APPLICABILITY OF RCRA DISPOSAL REQUIREMENTS
TO
LEAD-BASED PAINT ABATEMENT WASTES
Final Report
March, 1993
Technical Programs Branch
Chemical Management Division
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, DC 20460
-------
TABLE
OF CONTENTS
Page
EXECUTIVE SUMMARY '. . . !
1.0 INTRODUCTION 4
1.1 Purpose of the Report 4
1.2 RCRA Fackground, 5
1.3 Study Design and Results : e
1.4 Report Organization „ 9
2 . 0 TOXICITY TESTING DATA FROM THE HUD DEMO . 10
3.0 TCLP TESTING PROGRAM FOR SOLIDS
AND PLASTIC SHEETING 14
3 .1 TCLP Testing of Solids , 14
3.2 Results of TCLP Testing of Solids ..... 15
3.3 TCLP Testing of Plastic Sheeting 20
3.4 Analysis of TCLP Testing Results
from Plastic Samples 21
4.0 WASTE DISPOSAL EXPERIENCE FROM THE HUD DEMO 30
5 . 0 CONCLUSIONS 32
APPENDIX A: EP-TOX Testing Data from the HUD Demo,
Sorted by Waste Category and Lead
Concentration in the Extract. 34
APPENDIX B: Solids Samples from the HUD Demo
Analyzed by TCLP. . 38
APPENDIX C: Plastic Samples from the HUD Demo 40
APPENDIX D: Regression Analysis for Plastic Samples... 42
-------
EXECUTIVE SUMMARY
A study was conducted by the U.S. Environmental Protection
Agency's (EPA's) Office of Pollution Prevention and Toxics, in
response to a request from Congress (FY 1990 Appropriations
Conference Report on HUD and Independent Agencies rConference
Report 101-297f p. 30]) that the Agency prepare a report assuring
that Resource Conservation and Recovery Act (RCRA) hazardous waste
requirements would not be applied to debris from lead-based paint
(LBP) abatement projects. (The term "hazardous" is used throughout
this report as a legal definition under RCRA Section 3001 (40CFR
Part 261), not as a qualitative toxicological description). The
study was conducted in three parts. First, data on waste testing
from the Department of Housing and Urban Development's (HUD's)
nationwide abatement demonstration project was evaluated to
determine the hazardousness as defined under RCRA of various
categories of abatement waste. Second, EPA designed and conducted
a detailed testing program for two important categories of waste
not adequately tested in the HUD demonstration (large solid debris,
and protective plastic sheeting) . Third, waste disposal experience
of HUD's contractor for the demonstration project was examined to
obtain preliminary estimates of the volume of hazardous waste
generated, and the cost of disposing of these wastes.
There are three categories of waste produced in large volume during
lead-based paint abatement. These are:
*1. Filtered wash-water;
2. Solid debris, such as old woodwork, plaster, windows,
doors, and similar bulky components;
3. Plastic sheets and tape used to cover floors and other
surfaces.
Filtered wash-water was found to be non-hazardous, and may be
disposed of according to State and local requirements. Solid
-------
debris was generally found to be hazardous only if the lead level
in the paint, as measured in the laboratory by Atomic Absorption
Spectrometry (AAS), exceeded approximately 4.0 milligrams per
square centimeter (mg/cm2). Specifically, 5 of 6 samples tested
(83%) whose paint AAS lead level exceeded 4.0 mg/cm2 failed the
Toxicity Characteristic Leaching Procedure (TCLP) test for lead
toxicity. Conversely, only 1 of 14 (6%) with paint AAS lead level
below 4.0 mg/cm2 failed the TCLP test. Measurements of paint lead
level by field X-Ray Fluorescence (XRF) were poorly correlated with
TCLP results. Due to the limited number and non-random selection
of samples tested, the results for solid debris are suggestive
only, and require confirmation in a larger study before they can be
used as a basis for EPA policy.
Plastic sheeting was found to be hazardous if certain abatement
methods were used. Based on the samples available to us, when a
,heat gun was used for paint removal, the plastic sheeting was
hazardous. However, it should be noted that, because of the
limited availability of samples, all plastic sheeting samples
tested came from a single dwelling; it would be desirable to
confirm the above finding by testing additional plastic samples
from other dwellings where a heat gun was used. When encapsulation
(use of flexible wall covering systems of a reinforced fiber type
that form a secure bond with the =ubstrate) or enclosure (covering
LBP-contaminated surfaces with wood paneling, gypsum board, or
fabricated exterior enclosure systems of aluminum, vinyl or wood)
were used, the plastic was sometimes marginally hazardous. When
other methods were used, in particular removal and replacement of
contaminated components, the plastic was non-hazardous. It should
be noted that, although the above findings are indicative of
whether or not various types of abatement waste are likely to be
hazardous, waste generators are ultimately responsible under EPA
regulations for the proper characterization and disposal of their
waste.
-------
There are several other categories of waste commonly produced
during abatement, such as sludges, paint chips, mops and rags, etc.
Many of these categories, are often hazardous. However, in many
cases the volumes of waste involved may be sufficiently small that
it is cost effective to dispose of them as hazardous, rather than
incurring the expense of testing for lead toxicity. This trade-off
should be made by abatement contractors on a case-by-case basis.
In HUD's demonstration project, an average of 217 ibs of hazardous
waste was generated per housing unit in the three cities for which
data was available, with an average disposal cost of $255 per unit.
These estimates do not include hazardous-waste-related costs
incurred directly by the abatement contractor, such as management
time and TCLP testing costs, They_.are also low to the extent that
large solid debris and plastic sheeting were not generally treated
as hazardous by the contractor. The impact of this cannot be
quantified at present, because records were not kept on volumes of
solid debris and plastic sheeting generated in the HUD
demonstration. The estimates of hazardous waste disposal costs are
high to the extent that unrealistically large amounts of paint were
stripped in the demonstration. This is because the demonstration
was designed to evaluate a range of potential abatement methods,
and not necessarily to mirror -real-world abatement practices.
Because stripping of paint is extremely labor-intensive and costly,
and not often necessary, it is unlikely to be adopted in practice
on a large scale, except for historical properties.
-------
1.0 INTRODUCTION AND METHODS
1.1 PURPOSE OF THE REPORT
Under the Lead-Based Paint Poisoning Prevention Act (LBPPA), as
amended by the Housing and Community Development Act of 1987, and
the Stewart B. McKinney Homeless Assistance Amendment Act of 1988,
the Department of Housing and Urban Development (HUD) is mandated
to require inspection, by 1994, for lead-based paint (LBP) in a
random sample of dwellings and common areas in pre-1978 public and
Indian family housing. Under the statute, paint with lead levels
exceeding 1.0 mg/cm2 constitutes a hazard requiring abatement. In
addition, HUD was required to submit to Congress a "comprehensive
and workable plan" for addressing the LBP hazard in privately-owned
housing (Comprehensive and Workable Plan for the Abatement of Lead-
Based Paint in Privately Owned Housing: A Report to Congressf HUD
(December 7, 1990).
The large scope of the LBP problem, combined with increasing public
health concern about the effects of even low-level lead exposure to
young children, indicates the potential for a major nationwide
abatement effort in the next 10 years. Many LBP abatements,
especially those involving removal and replacement of doors,
windows and trim painted' with LBP, produce large quantities of
solid waste containing lead in varying concentrations. Congress
is concerned that the potential applicability of hazardous waste
requirements under the Resource Conservation and Recovery Act of
1976 (RCRA) to LBP abatement waste may substantially increase the
cost of abatement, and has requested a report from EPA "assuring
that hazardous waste requirements will not be applied to debris
from lead-based paint abatement projects".
The purpose of the present report is -to describe research conducted
by EPA to:
-------
(1) determine which "typical" forms of lead-based paint debris
would require classification as RCRA hazardous waste; and,
(2) provide guidance for persons conducting lead paint
abatements as to which types of waste may be disposed of
as normal construction debris, and which must be presumed
to be hazardous waste unless specifically tested and found
not to meet RCRA criteria.
1.2 RCRA BACKGROUND
RCRA is the basic Federal law governing waste disposal. A key
distinction under RCRA is that between solid waste and hazardous
waste. Solid waste is regulated by the States under RCRA, subject
to minimum Federal standards. By contrast, RCRA establishes a
"cradle-to-grave" system for the management of hazardous waste from
generation to ultimate disposal. Packaging requirements for
hazardous waste under RCRA are described in 40CFR, Parts 173, 178,
179, and 262 Subpart(c) .,
Under RCRA, a waste may be hazardous either because of its
characteristics or because it is specifically listed as hazardous.
Listed hazardous wastes are unlikely to be generated in lead-based
paint abatements. The four, hazardous characteristics are
ignitability, corrosivity, reactivity, and toxicity. Of these>
cprrosivity and toxicity are of most concern in lead paint
abatements. Chemicals used for paint stripping are typically
corrosive. With regard to toxicity, lead is the constituent of
concern in the waste. A waste is defined as exhibiting the
toxicity characteristic for lead if a standard testing procedure
results in the extraction of lead from the waste at a concentration
equalling or exceeding 5 milligrams per liter (parts per million).
This level is 100 times the National Interim Primary Drinking Water
Standard for lead.
-------
Prior to March 1990, the Extraction Procedure Toxicitv Test f EP-
TOX) was used to determine whether a waste was hazardous under
RCRA. This testing procedure was designed to mimic the leaching
action in a landfill. The March 1990 revision of the RCRA toxicity
characteristic (TC) replaced the EP-TOX with the Toxicity
Characteristic Leaching Procedure fTCLP) . The TC became effective
on September 25, 1990. However, small-quantity waste generators
could continue to use the EP-TOX until March 1991.. The TCLP is
considered to be a more reliable and reproducible test than the EP-
TOX.
Under RCRA, generators of waste are allowed to rely on the results
of prior testing or experience, or knowledge of the waste or
process generating the waste, in evaluating their waste to
determine if it is hazardous. Thus, the research reported here is
a first step towards a determination whether the wastes generated
in lead-based paint abatement projects fall under the RCRA
definition of hazardous waste.
1.3 STUDY DESIGN AND RESULTS
The waste samples evaluated in this study, and some of the testing
results, were provided to EPA by HUD from its Demonstration Project
("the Demo") on lead-based paint abatement methods. The HUD Demo
involved the application of a wide range of abatement methods to
vacant Federal .Housing Administration (FHA) housing in 5 cities
nationwide (Washington, D.C./Baltimore; Indianapolis; Denver;
Birmingham AL; and, Seattle/Tacoma). The purpose of the Demo was
to gather reliable data on the cost and applicability of existing
abatement techniques to public and private housing. As required
under RCRA, the contractor for the Demo, Dewberry & Davis,
evaluated waste from the abatements for lead toxicity. Results of
EP-TOX testing conducted in the Demo were made available to EPA,
and used ..as the 'basis for an initial evaluation of the
hazardousness of abatement waste.
-------
The generally non-hazardous categories of waste were: filtered
wash-water; disposable work clothes and respirator filters; and
rugs and carpets. The categories that were mixed, with both
hazardous and non-hazardous samples, were: paint chips; HEPA
vacuum debris, dust from air filters, and paint dust; sludge from
stripping; unfiltered liquid waste; and rags, sponges, mops, HEPA
filters, air monitoring cartridges, scrapers, and other materials
used for testing, abatement, and clean-up. The volume of these
wastes is expected to be relatively small. A cost effective
approach may be to treat all wastes-in the mixed hazard categories
as subject to RCRA requirements; a discussion of the trade-off
involved is presented in Section 2, Finally, two categories,
solid components such as doors and window frames, and, plastic
sheeting used to protect floors and contain dust during abatements,
had insufficient testing data from the Demo.
The second stage of the research was a carefully designed testing
program to evaluate solids and plastic sheeting. These categories
of waste are the highest volume categories produced during
abatement, with the exception of filtered wash-water. Because of
the change from EP-TOX to TCLP, EPA decided to test the solids and
plastic sheeting using the TCLP only.
A total of 30 solids and 30 plastic samples were selected by EPA
from preliminary lists of available waste from HUD Demo sites. A
quota of solids samples was specif ied in each of 4 classes based on
available measurements of paint lead levels taken in the field by
Dewberry & Davis. In this way, a representation of the range of
paint lead levels encountered in practice was obtained. The
plastic sheeting samples were selected to represent both the range
of abatement methods used, and the range of lead levels encountered
in the rooms from which the samples were taken. In practice, it
was not possible to obtain all the samples requested by EPA, due to
the constraints of the Demo schedule. The final sample consisted
-------
of 20 solids and 32 plastic samples, including substitutions made
by HUD. It was not possible to randomly assign samples to dwelling
units. For example, all plastic samples for the "Heat Gun" method
of abatement came from a single dwelling. The confounding of
dwelling unit with abatement method adds an unknown bias to the
study. Nevertheless, EPA evaluated the final set of samples and
concluded that the design objectives had been adequately met for
the study to proceec..
Six out of twenty solid samples (30%) , and twelve out of thirty two
plastic samples had TCLP test results exceeding the RCRA standard
of 5 mg/1 for lead. The data indicated that wood debris samples
are expected to exceed the 5 mg/1 level on the TCLP whenever the
lead level in the paint on the surface exceeds 4 mg/cm2, as measured
by a laboratory test. Paint lead measurements using an X-Ray
Fluorescence detector (XRF), the usual method of field screening
for lead, were not found to reliably predict the TCLP result. For
plastic samples, the data indicated that plastic sheeting used in
abatements conducted by the "Heat Gun" method of paint removal
generally fail, the TCLP test. Some samples from the "Enclosure"
(covering LBP-coritaminated surfaces with wood paneling, gypsum
board, or fabricated exterior enclosure systems of aluminum, vinyl
or wood) and "Encapsulation" (use of flexible wall covering systems
of a reinforced fiber type that form a secure bond with the
substrate) methods produced TCLP results slightly in excess of 5
mg/1, but never above 5.4 mg/1. These marginally hazardous results
do not suffice to definitively determine the hazardous waste status
of plastic sheeting from the "Enclosure" and "Encapsulation"
abatement methods; considerably more data would be required to do
so. Finally, no plastic sheeting samples from the "Chemical
Removal", "Remove/Replace", or "Abrasive Removal" abatement methods
triggered the TCLP test.
-------
The third stage of the research was a brief evaluation of volumes
and disposal costs of hazardous wastes in the HUD Demo. In the 3
cities for which data is available (Birmingham, Denver, and
Seattle/Tacoma), an average of 217 Ibs of hazardous waste was
generated per housing unit, with an average disposal cost of $255
per unit. These per-unit estimates are low to the extent that
large solid debris and plastic sheeting were not generally treated
as hazardous by the contractor. They were disposed of in a solid-
waste landfill and hence no hazardous waste disposal costs were
incurred. The per-unit hazardous waste disposal cost estimates are
high to the extent that far more stripping of paint was carried out
in the Demo than would likely occur in practice (stripping of paint
is extremely labor-intensive, costly, and not often necessary).
1.4 REPORT ORGANIZATION
Section 2.0 of the report presents a detailed discussion of the
analysis of EP-TOX results from the HUD Demo. Section 3.0
describes the follow-on testing program for solids and plastic
sheeting designed and conducted by EPA. Section 4.0 presents a
brief discussion of waste volumes and disposal costs encountered in
the Demo. Section 5.0 presents the conclusions of the study.
-------
2.0 TOXICITY TESTING DATA FROM THE HUD DEMO
As part of the HUD Demo, 80 abatement waste samples from 5 cities
were analyzed for lead by EP-TOX. Appendix A presents the test
data, sorted by waste category and lead concentration in the
extract. For each sample, the following data is given:
1. Numerical sample identifier (SAMPLE ID)
2. Sample description
3. Lead concentration in EP-TOX extract (ppm) (LEAD)
4. Waste category (CAT)
5. City identifier (CITY)
6. Location (address) identifier (LOG).
The waste categories are based on the nine waste categories in
Chapter 11 of "Lead-Based Paint; Interim Guidelines for Hazard
Identification and Abatement in Public and Indian Housing" ("the
Guidelines"), published by HUD in the Federal Register on April 18,
1990. For this report, the waste category "rugs and carpets" has
been added, and the Chapter 11 category "liquid waste" has been
divided into "filtered wash-water" and "unfiltered liquid waste."
Thus, there are eleven waste categories in this report. The
purpose of waste categorization is to facilitate the evaluation of
waste as hazardous/non-hazardous, and to reduce the amount of
testing which needs to be performed. RCRA regulations do not
require testing of waste when prior experience or knowledge of the
generator or disposal firm is sufficient to make a determination of
hazardousness. The use of appropriate waste categories makes
prior testing experience much easier to apply.
Each of the eleven waste categories may be classified in one of
three ways based on the EP-TOX lead testing data from the HUD Demo.
The following waste categories were found to be generally non-
hazardous;
10
-------
a. Filtered wash-water.
b. Disposable work clothes and respirator filters.
c. Rugs and carpets.
Therefore, abatement contractors may dispose of waste in these
categories according to State and local solid waste regulations.
However, contractors should be sure to check with local authorities
before flushing filtered wash-water down storm sewers.
The following categories were found to be hazardous in at least SQ%
of tested cases;
a.
b.
c.
d.
Paint chips.
High Efficiency Particle Air (HEPA) vacuum debris, dust
from air filters, paint dust.
Sludge from stripping.
Unfiltered liquid waste such as wash water from general
cleanup or from decontaminating surfaces after solvents
have been used; unfiltered liquid waste from exterior
blasting.
Rags, sponges, mops, HEPA filters, air monitoring
cartridges, scrapers, and other materials used for
testing, abatement and cleanup.
Abatement-specific conditions such as, for example, the level of
lead in the paint or the matrix holding the lead, will affect
whether or not waste in these categories must be treated as
hazardous waste under RCRA regulations. The abatement contractor,
or disposal firm, may choose to test the specific waste from their
project to determine whether it is hazardous. However, the above
categories will generally contain relatively small volumes of
waste, so that a more cost-effective solution may be to simply
treat all the waste in these categories as subject to RCRA
hazardous waste requirements.
11
-------
The trade-off between TCLP testing and simply disposing of the
waste as hazardous without testing depends on the cost of testing,
the cost of disposal, and the likelihood that the waste will fail
the TCLP test. As an example, EPA's Office of Solid Waste
estimates the cost of a TCLP test as $175. The average cost of
hazardous waste disposal in the HUD Demo was $1.18 per Ib, see
Section 4 of this report. Finally, the probability that the above
wastes will fail the TCLP is approximately 50%. Thus, if W is the
number of Ibs of waste expected in a particular category, TCLP
testing will save money provided 175 + (0.5)*(1.18)*W < 1.18*W,
i.e., provided W >297. Thus, for the hazardous waste disposal
costs experienced in the Demo, the break-even point for testing is
approximately 300 Ibs of waste in a single category (e.g., paint
chips). The break-even point depends on the cost of hazardous
waste disposal. Independent estimates developed by EPA's Office of
Solid Waste (OSW) show a range of disposal costs from $0.18 to
$1.03 per pound for off-site immobilization of the waste followed
by disposal (the most likely disposal method for waste from lead-
paint abatements). These estimates translate to a break-even point
for testing of between 340 pounds and 1940 pounds of hazardous
waste. OSW's disposal cost estimates are lower than the actual
costs experienced by HUD in the Demo. The differences between the
costs are discussed in Section 4.
The category "solvents and caustics" was not tested. However, the
wastes in this category are likely to be hazardous by virtue of
corrosiveness. Quantities of waste in this category are expected
to be relatively small, and again a cost effective solution may be
to treat these wastes as hazardous. The abatement contractor
should evaluate this trade-off on a case-by-case basis, using the
methodology outlined above.
The following categories had insufficient testing information from
the HUD Demo:
12
-------
a.
b.
"Solids", i.e., old woodwork, plaster, windows, doors, and
similar bulky components removed from the building.
Plastic sheets and tape used to cover floors and other
surfaces during LBP removal.
Aside from filtered wash-water, these categories contain the
largest-volume wastes expected to be encountered in LBP abatements.
In practice, the abatement strategy of choice for lead-painted
windows, doors, and wooden trim will often be removal and
replacement with new components, particularly when abatement is
carried out as part of a larger jrenovation project. Only for
houses of historic value is the removal of the paint from these
components by stripping likely to be adopted as an abatement method
on a large scale. Whatever method of abatement is used, large
volumes of plastic sheeting will be used to contain dust and
debris.
Because of the importance of these waste categories, EPA designed
and carried out a testing program, described in the next section of
the report. Since the TCLP replaced the EP-TOX test for all
generators in March 1991, only TCLP testing was conducted on these
additional solid and plastic samples.
13
-------
3.0 TCLP TESTING PROGRAM FOR SOLIDS AND PLASTIC SHEETING
3.1 TCLP TESTING OF SOLIDS
The original design for TCLP testing of solids called for testing
a total of, 30 samples, drawn from specified ranges of XRF lead
levels, as shown in Table I. The goal of this design was to
conduct TCLP testing on samples with a wide range of XRF levels, in
order to facilitate the determination of the XRF level which may
trigger a positive TCLP. Unfortunately, EPA was able to obtain
from the HUD Demo only 20 samples from a total of 10 dwellings, as
described in detail in Appendix Bv The inability to select samples
randomly from dwellings adds an unknown bias to the evaluation.
The selected samples are summarized by XRF level in Table I. Even
though only two thirds of the requested samples were obtained, EPA
decided to proceed with the TCLP testing for two reasons. First,
there was an urgent need for information, even if incomplete, on
the applicability of RCRA requirements to solids. Second, the
available samples were concentrated in the higher ranges of XRF
lead levels which are most likely to trigger a positive TCLP.
TABLE I; DESIGN FOR TCLP TESTING OF SOLIDS
XRF Level (mg/cm2)
1-2
2-4
4-8
>8
N/A
Number of Samples
(Original Design)
5
10
10
5
0
Number of Samples
Available
0
5
8
5
2
Appendix B presents the following information for each sample:
a. Numerical sample identifier
14
-------
b. Unit- address
c. Substrate from which the sample was taken
d. Location of the sample within the unit
e. XRF lead measurement on the substrate from which the
sample was taken (where available).
In 2 cases, no XRF measurement was availab.le. Thus, 18 of the 20
samples have an XRF lead measurement. In addition to the TCLP
test, chemical analysis of a sample of paint from each solid was
conducted to determine the concentration of lead in the paint, both
by weight and on an area basis.
3.2 RESULTS OF TCLP TESTING OF SOLIDS
Table II shows the.TCLP testing results for the 20 solids samples
tested, sorted in decreasing order of the lead concentration in the
TCLP extract. The data is taken,from the report "Hazardous Waste
Support; Report for Work Assignment No. 21. Lead Abatement Waste
Analysis", prepared for EPA's Office of Solid Waste by Science
Applications International Corporation (SAIC) under EPA Contract
No. 68-W9-0011 (July 31, 1990). The first column of the table
shows the sample identifier from Appendix B. The second column,
labelled "LAB LEAD (mg/cm2)", shows the area concentration of lead
in the paint on the sample, in units of milligrams per square
centimeter, as measured in the laboratory. The area of the painted
surface was first measured and then the concentration of lead on
the sample was measured using AAS. For one sample, ID # 105, the
areji concentration is reported as a range. The analytical
laboratory experienced difficulty determining the area of paint on
this sample, due to extreme weathering of the sample. The third
column, labelled "LAB LEAD (mg/kg)», shows the concentration of
lead in the paint by weight, as measured in the laboratory. The
units of this measurement are milligrams per kilogram, i.e., parts
per million. The fourth column of the table, labelled "XRF LEAD
15
-------
(mg/cm2)11, shows the area concentration of lead in the paint, as
measured in the field by an X-Ray Fluorescence instrument. For two
samples, ID #'s 102 and 104, no XRF measurement was available.
Finally, the last column presents the result of the TCLP test on
the sample, as the concentration of lead in the TCLP extract, in
units of milligrams per liter (parts per million) . Two of the
results, for sample ID #«s 102 and 108, were below the detection
limit of the procedure.
TABLE II; TCLP TESTING RESULTS FOR SOLIDS SAMPLES
SAMPLE ID
105
101
112
120
119
107
109
113
111
110
114
118
104
106
116
115
103
117
108
102
LAB LEAD
(mg/cm2)
3.4 - 5.1
9.3
5.3
19.4
19.4
3.9
6.2
0.86
0.72
0.028
2.8
3.9
0.38
3.4
2.9
0.54
2.2
1.1
0.016
0.12
LAB LEAD
(mg/kg)
94800
103000
50400
171000
174000
24300
96500
32700
5360
580
40400""""""
43800
21700
98200
27400
6240
20400
15300
670
2430
XRF LEAD
(mg/cm2)
5.9
6.1
9.7
9.4
9.4
2.8
8
6.8
5.3
8.2
4.6
4.8
N/A
8.7
4.8
3.1
3.7
3.1
3.1
N/A
TCLP LEAD
(mg/1)
21
13
12
10
9.5
5.4
4.8
4.5
4.1
3.4'
2.5
2.3
2.1
1.2
1.0
0.9
0.8
0.4
<0.3
<0.3
16
-------
Six of the twenty samples analyzed (30%) exceeded the RCRA toxicity
characteristic limit of 5 rag/1 lead in the TCLP extract. It is
clear from the table that the higher TCLP results are associated
with higher area concentrations of lead, as measured in the
laboratory. Figure I is a plot of area lead concentration versus
TCLP result. For plotting purposes, the area lead concentration
for sample # 105 has been treated as the average of the reported
range, and the TCLP results below the detection limit of 0.3 mg/1
have been set to 0.3 mg/1. The plot confirms the impression given
by the table. Of the 6 samples with paint AAS lead level exceeding
4.0 mg/cm2, 5 (83%) failed the TCLP. Conversely, only 1 of 14 (6%)
with AAS level below 4.0 mg/cm2 failed TCLP. The data suggests the
following rule Of thumb: TREAT AS HAZARDOUS WASTE ALL SOLID DEBRIS
WITH AREA LEAD CONCENTRATION. AS MEASURED IN THE IABORATORY.
EXCEEDING 4.Q mg/cm2. However, this rule is based on limited data
of doubtful representativeness. It requires confirmation in a
larger study before being used as a basis for EPA policy with
respect to debris from lead paint abatements.
For field testing purposes, it would be very convenient if XRF
measurements of lead concentration, rather than laboratory
measurements, could be used as a predictor of hazardous waste
status. This is because extensive XRF testing is routinely
conducted prior to abatement. Unfortunately, the data reported
here does not indicate a sufficiently strong relationship between
XRF and laboratory lead measurements for practical use. Figure II
is a plot of TCLP result versus XRF lead measurement for the 18
samples for which an XRF measurement was available. Clearly, the
relationship between the two variables is weak. For example, XRF
measurements as high as 8.7 mg/cm2 are associated with low TCLP
results. Conversely, one XRF measurement of 2.8 mg/cm2 has a TCLP
result of 5.4 mg/1. There are two possible causes for the weaker
relationship observed between XRF measurements and the TCLP than
between laboratory measurements and the TCLP. First, XRF
17
-------
TC
25-
20-
15-
10-
5-
o •
XP Cma/D
1
.
I
I
I
1
i
II •
I I III
0 5 10 15 20
AREA LEAD CONCENTRATION Cn>9/cm23
Figure I. Plot of TCLP Result for Solids Samples Against Paint
Lead Concentration on an Area Basis, as Determined by Laboratory
Analysis .
18
-------
TCLP Cmg/ID
25 -
20 •
15
10
~l
A
XRF LEAD
T
6
~T
10
Figure II. Plot of TCLP Result for Solids Samples Against Paint
Lead Concentration on an Area Basis, as Determined by XRF
Analyzer
19
-------
measurements are known to be much less accurate than laboratory
measurements of lead in paint. Second, the XRF measurement
reported for a sample was not necessarily taken on the specific
piece of debris analyzed by the TCLP. For example, the TCLP
analysis for debris from a door might have been conducted on a
piece of the door some distance from where the XRF measurement was
taken. Variations in paint thickness therefore contribute further
to inaccuracies in x.he XRF measurement for our purposes.
3.3 TCLP TESTING OF PLASTIC SHEETING
A large choice of potential samples of 6-mil plastic floor covering
from units in the HUD Demo was made available to EPA. From these,
EPA specified 30 sampling locations from each of which 10 square
feet of plastic, weighing approximately 100 grams, were to be
taken. Of the specified samples, 20 were actually available. An
additional 12 were provided by HUD as substitutes. Appendix C
presents a summary of the plastic samples. For each sample, the
following information is given:
a.
b,
c,
d.
e.
f.
g.
Numerical sample ID (ID)
Address of the unit
Selection code (SEL); ORG = original sample actually
obtained; SUB = HUD substitution; N/A = originally
requested sample which was unavailable
Room from which sample was taken (ROOM)
Average lead level of tested surfaces in the room (mg/cm2)
(MEAN)
Number of surfaces tested in the room (N)
Abatement method.
The specification of the 30 sampling locations was guided by two
considerations. First, EPA wanted to obtain samples from each of
the abatement methods used in the Demo, but especially from
20
-------
chemical removal and heat gun methods, because it was felt that
these were the methods most likely to contaminate the plastic and
trigger a positive TCLP result. Second, EPA wanted to take samples
from rooms with higher levels of lead. The average XRF measurement
taken in the room was used as a proxy for the amount of lead in the
room. Rooms with the highest average level within each abatement
method were specified for sampling.
i
The 32 samples actually obtained represent an acceptable
approximation to the design criteria stated above. Of the
abatement methods, only removal and replacement is poorly
represented. There is adequate representation of the higher room
lead levels, even though the substitutes generally come from rooms
with less lead. However, the abatement method used is confounded
with the dwelling, in the sense that the samples for most methods
come from only one or two dwellings. For example, all heat gun
samples come from the same dwelling. This confounding introduces
an unknown but unavoidable bias into the study results.
*
3.4 ANALYSIS OF TCLP TESTING RESULTS FROM PLASTIC SAMPLES
Table III shows the results of the TCLP testing of the plastic
samples, again taken from the SAIC report previously referenced.
The variables reported are as follows. "ID" is the sample
identifier given in Appendix C. "TCLP" is the result of the TCLP
test procedure, in milligrams of lead per liter in the extract.
"ROOM LEAD" is the arithmetic mean of all XRF lead measurements in
the room from which the plastic sample was taken, in milligrams per
square centimeter. "PRIMARY METHOD" is the primary method of
abatement applied to surfaces in the room, as reported by Dewberry
& Davis, the contractor for the HUD Demo. "SECONDARY METHOD" is
the secondary method of abatement applied, whenever more than one
method was used in the room. The determination of primary versus
secondary method was based on the number of surfaces abated by each
method.
21
-------
TABLE III; TCLP TESTING RESULTS FOR PLASTIC SAMPLES
ID
i
2
3
4
5
$
7
8
1 9
10
11
12
13
14
' 15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
TCLP
fmcr/1)
< 0.3
0.47
5.4
1.2
<0.3
0.41
<0.3
0.67
<0.3
8.4
15
17
28
12
46
40
5.9
3.1
60
0.78
0.44
2,1
0,98
0.67
0.69
8.3
0.41
< 0.3
< 0.3
< 0.3
5.4
0.5
ROOM LEAD
(ma /cm2)
2.6
3.2
5.6
9.4
3.4
8.6
5.4
9.4
9.6
2.6
6.9
7.4
8.7
8.8
8.9
9.3
0.4
1.1
8.3
1.3
1.4
1.5
1.6
1.8
2.8
1.3
2.2
7
2.8
3.6
6.6
2.2
PRIMARY
ABATEMENT METHOD
Encapsulation
Encapsulation
Encapsulation
Encapsulation
Encapsulation
Encapsulation
Abrasive Removal
Abrasive Removal
Abrasive Removal
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Enclosure
Heat Gun
Heat Gun
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Enclosure
Enclosure
Enclosure
Enclosure
Enclosure
Encapsulation
Encapsulation
SECONDARY METHOD
Enclosure
Enclosure
Chemical Removal
Enclosure
Heat Gun
Encapsulation
Remove / Rep lace
Remove /Replace
22
-------
Twelve of the thirty two samples tested (38%) exceeded the TCLP
toxicity characteristic limit of 5 mg/1 for lead. Of these, 8 had
"Heat Gun" as primary method of abatement, 2 had "Enclosure", and
2 had "Encapsulation". The 8 "Heat Gun" samples were the highest
values reported. The higher of the two "Enclosure" samples, with
TCLP level of 8.3 mg/1, had "Heat Gun" as secondary method of
abatement. The 8 "Heat Gun" samples 'and the 2 "Enclosure" samples
all came from the same housing unit, whereas the 2 "Encapsulation"
samples were from distinct units different from the unit for the
other 10. The TCLP result for each of these, 5.4 mg/1, was the
smallest of those exceeding 5 mg/1. The tentative conclusion to be
drawn from the data appears to be that plastic sheeting should be
treated as hazardous waste whenever the "Heat Gun" method of
abatement is used, and that there is a potential for triggering
RCRA hazardous waste requirements when "Enclosure" or
"Encapsulation" are used. Contrary to prior expectation, the
"Chemical Removal" abatement method resulted in very low TCLP
results for the plastic sheeting.
Further analyses were conducted to explore the quantitative
relationship between the level of lead in the room, the abatement
method used, and the result of the TCLP test on the plastic
sheeting. Figure III shows a plot of TCLP result in mg/1 versus
"ROOM LEAD" in mg/cm2. TCLP results below the detection limit of
0.3 mg/1 have been set to 0.3 mg/1. The plot is coded according to
the primary method of abatement: 1 = Abrasive removal; 2 = Chemical
Removal; 3 = Encapsulation; 4 = Heat Gun; 5 = Enclosure. The plot
indicates that the "ROOM LEAD" is a poor predictor of TCLP level.
Accordingly, a more refined measure of lead in the room was
"developed from so-called "Part C» data sheets supplied by Dewberry
& Davis. The Part C sheets provide, for each component abated, the
XRF lead level in the paint (or a laboratory measurement if XRF was
not used), and a measure of the total quantity of paint present, in
either square feet or lineal feet. The Part C data was used to
23
-------
T(
60
str
40"
30"
20"
10"
0"
:LP Cmo/ O
4
1 = Abrasive Removal
2 = Chem i ca I Remova I
3 = Encapsulation
4 = Heat Gun 4
5 = Enc I osure
4
4
4
4
4
....
5.4
5 3 3
4
2 3 2
3222 5 35 335 1 5 3 ^1
I > II I |
3 2 4 6 8 10
MEAN XRF LEAD IN ROOM Cmg/cm2D
Figure III. Plot of TCLP Result for Plastic Samples Against
Average Area Lead Concentration of Abated Surfaces in the Room as
Determined by XRF Analyzer (Coded by Primary Abatement Method)
24
-------
develop an estimate of the total mass of lead abated in each room
by each abatement method. The calculations are described below.
The first step was to convert lineal feet to estimated square feet.
Table IV shows the conversion factors for lineal feet to square
feet developed in discussions with Dewberry & Davis, it should be
emphasized that these conversion factors are approximate. The next
step was to develop an estimate of total mass of lead on each
abated component. The formula used was:
LEAD (kg) = [AREA (sq. ft. )] *[XRF LEAD (mg/ cm2) ] *0. 000929.
In calculating component areas, it was assumed that doors were
painted on both sides, that shelves were painted on one side only
and that 20% of the total area of a window was painted surface!
The estimated mass of lead was then totalled in each room by
abatement method used. Table V shows the results of these
calculations. The columns show the total mass of lead abated in
FACTORS FOR LI
CONVERSION FACTOR
Baseboards
1.0 (Baltimore)
0;33 felsewherel
Window Sills
Window Trim
Stringers
the room by each abatement method. The last column of the table
gives the total mass of lead abated in the room by all methods
used. Figure IV shows a plot of TCLP result versus total mass of
lead abated for the heat gun abatement method. The plot indicates
a strong linear relationship (R* = 0.78) between TCLP result and
25
-------
TCLP
60
50
40
30
10
I
0
r
-i
TOTAL ROOM LEAD
Figure IV. Plot of TCLP Result for Plastic Samples, for "Heat
Gun" Abatement Method, Against Total Mass of Lead Abated in the
Room, as Determined from XRF Measurements and Surface Dimensions
26
-------
total lead abated. Figure V is similar to Figure IV, but includes
only those samples where "Heat Gun" was not the primary method of
abatement. As noted before, abatement methods 3 (encapsulation)
and 5 (enclosure) produce higher TCLP levels than methods 1 and 2,
but there appears to be only a weak relationship to total lead
abated, in contrast to method 4 (heat gun).
To further explore the relationship between mass of lead abated and
TCLP result, several regression analyses were run. The results are
presented in Appendix D. Two of the regression models fitted have
promise as tools for predicting the hazardous waste status of
plastic sheeting used in abatements. However, much further testing
and refinement would be needed before practical application of any
of the regression models.
27
-------
T
10"
8"
6"
4 "
2"
0"
CLP Cn^/ O
1 = Abras I ve Remova I
5
2 = Chem i ca I Remova I
3 = Encapsulation
5 = Enc I osure
5
3 3
•
2
3
212
2^ 3 3 3
3g 1 5 5 3
I I I I I I
0 1 23 4 5
TOTAL ROOM LEAD C^QD
Figure V. Plot of TCLP Result for Plastic Samples Against Total
Mass of Lead Abated in the Room, Determined from XRF Measurements
and Surface Dimensions, for Samples for which Heat Gun was not
the Primary Abatement Method (Coded by Primary Abatement Method)
28
-------
V; MASS OF LEAD ABATED (Ka\ . BY
29
-------
4.0 WASTE DISPOSAL EXPERIENCE FROM THE HUD DEMO
In Birmingham, Denver, and Seattle/Tacoma, the HUD contractor was
required to conduct the EP-TOX test on abatement waste, and treat
as hazardous any waste failing the test. Requirements in other
states may vary since the RCRA program is delegated to States to
administer. Accordingly, individual states may interpret the
regulations somewhat differently in certain situations or have
their own state regulations regarding specific wastes. Further,
municipal regulations may impose additional requirements. The
contractor reports the following summary experience of volumes of
hazardous waste generated and associated disposal costs fThe HUD
Legd-Based Paint Abatement Demonstration (FHA) , HUD, Office of
Policy Development and Research (August 1991)):
TABLE VI; VOLUME OF HAZARDOUS WASTE AND DISPOSAL COSTS FOR THREE
CITIES IN THE HUD DEMO
CITY/(# UNITS)
Birmingham (23)
Denver (57)
Sea/Tac (16)
3 CITIES (96)
QUANTITY OF HAZ
WASTE (Ibs)
TOTAL
11,900
7,350
1,550
20,800
PER UNIT
517
129
97
217
COST OF DISPOSAL
TOTAL
$10,221
$9,625
$4,675
$24,521
PER UNIT
$444
$169
$292
$255
PER LB
$0.86
$1.31
$3.02
$1.18
Thus, the per-unit average in the 3 cities was 217 Ibs of hazardous
waste, with an average disposal cost of $255 per unit. Disposal
cost averaged $1.18 per Ib. These costs include waste stream
analysis by the disposal contractor, pick-up, and disposal but do
not include TCLP testing costs incurred by the abatement contractor
to classify waste, or the cost of abatement contractor management
time devoted to the hazardous waste issue. More detailed cost
breakdowns, e.g., transportation costs, were not available. The
30
-------
hazardous waste disposal costs experienced in the Demo are higher
than independent estimates developed by EPA's Office of Solid Waste
(OSW). For off-site immobilization followed by disposal (the most
likely disposal method for waste from lead-based paint abatements),
OSW estimates a cost of between $0.18 and $1.03 per pound. This is
lower than the average cost of $1.18 experienced in the Demo,
especially when we note that the Demo waste was not treated. The
likely reason for the difference is the small volume of waste
disposed of in the Demo. As Table VI shows, the per-pound disposal
costs in the Demo were inversely related to the volume of waste
disposed of.
Caution should be exercised in extrapolating the experience of the
Demo to larger abatement programs. First, it must be remembered
that solids and plastic sheeting were generally treated as ordinary
solid waste in the Demo. A requirement to treat these items as
subject to RCRA requirements could substantially increase the cost
of disposal. However, no estimates are available to quantify this.
Second, the mix of abatement strategies used in the Demo is
unlikely to reflect future practice. For example, a considerable
amount of paint stripping was conducted in the Demo to test out
different approaches. in practice, however, paint removal is
unlikely, except in special circumstances such as historic
properties, because it is extremely labor-intensive. Thus, future
abatements will probably generate much less sludge from stripping
and heat gun operations than did the Demo. This factor would tend
to reduce hazardous waste disposal costs as compared to the Demo.
31
-------
5.0 CONCLUSIONS
The major conclusions of this study are as follows:
1. Waste from lead-based paint abatements in the categories:
filtered wash-water, disposable work clothes and
respirator filters, and rugs and carpets, is non-
hazardous under RCRA, and may be disposed of as solid
waste.
3.
Waste in the categories: paint chips; HEPA vacuum debris,
dust from air filters, and paint dust; sludge from
stripping; unfiltered liquid waste; and, rags, sponges,
mops, HEPA filters, air monitoring cartridges, scrapers,
and other materials used for testing, abatement, and
cleanup, may be either hazardous or non-hazardous,
depending on abatement conditions. Waste in these
categories may either be tested to determine whether it
is hazardous, or may be disposed of as hazardous waste
without testing. Since relatively small volumes of waste
are expected in these categories, disposal as hazardous
waste may be the cost-effective solution for the
generator, depending on the cost of hazardous waste
disposal, the volume of waste involved, the cost of TCLP
testing, and the estimated probability that the waste
will fail the TCLP test anyway.
Waste in the "solids" category, i.e., old woodwork,
plaster, doors, and similar bulky components removed
during abatement, was found to be generally hazardous
when the lead level in the paint exceeded 4 milligrams
per square centimeter, as determined by a laboratory
analysis. This suggests that waste with paint at lower
lead levels may be disposed of as solid waste. However,
this study examined only a limited number of solid debris
32
-------
samples selected in a non-random fashion. Thus, the
results are suggestive only, and require confirmation in
a larger study before being used as the basis for EPA
policy with respect to waste from lead paint abatements.
Field X-Ray Fluorescence (XRF) measurements of lead in
paint are not sufficiently accurate to permit an accurate
determination of the hazardous waste status of "solids".
Waste in the category of plastic sheeting and tape used
to cover floors and other surfaces during abatement is
hazardous when the "Heat Gun" method of paint removal is
used. However, this conclusion is based on a set of
samples all taken from a single dwelling. When "Chemical
Removal", "Abrasive Removal", or "Remove/Replace" are
used, plastic sheeting is not hazardous. When
"Encapsulation" or "Enclosure" are the abatement methods
used, plastic sheeting sometimes slightly exceeds the
TCLP limit for lead in waste.
The total quantity of lead abated in a room by each
abatement method used has potential as a predictor of the
hazardousness of plastic sheeting waste. More research
would be needed to develop a predictive model, however.
In the HUD Demo, the cost of waste stream analysis, pick-
up, and disposal of hazardous waste was $255 per unit.
This does not include TCLP testing costs incurred by the
abatement contractor to classify waste, or the cost of
abatement contractor management time devoted to the
hazardous waste issue. For this reason and because
solids and plastic sheeting were not treated as hazardous
waste in the Demo, hazardous waste disposal costs may be
higher in practice. Further data is needed to quantify
this.
33
-------
APPENDIX A
EP-TOX TESTING DATA FROM THE HUD DEMO. SORTED BY WASTE CATEGORY AND
LEAD CONCENTRATION IN THE EXTRACT
34
-------
SAMPLE ID
SAMPLE DESCRIPTION
LEAD CAT CITY LOG
(PP»)
89-5010:
89-5024:
89-5023:
89-5019:
89-5017:
89-5005:
89-5004:
KTA - 5:
89-5015:
89-5133:
89-5161:
89-5029:
89-5016:
89-5020:
89-5026:
89-5135:
89-5127:
KTA - 1:
KTA - 3:
90-146:
KTA - 4:
KTA - 2:
90-147:
90-145:
89-5301:
90-157:
89-5163:
90-154:
90-150:
90-153:
89-5160:
89-5168:
89-5030:
89-5167:
89-5006:
89-5003:
89-5169:
89-5165:
89-5008:
89-5012:
89-5166:
89-5011:
89-5136:
89-5129:
89-5132:
89-5159:
Paint from bathroom tile and master
Paint from baseboard trim
Heat gun paint debris
Debris from use of heat gun
Paint debris - "Peel-Away"
Paint debris - "Peel-Away"
Paint debris - heat gun
Paint chips r
Paint from cedar shake
Paint chips
Exterior heat gun chips (paint)
HEPA vac debris
HEPA vac debris
HEPA vac debris
Negative air prefilter impregnated dust
HEPA vacuum contents
HEPA vacuum contents
Drywall with 4 sq.in 1.5 mg/sq.cm paint
Painted wood
Window, attic vent, door frame
KTA#2, plus 4 sq.in 1.5 mg/sq.cm paint
Painted wood
Window frame and trim
Cedar shake, wall 1, 2nd level
Plastic from floor and bags
Chemical treatment sludge
it Peel -Away" sludge w/rag dye, paper
Chemical treatment sludge
Heat gun sludge
Heat gun sludge .......
Vinegar/chem neutralizer Peel-Away wash
Wash used before neutralizer (Peel-Awav)
TSP wash water
TSP and neutralizer wash (Peel-Away)
Liq on poly below chem stripper cleanup
Non-filtered wash/rinse water
5 mic filt rinse water-sanding/Peel-Away
5 mic filtered rinse water - "Peel-Away"
20, 5 micron filtered hand wash water
5 micron filtered hand wash water
5 mic filtered rinse water - heat gun
Non-filtered hand wash water
Filtered (5 micron) waste water
Rinse/wash water supernatant
Rinse/wash water sludge
5 mic filtered rinse water - "Peel-Away"
0
1.3
1.4
2.4
5.6
5.8
9.7
11.4
18
18
52
0
1.3
2.6
7.5
13
22
0.5
0.7
0.72
1.7
2.1
8
140
34
0.64
1
1.22
12
40
620
5.7
45
50
69
77
0
0
0
0
0
0
0
0
0.7
4
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
3
4
6
6
6
6
6
6
7
7
7
7
7
8
8
8
8
8
8
8
8
8
8
1
1
1
1
1
1
1
4
1
JL
2
3
i
JL
1
JL
1
JL
1
2
&•
2
4
4
5
4
4
5
1
JL
3
5
4
5
5
3
4
1
4
1
1
3
4
1
JL
1
4
1
2
ft
2
2
3
1
3
2
2
ft
3
1
1
JL
14
1
JL
5
*J
6
3
^
3
fj
3
5
+J
4
14
14
JL*V
9
14
14
JL"V
1O
JL w
1
JL
6
V
13
7
12
10
1 1
JL J. •
6
7
3
«J
7
1
1
8
7
0
£*
3
•J
/
2
£*
4
^
4
6
35
-------
SAMPLE ID
SAMPLE DESCRIPTION
LEAD CAT CITY LOC
89-5013:
89-5018:
89-5137:
89-5164:
89-5162:
89-5027:
89-5134:
89-5131:
89-5138:
89-5007:
89-5302:
90-152:
89-5172:
90-155:
90-149:
89-5014:
89-5021:
89-5022:
90-156:
90-148:
90-151:
90-158:
89-5170:
90-159:
89-5009:
89-5128:
89-5130:
89-5171:
89-5300:
89-5032:
89-5031:
89-5025:
89-5033:
89-5028:
89-5139:
5 micron filter used during hand washing
Sand paper from use of HERA sander
5 micron filter with waste
5 mic rinse water filter - "Peel-Away"
5 mic rinse water filter - "Peel-Away"
Negative air filter
HEPA cartridge
HEPA filters from vacuum
Wiping of HEPA vac filter
Paper towels - chem stripper cleanup
Rag - TYVEK suit used to wipe floor
TYVEK suit and rubber gloves - heat gun
PAPR filter - Peel Away
Respirator filter - chemical treatment
TYVEK suit and rubber gloves - heat gun
Respirator filters - heat gun use
Protective suit - chem stripper use
Respirator filters (3 weeks use)
TYVEK suit and rubber gloves-chem treat
Heat gun - respirator filter
Heat gun - respirator filter
Respirator filter - chemical treatment
PAPR filter - mainly sanding
TYVEK suit and gloves
TYVEK suit used during heat gun use
Respirator filters
Composite poly & TYVEK suits
PAPR filter - mainly heat gun use
Rug and pad
Orange foam back and fiber back carpet
Red & black foam back carpet
Red carpet
Corrugated foam pad; fiber back carpet
Green carpet
Blank of wipe "Diaparene"
0
0
0.5
0.7
1.4
4.7
27
95
97
110
220
0
0
0
0
0
0
0
0.53
0.54
0.55
0.68
0.9
0.99
1.2
2.1
3.2
3.9
0
0
0
0
0
1.4
0
9
9
9
9
9
9
9
9
9
9
9
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
11
11
11
11
11
11
12
1
1
2
4
3
1
2
2
2
1
3
5
4
5
5
1
1
1
5
5
5
5
3
5
1
2
2
4
3
1
1
1
1
1
2
3
3
5
7
6
2
5
4
5
1
6
11
7
12
10
1
3
3
12
10
11
13
8
13
1
4
4
7
6
1
1
2
1
2
5
LOCATION CODES
1 = 4033 Vallejo (Denver)
2 = 4320 Zuni (Denver)
3 = 2921 Curtis (Denver)
4 - 615 Udel (Indianapolis)
5 = 922 E. 42nd Street (Indianapolis)
6 = 905 Drum (Washington, D.C.)
36
-------
7
8
9
10
11
12
13
14
6155 Parkway (Baltimore)
5716 Sheridan (Washington, D.C.)
3425 38th Place (Birmingham)
1778 Jefferson (Birmingham)
4104 Main Street (Birmingham)
1415 30th Street (Birmingham)
4340 Greenwood (Birmingham)
KTA samples - Baltimore, address unknown
CITY CODES
1 = Denver
2 = Indianapolis
3 = Washington, D.C.
4 = Baltimore
5 = Birmingham, Alabama
CATEGORY CODES
Paint chips
HEPA vac debris, dust from air filters, paint dust
Old woodwork, plaster, windows, doors, and similar bulky
components removed from the building
Plastic sheets and tape used to cover floors and other surfaces
during LBP removal
Solvents and caustics used during stripping
Sludge from stripping
Unfiltered liquid waste such as wash-water from general cleanup
or from decontaminating surfaces after solvents have been used;
unfiltered liquid waste from exterior blasting
8 = Filtered wash water
9 = Rags, sponges, mops, HEPA filters, air monitoring cartridges,
scrapers, and other materials used for testing, abatement and
cleanup
10 = Disposable work clothes and respirator filters
11 = Rugs and carpets
12 = Blanks
4 =
5 =
6 =
7 =
37
-------
APPENDIX B
SOLIDS SAMPLES FROM THE HUD DEMO ANALYZED BY TCLP
38
-------
Sample
ID
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
Unit Address
5230 16th
17 Tacoma
3924 E. 30th
3421 N. Gale
2739 Mura
1321 E. 27th
2931 Riggs
1422 E. 34th
•1422 E. 34th
1422 E* 34th
1422 E. 34th
1422 E. 34th
1422 E. 34th
1422 E. 34th
3449 Kinnear
3449 Kinnear
3449 Kinnear
3449 Kinnear
1649 Temple
1649 Temple
Substrate
WTM
BSB
DRF
FENCE
DOOR
WDW
DOOR
DRF
WSL
WDW
DRF
COL
DOR
WTM
DOR (FRONT)
DOR (FRONT)
DOR WLA
DRF WLA
DRF (FRONT)
DRF (REAR)
Location XRF
(rag/cmz)
Garage WL4
BD # 2
EXT (LV RM)
EXT
EXT
BSM
GAME # 2
KIT
EXT
EXT
BAT
EXT
EXT
EXT
EXT
EXT
EXT
EXT
EXT
EXT
6.1
N/A
3.7
N/A
5.9
87
• /
2.8
3.1
ff • JL.
8.0
8.2
5.3
9.7
6.8 •
4C
• u
3.1
4.8
3 1
••* • J_
4.8
9.4
9.4
39
-------
APPENDIX C
PLASTIC SAMPLES FROM THE HUD DEMO
40
-------
ID
ADDRESS
SEL ROOM
MEAN N
(mq/cm2)
METHOD
9 4895 Vallejo ORG
8 4895 Vallejo ORG
7 4895 Vallejo ORG
20 1304 Walters ORG
2.1 1304 Walters ORG
25 1304 Walters ORG
24 1304 Walters ORG
22 1304 Walters ORG
23 1304 Walters ORG
5 1665 Macon ORG
6 2516 N. 8th ORG
4 1308 Wallace ORG
3 1308. Wallace ORG
29 1422 East 34th ORG
30 1422 East 34th ORG
16 2931 Riggs ORG
11 2931 Riggs ORG
14 2931 Riggs ORG
15 2931 Riggs ORG
13 2931 Riggs ORG
2 1308 Wallace SUB
32 3745 Eudora SUB
1 1665 Macon SUB
31 338 Chester SUB
26 2931 Riggs SUB
27 4895 Vallejo SUB
28 3565 Krameria SUB
10 2931 Riggs SUB
12 2931 Riggs SUB
18 2931 Riggs SUB
17 2931 Riggs SUB
19 2931 Riggs SUB
N/A 3311 West Walsh N/A
N/A 617 Elk N/A
N/A 1321 East 27th N/A
N/A 1321 East 27th N/A
N/A 407 South 30th N/A
N/A 407 South 30th N/A
N/A 407 South 30th N/A
N/A 407 South 30th N/A
N/A 3311 West Walsh N/A
N/A 1422 East 34th N/A
KIT 9.6 1
LVG ROOM 9.4 1
HALL (BLV) 5.4 2
BED 2 1.3 2
. BED 1 1.4 2
BASEMENT 2.8 2
BATH (2LV) 1.8 2
KIT 1.5 2
LVG ROOM 1.6 2
LVG ROOM 3.4 5
BATH 8.6 8
BATH 9.4 13
KIT 5.6 16
PANTRY 2.8 7
BATH 3.6 2
HALL (LVL 2) 9.3 2
DIN RM 1 6.9 2
BED 3 8.8 2
LVG ROOM 8.9 8
BED 1 8.7 5
GAME 3.2 1
EXT WALL 1.2 10
HALL 2.6 2
BED 2 2.1 12
KIT 1.3 5
HALL (LV1) 2.2 1
KIT 7 2
LAUNDRY 2.6 2
BED 2 7.4 2
DIN RM 2 1.1 2
GAME #1 0.41
BATH 8.3 10
BED 3 6.1 1
PANTRY 5.6 8
BATH 16 7
KIT 11.3 6
BATH (LVL 1) 9.9 6
HALL (1LV) 4.1 8
KIT (LVL 1) 4.7 4
LAUNDRY (BLV 4.5 1
HALL (BLV) 7.5 4
KIT 4 8
Abrasive Removal
Abrasive Removal
Abrasive Removal
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Chemical Removal
Encap, Enclose
Encap, Enclose
Encapsulation
Encapsulation
Enclosure, R/R
Enclosure, R/R
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Heat Gun
Encapsulation
Encapsulation
Encapsulation
Encapsulation
Enclose, Heat Gun
Enclosure
Enclosure, Encap
Heat Gun
Heat Gun
Heat Gun, Chem
Heat Gun, Chem
Heat Gun, Enclose
Chemical Removal
Encap, Enclose
Encapsulation
Encapsulation
Enclosure
Enclosure, R/R
Enclosure, R/R
Remove/Replace
Remove/Replace
R/R, Encl
CODES: MEAN
N
SEL
ORG
SUB
N/A
Arithmetic average of XRF measurements in the room
Number of XRF measurements in the room
Selection code
original sample actually obtained
HUD substitution
originally requested sample which was unavailable
41
-------
APPENDIX D
REGRESSION ANALYSIS FOR PLASTIC SAMPLES
Several regression analyses were conducted to further explore and
quantify the relationship between the mass of lead abated and the
TCLP results for plastic samples. The first analysis had TCLP as
dependent variable, and estimates of the total mass of lead abated by
each abatement method (see Section 3.4) as independent variables.
The fitted model was
TCLP (mg/1) =
0.23 + 0.65*[ABR REM] + 2.40*[CHEM REM]
+ 0.64*[ENCAP] + 40.96*[HEAT GUN]
+ 13.56*[ENCL] - 31.83*[REM/REP],
where each dependent variable represents the total kilograms of lead
abated in a room by the corresponding abatement method. This model
explained 85.1% of the variability in the TCLP results. However,
only the variables HEAT GUN and ENCL were significant in the model.
Accordingly, a second regression, with only these two variables as
independent variables, was run, resulting in the following fit:
TCLP (mg/1) = 0.33 + 40.9*[HEAT GUN] + 13.3*[ENCL].
This model explained 84.8% of the variability in TCLP, so there was
essentially no loss of explanatory power versus the full model.
Finally, a model with just HEAT GUN as independent variable was run,
with the following result:
TCLP (mg/1) = 1.79 + 41.4* [HEAT GUN].
This model explained 78.4% of variability.
The second and third models have potential use in developing decision
rules for determining whether plastic sheeting from an abatement
needs to be treated as hazardous waste. For example, the second
model predicts .that 10 of the 32 samples would fail the TCLP test.
Of these 10, 8 actually did. Thus, 4 of the 12 samples actually
failing the TCLP test are not predicted to do so by the model, while
2 samples which did not fail are predicted to fail. The 4 samples
not predicted to fail by the model are those with the 4 smallest TCLP
levels above the 5 mg/1 cutoff. The 2 samples incorrectly predicted
to fail both had TCLP levels below the detection limit of 0.3 mg/1.
The predicted values from the model were 11.2 and 12.7 mg/1
respectively. The predictive power of this model, based on currently
available data, is clearly limited. Further data collection and
model refinement would be necessary before practical application.
The third model predicts 7 failures of the 12 that actually occurred.
It fails to predict correctly for the 5 samples with the smallest
TCLP levels exceeding 5 mg/1. However, the model always predicts
correctly for those samples whose TCLP level was less than 5 mg/1.
This simpler model performs slightly better than the two-variable
42
-------
43
-------
------- |