United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/SR-01/055   January 2002
Project Summary

Field   Demonstration  of
Lead-Based  Paint  Removal  and
Inorganic  Stabilization
Technologies
John R. Kominsky
  A study was conducted to demon-
strate the effectiveness of a wet abra-
sive  blasting technology to remove
lead-based paint from exterior wood sid-
ing and brick substrates, and the effec-
tiveness of two  Best Demonstrated
Available Technologies (BOAT) to sta-
bilize the resultant blasting media (coal
slag and mineral sand) paint debris to
reduce the leachable lead  content. The
average lead loading of the paint coat-
ing on the wood and brick substrates
was 6.9 and 51.9 mg/cm2,  respectively.
The effectiveness of  the lead-based
paint removal technology was  deter-
mined using an  X-ray fluorescence
(XRF) spectrum analyzer  (L&K shell).
The XRF measurements were corrobo-
rated by analysis of substrate samples
using inductively-coupled plasma
atomic  emission  spectroscopy (ICP-
AES). The  effectiveness  of the tech-
nologies to stabilize  the debris was
evaluated through the Toxicity Charac-
teristic  Leaching Procedure (TCLP).
Aerodynamic particle size  distributions
of lead particulate  generated  during
paint removal were  measured using a
multi-stage personal cascade impactor.
Personal and area  air samples  were
collected to evaluate  the potential  of
the wet abrasive blasting technology to
generate exposure levels of lead above
the OSHA Permissible Exposure Limit
(PEL) of 50 ug/m3, 8 hour time-weighted
average.
  Wet abrasive blasting effectively re-
moved  the lead-based paint coating
from both the wood and brick substrates
to below the U.S.  Department of Hous-
ing and Urban Development Guideline
(1 mg/cm2) with minimal or no damage
to the underlying substrates (p<0.0001).
The mean area air levels of lead-con-
taining  particulate  generated during
paint removal were significantly below
the PEL (p<0.001), whereas the mean
personal  breathing zone lead  levels
were approximately three times higher
than the PEL. Neither of the two stabili-
zation technologies consistently stabi-
lized the abrasive media paint debris to
achieve a leachable lead content below
the RCRA regulatory threshold (< 5 mg/
L).
  This project Summary was developed
by EPA's National Risk  Management
Research Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a sepa-
rate report of the same title (see Project
Report ordering information at back).

Introduction
  The Lead-Based Paint Poisoning  Pre-
vention Act of 1971, as amended by the
Housing  and Community  Development
Act of 1987,  established 1.0 milligram of
lead per square centimeter of surface  area
(mg/cm2) as the federal threshold requir-
ing abatement of lead-based paint on ar-
chitectural components in public  and
Indian housing developments nationwide.
The Residential Lead-Based Paint Haz-
ard Reduction Act of 1992 (commonly
referred to as  "Title X")  mandated the
evaluation and reduction  of lead-based
paint hazards  in  the nation's existing
housing. Title X also established 0.5  per-
cent lead by  weight as an alternative to

-------
the 1.0  mg/cm2 threshold. An U.S. Envi-
ronmental Protection Agency (EPA) study1
found that a level  of 1.0 mg/cm2  was
roughly  equivalent to  1.0 percent by
weight  and a level  of 0.5 percent by
weight was roughly equivalent to 0.5 mg/
cm2.
  The management of wastes generated
from  lead-based paint abatement activi-
ties are governed by the Resource Con-
servation  and Recovery  Act (RCRA)  of
1976  and provisions contained in 40 CFR
Parts  260-268. RCRA classifies any waste
that leaches 5 milligrams per liter (mg/L)
of lead or more (as determined by a Tox-
icity Characteristic Leaching Procedure2)
a  hazardous waste.  The leachability  of
lead is affected by various factors, includ-
ing speciation of the metal,  pH of the
leachate,  particle size, acid flux through
the waste, and time  of contact with the
leachant. The U.S. Environmental Protec-
tion Agency (EPA) has promulgated a list
of Best Demonstrated Available Technolo-
gies (BOAT) for the  inorganic  stabiliza-
tion   of  hazardous   wastes including
lead-containing  wastes. Stabilization in-
cludes those techniques that limit the solu-
bility  of hazardous constituents in the
waste.3   Much of the  inorganic  stabiliza-
tion that occurs  in the  United States  is
based on the chemistry  of lime or ordi-
nary Portland cement.

Objective
  The overall objective of this study was
to demonstrate the effectiveness  of a wet
abrasive  blasting technology combined
with an inorganic-based stabilization tech-
nology to remove lead-based paint from
exterior  substrates (wood  and brick) and
to generate a non-hazardous waste for
disposal.

Study  Design
  This study evaluated the effectiveness
of a  wet abrasive  blasting technology
(Torbo®) combined  with two inorganic-
based stabilization technologies  (Blastox®
and  PreTox  2000 Fast Dry) to  remove
lead-based paint from exterior substrates
(brick and wood) and  to generate a non-
hazardous waste for disposal. Each tech-
nology  combination  (e.g., Torbo®  with
PreTox 2000 Fast Dry) was demonstrated
on the two substrates (brick and wood) to
yield two treatments. Each treatment was
replicated three times  to yield six experi-
ments per technology combination.
  Brick- A single building wall  (approxi-
mately 28' H x 157' L) was used as the
exterior  painted brick substrate.  The  lead
loading  on the brick  ranged from 1.5  to
15.2  mg/cm2 (average 6.9 mg/cm2,  std.
dev. 3.2 mg/cm2) using a NITON Model
703-A X-ray fluorescence (XRF) spectrum
analyzer (K &  L Shell  Combined). The
masonry wall was divided into six  areas
that ranged from 556 to 756 ft2 (average
627 ft2). The  differences in surface area
are due to the presence of varying num-
bers of windows on the  wall; the respec-
tive areas were subtracted from  each of
the test  areas. Each technology combina-
tion was assigned at  random  to the six
test areas.
   Wood — Five buildings with 4-inch pop-
lar wood siding were used  as the exterior
painted  wood substrate.  The buildings
were located  on the same property, had
an identical architectural design,  and re-
portedly had similar painting histories. The
lead loading on  the wood siding ranged
from 13.1 to 51.9 mg/cm2 (average 33.3
mg/cm2, std. dev. =  10.1 mg/cm2) using a
NITON Model 703-A XRF  spectrum ana-
lyzer (K &  L  Shell Combined). Two test
areas were selected  from one  building
and one test area was  selected from each
of the remaining four buildings, yielding a
total of six test areas.  The six  test  areas
ranged from 294 to 431  ft2 (average 363
ft2). The technology combinations  were
randomly assigned to the test areas.

Technologies Evaluated

Torbo0 Wet Abrasive Blasting
System
  The  Torbo® Wet Abrasive  System is
manufactured by  Keizer Technologies of
Americas, Inc.  in Euless, Texas. The sys-
tem uses conventional blasting abrasives
mixed with  water (80%  abrasive to 20%
water) in a pressure vessel. During this
study, mineral slag  was used to remove
the paint from  the  brick  and  coal slag
(Black Beauty®) was used  to remove the
paint from the  wood.
  The  system  combines the  abrasive
media and  water to create a slurry-mix-
ture that is fed to a blast nozzle much like
a  conventional  blasting  system.  In con-
cept, each  particle of the abrasive  is en-
cased in a  thin layer of water. It utilizes
this coating to both  reduce the heat gen-
erated  by friction and form a  cohesive
bond for the dust created  by the blasting
process that reduces the fugitive particu-
late emissions.

Blastox®
  Blastox® is manufactured by TDJ Group
Inc. in Gary, Illinois.  Blastox®, an abrasive
additive, is  a di- and  tri-calcium  silicate-
based material  similar in  chemical  com-
position  to  Type I  cement. Typically, for
lead-based  paint removal, it is added at a
20-25 weight  percent ratio to the  non-
recyclable blasting  media  such  as min-
eral sand or coal slag. For this study, the
supplier  of  the  abrasive  reportedly
premixed  the  Blastox® additive at a 20
and  15  percent weight ratio to the abra-
sive  (mineral  sand or coal slag) for paint
removal from  the  wood  and brick  sub-
strates,  respectively.
  A  U.S. Army Corps of Engineers study4
concluded that Blastox®  stabilizes lead-
containing paint blast media wastes (i.e.,
reduces the leachability of lead) by a se-
ries  of simultaneous  reactions that result
in an  encapsulated  lead  silicate com-
pound, which is insoluble  at all pH levels.
The  first reaction is a pH  adjustment that
simultaneously stabilizes  the lead by ad-
justing the pH range (8.0-11.5) where
there is limited leachability for lead. Sec-
ondly, the chemical  form of  the  lead is
changed from a lead  oxide, carbonate, or
hydroxide, to  a lead  silicate, which is in-
soluble. A U.S. EPA study5 concluded that
Blastox® appears  to stabilize the  lead
through an  immobilization  mechanism,
rather than by chemical  reaction  of lead
oxide, to form a lead silicate. Lastly, hy-
dration  reactions  encapsulate the waste
into  a cementitious material, which limits
the gravitational flow  of water through the
waste.

PreTox 2000 Fast Dry
  PreTox  2000  Fast Dry (hereafter re-
ferred to  as  PreTox 2000)  is manufac-
tured by NexTec, Inc. in  Dubuque, Iowa.
PreTox  2000  is a  cementitious paint-like
mixture  (i.e., treatment layer) designed to
be applied to  lead-based paint surfaces
and  allowed to cure and adhere to  the
paint coating; it then is removed  in  con-
junction with  the  underlying  lead-based
paint coating using abrasive blasting  or
other standard techniques. PreTox 2000
is composed  of materials from the com-
pounds of sodium and potassium silicates,
sodium  and  potassium  phosphate, and
calcium silicate,  iron and aluminum sul-
fates, and an  alkali  metal salt.6   It also
contains toluene,  acetone, and  VM&P
naptha  as carrier solvents.  Typically,
PreTox 2000 is designed  to be applied to
a 10- to 60-mil (wet) thickness depending
on substrate and  paint condition; the av-
erage application  is  40-mil (wet) thick-
ness. For this study,  the  manufacturer's
representative used an airless sprayer to
apply PreTox 2000 to a surface of 40 mil
(wet) thickness.
  The  manufacturer reports that the
PreTox  2000 system stabilizes the  lead
through two mechanisms.  The first mecha-
nism is  chemical stabilization through pH
adjustment, which instantaneously stabi-
lizes the lead by  adjusting the pH range
(8.0-11.5)  where  there is limited leach-
ability for  lead. The  second is chemical

-------
fixation that changes the soluble ionic
form of lead to an insoluble metallic form.
Test data  provided by  NexTec,  Inc.
showed that  PreTox 2000 successfully
stabilized lead-based paint debris, yield-
ing  a leachable  lead content of <5 mg/L
using both the TCLP and Multiple Extrac-
tion Procedure (MEP).

Sampling and Analytical
Methods

Thickness of Dry Paint Film
  The measurement of dry film thickness
of the  paint  was  made using  ASTM
Method D 4138-88.

Lead in Dry Paint Film
  Lead in paint measurements (XRF and
ICP-AES)  were  made  before paint  re-
moval to establish  the  lead  loading on
the  test panel. The  measurements were
made at approximately the same five lo-
cations as the paint film thickness mea-
surements. The measurements  were
made  in  accordance with  Chapter 7
"Lead-Based Paint Inspection" (1997 Re-
vision)  of the HUD  Guidelines.

XRF Measurements
  A  NITON XRF  Spectrum  Analyzer
(Model  703-A) running software Version
5.1 was used to  determine the lead load-
ing  on  the brick and  wood  substrates.
The instrument was  operated  in the vari-
able-time paint test  mode "K & L + Spec-
tra"  using the  "Combined Lead Reading?
with the instrument display of a 95% con-
fident (2-sigma)  positive or negative de-
termination versus the threshold-level (1
mg/cm2) as the stopping point of the mea-
surement.  There is  no inconclusive clas-
sification  when  using the threshold  for
this instrument running  software version
5.1.11   Results are  classified  as positive
(i.e., > 1.0 mg/cm2), if greater than or equal
to the threshold, or  negative  (i.e., <  1.0
mg/cm2) if less  than the threshold.  The
instrument reads until  a 95% confident
reading of "Positive"  or "Negative" versus
the threshold (1  mg/cm2) is achieved.

Paint Chip  Sampling
  A paint chip sample for ICP-AES analy-
sis  was obtained  at approximately the
same location as three of the five XRF
measurements.  Each  sample was ob-
tained from a 1Vi-inch  by 1%-inch  (ap-
proximately 3.17-cm by  3.17-cm) square
area. The samples were prepared  for
analysis in accordance with EPA SW-846
Method 3050  and analyzed by ICP-AES
in accordance with EPA SW-846 Method
6010. The analytical limit of detection was
reported as 5  ug/sample.
Lead on Bare Substrate
  Lead on bare substrate measurements
(XRF and ICP-AES) were made after paint
removal to establish the residual lead load-
ing in the test area. The six wood siding
test areas and the six brick wall test ar-
eas were each equally dimensioned into
25 areas (i.e., grid squares). The  mea-
surements were made at the approximate
center point of each grid square. An XRF
measurement was made in each of the
25 grid squares. A bare substrate sample
for ICP-AES analysis  was  collected from
five of the 25 squares; the  test locations
were randomly selected.

Lead in Airborne Particulate

Personal Breathing Zone Samples
  Personal breathing zone  and  work area
air samples  were collected during each
technology  demonstration.   The samples
were collected and prepared for analysis
by ICP-AES in  accordance with NIOSH
Method 7300.

Lead Particulate Aerodynamic
Particle Size Distribution
  An 8-stage Marple  Personal  Cascade
Impactor (Model 298)  was  used to deter-
mine the aerodynamic particle size distri-
bution of the lead particulate  generated
during each technology demonstration.
The samples were collected and prepared
for analysis  by  ICP-AES in accordance
with NIOSH Method 7300.

Characterization  of Abrasive
Media Paint Debris
  Representative samples  of  the  abra-
sive media  paint debris (spent abrasive,
stabilization product, paint chips/particles)
were collected to determine whether the
material  generated from  a technology
combination was a RCRA  (40  CFR  Part
261) hazardous waste based on the Tox-
icity  Characteristic Leaching  Procedure
(TCLP).  The samples  were extracted in
accordance with EPA SW-846  Method
1311,  digested in accordance  with EPA
SW-846  Method  3015,  and analyzed in
accordance with EPA SW-846  Method
6010.

Statistical Methods
  All comparisons of two sample means
were made using a standard two-sample
t-test.  If  the distributional assumption of
normality was  not reasonable,  then the
corresponding nonparametric distribution-
free  method  was used (i.e.,  Wilcoxon
Rank  Sum  Test). All  one-sample  com-
parisons to a regulatory action level (1
mg/cm2) were made using a standard one-
tailed t-test. Again, if the distributional as-
sumption of normality was  not  reason-
able, then  the  corresponding nonpara-
metric  method  was  used  (i.e.,  Signed
Rank Test). All  of these statistical  com-
parisons were made at the 0.05  level of
significance. The upper  limit of  the  80
percent confidence  interval for the mean
concentration  of leachable  lead in the
abrasive media  paint debris was calcu-
lated to determine if the material was a
RCRA  hazardous waste.7  If the mean
concentration of leachable lead  plus the
80 percent  confidence interval is greater
than  the  regulatory threshold (5 mg/L),
the material was considered to be a haz-
ardous waste.

Results and Discussion

Effectiveness of Paint Removal

XRF Measurements Before and
After Paint Removal
  Tables 1 and 2 present descriptive sta-
tistics for the XRF measurements obtained
before  and  after paint removal on wood
and brick substrates, respectively,  for
each technology combination. A one-tailed
t-test was used to determine whether the
mean  lead  concentration  after paint  re-
moval was  significantly  less than 1  mg/
cm2 both by substrate  (i.e.,  wood  and
brick)  and  overall  for each  technology
combination. In every case, both  by sub-
strate  and overall, the results show that
both Torbo®-Blastox®  and Torbo®-PreTox
2000  reduced   lead  concentrations  on
wood  and  brick to a level  significantly
below 1 mg/cm2.

Comparison of XRF
Measurements and ICP-AES
Analysis
  Tables 3 and 4 present descriptive sta-
tistics for the XRF measurements obtained
before  and  after paint removal on wood
and brick substrates, respectively,  for
each  technology  combination.  The
Wlcoxon Rank  Sum  Test was  used to
compare lead concentrations  measured
by XRF and ICP-AES on the wood and
brick substrates  both before and  after
paint removal.  The  lead concentrations
determined  by  ICP-AES  and XRF  mea-
surements before paint removal on wood
were not significantly different (p=0.1055);
however, the measurements  before paint
removal on  brick were significantly differ-
ent (p=0.0001).  The lead concentrations
determined  by  ICP-AES  and XRF  mea-
surements after paint removal on wood
were significantly different  (p=0.0331);
however, the  measurements  after  paint
removal on brick were not  significantly
different (p=0.5504).

-------
Table 1. Descriptive Statistics for XRF Measurements (K & L Shell Combined)
        Collected Before and After Paint Removal on Exterior Wood Siding
Technology
Combination
Lead Concentration (mg/cmi2)
N
Mean
Std. Dev.
Minimum
Maximum
Before Paint Removal
Torbo®with Blastox®
Torbo®with PreTox 2000
15
15
36.9
29.7
9.52
9.66
15.5
13.1
51.9
41.4
After Paint Removal
Torbo®with Blastox®
Torbo®with PreTox 2000
75
75
0.24
0.16
0.22
0.16
0
0
1.1
0.70
Table 2. Descriptive Statistics for XRF Measurements (K & L Shell Combined)
        Collected Before and After Paint Removal on Exterior Brick
Technology
Combination
Lead Concentration (nig/cmi2)
N
Mean
Std. Dev.
Minimum
Maximum
Before Paint Removal
Torbo®with Blastox®
Torbo®with PreTox 2000
15
15
5.59
8.18
1.78
3.71
1.5
3.9
9.7
15.2
After Paint Removal
Torbo®with Blastox®
Torbo®with PreTox 2000
75
75
0.14
0.11
0.09
0.14
0
0
0.4
1.1

-------
Table 3.  Lead Concentrations in Paint and on Wood Measured by ICP-AES and
        XRF (K & L Shell Combined)
Method of
Measurement
Lead Concentration (mg/cm2)
N
Mean
Std. Dev.
Minimum
Maximum
Before Paint Removal
ICP-AES
XRF (L& K Shell)
18
30
28.2
33.3
12.8
10.1
9.1
13.1
51.6
51.9
After Paint Removal
ICP-AES
XRF (L& K Shell)
30
150
0.37
0.20
0.50
0.20
0.01
0
2.68
1.10
Table 4.  Lead Concentrations in Paint and on Brick Measured by ICP-AES and
        XRF (K & L Shell Combined)
Method of
Measurement
Lead Concentration (mg/cm2)
N
Mean
Std. Dev.
Minimum
Maximum
Before Paint Removal
ICP-AES
XRF (L&K Shell)
18
30
2.93
6.89
2.11
3.15
0.20
1.5
9.1
15.2
After Paint Removal
ICP-AES
XRF (L&K Shell)
30
150
0.20
0.13
0.30
0.12
0.005
0
1.39
1.10

-------
Characterization of Abrasive
Media Paint Debris

Coal Slag Paint Debris from
Wood Substrate
  Table 5 presents the  mean  leachable
lead levels and corresponding upper con-
fidence  limits for the abrasive  paint de-
bris by site and overall for both technology
combinations. Overall,  the abrasive paint
debris from both technology combinations
was determined to be a hazardous waste.
If examined on a  site-by-site basis, the
debris is also determined to  be  a  hazard-
ous waste. The mean leachable lead lev-
els  in abrasive media debris generated
from the removal of paint from wood by
the  two technology combinations were
compared by using a standard two-sample
t-test. The mean leachable  lead  level  in
the  debris  generated  from  the  Torbo®-
Blastox® combination (21.3 mg/L) was not
significantly different (p=0.4459) from the
mean leachable lead level in the debris
generated from the Torbo®-PreTox 2000
combination  (14.8 mg/L).

Mineral Sand Paint Debris from
Brick Substrate
  Table 6 presents the  mean  leachable
lead levels and corresponding upper con-
fidence  limits for the abrasive  paint de-
bris by site and overall for both technology
combinations. Overall,  the abrasive paint
debris from both technology combinations
was determined to be a hazardous waste.
If examined on a  site-by-site basis, the
debris is also determined to  be  a  hazard-
ous waste,  with one exception. The two
samples collected from  debris  at Site 1
(Torbo®-Blastox®) showed an 80% UCL
of 3.9, which by itself would not  be classi-
fied  as  a hazardous waste. The mean
leachable lead levels in abrasive media
debris  generated from  the  removal  of
paint from  brick  by the two technology
combinations were  compared by  using a
standard two-sample  t-test. The mean
leachable lead level in the debris gener-
ated from the Torbo®-Blastox®  combina-
tion  (8.1  mg/L) was  not  significantly
different (p=0.9555) from the mean leach-
able lead  level in  the debris generated
from the Torbo®-PreTox 2000  combina-
tion (7.8 mg/L).
  Overall, the abrasive  media  paint de-
bris characterization results  (Tables 5-6)
are  somewhat surprising. The leachablility
of lead is affected by many factors includ-
ing, type of lead  in  paint, resins  used  in
the  paint, age of the paint,  particle size,
and others.8'9  The manufacturers of the
stabilization  technologies postulate that
the  ineffectiveness of their respective
products in this study was due  to insuffi-
cient product added or applied to stabi-
lize the concentration  of lead present  in
the paint. The reason(s) why these stabi-
lization technologies were ineffective un-
der the conditions  of  this  study  is
equivocal.
  Blastox®—The material  supplier pro-
vided  a  20%  and  15%  blend ratio  of
Blastox® with the coal slag and  mineral
sand abrasives  for use on the wood and
brick substrates, respectively. A 30% and
20% blend ratio of Blastox® with  the re-
spective abrasives would have been pre-
ferred  by the manufacturer. Hence, the
optimum blend  ratio was not used in the
demonstration.  Mis-communication be-
tween the manufacturer and the abrasive
supplier resulted in the  incorrect blend-
ing ratio of Blastox® with the  abrasive.
Subsequently, the manufacturer issued a
technical bulletin to minimize the  prob-
ability of this blending error occurring  in
the future.12
    PreTox 2000—The manufacturer  of
PreTox  2000 recommends  a 10-40 mil
(wet) thickness application; a 40 mil  (wet)
thickness was applied to  both the wood
and brick substrates. A 60 mil (wet) thick-
ness application for the wood substrates
would have been preferred by the manu-
facturer.  Hence, the optimum application
mil thickness was not used in the demon-
stration.

Air Measurements

Personal and Area Air
Measurements
  In all cases,  the mean airborne lead
levels measured by the personal  breath-
ing zone  samples were  significantly
greater  than the 50 ug/m3  8-hour  TWA
(Table  7).

Lead Particulate Aerodynamic
Particle Size  Distribution
  Figure 1 shows the average differential
lead particle size distribution for the two
samples. This graph provides the  particle
mass concentration ( C) in each particle-
size band versus the geometric mean di-
ameter (GMDj),  where GMDt =  D: x DM.
The lead particles generated by the wet
abrasive blasting of the surface  coating
covers  a wide-size spectrum, where the
larger particles  account for the  greatest
mass of lead. The corresponding cumu-
lative particle size distribution was deter-
mined  by  preparing a  log-probability plot
of the  particle size cut-point (Dp) versus
the cumulative  percent of mass (mg/m3)
less than  the  Dp.  The  distribution  of
sample weights appeared to approximate
a lognormal distribution with a mass me-
dian diameter (MMD) of 8.3 urn. That is,
50% of the particle mass  is  borne  by
particles  larger  than  50 urn. The calcu-
lated geometric  standard deviation (GSD)
was 3.4.  By comparison, a GSD of 1  rep-
resents a monodisperse aerosol.

Cost Analysis
  A cost analysis of the Torbo®-Blastox®
and  Torbo®-PreTox  2000  technology
combinations to remove lead-based paint
from wood and  brick substrates  is  pre-
sented in the  Project Report.

Conclusions

  •_  Wet abrasive  blasting effectively
      removed the lead-based paint from
      both exterior wood siding and brick
      masonry  with minimal damage to
      the underlying substrates (only light
      sanding of the wood was required
      prior to painting or tuck pointing of
      the mortar joints). The residual  lead
      levels were significantly below the
      HUD  Guideline  of  1  mg/cm2
      (p<0.0001). The average paint re-
      moval rates  were 76.4 and 119.8
      ft2/hr on  wood  and brick,  respec-
      tively.

  •_  The lead concentrations  deter-
      mined  by ICP-AES analysis  and
      determined by XRF measurements
      before paint removal on wood were
      not    significantly    different
      (p=0.1055); however, these deter-
      minations  before paint removal on
      brick were  significantly different
      (p=0.0001).  The lead  concentra-
      tions  determined  by ICP-AES
      analysis  and determined  by XRF
      measurements after paint removal
      on  wood were  significantly differ-
      ent (p=0.0331); however,  these
      determinations after paint removal
      on  brick were  not significantly dif-
      ferent (p=0.5504).

  •_  The wet  abrasive  slurry-mixture
      appears to reduce the fugitive emis-
      sions of lead-containing particulate,
      which serves to enhance the level
      of  environmental protection as well
      as worker health and  safety.  The
      mean  area air levels of lead-con-
      taining  particulate  generated  dur-
      ing paint removal were significantly
      below the OSHA PEL  of 50 ug/m3
      (p<0.001), whereas the mean  per-
      sonal breathing zone levels of  lead
      were  approximately three times
      higher than the  PEL. The personal
      breathing zone levels  of lead did
      not vary significantly with substrate
      (p=0.6396);  the area  samples
      showed higher  levels of lead  dur-
      ing removal of paint from brick  than
                                                           6

-------
Table 5. Characterization of Coal Slag Paint Debris from Wood Substrates
Technology
Combination
Torbo®-Blastox®
Torbo®-PreTox
2000
Substrate
Wood
Wood
Site
1
2
5
Overall
3
4
6
Overall
N
2
2
2
6
3
3
3
9
Leachable Lead Level
Mean (mg/L)
12.4
15.5
36.0
21.3
7.7
29.7
7.1
14.8
80% UCL for Mean
39.0
47.9
85.2
31.9
20.2
39.2
17.5
21.4
Table 6. Characterization of Mineral Sand Paint Debris from Brick Substrates
Technology
Combination
Torbo^-Blastox®
Torbo®-PreTox
2000
Substrate
Brick
Brick
Site
2
4
6
Overall
1
3
5
Overall
N
2
2
2
6
2
2
2
6
Leachable Lead Level
Mean (mg/L)
1.1
19.5
3.6
8.1
9.4
5.9
8.3
7.8
80% UCL for Mean
3.9
21.0
9.6
13.5
11.4
11.9
9.5
9.1

-------
Table 7.  Comparisons of Personal and Area Air Concentrations to OSHA PEL
Technology
Combination
Torbo® with
Blastox®
Torbo®with
PreTox 2000
Substrate
Wood
Brick
Wood
Brick
Type of
Sample
Personal
Area
Personal
Area
Personal
Area
Personal
Area
N
3
9
6
18
3
12
6
16
Mean 8-hr TWA
(Mg/m3)
70.9
20.5
68.4
21.2
55.1
26.9
81.5
24.9
t statistic
0.8958
-6.40
1.03
-3.36
0.3163
-6.53
5.63
-3.60
p-value
0.7675
0.0001
0.8257
0.0018
0.6091
0.0001
0.9975
0.0013
                                  8

-------
      for  paint removal  from  wood
      (p=0.0463).

  • _  Neither of the two stabilization tech-
      nologies (Blastox® and  PreTox
      2000) consistently stabilized  the
      abrasive  media  lead-based paint
      waste to reduce the leachable lead
      content.  The  80 percent  upper
      confidence interval for  the  mean
      leachable lead  concentration in the
      debris  consistently exceeded  the
      RCRA regulatory threshold (5  mg/
      L). Failure of the technologies to
      stabilize the  lead most  likely was
      due  to  an inadequate chemical
      stabilizer-abrasive  blend  ratio or
      insufficient application mil thickness
      of the pre-paint removal  coating
      treatment in  the case of  Blastox®
      and PreTox 2000, respectively.

Recommendations
  •_  Although wet abrasive blasting re-
      duces fugitive  emissions  of lead-
      containing particulate  generated
      during  removal  of paint film from
      exterior wood or brick masonry,  it
      should be conducted in at least  a
      Class 4 Containment System as
      specified in SSPC Guide 6, "Guide
      for Containing  Debris  Generated
      During Paint Removal Operations."
      Air monitoring should be conducted
      at the perimeter work area to  de-
      termine  the extent that lead-con-
      taining  particulate are escaping
      from  the work area.

  •_  To maximize  the performance of
      these technologies the user should
      understand the various factors  that
      may  affect the  effectiveness  of the
      product  to  reduce the  leachable
      lead  content of the debris.  Included
      are  paint film  lead content, paint
      film thickness,  paint film condition,
      type  of substrate (e.g., wood, brick,
      metal), variant  particle  size,  and
      other potentially significant factors.

      PreTox 2000-The user of this tech-
      nology should  follow the  applica-
      tion   optimization  procedure
      specified in the technical guidance
      provided by the manufacturer. This
      will  ensure that the  optimum  mil
      thickness  application  rate  of
      PreTox 2000 is applied to the lead-
      based paint coating to be  abated.

      S/asfox®-Subsequent to complet-
      ing this study, the manufacturer of
      Blastox® revised their technical
      guidance  regarding the  proper
      blend ratios of abrasive to chemi-
      cal-stabilizer. The user of this tech-
      nology should verify that the blend
      ratio provided by the material sup-
      plier is consistent with the recom-
      mended blend  ratio for  a given
      lead-based paint  coating  to  be
      abated.

  •_  Due to the inability  of these tech-
      nologies to consistently reduce the
      leachable lead content in the abra-
      sive media paint debris during this
      demonstration, all debris should be
      tested by  TCLP  prior to disposal.
      The sampling strategy  should  be
      consistent  with  Chapter 9  "Sam-
      pling Plan" of SW-846 " Test Meth-
      ods of Environmental Testing of
      Solid Wastes."

References
1.   U.S.   Environmental Protection
    Agency, Office  of Pollution Preven-
    tion and Toxics.  A Field Test of Lead-
    Based Paint Testing  Technologies:
    Technical  Report, EPA 747-R-95-
    002b.  May 1995.

2.   U.S.   Environmental Protection
    Agency, Office  of Pollution Preven-
    tion and Toxics.  Toxicity Character-
    istic Leaching  Procedure,  Method
    1311.  Test Methods  of Environmen-
    tal Testing Solid Wastes  -  Physical
    and Chemical Methods, SW-846 (3rd
    ed.).  November 1986.

3.   U.S.   Environmental Protection
    Agency, Office  of Pollution Preven-
    tion and Toxics.  Stabilization/Solidi-
    fication of CERLA and RCRA Wastes.
    EPA/625/6-89/022.  May 1989.

4.   Hock, V., C. Gustafson, D. Cropek,
    and S. Drozdz.  U.S.  Army Corps of
    Engineers, Construction Engineering
    Research Laboratories. Demonstra-
    tion  of  Lead-Based  Paint Removal
    and  Chemical  Stabilizer  Using
    Blastox. Technical Report FEAP-TR-
    96/20. October 1996.

5.   U.S.   Environmental Protection
    Agency, Office  of Pollution Preven-
    tion and Toxics.  Stabilization of Lead-
    Based Paint Waste.   October 1998
    (Draft).

6.   U.S.  Patent Number 5,674,108.
    Method for Removing Coatings Which
    Create Hazardous  Products.  Octo-
    ber 7, 1997.
7.   U.S.  Environmental   Protection
    Agency, Office  of Pollution Preven-
    tion and Toxics.  Chapter Nine Sam-
    pling Plan.    Test  Methods  of
    Environmental Testing Solid Wastes
    -  Physical  and  Chemical  Methods,
    SW-846 (3rd ed.).  November 1986.

8.   Bishop, P. L. 1988.  Leaching of inor-
    ganic hazardous  constituents  from
    stabilized/solidified    hazardous
    wastes.   Hazardous  Waste &  Haz-
    ardous Materials 5(2): 129-143.

9._ Buskowski, J. M., J. H. Boy, X. Zhu, T.
    D. Race,  and K. A. Reinhold. Immo-
    bilization  chemistry  in  Portland ce-
    ment stabilized paint  blast media
    wastes.   Environmental and Waste
    Management Issues in  the Ceramic
    Industry II 45: 155-163.   1994.

10. Technical  Bulletin:  Determining
    Proper Blastox®  Blend  Ratios.   TB-
    007 (Issued 10-98),  The TDJ Group
    Inc.,  Gary, Illinois 60013.

  The full report was submitted in partial
fulfillment  of Contract No.  DACW88-97-
D-0017  by Environmental Quality Man-
agement, Inc.  This  work was conducted
under sponsorship  of the U.S. Environ-
mental Protection  Agency and the  U.S.
Army  Construction Engineering Research
Laboratories (USACERL).
  This Project Summary was written by
John R. Kominsky of Environmental Qual-
ity  Management,  Inc.,  Cincinnati,  OH
45240.

  Alva  Edwards-Daniels  is  the  EPA
Project Officer and can be contacted at:

  National Risk Management  Research-
  Laboratory _
  U.S. Environmental Protection Agency_
  26 W. Martin  Luther King  Drive_
  Cincinnati, OH 45268_
  Telephone:  513 569-7693_

  The complete report,  entitled "Field
Demonstration of Lead-Based  Paint  Re-
moval and Inorganic Stabilization Tech-
nologies," can be located on the Internet
at:

  www.epa.gov/ORD/NRMRL/Pubs/
  600R01055/600R01055.pdf

  It is also available from  NTIS (Order
No. PB2002-102037)

  National Technical Information Service_
  5285 Port Royal  Road_
  Springfield,  VA 22161 _
  Telephone:.
  (703)  605-6000  (worldwide) _
  (800) 553-6847 (U.S. only)_

-------
  1000
^ 100  =
c      ._.,-
s
•
 0.


Q

 o
    10  -
0.35      0.67
                             1.34     2.65     4.58     7.75      12.2     17.7


                                    Particle Geometric Mean Diameter (urn)
                                                                      32.4
 Figure 1.  Differential Lead Particle Size Distribution During Wet Abrasive Blasting of Brick.

-------