|
Issues Impacting Bridge Painting: an Overview
Chapter 4. Task C - Evaluation
of Procedures for Analysis and Disposal of Lead-Based Paint-Removal
Debris Table of Contents
INTRODUCTION
The objective of this task was to review current waste
analysis, handling, and treatment procedures especially with
respect to the toxicity characteristic leaching procedure
(TCLP) and the effect that iron has on lead leachability.
The objective was achieved by: (1) analyzing published reports,
(2) discussing results with other investigators, including
State DOTs, and (3) by analyzing paint debris samples obtained
from removal operations, including operations where recyclable
steel-grit blasting was used.
Our results are consistent with observations
made by others, wherein lead was found to be stabilized against
leaching
in the TCLP test by using steel blasting grit. The steel
(iron) may be present as an additive or as the primary
blasting media. The effect appears to be due to the fact
that iron
is more electropositive than lead, thus reducing the lead
ions in solution to metallic Pb that plates out on the
iron particles and are subsequently removed by filtration
from
the leachate being analyzed. Initially, this dramatically
reduces solubility, but the permanency of the stabilization
of lead-paint debris by iron is questionable on long-term
exposure to commercial dump environments. Repeated leaching
of the same debris has shown that the rate of leaching
increases to where the lead is sufficiently solubilized to
fail the
TCLP test. As a result, some state DOTs have chosen to
treat all paint debris as hazardous waste regardless of whether
it passes the TCLP test.
Additives to blast media other than iron are
being developed and evaluated. Lead-based paint debris stabilized
against
leaching by proprietary silicate-based materials such as
BlastoxTM have proven to be more resistant to
the TCLP test conditions than the iron-stabilized debris.
This is true both in terms of the initial degree of leachability
as well as for repeated leaching. While the long term stability
in commercial waste dumps remains to be demonstrated, work
is under way at BIRL and other laboratories to try to define
test procedures more nearly representative of natural exposures.
Other candidate additives undoubtedly will be forthcoming.
These additives will be based on their ease of reaction
with the soluble lead compounds in paints to convert them
to highly
insoluble products.
In the future, environmental regulators may
increase the aggressiveness of the TCLP test environments.
Consideration
is being given to lowering the pH of the test from 5.0
to 1.5, substituting mineral acids for the acetic buffer
solution,
and lowering the permissible limit of leachability from
5 ppm to 1.5 ppm. But since neither the present TCLP test
nor
the proposed changes represent long-term commercial dump
exposures, the logic behind these changes is questionable.
It may be much more realistic to specify a reduction of
the debris to a smaller particle-size distribution and to
perform
longer term, repetitive leaching. An obvious difference
between the TCLP test conditions and those present in disposal
sites
is the availability of oxygen. The disposal site environment
is likely to be less oxidizing and this may effect the
chemistry that controls lead solubility. Any proposed changes
in the
TCLP test need to be shown to correlate with containment-site
environments before they are implemented.
DISCUSSION
Waste Analysis Methods
Analysis of
the Existing Paint
Paint Analysis should begin with the existing paint
to determine if it contains toxic metals such as lead,
chromium, etc. The composition of the existing paint
may be determined by:
- DOT paint history records, or as defined in 15 USC
2601, Section 401.
- Onsite x-ray fluorescence analysis.
- Laboratory paint chip analysis by atomic absorption
(AA) or inductively coupled plasma atomic emission
spectroscopy (ICP or ICP-AES).
Each of the above methods has its advantages
and disadvantages. Examination of the records should be sufficient
provided
accurate records have been kept. But since the records
probably do not tell how much lead existed in what originally
was called a non-lead-based paint or what subsequent
contamination from maintenance painting may have occurred,
actual analysis is necessary in order to know what precautions
need to be taken in removing, handling, and disposing
of the paint. Since lead was a common contaminant in
zinc, the use of high levels of zinc in paints for corrosion
inhibition has been examined to estimate whether or not
removal of these paints might lead to a debris which
would be a hazardous waste (Journal
of Protective Coatings and Linings, 3/93, pp.
24-36). It was concluded that the lead in zinc-rich paints
does not pose a significant
environmental or public health hazard.
Paint chip analysis is an inexpensive
means of determining the amount of lead present. The
analysis is done by atomic
absorption or ICP after acid digestion to dissolve
the lead, and the results are generally expressed
as total
weight percentage lead in the paint. The disadvantages
of paint chip analysis are that the analysis is done
off-site and that care must be taken to ensure that
the paint is uniformly removed down to the steel substrate
without inclusion of any significant amount of rust.
We have found energy dispersive x-ray (EDX)
analysis to be a valuable laboratory tool when used in combination
with a scanning electron microscope (SEM) to analyze
paint chips. The method, although semi-quantitative,
is rapid and very informative. Elements heavier than
fluorine can be determined by this method. Thus, the
presence of elements found in the most common pigments
and contaminants can be determined. These include sodium,
potassium, chlorine, calcium, silicon, titanium, iron,
lead, zinc, copper, aluminum, etc. By analyzing both
sides of the paint chip, separate analyses can be obtained
for the topcoat and the primer.
Samples of Existing
Paint From I-55
Samples of paint were taken for analysis prior
to paint removal. These samples were taken from three different
locations.
The first sample (no. 414-1) was taken from an area beneath
the center of a span. This paint was gray in color and came
from an area that later was used to demonstrate vacuum power-tool
paint removal. The gray paint was a three-coat system consisting
of a red-lead/alkyd primer, a leafing aluminum/phenolic intermediate
layer, and a non-leafing aluminum/phenolic topcoat. The second
sample (no. 414-2) was taken from the same span, but from
an outer beam that was painted green for aesthetics. This
paint system also had a red-lead/alkyd primer followed by
an intermediate coating that contained aluminum. The green
topcoat, however, contained significant amounts of lead and
titanium and much less aluminum. The third sample (no. 414-3)
was taken from a badly deteriorated area between spans on
eastbound I-55 where heavy rusting and loss of adhesion were
obvious. Its composition was the same as that of Sample 414-1
as described above.
Photographs of the inner (next to the steel)
surfaces of the paint chips at x 20 magnification are shown
in FIGURES 50a, 50b, and 50c.
All samples of this surface show the red-lead primer, but
to a different degree. Sample 414-1 contains a thick, nearly
continuous layer of primer, indicating that the primer
was more strongly adhered to the rest of the paint than to
the
steel beneath. Sample 414-2 has a large amount of what
appears to be rust attached to the primer, indicating that
corrosion
had taken place beneath the primer. Sample 414-3 has little,
if any, iron or rust attached to the inner surface and
has less red-lead visible, indicating that the primer was
more
strongly adhered to the steel and separated cohesively.
Existing Paint Analysis by
Energy-Dispersive X-Ray Analysis
The three paint chip samples were analyzed
by energy dispersive x-ray (EDX) analysis and the results
are given in TABLE 6. Both the outer (exposed to the atmosphere)
and the inner (next to the steel) surfaces of each paint
chip were analyzed. EDX analysis is a relatively quick and
very informative technique that analyzes surfaces to a depth
of approximately 1 µm. Since EDX analyzes only a small
area at one time and old paint surfaces are not homogeneous,
it is important to try to analyze as representative a portion
of the surface as possible. Also, it is important to note
that only elements with an atomic number greater than 10
(sodium and above) can be determined by EDX. Thus, elements
such as C, H, O, and N are not detected by EDX. The results
shown in TABLE 6 for two portions
of the outer surface of Sample 414-2 indicate the general
degree of variability in going from one spot of the surface
to another.
The data presented in TABLE
6 were
calculated by the Kevex EDX unit using standardized weighing
factors. Trace amounts of other elements were found, but
are not listed in TABLE 6.
In addition, Sample 414-3 had potassium (7.2 percent) on
its
outer surface. Although the EDX analyses reported in TABLE
6 were single-spot results taken on surface areas of
approximately 1 mm2, they were representative
of several such analyses made at different spots on the
paint chip surface.
From the photographs in FIGURES
50a, 50b, 50c and the EDX
results in TABLE 6, the following
observations and conclusions can be drawn.
- The use of a lead-containing primer is confirmed for
all samples. This is consistent with the orange color and
also was consistent with IDOT records that identified that
a red-lead/alkyd primer was used.
- The top coats of all of the paint samples contained aluminum,
but the relative concentration in Samples 414-1 and 414-3
was more than three times that of Sample 414-2. Again,
this is consistent with IDOT records which show that the
gray paint (Samples 414-1 and 414-2) consisted of a leafing
aluminum/phenolic intermediate layer with a non-leafing
aluminum/phenolic topcoat.
- The composition of the green paint, Sample 414-2, clearly
was different from the other two samples. It appeared to
contain titanium dioxide in addition to aluminum. Furthermore,
its lead content was nearly the same on the outer and inner
surfaces, indicating that both the primer and the topcoat
contained lead. These conclusions also are in agreement
with the IDOT records.
- Between 5 and 10 percent of the elements found on the
outer (topcoat) surfaces of the chips was iron. The source
of this iron is not known. It could have come from the
pigments, but probably was surface contamination by rust
and dirt.
- The amount of iron found on the inner (primer) surface
varied greatly. A high concentration of iron (63 percent)
was found on the inner surface of Sample 414-2. Iron apparently
was pulled from the steel along with the chip, suggesting
that rusting beneath the primer may have occurred. This
was confirmed visually as seen in FIGURES 50a, 50b, 50c.
- Chloride was found in significant quantities on the outer
surfaces of the paint chips, but generally its concentration
on the inner surfaces was quite low. These data indicate
that salt, probably from deicing, is mainly on the outer
surface and that very little diffusion of chloride through
the topcoats to the primer occurred.
- The most likely source of silicon, which was found in
rather large amounts on the outer paint chip surfaces,
is dirt.
- Lead chloride crystals were found on the topcoat surface
of Sample 414-2. Of the paint chip samples taken, this
sample would have had the greatest exposure to vehicle
exhaust fumes. It is possible that the source of lead in
these crystals was tetraethyl lead from prior use of leaded
gasoline. Alternatively, this may be the result of interaction
of lead in the topcoat with deicing salt.
TCLP Analysis of the Paint Debris
An analysis of the paint debris to determine
if it should be classified as hazardous waste was done by
the toxicity
characteristic leaching procedure (TCLP) test (EPA Method
1311). It is the required regulatory test to determine if
the debris contains sufficient leachable toxic materials
to classify the waste as hazardous. Briefly, the test consists
of extracting 100g of the solid debris with 2000g of aqueous
solution, pH adjusted to 5 with acetic acid. The solid sample
and the extractant liquid are placed in a bottle and rotated
end over end at 30 rpm for 18 h at 23°C. The mixture
is filtered and the filtrate analyzed. If the filtrate contains
5 or more ppm lead, the waste is classified as hazardous.
Although the TCLP test was designed to try
to simulate dump-site conditions, it suffers from several
drawbacks.
The test is sensitive to debris particle size, shape, and
surface area. Particle size and shape will depend on the
method of removal, the type, and the age of the paint.
Although grit blasting will tend to produce fine debris particles,
the actual size distribution will depend on the type of
media
used. Vacuum power-tool removal will produce an intermediate
size debris, and handtool removal will produce the largest
particle sizes. Although the TCLP test specifies a maximum
particle size, it probably would be improved by ball milling
the debris to a definite screen size range prior to extraction.
This would make the test samples more uniform in surface
area as well as to increase the surface area. The more
uniform surface area from one sample to another should make
the results
more reproducible and, therefore, allow a more precise
comparison of sample leach rates. The greater surface area
should have
two significant effects. The first would be to increase
the leach rates because of the increase in surface area exposed
to the leachant. The second would be to break up lead-containing
particles that have been encapsulated or complexed by additives
such as iron or BlastoxTM, thus simulating long-term
leaching where natural events such as ground shift and
chemical interactions occur.
The TCLP test was designed to be mildly aggressive
by specifying that the leachant have a pH of 5, which is
slightly acidic.
Actual ground conditions in the landfill site, however,
may be quite different, not only in terms of pH, but in terms
of the soil composition. For example, the TCLP test specifies
an acetic acid leachant, whereas the soil may contain other
acids and salts and possible chelating agents.
The TCLP test results depend not so much on
the amount of lead present as on the form of the lead. Certain
lead
salts are much more water soluble than others and lead
metal is very insoluble. Conversion of more soluble lead
salts
to less soluble forms will reduce the amount and rate of
leaching. It is for these reasons that treatments with
iron and proprietary compositions are effective. On the other
hand, it is possible that certain contaminant ground components
may convert the lead to more soluble forms after disposal.
The TCLP test is widely used, being the best
test currently available, and it is the EPA-specified standard.
It will
continue to be used until a better test is specified. In
general, the TCLP test is more aggressive than natural
conditions and, therefore, has been thought to represent
a worst-case
scenario. Recently unreported work by Lloyd Smith and Gary
Tinklenberg, however, indicates that a more severe and
perhaps more realistic test procedure might be to periodically
percolate
the leachant solution through a bed of the debris and analyze
the effluent for lead concentration. This simulates the
effects of natural changes in groundwater flow. Under these
conditions,
stabilization by iron ultimately fails.
Changes in the TCLP test are being considered
by the EPA, both at the State and federal levels, that would
make the
test more aggressive. Changes under consideration are:
(1) to lower the pH from 5.0 to 1.5, (2) to use mineral acids
rather than acetic acid, and (3) to lower the permissible
limit for non-hazardous classification from the current
5
ppm to 1.5 ppm or lower. Lowering the pH makes the test
more aggressive for two reasons. First, most lead salts become
more soluble as the pH is lowered (acidity is increased).
Second, the protective nature of lead-stabilizing additives
may be reduced because they may dissolve in the acid and
may no longer be available to react with the lead salts.
Mineral acids being considered are sulfuric and combinations
of sulfuric and nitric acids. These acids are strong oxidizers
and are very aggressive in terms of their reactions with
metals, salts, and the paints themselves. Commercial landfill
environments where the pH is below 5 and where large amounts
of mineral acids are present are extremely rare and would
only occur in the case of an acid spill or if the site
was
built on an acid runoff area. Lowering the permissible
lead level standard for hazardous waste has been implemented
by
some States. Illinois and North Carolina classify debris
having TCLP leachable lead levels between 0.5 and 5 as
special wastes which have to be handled much like hazardous
waste.
Waste Treatment
Paint debris that is determined to be a hazardous
waste by the TCLP test must be stabilized against leaching
prior
to disposal. It is illegal to simply dilute the waste to
pass the TCLP test. Stabilization can be accomplished by
several means. Onsite post-treatment is legal, but a waste-site
analysis plan must be provided and a license must be obtained
that may be very difficult to get. On-site post-treatment
methods, therefore, can be avoided by including iron or other
materials such as BlastoxTM in the blast media
which renders the lead much less leachable. The mechanism
of iron stabilization of lead-based paint debris is discussed
in the next section. The mechanism of stabilization appears
to be conversion of the lead compounds in the paint debris
to less soluble forms. As the mechanism of stabilization
by other materials becomes better understood, it is expected
that new stabilizers will become commercially available.
Post-treatment methods include encapsulation
in portland cement, and treatment with lime, lime/fly ash,
cement/kiln
dust, and proprietary silicates. The most common method
is the one based on portland cement. Care must be taken to
be
sure these treatments are done properly. For example, if
the cement is underhydrated, its effectiveness will be
greatly reduced. Also, lime stabilization may not be permanent.
An
excellent discussion of fixation of metals by cement-based
processes is given in Chemical
Fixation and Solidification of Hazardous Wastes by Jesse R. Conner, Van Nostrand
Reinhold, New York, 1990.
Iron Stabilization of Lead-Based
Paint Debris
The mechanism by which lead is stabilized
against leaching when steel-grit blasting media is used has
been reviewed
with personnel from various DOT and independent laboratories
as well as by suppliers and users of blast media. The users
and suppliers of blasting media tend not to have any explanation
for the reduced leachability in the TCLP test. They view
it as good and do not care why it is so. DOT laboratory personnel,
such as Rich Kramer of IDOT (private communication), suggest
that the effect steel (iron-based) blast media has on leachable
lead in paint debris appears to be due to a reduction of
lead ions to lead metal by reaction with metallic iron. Others
call this a plating-out effect. Since the lead metal is not
soluble in the acetic acid leachant, the portion analyzed
in the TCLP test, the measured soluble lead is reduced. It
should be noted that all of the lead need not be made insoluble
to pass the TCLP test. If the exposed surface of the lead
in the paint pigment becomes converted to the insoluble metallic
form, it will protect the remaining unexposed lead from the
leachant. As this protective outer layer erodes away, however,
the material may again become soluble and hazardous.
The oxidation potentials for lead and iron
are -0.126V and -0.441V, respectively, so the following oxidation
reduction
reaction occurs:
Pb2+ + Fe0 ---- Pb0 + Fe2+
(ion) (metal) (metal) (ion)
Since this reaction does not remove lead from
the debris, but only changes its solubility state, there
exists the possibility
that some later reaction will allow the lead to redissolve
and therefore become hazardous. If this happens, the co-generators
(State/contractor), treater, and anyone in the disposal process
becomes responsible.
Other factors that may affect the leachability
of lead in paint/abrasive wastes are as follows.
- The chemical form of the lead in the paint being removed
may have an effect. For example, red lead (Pb3O4),
which is the most common form of lead in the primer, may
dissolve in the leachant more or less slowly than lead
silicochromate or white lead (2PbCO3 · Pb(OH)2)
which may be found in the topcoat of older paints. Only
the soluble surface lead is available to readily react
with the iron.
- The composition, shape, size, hardness, and oxidative
state of the iron-based blast media may play a role in
how readily the iron and lead interact.
- The effects of other oxidizing and reducing chemicals
that may be in the waste being disposed of may determine
the degree of conversion of ionic lead to metal as well
as how long the lead will remain unleachable.
In 1992, North Carolina removed lead-containing paint from
two bridges. For one bridge, 10-percent steel grit was used
in their blast media and for the other bridge, no steel was
added. The debris generated from the mineral-steel abrasive
contained 0.3 percent to 3.7 percent total lead, but the TCLP
leachate contained only 0.3 to 6.3 ppm. The debris where no
steel was added in the blast media contained between 0.5 percent
and 1.5 percent total lead and the TCLP leached lead was between
124 and 202 ppm. While the lead may or may not be permanently
fixed (made insoluble) by the addition of steel, the initial
reduction of lead was impressive. This effect of iron on the
reduction of leachable lead has been documented by many investigators,
but the long-term insolubility in commercial containment site
environments is unknown.
TCLP and Optical Analysis
of Paint Debris Samples
TCLP test analysis of paint debris obtained via different
removal methods results are summarized in TABLE
7. The amount of lead found by the TCLP test and the
amount of lead remaining in the paint debris after the TCLP
testing (residual lead) are reported. The total lead in each
sample is the sum of the TCLP lead analysis and the residual
lead after TCLP testing. The first four samples in TABLE
7 contain both paint debris and steel grit that was not
removed by the vacuum systems. Photos at x 20 magnification
of the paint-removal debris for the first, fourth, fifth,
and sixth samples in TABLE 7 are shown in FIGURES 51a, 51b, 51c,
and 51d.
The data clearly show that use of steel-grit blast media
within full containment areas (the first three samples in TABLE 7) produces debris that is classified as a non-hazardous
waste according to the TCLP test. Even though these samples
contain a few thousand parts per million lead, the amount
leached was less than the detectable limit of 0.1 ppm. Thus,
for the first three samples in TABLE
7, the residual lead determined after TCLP testing is
essentially the same as the total lead in the sample. It
should be noted that much of the remaining (unleached) lead
was encapsulated by paint resin and was not subject to dissolution.
Furthermore, use of a non-iron-containing abrasive was not
tested in this removal demonstration. Thus, the magnitude
of the reduction in leachable lead resulting from the use
of steel grit as compared to non-iron-containing grit is
not available. Debris picked up on the tarp around the vacuum
blast area was found to be a hazardous waste even though
the blast media was steel. Debris from the vacuum power-tool
operations contained over 200 ppm leachable lead. These paint
debris samples are very high in total lead since they are
not diluted by blast media (abrasive). Paint debris from
handtool removal was high in total lead and leachable lead,
but the leachable lead was not as great a percentage of the
total lead as it was in debris produced by the power tools.
This difference is due to the power tools breaking up the
paint to a greater extent than did the handtools, thus exposing
more of the paint surface to the TCLP leachant.
CONCLUSIONS AND SUGGESTIONS
The results of this task study lead to the following conclusions
and suggestions. Good paint analysis techniques are available
to determine both qualitatively and quantitatively the presence
of lead and other hazardous metals in existing paints and
in the debris produced by various paint removal methods.
They include: (1) x-ray fluorescence (XRF) as an onsite,
non-destructive means of determining whether or not the paint
should be classified as lead-containing and (2) paint chip
analysis done by classical laboratory methods such as atomic
absorption (AA and ICP-AA) or energy dispersive x-ray (EDX)
analysis to determine total and leachable Pb, Cr, Fe, etc.
- The best paint debris analysis method for determining
total metal content, including lead, is acid digestion
and atomic absorption. This is the accepted regulatory
method.
- Although the TCLP test has its deficiencies, it is currently
the best method and the regulatory specified method of
determining if paint debris is to be classified as a hazardous
waste.
- One of the main deficiencies of the TCLP test is that
repeated leach testing often gives different results than
originally obtained, such that a waste originally passing
the test may later show high soluble lead levels depending
upon the waste exposure environment history.
- It is suggested that the reproducibility of the TCLP
test results could be improved by: (1) specifying reduction
of the paint debris to a narrower range and finer particle
size distribution, and (2) introducing a periodic percolation
of leachant through the debris, either instead of or after
the presently specified 18-h leaching procedure.
- Despite the fact that steel (iron-based) blast media
has the advantages of recyclability and stabilization of
lead in paint-removal debris, and that BlastoxTM works
well as an additive to reduce leachable lead, the search
should continue for materials that will stabilize lead
efficiently with long-term resistance to aggressive waste
containment site environments.
- The production of hazardous paint-removal waste should
be minimized by the use of recyclable abrasive and the
waste generated should be treated by effective methods
to ensure its stability in waste containment sites
|
