&EPA
United States
Environmental Protection
Agency
September 1992
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
Technology Demonstration
Summary
Babcock & Wilcox Cyclone
Furnace Vitrification
A Superfund Innovative Technology
Evaluation (SITE) Demonstration of the
Babcock & Wilcox Cyclone Furnace
Vitrification Technology was conducted
in November 1991. This Demonstration
occurred at the Babcock & Wilcox
(B&W) Alliance Research Center (ARC)
in Alliance, OH. The B&W cyclone fur-
nace may be used for thermal treat-
ment of soils contaminated with organ-
ics, metals, and radionuclides. The cy-
clone furnace is designed to destroy
organic contaminants and to immobilize
metals and radionuclides in a vitrified
soil matrix (slag).
For the SITE Demonstration, the cy-
clone furnace was used to treat a syn-
thetic soil matrix (SSM) spiked with
semivolatile organics, metals, and
simulated radionuclides. Demon-
stration Test results indicate that the
majority of the nonvolatile metals and
radionuclides were retained in the slag.
The leachate from the slag complies
with the Toxicity Characteristic Leach-
ing Procedure (TCLP) regulatory, re-
quirements. The slag also demonstrates
extremely low teachability for the spiked
simulated radionuclides. Destruction
and Removal Efficiencies (DREs) for
both spiked organic contaminants ex-
ceeded 99.99% with stack gas concen-
trations below detection limits.
The soil, metals, and simulated ra-
dionuclides not retained in the slag
exited the furnace with the flue gas
and were captured in the baghouse as
flyash. The leachate from the flyash
did not meet the TCLP regulatory re-
quirements and therefore was consid-
ered a hazardous waste. Wash water
and rinse water which came in contact
with the SSM were also disposed of as
hazardous wastes. The quench water
used to cool the slag was analyzed and
found suitable for discharge to the lo-
cal POTW.
An economic analysis was conducted
to estimate costs for a commercial
treatment system utilizing the B&W cy-
clone furnace vitrification system. This
analysis was based on the pilot-scale
results from the SITE Demonstration.
The economic analysis is for a pro-
jected commercial unit capable of
treating approximately 3.3 tons per hour
(tph) of contaminated soil. The cost to
remediate 20,000 tons of contaminated
soil using this commercial unit is esti-
mated at $465 per ton if the system is
Printed on Recycled Paper
-------
on-line 80% of the time or $529 per ton
if the system is on-line 60% of the time.
Treatment costs appear to be competi-
tive with other available technologies.
This Summary was developed by
EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the SITE program dem-
onstration that is fully documented in
two separate reports (see ordering in-
formation at back).
Introduction
In response to the Superfund Amend-
ments and Reauthorization Act of 1986
(SARA), the U.S. Environmental Protection
Agency's (EPA's) Office of Research and
Development (ORD) and Office of Solid
Waste and Emergency Response
(OSWER) have established a formal pro-
gram to accelerate the development,
demonstration, and use of new or innova-
tive technologies as alternatives to current
containment systems for hazardous
wastes. This new program is called Su-
perfund Innovative Technology Evaluation
or SITE.
The major objective of the SITE Pro-
gram is to develop reliable performance
and cost information for innovative tech-
nologies. One such technology is the
Babcock & Wilcox (B&W) Cyclone Furnace
Vitrification Technology, which was dem-
onstrated over a 2-wk period beginning
November 4, 1991 and ending November
16, 1991. The Demonstration was con-
ducted at B&W's Alliance Research Cen-
ter (ARC) in Alliance, Ohio.
The cyclone furnace system was dem-
onstrated on a SSM provided by the EPA's
Risk Reduction Engineering Laboratory
(RREL) in Edison, New Jersey. SSMs are
well-characterized, clean soils which are
spiked with known concentrations of con-
taminants of concern. For this study, the
SSM was spiked with contaminants similar
to those found at Superfund sites, De-
partment of Defense (DOD) facilities, and
Department of Energy (DOE) facilities:
heavy metals, semivolatile organics, and
simulated radionuclides. (Simulated radio-
nuclides are non-radioactive metals whose
behavior in the cyclone furnace will simu-
late true radionuclide species.)
The B&W cyclone furnace process is
designed to destroy semivolatile organics
and immobilize heavy metals and simu-
lated radionuclides in a nonleachable slag.
The critical and noncritical objectives were
established to evaluate the effectiveness
of the process. Critical parameters pro-
vided data to support the developer's
claims. Noncritical measurements provided
additional information on the technology's
applicability to other Superfund sites and
allowed observation and documentation
of any process performance anomalies.
The critical objectives of this SITE
Demonstration were to assess the ability
of the B&W cyclone vitrification furnace
technology to:
produce a vitrified material (slag) that
does not exceed TCLP regulatory
levels for cadmium, chromium, or lead
achieve at least a 10 to 1 ratio (dry
weight basis) of slag to flyash
capture at least 60% (by weight) of
the chromium from the dry, untreated
SSM in the vitrified slag
achieve at least a 25% volume re-
duction in solids when comparing
product solid to the untreated SSM
achieve 99.99% DREs for each or-
ganic contaminant spike
comply with emission limits for nitro-
gen oxide (NOX), carbon monoxide
(CO), total hydrocarbons (THC), and
particuiates from the stack as stipu-
lated by 40 CFR 264
The non-critical objectives of this Dem-
onstration were to determine the
technology's general applicability and to
document process performance by ana-
lyzing:
simulated radionuclide concentrations
in stack emissions
metals concentrations in the baghouse
solids, furnace outlet, and slag quench
water
semivolatile organics concentrations
in the slag, baghouse solids, and
quench water
teachability of metals in the baghouse
solids
leachability of simulated radionuclides
in the slag
emissions of products of incomplete
combustion (PICs)
ultimate analysis, proximate analysis,
ash content, ash fusion temperature,
and particle size distribution of the
SSM
Process and Facility
Description
The Demonstration of the B&W cyclone
furnace vitrification technology was per-
formed at B&W's ARC located on a 37.5
acre site containing 230,000 sq ft of of-
fice, laboratory, and service facilities.
A 6-million Btu/hr pilot-scale cyclone
furnace located at the ARC was used for
the SITE Demonstration. This non-mobile
cyclone furnace is a scaled-down version
of B&W's commercial cyclone boiler and
is capable of firing natural gas, oil, or
coal. The cyclone furnace is watercooled
and simulates the geometry of B&W's
single cyclone, front-wall-fired cyclone
boiler. The furnace has a horizontal cylin-
der (barrel) Jined with a refractory layer
suitable for operation at high temperatures.
This unit is designed to achieve very high
release rates, temperatures, and turbu-
lence. The cyclone test facility is shown in
Figure 1.
The SSM was contained in 55-gal
drums. A drum tumbler was used to mix
each drum before it was transferred into
the feeder tank. The feed SSM was intro-
duced at a nominal feed rate of 170 Ib/hr
via a soil disperser (atomizer) at the cen-
ter of the cyclone. The cyclone furnace
was fired with natural gas during the
Demonstration and preheated combustion
air (nominal 800°F) entered the furnace
tangentially. Particulate matter from the
feed soil is retained along the walls of the
furnace by the swirling action of the com-
bustion air and is incorporated into a mol-
ten slag layer. Organic material in the soil
is incinerated in the molten slag or in the
gas phase. The slag exits the furnace
from a tap at the cyclone throat at a tem-
perature of approximately 2400°F, then
drops into a water-filled quench tank,
where it cools and solidifies. I
The gas residence time in the furnace
is approximately two seconds. The gas
exits the cyclone barrel at a temperature
of over 3000°F and exits the furnace at a
temperature of over 2000°F. A heat ex-
changer cools stack gases to approxi-
mately 200°F before they enter the pulse-
jet baghouse. A small portion of the soil
exits as flyash in the flue gas,and is col-
lected in the baghouse. The cyclone facility
is also equipped with a scrubber (a lime
spray dryer) to control any acid gases that
may be generated. The scrubber was not
used for the Demonstration since no
chlorinated compounds were spiked in the
SSM. The scrubber and baghouse are
followed by an induced draft (ID) fan, which
draws flue gases into a process stack for
release to the ambient air. '
SSM Description
The SSM feed was spiked with heavy
metals (lead, cadmium, and chromium),
semivolatile organics (dimethylphthalate
and anthracene), and simulated radionu-
clides (bismuth, strontium, and zirconium).
These spikes were chosen to allow for
proper evaluation of the technology without
risk to personnel safety while limiting the
generation of hazardous products. Con-
taminant concentrations in the SSM are
summarized in Table 1. ',
Sampling and Monitoring
A total of four runs were; conducted
over a 3-day period and included a back-
-------
SSMFeed
System
Continuous Emissions
_- ,. . .. Monitor (OEM)
Sampling Location Sampling Location
Stack Paniculate
SSM
Sampling
Location
Slag and
Quench water j
Sampling
Location
'ID Fan
Scrubber
" (Not in Use)
Slag
Trap
Figure 1. B&W Cyclone test facility.
Natural Gas
I I I I lnJectors
Furnace
Stack
Natural
Gas
Soil
Injector
Cyclone
Spout Barre,
.Slag
Quenching
Tank
ground run (Run 0) and three test runs
(Runs 1A, 2, and 3).
Prior to the firing of SSM, a background
run was conducted to establish baseline
conditions. For this run, the furnace was
fired with natural gas. Air sampling at the
stack and furnace outlets was used to
Table 1. Spiked Components in the SSM
Cone, (mg/kg)
Analyte
Average Range
Heavy Metals
Lead
Cadmium
Chromium
6350 3880-7510
1250 1000-1800
4300 3800-4630
Simulated Radionuclides
Bismuth 4140 2810-7205
Strontium 3690 3300-4080
Zirconium 4030 3660-5000
Organic Compounds
Anthracene 4340 920-7800
Dimethylphthalate 8340 4800-10000
determine background concentrations of
metals and particulates. Water samples
from the quench tank were used to deter-
mine background levels of metals. Flyash
was collected from the baghouse to de-
termine metals concentrations.
The cyclone furnace was fired with
natural gas for approximately two hr at
the beginning of each test run to preheat
the barrel. Soil was then added and the
test was begun after the furnace reached
steady-state, slag-production conditions
(approximately an hour later). During the
test runs, which lasted an average of about
four hours, sampling of SSM feed, slag,
and air emissions was conducted. Quench
water was sampled at the beginning and
end of each test run. Baghouse solids
were sampled after completion of each
test run.
Each drum of SSM was sampled imme-
diately after mixing and was then trans-
ferred to a screw feeder. The hopper was
accessed through an opening sealed by a
plexiglas cover and samples of the SSM
were drawn out through this opening.
The flue gas was sampled in two loca-
tions: immediately downstream of the fur-
nace (prior to the air pollution control
equipment) and in the process stack.
A composite sample of flyash was col-
lected from drums located below the bag-
house after the completion of each test
run. Quench water samples were collected
from the quench tank located below the
cyclone furnace before and after each test
run.
A perforated, galvanized tub ("colander")
mounted on wheels was used to collect
slag samples from the quench tank located
below the furnace slag tap. At the start of
each test run, the slag sampling device
was placed beneath the furnace slag tap.
This was to ensure that the slag produced
before the start of the run could be re-
moved from the quench water. Segregation
of test slag and non-test slag was essen-
tial since total mass of slag produced
during each test was needed for several
calculations.
Slag samples from the perforated, gal-
vanized tub were composited into galva-
nized transfer pails, which were covered
with aluminum foil to prevent contamina-
tion by dust. The slag samples were then
transferred from the galvanized buckets
to aluminum pans and were air-dried in
ovens at ARC'S Coal Preparation Labora-
tory. All slag was air-dried so that the total
slag mass generated could be accurately
determined. After the dried weight of each
sample was determined and recorded, the
slag was separated into two categories:
slag to be sent to a laboratory for analysis
("analytical slag") and slag to be used for
bulk density measurements ("bulk density
slag").
Three analytical composite slag samples
were collected for each run. After air dry-
ing, the analytical slag samples were
separately crushed inside cotton/polyes-
ter sacks to pass a %-inch screen. The slag
was then passed through a splitter to ob-
tain representative samples.
Slag not collected for laboratory analy-
ses was used for bulk density measure-
ments. After the slag was dried, nine cy-
lindrical pieces of slag were handpicked
from the array of trays. The length-to-
width ratios of the samples ranged from
approximately 1.6 to 3.7. These samples
were analyzed for leachability of the
simulated radionuclides (bismuth, stron-
tium, and zirconium) according to the
American Nuclear Society (ANS) 16.1 test
method. After sample selection, several
bulk density measurements were made
for each run.
A maximum number of bulk densities
were recorded by overfilling a previously
-------
weighed box of known dimensions, strik-
ing off the excess with a straightedge,
and reweighing the filled box. An average
was then determined from the bulk density
measurements taken for each run.
Operating conditions related to the effi-
cient performance of the cyclone furnace
were controlled and monitored by B&W
during all tests, including the background
run. Much of the data was acquired and
downloaded via a computerized data ac-
quisition system every five min. Param-
eters monitored in this manner included
combustion air flow rate and temperature;
soil dispersing air flow rates; oxygen (O2),
CO, and NOX concentrations at the con-
vection pass; baghouse inlet and outlet
temperature; pressure drop across the
baghouse; and total heat input. Operating
data for the three test runs is summarized
as follows:
cyclone slag temperature ranged from
2360 to 2470 °F.
total heat input ranged from 4.9 mil-
lion to 5.1 million Btu/hr.
nominal soil feed rate remained con-
stant at 170 Ib/hr.
combustion air temperature ranged
from 811 to 832 °F.
temperature of the gas entering the
baghouse ranged from 196 to 215 °F;
Table 2. B& WSITE Demonstration Data and
Potential Incineration ARARs
SITE Demonstration
ResultsRange ARARs
TCLP(mg/L)
Cadmium
Chromium
Lead
0.07-0.18
0.15-0.61
0.22-0.39
1.0
5.0
1.0
DRE(%)
Anthracene >99.996->99.998 99.99
Dimethyl- >99.996 for each
phthalate
Particulate 0.004-0.0017 0.08
matter (gr/dscf
at7%OJ
Carbon 4.8-54.1 <100
Monoxide
(ppm)
Total <5.9-18.2 <20
Hydrocarbons
(ppm)
Metals (Ib/hr)
Cd 9.4x10^-1.5x10^ a
Cr 2.1 x 10-s-1.9x10'< a
Pb 4.8x10-s-7.1x10-4 a
NOf(ppm) 310-435 b
a Less those established by EPA Guidance on
Metal Emissions from Hazardous Waste In-
cinerators
b Allowable emissions limits established on a
case -by-case basis as per the requirements of
the Clean Air Act.
temperature of the gas exiting the
baghouse ranged from 185 to 206 °F.
Results and Discussion
Table 2 illustrates the cyclone furnace's
performance as it relates to Applicable or
Relevant and Appropriate Requirements
(ARARs).
TCLP analyses were performed on both
the feed SSM and the slag to determine if
the B&W technology produces a non-
leachable slag from a leachable soil.
The TCLP conducted on the feed SSM
indicated that cadmium and lead leached
from the SSM in quantities significantly
above regulatory levels (the cadmium and
lead concentrations in the leachate were
49.9 mg/L and 97.3 mg/L, respectively).
Following cyclone vitrification, the slag
passed the TCLP for these metals. TCLP
results are presented in Table 3.
The teachability of the simulated radio-
nuclides from the slag was determined
according to ANS 16.1- "American Na-
tional Standard Measurement of the
Leachability of Solidified Low-Level Ra-
dioactive Wastes by a Short-Term Test
Procedure." The method used to quantify
the external surface area of the slag was
modified to account for the irregular shape
of the slag material. Although all other
equations and data reduction procedures
remain the same, the method has not
been validated for the material in question
and the data are, therefore, suspect. The
test results suggest, however, that the cy-
clone furnace may be able to effectively
immobilize radionuclides in low-level ra-
dioactive wastes.
The percent volume reduction experi-
enced during the Demonstration was de-
termined by comparing the volume of dry
SSM introduced into the furnace* to the
volume of dry slag produced by the fur-
nace. Percent volume reductions were
calculated using a method developed by
B&W. This method determines bulk density
by weighing the soil in a box of known
volume. Since this method was used to
determine bulk density for the both the
SSM and slag, comparisons between the
SSM and slag data provide reliable results
which agree with field observations and
Demonstration Test objectives. These re-
sults confirm B&W's claim that an average
of 25% reduction in the volume is experi-
enced during treatment.
As the SSM goes through the cyclone
furnace, metals partition to the flyash and
the slag. Their fates are primarily depen-
dent on their relative volatilities. The non-
volatile metals, such as chromium, stron-
tium, and zirconium, primarily remain in
the slag. The more volatile metals, such
as bismuth, cadmium, and lead, partition
Tables. TCLP Results (mg/L) ':
Cadmium Chromium [ Lead
Regulatory 1.0
Limits
5.0
1.0
Feed
Run 1
Run 2
Run 3
Slag
Run 1
Run 2
Run 3
52.0
63.6
34.2
0.11"
0.19
0.07
2.29
1.77
3.87
0.15
0.37
0.15
' 90.8
75.6
; 125
' <0.25
0.39*
: 0.29*
* Worst case scenario
to the flue gas and are collected by the
baghouse. During the Demonstration, over
75% (by weight) of the chromium in the
SSM was incorporated in the vitrified slag.
Approximately 85 and 95% of the stron-
tium and zirconium, respectively, remained
in the slag. The more volatile metals (bis-
muth, cadmium, and lead) had lower re-
tentions of 27, 12, and 29%, respectively.
(Data regarding the simulated radionu-
clides are suspect because the analytical
methods have not been validated for these
metals. Data are used for informational
purposes only.)
Almost all of the metals which partition
to the flue gas are captured by the bag-
house. A very small portion of the metals
pass through the baghouse and out the
stack. As long as these levels do not
exceed the furnace's permit limits (as de-
termined by a site-specific risk assess-
ment) no significant changes to emissions
treatment need be employed.
Particulate emissions were! measured
at the cyclone furnace outlet and stack for
all Demonstration Tests. Particulate emis-
sions out of the stack averaged 0.008 gr/
dscf (corrected to 7% OJ, or 0.001 Ib/hr,
which is well under the RCRA regulatory
limit of 0.08 gr/dscf. Average emissions
from the furnace outlet were 0.806 gr/dscf
(corrected to 7% O ), or 6.07i Ib/hr. The
average removal efficiency of the bag-
house was 99.8%.
A slag-to-flyash ratio can be determined
by comparison of the stack ;particulate
emissions with the amount of slag pro-
duced per hour by the cyclone furnace.
The average slag-to-flyash ratio from the
Demonstration was 13.7. The slag-to-
flyash ratio illustrates that the cyclone fur-
nace is capable of converting the vast
majority of a contaminated soil to a non-
hazardous slag.
The cyclone furnace achieved greater
than 99.99% DREs for the two organic
spikes in the feed SSM. Because anthra-
-------
cene and dimethylphthalate are relatively
difficult organics to destroy, it is projected
that the commercial-scale cyclone furnace
will be capable of achieving DREs of
99.99% or greater for all, or nearly all,
organics.
The cyclone furnace formed PICs but
only at very low concentrations (in the
parts per trillion range). Several chlorinated
PICs were detected, although no chlori-
nated organics were spiked in the SSM.
Potential sources of chlorine include trace
levels in the SSM and in the furnace from
previous tests. These trace amounts of
chlorine probably caused the formation of
the chlorinated volatile organic compounds
(VOCs). Higher concentrations of chlori-
nated VOCs will be present if a feed soil
contains chlorinated compounds. Soils
contaminated with chlorinated organics
would also form hydrogen chloride (HCI)
gas, which would have to be controlled by
a scrubber.
Throughout each of the Demonstration
Tests, CO, CO2, THC, O2, and NOx were
monitored continuously to present a real-
time image of the combustion process and
to determine if regulatory standards were
being exceeded. Exhaust gas data is
summarized in Table 4.
The O2 and CO2 values obtained reflect
typical excess air values for a natural gas-
fired furnace. The NO concentrations in
the emissions from the Demonstration
were relatively low.
Quench water samples collected before
and after each run were analyzed to de-
termine if any of the metals present in the
slag or infusible matter leached into the
quench water. Analyses of the quench
water from the baseline run and the three
test runs indicated minimal increases in
the concentrations of certain metals during
the test runs. Concentrations of cadmium,
chromium, lead, and strontium were so
close to the detection limits it cannot be
determined if any increases or decreases
are a result of the process. Concentra-
tions of bismuth and zirconium remained
Table 4. Summary of Exhaust Gas Data
below detection limits throughout the test-
ing period.
Quench water samples collected before
and after the second and third test runs
were analyzed for anthracene and
dimethylphthalate to determine whether
these chemicals leached into the quench
water. Concentrations of both chemicals
remained below method quantitation limits
throughout both test runs.
When the Demonstration Tests were
complete, the quench water was found to
be suitable for discharge to a sanitary
sewer and was disposed of in accordance
with the terms of B&W's wastewater dis-
charge agreement with its local POTW.
Water that came in contact with the SSM
(wash and rinse water from Demonstration
equipment cleanups) was collected, stored
apart from other wastes, and disposed of
as a hazardous waste.
An economic analysis has been devel-
oped to estimate costs (not including
profits) for a commercial treatment system
utilizing the B&W cyclone furnace vitrifica-
tion process. This analysis is based on
the results of the SITE Demonstration,
which utilized a pilot-scale cyclone furnace
vitrification system. The pilot-scale unit
operated at a feed rate of 170 Ib/hr of
contaminated soil and utilized energy at a
rate of 5-million Btu/hr. It is projected the
commercial unit will be capable of treating
approximately 3.3 tph of contaminated soil
and will require an energy input of 100-
million Btu/hr. The daily feed rate for the
pilot-scale system was approximately 2
tpd, while it is projected the commercial
system will be capable of treating 80 tpd.
Treatment costs appear to be competi-
tive with other available technologies. The
cost to remediate 20,000 tons of contami-
nated soil using a 3.3 tph cyclone furnace
vitrification system is estimated at $465
per ton if the system is on-line 80% of the
time or $529 per ton if the system is on-
line 60% of the time. Projected unit costs
for a smaller site (less than 20,000 tons of
contaminated soil) are slightly higher; pro-
Concentration (ppm - dry basis) Concentration, %
Run No.
1
2
3
Value
Average
Low
High
Average
Low
High
Average
Low
High
Nox
357
328
373
338
310
423
383
311
435
CO
>6.1
4.8
>54.1
6.9
6.3
7.4
5.0
4.9
5.2
THC as C3H3
<7.4
<6.9
8.4
11.3
8.9
18.2
<6.4
<5.9
8.1
cos
9.2
8.8
9.5
8.9
8.2
11.8
9.6
9.6
9.7
02
4.9
4.6
6.5
4.9
4.4
5.2
4.9
4.8
5.1
gr/dscf
0.0016
0.0009
0.0003
dscfm
1259
1208
1291
jected unit costs for a larger site are slightly
lower.
Conclusions and
Recommendations
The B&W cyclone furnace transforms
the majority of the hazardous feed with a
high inorganic content (such as soils,
sludges, and sediments) to a nonleach-
able, nonhazardous slag. A small portion
exits the stack as flyash.
TCLP analyses were performed on both
the feed SSM and the slag and it was
determined that the B&W technology
achieves its critical objective of producing
a slag that does not exceed TCLP regula-
tory levels for cadmium, chromium, or lead.
The method used to determine the
teachability of the simulated radionuclides
was altered slightly for application to the
slag. The altered method has not been
validated and the data are therefore sus-
pect. The test results suggest, however,
that the cyclone furnace may be able to
effectively immobilize radionuclides in low-
level radioactive wastes.
A comparison of the volume of the dry
untreated SSM and the slag confirms
B&W's claim that an average volume re-
duction of 25% is experienced during cy-
clone furnace vitrification.
The metals (including simulated radio-
nuclides) contained in the SSM prior to
cyclone furnace vitrification were primarily
contained in the slag and in the baghouse
solids after treatment. The majority of the
nonvolatile metals remained in the slag,
while the majority of the more volatile
metals were found in the baghouse solids.
In particular, approximately 75% of the
chromium from the SSM is retained in the
slag, exceeding the critical objective of
retaining at least 60% of the chromium.
Almost all of the metals which partition
to the flue gas are captured by the bag-
house. A very small portion of the metals
pass through the baghouse and out the
stack. As long as these levels do not
exceed the furnace's permit limits (as de-
termined by a site-specific risk assess-
ment) no significant changes to emission
treatment need be employed. Modifications
have been proposed which would
recirculate the baghouse solids through
the furnace, allowing the system additional
opportunities to trap the metals within the
slag. These modifications would eliminate
the need to dispose of or treat the flyash
as a hazardous waste.
Because the radionuclides and heavy
metals in the slag are nonleachable, the
flyash from the baghouse is the primary
hazardous waste produced by this pro-
-------
cess. If a radioactive feed is employed,
both the slag and baghouse solids would
be radioactive. However, since the radio-
nuclides are expected to be nonleachable,
this technology may be used to treat ra-
dioactive soils to prevent the migration of
radionuclides from a site.
Particulate emissions out of the stack
during the Demonstration Tests averaged
0.008 gr/dscf (corrected by 7% O), or
0.001 Ib/hr, which is well under the RCRA
regulatory limit of 0.08 gr/dscf.
The average slag-to-flyash ratio from
the Demonstration was 13.7, indicating
that the cyclone furnace is capable of
converting the vast majority of a contami-
nated soil to a nonhazardous slag. This
also means that the cyclone furnace ex-
ceeds its critical objective of achieving a
slag-to-flyash ratio of at least 10 to 1.
The cyclone furnace met another criti-
cal objective by achieving greater than
99.99% DREs for the two organic spikes
in the feed SSM. Because anthracene and
dimethylphthalate are relatively difficult
organics to destroy, it is projected that the
commercial-scale cyclone furnace will be
capable of achieving DREs of 99.99% or
greater for all or nearly all organics.
Throughout each of the Demonstration
Tests, CO, CO2, THC, O2, and NOx were
monitored continuously to present a real
time image of the combustion process and
to determine if regulatory standards were
being exceeded. Results indicate that the
cyclone furnace should not have difficulty
meeting the RCRA limit of 100 parts per
million (ppm) for CO. THC emissions!, how-
ever, are close to the RCRA limit of 20
ppm. Careful monitoring of THC emissions
for the cyclone furnace will be required for
the unit to operate in compliance. The
THC and CO levels measured indicate
that effective thermal destruction of the
organic compounds was occurring; the O?
and CO2 values obtained reflect typical
excess air values for a natural ga.s-fired
furnace. The NO concentrations in the
emissions from the Demonstration were
relatively low. A larger unit, however, may
emit significant levels of NOx and may be
designated a major source under the Clean
Air Act. Allowable emissions of NOX will
be established on a case-by-case basis.
Particulate emissions were measured
at both the cyclone furnace outlet and the
stack. Emissions out of the stack easily
met the RCRA emissions limit of 0.08 gr/
dscf corrected to 7% O2. The furnace
achieved its critical objectives for emissions
by complying with emission limits for car-
bon monoxide, THC, and particulates from
the stack.
Analyses of the quench water from the
baseline run and the three test runs indi-
cated minimal increases in the concentra-
tions of certain metals during the tesit runs.
Concentrations of both anthracene and
dimethylphthalate remained below method
quantitation limits. When the Demonstra-
tion Tests were complete, the quench wa-
ter was found to be suitable for discharge
to a sanitary sewer; it is projected that the
quench water from the commercial-scale
system will also be suitable for discharge
to a sanitary sewer, but this must be de-
termined on a site-specific basis.
Wash water and rinse water from the
Demonstration was disposed of as a haz-
ardous waste. The nature of the wash
water and rinse water during commercial-
scale treatment will be site-specific. It may
be a hazardous or radioactive waste at
some sites; at other sites it may be suitable
for discharge to a sanitary sewer. Note
that any wash water, rinse water, or
quench water used in the commercial-
scale system will create only occasional
discharges. It is projected that the com-
mercial-scale system will continuously
discharge water from a quench tower,
which will use water to cool the flue gas
(the pilot-scale system did not include a
quench tower). The water from the quench
tower should be suitable for discharge to
a sanitary sewer.
An economic analysis was con-ducted
to estimate costs for a commercial treat-
ment system utilizing the B&W cyclone
furnace vitrification system. The economic
analysis is for a projected commercial unit
capable of treating approximately 3.3 tph
of contaminated soil. Projected treatment
costs appear to be competitive with other
available technologies.
US. Government Priming Offlca: 1992645-080/60131
-------
-------
The EPA Project Manager, Laurel Staley, is with the Risk Reduction Engineer-
Ing Laboratory, Cincinnati, OH 45268 (see below)
The complete report, entitled "Technology Evaluation Report: Babcock & Wilcox
Cyclone Furnace Vitrification Technology," consists of two volumes:
"Volume I", (Order No. PB92-222215AS; Cost: $26.00, subject to change)
"Volume II", (Order No. PB92-222223 AS; Cost: $43.00, subject to change).
Both volumes of this report will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
A related report, entitled "Babcock & Wilcox Cyclone Furnace Vitrification
Technology: Applications Analysis Report," which discusses application and
costs is under development.
The EPA Project Manager can be contacted at:
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penally for Private Use
$300
BULK RATE
POSTAGES FEES PAID
EPA
PERMIT No. G-35
EPA/540/SR-92/017
------- |