SEPA
United States
Environmental Protection
Agency
EPA/540/SR-93/507
May 1993
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
Emerging Technology
Summary
Babcock & Wilcox Cyclone
Vitrification
The Babcock & Wilcox 6 million Btu/
hr pilot cyclone furnace was success-
fully used in a 2-yr Superfund Innova-
tive Technology Evaluation (SITE)
Emerging Technology project to melt
and vitrify an EPA Synthetic Soil Matrix
(SSM) spiked with 7,000 ppm lead, 1,000
ppm cadmium, and 1,500 ppm chro-
mium. An advantage of vitrification over
other thermal treatment technologies
is that in addition to destruction of or-
ganic wastes, the resulting vitrified
product captures and does not leach
non-volatile heavy metals. Indeed, when
operated at 50 to 150 Ib/hr of dry SSM
feed, and from 100 to 300 Ib/hr of wet
SSM feed, the cyclone technology was
able to produce a non-leachable prod-
uct (as measured by TCLP) from the
hazardous soil. From 95% to 97% of
the dry input SSM was incorporated
within the slag. Stable cyclone opera-
tion was achieved during the 2-yr
project which processed over 6 tons of
clean, unspiked SSM and 5 tons of
spiked SSM. During the thermal vitrifi-
cation process, the heavy metals parti-
tion between the vitrified slag and the
stack flyash. The capture of heavy met-
als in the slag was found to increase
with increasing feed rate and with de-
creasing metal volatility. The treatment
of the synthetic soil matrix resulted in
a volume reduction of 25% to 35% (dry
basis). Vitrification results in an easily-
crushed, glassy product.
This summary was developed by
EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the SITE Emerging Tech-
nology program that is documented in
a separate report (see ordering infor-
mation at back).
Introduction
Organization of this Project
Summary
The Babcock & Wilcox (B&W) cyclone
vitrification process has been developed
and tested for treatment of a U.S. EPA-
developed Synthetic Soil Matrix (SSM)
contaminated with heavy metals. This tech-
nology is significant because it combines
incineration with the production of a non-
leachable (heavy metals and radionuclides)
soil residue. Organics are combusted and
destroyed in the cyclone furnace while
melting the soil to form a vitrified slag.
Non-volatile metals (e.g., chromium) and
non-volatile radionuclides (e.g., strontium
and zirconium) partition mainly to the vitri-
fied slag where they are rendered non-
leachable. The process description, op-
Printed on Recycled Paper
-------
eration, and applicable wastes are dis-
cussed in the sections below.
In this Summary, results from two sepa-
rate cyclone vitrification projects occurring
between 1990-1992 are discussed. The
projects (Phases I and II) were sponsored
underthe U.S. EPA SITE Emerging Tech-
nologies Program. Both projects were per-
formed on the B&W 6 million Btu/hr pilot
cyclone test facility. Brief descriptions of
the two projects are given below:
• Phase I Emerging Technologies
Project - In this project, a dry synthetic
soil matrix spiked with lead, cadmium,
and chromium was vitrified to
determine cyclone furnace operation
conditions (process feasibility), heavy
metals leachability in the vitrified slag,
volume reduction, and preliminary
heavy metals mass balance (i.e., how
much of each metal was retained in
the vitrified slag and how much
volatilized).
• Phase II Emerging Technologies
Project - Once feasibility was
established in Phase I, a wet feed
system was constructed and furnace
modifications performed to optimize
the throughput of soil and increase
heavy metals capture in the slag.
Wet SSM (wet soil is often
encountered at Superfund sites)
spiked with lead, cadmium, and
chromium was vitrified to determine
heavy metals leachability, volume
reduction, and detailed heavy metals
mass balance.
On the basis of the results of Phases I
and II, Babcock & Wilcox was asked to
perform a SITE Demonstration. The Dem-
onstration results may be obtained from
the EPA Project Officer.
Technology Development at
Babcock & Wilcox
The Babcock & Wilcox cyclone furnace
is a well-established design (over 26,000
MWe installed electrical capacity) for the
combustion of high inorganic content (high
ash) coal. The combination of high heat
release rates (450,000 Btu/cu ft for coal)
and high turbulence in cyclones assures
the high temperatures required for melting
the high-ash fuels. The inert ash exits the
cyclone furnace as a vitrified slag.
Taking advantage of the ability of the
cyclone furnace to form a vitrified slag
from waste inorganics, the cyclone fur-
nace was used in a research and devel-
opment project to vitrify municipal solid
waste (MSW) ash containing heavy met-
als. The cyclone furnace produced a vitri-
fied MSW ash which was below EPA
leachability limits for all eight RCRA met-
als. The successful treatment of MSW ash
suggested that the cyclone vitrification
technology would be applicable to high
inorganic content hazardous wastes and
contaminated soils that also contain or-
ganic constituents. These types of materi-
als exist at many Superfund sites, as well
as sites where petrochemical and chemi-
cal sludges have been disposed. Our ap-
proach for establishing the suitability of
the cyclone vitrification technology relies
on the premise that for acceptable perfor-
mance in treating contaminated soils con-
taining organic and heavy metal/radionu-
clide constituents, the cyclone furnace
must melt the soil matrix while producing
a non-leachable slag and must achieve
the destruction and removal efficiencies
(DRE's, currently 99.99%) for organic con-
taminants normally required for FICRA haz-
ardous waste incinerators. The high tem-
perature (>2,500 to 3,000 °F), turbulence,
and residence time in the cyclone and
main furnace are expected to result in
high organics destruction and removal ef-
ficiencies (DRE's).
Process Description
The Babcock & Wilcox 6 million Btu/hr
cyclone furnace located in Alliance, OH,
was used to perform all pilot-scale vitrifi-
cation tests. The furnace is water-cooled
and simulates the geometry of B&W's
single cyclone, front-wall fired cyclone coal-
fired boilers. This cyclone facility has been
proven to simulate typical full-scale cy-
clone units in regard to furnace/convec-
tion gas temperature profiles and residence
times, NOX levels, cyclone slagging poten-
tial, ash retention in the slag, unburned
carbon, and flyash particle size. It is im-
portant to note that this particular furnace
configuration, representative of a fossil
fuel-fired utility boiler, is likely to be modi-
fied significantly for a transportable unit
dedicated to soil vitrification.
The pilot cyclone furnace, shown in Fig-
ure-1, is a scaled-down version of a com-
mercial coal-fired cyclone with a restricted
exit (throat). The furnace geometry is a
horizontal cylinder (barrel). A schematic of
the process is illustrated in Figure 2. For
the present application, natural gas and
preheated combustion air (820°F) enter
tangentially into the cyclone burner. In dry
soil processing, the soil matrix and natural
gas enter tangentially along the cyclone
furnace barrel. For wet soil processing, an
atomizer using compressed air is used to
spray the soil paste directly into the fur-
nace. The soil is captured and melted,
and organics are destroyed in the gas
phase or in the molten slag layer formed
and retained on the furnace barrel wall by
centrifugal action. The soil melts, exits the
cyclone furnace from the tap at the cy-
clone throat, and is dropped into a water-
filled slag tank where it solidifies. A small
quantity of soil also exits as flyash with
the flue gas from the furnace and is col-
lected in a baghouse. In principle, this
flyash could be recycled to the furnace as
indicated in Figure 2 to increase the cap-
ture of metals and to minimize the volume
of the potentially hazardous waste stream.
The energy requirements for vitrification
were 15,000 Btu/lb. Given the much larger
surface-to-volume ratio of the relatively
small pilot unit and its cool ^surface, one
may expect a full-scale unit to achieve
lower energy requirements.
Particulate control is achieved by way
of a baghouse. To maximize the capture
of metals, a heat exchanger is used to
cool the stack gases to approximately
200°F before entering the baghouse. Al-
though the cyclone facility is equipped with
an acid gas scrubber, it was not used for
these tests because acid gas generation
(e.g., HCI) from the vitrification of the U.S.
EPA SSM was expected to be low.
Applicable Wastes and Soils
and Possible Technology
Configurations
An advantage of vitrification over other
thermal destruction processes is that in
addition to the destruction of organic con-
stituents, the resulting vitrified product cap-
tures and does not leach non-volatile
heavy metals or radionuclides. The cy-
clone vitrification technology would be ap-
plicable to high inorganic content hazard-
ous wastes, sludges, and contaminated
soils that contain heavy metals and or-
ganic constituents. The wastes may be in
the form of solids, a soil slurry (wet soil),
or liquids. To be treated in the cyclone
furnace, the ash or solid matrix must melt
and flow at cyclone furnace temperatures
(2400 to 3000 °F). Because of the
technology's ability to capture heavy met-
als in the slag and render these non-
leachable, an important application of the
technology is contaminated soils which
contain non-volatile radionuclides (e.g.,
strontium, transuranics).
The cyclone furnace can be operated
with gas, oil, or coal as the; supplemental
fuel. The waste may also supply a signifi-
cant portion of the required heat input.
Additional air pollution control devices,
such as NOX reduction technologies, can
be applied as needed. An acid gas scrub-
ber would be required, for example, when
chlorinated wastes are treated. HEPA/car-
bon/scrubbing towers would be used for
radioactive waste processing.
-------
Stack Paniculate
SSM Feed
System
Sampling
Location
Slag and
Quench
Water
Sampling
Location
ID Fan
Furnace
Stack
Natural
Gas
Soil
Injector
Slag
Trap
Cyclone
Barrel
Slag
Quenching
Tank
Figure 1. Cyclone furnace facility with detail of the cyclone furnace barrel.
Stack
Materials and Methods
Synthetic Soil Matrix
A synthetic soil matrix formulated by
EPA was used for all cyclone testing. Both
clean and spiked SSM were obtained from
the EPA Risk Reduction Engineering Labo-
ratory (RREL) Releases Control Branch in
Edison, NJ. SSM, used by EPA for treat-
ment technology evaluations, has been
well-characterized in previous studies (1).
Clean soil was used for furnace condi-
tions optimization. The spiked SSM used
in the Emerging Technologies projects con-
tained 7,000 ppm (0.7%) lead, 1,000 ppm
(0.1%) cadmium, and 1,500 ppm (0.15%)
chromium. The SSM used in the SITE
Demonstration contained 7,000 ppm lead,
1,000 ppm cadmium, 4,500 ppm stron-'
tium, and 4,500 ppm zirconium.
Typical Run Conditions
Typical run conditions for the Phase I
and Phase II tests are given in Table 1.
Sampling and Analysis
Sampling and analysis followed guide-
lines in the U.S. EPA SW-846 Manual,
and the Quality Assurance Project Plan
met RREL Category III requirements.
Phase I and Phase II sampling locations
for measurements and analyzers are
shown in Figure 3.
Results and Discussion
The Phase I and Phase II Emerging
Technologies projects were conducted us-
ing approximately 6 tons of unspiked SSM
and 5 tons of SSM spiked with heavy
metals. Phase I tests were conducted with
dry SSM and Phase II tests with we't SSM
(26% moisture). Stable cyclone operation
was achieved during the several pilot tests,
which ranged from 3 to 14 hr in duration.
Particulate loading data and materials
mass balance suggested from 95% to 97%
of the input SSM was incorporated within
the slag. Using natural gas as the fuel,
the CO, CO O2> and NOX stack emis-
sions gases from the process were within
acceptable ranges (<30 ppm, 11.5%, 1%,
and <400 ppm, all corrected to 3% oxy-
gen). The NOx levels can be readily re-
duced by NOx control technologies.)
The slag (vitrified soil) from the tests
appeared to be a black, glassy, obsidian-
like mass. Some large white glass particu-
lates are readily visible in the slag frag-
ments. When viewed under a low-magnifi-
cation microscope, both the slag (soil)
matrix and the embedded white particles
appeared to have completely melted. The
vitrified material can be easily crushed.
Figure 2. Cyclone vitrification process.
-------
Table 1. Typical Cyclone Furnace Test Conditions
Condition Typical Range of Values
Heat Input (natural gas tuel)
SSM Feed Rate
Excess Oxygen at the Stack
Primary and Secondary
Air Temperature
Slag Temperature
Furnace Exit Gas Temperature
Flue Gas Exit Temperature
Baghouse Temperature
5 million Btu/hr
50 to 300 Ib/hr
1.0%
83CPF
2370 to 2460 °F
2800 to 3000 °F
2100 to 2200 °F
<20CPF
TCLP Results
TCLP results for the Phases I and II
heavy metals tests are shown in Figure 4.
For both Phase I and Phase II, the, un-
treated SSM exceeded the TCLP limits for
lead and cadmium by >10X. Chromium,
spiked at a level similar to lead and cad-
mium, did not exceed the TCLP limits (a
similar phenomenon was reported in Ref.
1, and during the SITE Demonstration).
The treated SSM was about 10X lower
than TCLP limits for the three metals. The
results show that the cyclone vitrification
process always succeeded in producing a
non-Ieachable slag.
Volume Reduction
Approximately 35% and 25% volume
reduction (dry basis) was obtained by vit-
rification of dry SSM (Phase I) and wet
SSM, (Phase II), respectively. The volume
reduction is a combination of 22% mass
reduction, mainly attributed to the calcina-
tion of the limestone component of SSM,
and the increased bulk density from 80 Ib/
cu ft for SSM to 86 to 92 Ib/cu ft for the
slag.
Fate of Heavy Metals (Mass
Balance)
A mass balance for total ash, cadmium,
chromium, and lead was performed for
the cyclone furnace treatment process.
The purpose of the mass balance was to
determine the fate of the heavy metals
during soil treatment. The heavy metals
could be retained in the glass-like slag or
be volatilized and leave the cyclone with
the flue gas.
For the Phase I tests, the overall mass
balance accounted for 79% to 103% of
the total materials input to, the furnace
and the heavy metals mass ibalances ac-
counted for 65% to 77% of the lead, 56%
to 61% of the cadmium, and 141% to
145% of the chromium input to the fur-
nace. In the case of chromium, mass bal-
ances in excess of 100% were calculated.
The most likely source of excess chro-
mium was a newly-installed refractory
which contains 9.8% chromium oxide (Cr2
O3), with "bake-out" or abrasion of the
material causing the elevated stack chro-
mium levels. i
For the Phase lltests, the overall and
heavy metals mass balances were closer
to 100% of the materials input to the fur-
nace. The, overall mass balance achieved
102% to 107% of the total material input
to the furnace, and the heavy metals bal-
ances achieved 74% to 87,.5% for lead,
50.5% to 71.5% for cadmium, and 78.9%
to 96.8% for chromium input to the fur-
nace. (This time, the use of chromium
I
.£>
I
«j
CD
-J
•S3
to
§•
en
120
100
80
60
40
20
n
u
—
I I Lead
V/\ Cadmium
K\l Chromium
I
m
—
t
i
i
V^ ' Treated SSM .results shown
f, 1 with 10 times expanded scale
/j — ii'-h i. - ' -• i r7~rz-<3 1 r7T\N
Average Average Treated SSM, Treated SSM, EPA limits
SSM SSM Phase 1 Phase II
Phase 1 Phase II 141 Ib/hr , 200 Ib/hr
Figure 3. Toxiciiy Characteristic Leaching Procedure (TCLP) results.
-------
Heavy metals
TCLP
Cyclone
Furnace
Temperature
02
Comb, air
Ash melt properties
Major constituents
% moisture \ j
Bulk density '
Feed rate _. _
Proximate «g*™*
Ultimate
%ash
Ash elemental
Feed rate
Postf
Hea
Part
(as/-
Maj
Gas
i
\
Slag
Heav
TCLf
Bulk
Majo
Slag
•urnace
vy Metals
iculate loading
/slag split)
or constituents
velocity
\
^ Bagt
i
y metals
3
density
<• constituents
temperature
Flue Gas
CO
CO,
NO,
Temperature
Moisture
__t
>ouse
|
Fly Ash
Figure 4. Sampling locations and analyses.
refractory was minimized to prevent any
chromium contamination.)
A mass balance was calculated for each
test as follows: The total flyash and slag
streams were measured and normalized
to 100%. The non-normalized mass bal-
ance was calculated from the percent
flyash and the flyash metals concentration
and from the percent slag and the slag
metals concentrations. The mass balance
was also normalized (100% = amount of
heavy metals measured in the input SSM
calculated from the feed rate and the SSM
metals concentration). The amount of cad-
mium captured in the slag was 8% to 16%
and 12% to 23% (Phase I and II data,
respectively); lead captured in the slag
was 24% to 35% and 38% to 54%; chro-
mium captured.in the slag was 80% to
95% and 78% to 95%.
The heavy metals content in the slag
increases with increasing SSM feed rate
between 50 to 300 Ib/hr. Since fuel (natu-
ral gas) feed was relatively constant, this
suggests that increasing SSM feed rate-
reduces the solids residence time in the
furnace (and lowers measured slag tem-
peratures) and consequently reduces va-
porization of heavy metals into the flue
gas. This is a promising trend for full-
scale operation.
An attempt was also made to correlate
the different behavior of the metals during
cyclone treatment with their volatility. The
temperature at which the metal vapor pres-
sure was 100 mm Mercury was chosen as
the volatility parameter. The percentage
of heavy metals retained in the slag was a
function of volatility of the metal for the
Phase I, Phase II, and demonstration tests.
These results suggest that the cyclone
vitrification process will show high capture
for very low volatility contaminants such
as many radionuclides (e.g., zirconium,
uranium, thorium). Conversely, high vola-
tility metals are likely to be concentrated
in the flyash, which may then be suitable
for recycle to the cyclone furnace or pos-
sible metal recovery.
Conclusions
The Babcock & Wilcox 6 million Btu/hr
pilot cyclone furnace was used success-
fully to
• Vitrify an EPA Synthetic Soil Matrix
(SSM) spiked with 7,000 ppm lead,
1,000 ppm cadmium, and 1,500 ppm
chromium.
• Produce a non-leachable (TCLP)
product.
• Incorporate from 95% to 97% of the
input SSM within the slag.
• Maintain stable cyclone operation.
• Using natural gas as the fuel, produce
CO and NOx stack emissions gases
from the process within acceptable
ranges.
• Increase the capture of heavy metals
in the slag with increasing feed rate
and with decreasing metal volatility.
• Reduce the volume of the synthetic
soil matrix by 25% to 35% (dry basis).
Because the projects were conducted
on a pilot cyclone furnace configured as a
utility boiler for proof of concept of testing,
and by no means optimized for soil vitrifi-
cation, a unit designed for dedicated soil
vitrification may improve process through-
put and performance. In November of
1991, a SITE Demonstration was success-
fully conducted to provide performance
and cost information for mixed waste (or-
ganics, metals, and simulated radionu-
clides). The next steps in the technology
development include design, construction,
and field demonstration of a full-scale unit.
References
1. P. Esposito, J. Hessling, B. Locke,
M. Taylor, M. Szabo, R. Trumau, C.
Rogers, R. Traver, and E. Barth, 'Results
of Treatment Evaluations of a Contami-
nated Synthetic Soil," JAPCA, 39- 294
(1989).
•frUS. GOVERNMENT PRINTING OFFICE: nti - 750471/80112
-------
-------
-------
Jean M. Czuczwa, James J. Warhol, Hamid Farzan, and William F. Musiol are
with Babcock & Wilcox, Company, Research and Development Division,
Alliance, OH 44601.
Laurel Staley is the EPA Project Officer (see below).
The complete report, entitled "SITE Emerging Technologies Project: Babcock
& Wilcox Cyclone Vitrification," (Order No. PB93-163038; Cost: $9.00,
subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer 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
Penalty for Private Use
$300
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/540/SR-93/507
------- |