
0
0
0
0
0
.61260
.04010
.06290
.04240
.06030
.05030
.05510
.04520
.06020
.03520
.03510
.18240
.20520
.21280
.20330
.21610
.06150
.09340
.09655
.09850
.08270
.02920
.09820
.11210
.11290
.09500
.17060
.23590
.23970
.24290
.26440
.47870
.09920
.84880
.73755
.83240
.45880
.94410
.20770
.58075
.76640
.45670
.83770
.37640
.54440
.43530
0.
0.
- 0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
o..
0.
0.
0.
0.
0.
0.
0.
0.
0.
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
005026
000000
000000
000000
004923
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
005039
000000
008488
013746
016163
014443
000000
000000
010417
026098
004434
000000
000000
032664
004891
61
-------
OBS REPNUM PERIOD
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE
BEFORE.
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
REBUILD
AFTER
AFTER
METHOD
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
EPA
EPA
P.G.
P.G.
P.G.
P.G.
P.G.
EPA
EPA
LOCATION
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
118
118
118
118
118
117
117
117
117
117
118
118
118
118
118
110
110
110
110
110
103
103
103
103
103
110
110
110
110
110
103
103
103
103
103
110
110
110
110
103
103
103
103
103
110
110
SAMPLE
ID
05P-001R1
05P-002R1
05P-003R1
05P-004R1
05P-005R1
05E-001A
05E-002A
05E-003A
05E-004AD
05E-005A
05P-001A
05P-002A
05P-003A
05P-004A
05P-005A
06E-001B
06E-002BR
06E-003B
06E-004B
06E-005B
06P-001B
06P-002B
06P-003BD
06P-004B
06P-006B
06E-001D1
06E-002D1R
06E-003D1
06E-004D1
06E-005D1
06P-001D1
06P-002D1
06P-003D1
06P-004D1
06P-005D1
06E-001R1
06E-003R1
06E-004R1
06E-005R1
06P-001R1
06P-002R1
06P-003R1
06P-004R1
06P-005R1
06E-001A
06E-004A
SAMPLE
TYPE
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
AREA
TOTAL
STRUCTURE
CONC
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
2
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(S/CC)
.83140
.64690
.68880
.74930
.68080
.23650
.20810
.23770
.21285
.28770
.22010
.16370
.21200
.19760
.27390
.08630
.13150
.06090
.04830
.07230
.82370
.51480
.78080
.54110
.92950
.40020
.30685
.35850
.25530
.30260
.59040
.64740
.66010
.60600
.74810
.01490
.01010
.01960
.01470
.09630
.05450
.05820
.10550
.03410
.15540
.23900
CONC
FIBERS
> 5 /im
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
(F/CC)
008151
006342
000000
000000
000000
005142
000000
010336
005314
005047
000000
000000
005047
000000
005168
000000
000000
000000
000000
000000
000000
000000
000000
000000
018552
004881
004949
005272
005007
004880
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
000000
62
-------
TOTAL CONG
STRUCTURE FIBERS
SAMPLE SAMPLE CONG > 5 /xm
OBS REPNUM PERIOD METHOD LOCATION ID TYPE. (S/CC) (F/CC)
231 6
232 6
233 6
234 6
235 6
236 6
237 6
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
AFTER
EPA
EPA
P
P
P
P
P
.G.
.G.
.G.
.G.
.G.
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
110
110
103
103
103
103
103
06E-005A
06E-007A
06P-001A
06P-002A
06P-003A
06P-004A
06P-005A
AREA
AREA
AREA
AREA
AREA
AREA
AREA
0
0
0
0
0
0
0
.2507
.2314
.5195
.7073
.5576
.7725
.6458
0
0
0
0
0
0
0
.000000
.000000
.000000
.000000
.000000
.000000
.000000
63
-------
APPENDIX E. PERSONAL TEM SAMPLE STUDY DATA
65
-------
OBS REPNUM
1
2
3
4
5
6
7
8
9
1
2
6
3
4
5
6
6
6
PERIOD
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
METHOD
EPA
EPA
EPA
P.O.
P.G.
P.G.
P.G.
P.G.
P.G.
LOCATION
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
ROOM
111
113
110
109
114
118
103
103
103
SAMPLE
ID
01E-001D2
02E-001D2
06E-003D2
03P-003D2
04P-001D2
05P-002D2
06P-001D2
06P-002D2
06P-003D2
TOTAL
STRUCTURE
CONG
0
0
0
1
1
1
0
0
0
(S/CC)
.57292
.26013
.25585
.43297
.29861
.49079
.52130
.42346
.49087
0
0
0
0
0
0
0
0
0
FIBERS
> 5 (Jim
CONG
(F/CC)
.011458
.000000
.000000
.014188
.000000
.014474
.000000
.014946
.000000
66
------- |