xvEPA
United States
Environmental Protection
Agency
EPA/540/S5-89/008
May1989
SUPEFtFUND INNOVATIVE
TECHNOLOGY EVALUATION
Technology Demonstration
Summary
The American Combustion
Pyretron Thermal Destruction
System at the U.S. EPA's
Combustion Research Facility
Under the auspices of the
Superfund Innovative Technology
Evaluation, or SITE, program, a
critical assessment was made of the
American Combustion Pyretron™
oxygen enhanced burner system
during eight separate tests at the
United States Environmental
Protection Agency's Combustion
Research Facility (CRF) in Jefferson,
Arkansas. The report includes a
description of the Pyretron and of the
facilities used at the CRF, the tests
conducted as part of this
demonstration, the data obtained,
and an overall performance and cost
evaluation of the system.
Results show that Destruction and
Removal Efficiencies (DREs) of 99.99
percent were achieved for a series of
polycyclic aromatic hydrocarbons
found in decanter tank tar sludge,
RCRA listed waste K087, the organic
waste tested during this
demonstration. Particulate emissions
of less than 180 mg/dscm at 7 per-
cent 02 were measured for all tests.
The use of oxygen enhancement with
the Pyretron enabled the feed rate of
the waste to be doubled. All solid and
liquid residues generated during
these tests were contaminant free.
The costs associated with using the
Pyretron in place of an air-only
burner depend upon the relative
costs of oxygen and fuel and to some
extent the capital costs of the
burners themselves. For this
demonstration, operating the
Pyretron with oxygen enhancement
used oxygen worth between $3250
and S3870 (it was provided free of
charge) and roughly $2672 worth of
propane. Operation without oxygen
enhancement consumed $4000 worth
of propane. During this period 1820
kg of waste were treated using
oxygen and 1180 kg were treated
without oxygen. The Pyretron burners
used in this demonstration had an
estimated cost of $150,000 and
involved $50,000 of design and
development effort.
This Summary was developed by
EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to
announce key findings of the SITE
program demonstration that is fully
documented in two separate reports
(see ordering information at back).
Introduction
The SITE demonstration of the
American Combustion, Inc. Pyretron
-------
oxygen-enhanced burner system was
conducted from November 16, 1987 to
January 29, 1988 at the U.S.
Environmental Protection Agency's
Combustion Researcli Facility (CRF) in
Jefferson, Arkansas. The Pyretron was
installed on the CRF's Rotary Kiln
Incinerator System (RKIS). This
demonstration was conducted using a
mixture of decanter tank tar sludge from
coking operations (RCRA listed waste
K087) and waste soil excavated from the
Stringfellow Superfund site near
Riverside, California. These two wastes
were mixed together to provide a feed
stream that had high levels of organic
contamination and was in a soil matrix.
This was determined to be the best feed
material to use to evaluate the
performance of the Pyretron. The
purpose of the demonstration tests was
to provide the data to evaluate three ACI
claims regarding the Pyretron system.
These claims are as follows:
• The Pyretron system with oxygen
enhancement reduces the magnitude
of the transient high levels of organic
emissions, CO, and soot ("puffs") that
occur with repeated batch charging of
waste to a rotary kiln.
• The Pyretron system with oxygen
enhancement is capable of achieving
the RCRA mandated 99.99 percent
destruction and removal efficiency
(ORE) of principal organic hazardous
constituents (POHCs) in wastes
incinerated at a higher waste feedrate
than conventional, air-only, incineration.
• The Pyretron system is more
economical than conventional
incineration.
Process and Facility
Description
Two Pyretron burners were installed on
the RKS. One was installed on the kiln
and one on the afterburner. Valve trains
for supplying these burners with
controllable flows of auxiliary fuel,
oxygen, and air; and a computerized
process control system were also
provided. A schematic of the system as it
was installed at the CRF is shown in
Figure 1. The Pyretron burners use the
staged introduction of oxygen to produce
a hot luminous flame which efficiently
transfers heat to the solid waste fed
separately to the kiln. Oxygen, propane
and oxygen-enriched air enter the burner
in three separate streams each
concentric to one another. A stream of
pure oxygen is fed through the center of
the burner arid is used to burn propane in
a substoichiometric manner. This
produces a hot and luminous flame.
Combustion | is completed by mixing
these hot combustion products with the
stream of oxygen-enriched air.
All tests were performed in the RKS at
the CRF. A simplified schematic of this
system is given in Figure 2. The system
consists of an 880 KW (3MM BTU/hr)
rotary kiln incinerator, a transition section,
a fired afterburner chamber, a venturi-
scrubber and a packed-column scrubber.
In addition, a' backup air pollution control
system conjsisting of a carbon-bed
adsorber and a HEPA filter is installed
downstream bf the previously mentioned
air pollution |control devices. With the
exception of the carbon bed and HEPA
filter, the system is typical of what might
exist on an actual commercial or
industrial incinerator. The carbon bed and
HEPA filter are installed to ensure that
organic compound and particulate
emissions to the atmosphere are
negligible.
The waste incinerated during the SITE
demonstration was a mixture of 60%
decanter tank tar sludge from coking
operations, RCRA listed waste KO87, and
40% contaminated soil from the
Stringfellow Superfund site. The K087
waste was included in the test mixture to
provide high evels of several polynuclear
automatic hydrocarbon compounds. Six
of these, naphthalene, acenaphthylene,
fluorene, phenanthrene, anthracene, and
fluoranthene; were selected as the
Principal Organic Hazardous Constituents
(POHCs) for the test program. The
Stringfellow soil was included in order to
make the resulting feedstream more
closely resemble the type of waste that
would be incinerated using this
technology. For all tests the waste was
packed into 5.7 L (1.5 gal) fiber pack
drums. Each drum contained between 4.1
and 7.9 kg (9land 17 Ib) of waste.
Eight tests were performed. These
tests were designed to compare oxygen
enhanced incineration to air-only
incineration Using the Pyretron. Table 1
summarizes ^he test conditions for the
eight tests cohducted.
During each test the feed and all
effluent streams were sampled and
analyzed to determine levels of
contamination. In addition levels of
carbon monoxide, carbon dioxide,
oxygen, total unburned hydrocarbons,
and nitrogen oxides in the exhaust gas
were continuously measured and
recorded. Comparison of the stripchart
recordings of the oxygen-enhanced and
air-only operation made it possible to
determine whether or not the controlled
introduction of oxygen reduced transient |
emissions.
Results and Discussion
A detailed summary of the SITE
demonstration test results is presented in
Figures 3 and 4 that follow and in Tables I
2 and 3. Based on the test objectives |
outlined in the Introduction, the following
results were obtained and conclusions
drawn.
Transient Emissions
American Combustion claimed that the
Pyretron with oxygen enhancement could
reduce the levels of transient emissions
that occur when solid waste is batch
charged to a rotary kiln. Transient
emissions occur when organic
contaminants originally present in the
solid waste are volitalized in the hot kiln
environment. The rapid volatilization and
reaction of these organic contaminants
depletes the kiln environment of oxygen.
Pyrolysis occurs in the oxygen-depleted
kiln environment. This results in the
production and emission of soot and
other pyrolysis products.
The basis for American Combustion's
claim of reduced transient emissions is
based upon the belief that the timed
addition of oxygen would provide
sufficient oxygen to the kiln atmosphere
to oxidize the volatilized organic matter,
thus reducing pyrolysis and the resulting
emissions of pyrolysis products. The
demonstration tests were planned to
evaluate this claim by deliberately
feeding the kiln in a way that would
produce transient emissions and then by
measuring and recording those emissions
as carbon monoxide "spikes" on
continuous emission monitor stripcharts.
The stripcharts produced under air-only
and oxygen-enhanced operation would
be compared in order to determine any
statistically significant differences in
transient emissions between air-only and
oxygen-enhanced operation.
Figures 3 and 4 show the stripchart
data for tests 2 (air-only operation at 105
Ib/hr feedrate) and 5 (oxygen
enhancement at 210 Ib/hr feedrate).
These two figures are presented to
indicate the frequency and level of
transient emissions during the
demonstration. All of the continuous
emission monitor stripcharts obtained
during the demonstration are presented
in the Technology Evaluation Report.
Comparison of the carbon monoxide
emissions indicates that no significant
differences in transient emissions could
-------
Afterburner Pyretron Burner
Measured
Process
Parameters
Valve Train
(gas, oxygen, air)
Gas, Air, and
Oxygen Flows to
the Burners
Figure 1. Pyretron thermal destruction system process diagram.
be readily observed. Statistical analysis
of carbon monoxide peak height
indicated that test-to-test variation was
greater than the variation observed
between air-only and oxygen-enhanced
operation. Thus, it was not possible to
state as a result of these tests whether
the timed addition of oxygen reduced
transient emissions.
There are two possible explanations for
the inconclusive results of these tests.
First, in order to achieve throughput
increases with the Pyretron, water was
injected into the kiln. This was not part of
the original test plan, but was later
deemed necessary in order to
demonstrate throughput increases with
high heating value waste. Even though
water injection is a reasonable way to
achieve throughput increases with high
heating value wastes, the injection of
water made it impossible to assess the
ability of the Pyretron to reduce transient
emissions through the timed injection of
oxygen. An explanation of this is as
follows.
Without the water, temperature
excursions and other operational
problems would have resulted from
feeding high heating value waste (24.61
MJ/kg) to the kiln at elevated feed rates.
Higher kiln temperatures are believed to
increase transient emissions by driving
more organic material off of the solid
waste fed to the kiln. The use of oxygen
often results in higher kiln temperatures.
The added water reduced kiln
temperatures and may have reduced
them enough to reduce transient
emissions over what they would have
been had the Pyretron been used without
water. Thus, had there been any
statistically significant reduction in
transient emissions, it would have been
the result of the addition of water and not
the timed injection of oxygen into the kiln
atmosphere.
Secondly, Superfund wastes are very
heterogeneous in nature. Even though
these wastes were mixed prior to the
start of testing, it is likely that significant
variation in organic content existed from
batch to batch during feeding. Variation in
waste feed organic content may have
also affected the variations observed in
transient emissions. It was difficult to
separate variations of tdis nature from
variations resulting from the performance
of the Pyretron. Studies of transient
emissions are best carried out in a
laboratory setting using specially
prepared, and therefore uniform,
surrogate wastes. Such a study was
conducted on a smaller version of the
Pyretron at the U.S.EPA's Air and Energy
Engineering Research Laboratory.
Results indicate that the elevated
temperatures that result from the use of
oxygen enhancement result in an
increase in transient emissions despite
the additional oxygen present in the kiln
atmosphere. This study is described in
more detail in the Applications Analysis
Report.
Destruction and Removal
Efficiencies (DREs) and
Paniculate Emissions at
Elevated Feed Rates
Even at double the feedrate, no organic
contamination was detected in all but one
of the twelve stack samples taken. Table
2 summarizes the DREs achieved. It
should be noted that for the waste treated
in this demonstration, 136 L/hr of water
had to be added to the system at the
elevated feed rates in order to achieve an
increase in throughput. This is because
the waste feed had a heating value of
-------
Afterburner
Pyrethron
Burner
Propane -5
Alr—S
Venturi
Inlet Duct
Venturi
Scrubber
50: Caustic
Air0*
Propane
Solids
Feeder
Kiln,
Pyretrqn
Burner
Cyclone ' p*cked
Separator Tower
Scrubber
Fresh
r—Process
Water
HEPA
Filter
Regulation pecirculation
Pump Tank
To
Slowdown Collection
or Disposal
Figure 2. CRF rotary Kin system.
Table 1. Summary of Demonstration Test Conditions
Test
No.
1
2
3
4
5
6
7
8
Mode*
A
A
02
02
02
02
02
A
Feed Rate
kglhr
47.7
47.7
47.7
47.7
95.5
95.5
55
55
(Ib/hr)
(105)
(105)
(105)
(105)
(210)
(210)
(120)
(120)
Temp. "F
954
921
1035
963
979
979
1010
1010
Kiln
(C)
(1750)
(1690)
(1895)
(1765)
(1795)
(1795)
(1850)
(1850)
Afterburner
02, %
13.3
12.8
17.6
14.5
13.9
14.6
13.5
8.8
Temp. °F
1121
1121
1121
1121
1121
1121
1121
1121
(C)
(2050)
(2050)
(2050)
(2050)
(2050
(2050)
(2050)
(2050)
02, %
7.7
7.4
15.2
15.0
14.0
15.3
13.5
11.4
*A « Air only
O2 - Oxygen enhanced
24.61 MJflcg (10.400 Btu/lb). At the feed
rates obtained with oxygen enhancement,
this resulted in a total heat input of 640
KW (2.2MMBtu/hr). Without all of the
nitrogen provided by air-only operation,
an additional heat sink had to be
provided. This problem would be
alleviated when treating a lower heating
value waste.
As indicated by Table 2, the stack
gases were virtually free of organic
contamination. The solid and liquid
residues produced from these tests were
also free of contamination. The
composite scrubber blowdown liquor and
kiln ash samples from each test were
analyzed for the POHCs and other
Method 8270 semivolatile organic
hazardous constituents. No POHC was
detected in any blowdown sample at a
detection limit of 20 pg/L; no other
-------
I
.
P tC
960-
SOO
3,200 -
800-
0
T5-
70-
0
270-
90-
30
^~u^__~
11 Test
Start
13
Time of Day (hr)
15
Test
Stop
Figure 3. Kiln data, Test 2.
semivolatile organic hazardous
constituent was detected at detection
limits ranging ir,om 100 ng/L
(nitrophenols and pentachlorophenol) to
20 ng/L (all other Method 8270
constituents). No POHC analyte was
detected in any kiln ash sample at a
detection limit of 0.4 mg/kg ash. No other
semivolatile organic hazardous
constituent was detected at detection
limits ranging from 2.0 mg/kg
(nitrophenols and pentachlorophenol) to
0.4 mg/kg (all other Method 8270
constituents). This high level oi
decontamination is understandable given
that all tests were performed at relatively
high kiln and afterburner temperatures.
Particulate concentrations in the flue
gas at the two locations sampled are
summarized in Table 3. The two locations
sampled were the scrubber discharge,
and the CRF stack. Between the scrubber
discharge and the stack are the carbon
adsorber and the. HEPA filter. Particulate
levels were measured in the stack for all
tests. Sampling port availability limitations
precluded measuring scrubber discharge
flue gas particulate emissions for the
tests during which simultaneous MM5
sampling was performed (Tests 1, 2, 5,
and 6).
The data in Table 3 show that
particulate levels in the scrubber
discharge flue gas for three Pyretron
tests and one conventional incineration
test were in the 20 to 40 mg/dscm at 7
percent Oa range. All levels measured
were below the RCRA incinerator
performance standard of 180 mg/dscm at
7 percent O2.
Costs
Since the Pyretron is a burner and,
therefore, only one of many components
of an incineration system, the use of the
Pyretron can be expected to affect cost
only incrementally. Since the capital cost
for any burner is only a fraction of the
capital cost for the entire incinerator, the
majority of the costs associated with the
use of the Pyretron will be associated
with the costs of fuel and oxygen. Table 4
summarizes the costs for fuel and
oxygen during the SITE demonstration. A
range of costs is presented to give the
reader an estimate of the variability in
costs associated with using this
technology. More information on costs is
provided in the Applications Analysis
Report on this demonstration.
The capital costs for the Pyretron
system used in the SITE demonstration
was $150,000. In addition $50,000 was
spent in design and development work on
the system.
Since this demonstration was done at a
research facility and not under actual
field conditions, the incremental effect
that using the Pyretron has on the cost of
incinerating a ton of hazardous waste
cannot be directly determined. It is likely
that the major factor in determining the
cost effectiveness of the Pyretron will
remain the oxygen and fuel. These costs
-------
8
7,200 -
960-
?M
S*l
.
Hi
*
800
800-
200
0
70-
0
270-
90 _
30
9-
*W\!\f**J^^
nl li— II-K-I
,
MTty\fvvvww
-------
Table 2. Destruction and Removal Efficiencies (DREs)
Test No.
Naphthalene
Acenaphthylene
Fluorene
Phenanthrene
Anthracene
Hexachloroethane 1,3,5 Trichlorobenzene
Fluoranthene
la
*
2a
*
30
40
50
*
60
*
> 99.9988
> 99. 9989
> 99.9989
> 99.9940
> 99.9986
> 99.99970
> 99.99985
> 99.99987
> 99.99989
> 99.99987
> 99.9955
> 99.9962
> 99.9954
> 99.9739
> 99.9941
> 99.9987
> 99.99942
> 99.99952
> 99. 99956
> 99.99946
> 99.9900
>99.9975
> 99.9905
> 99. 947
> 99.9978
> 99. 9974
> 99. 9988
> 99.9990
> 99.9997
> 99.9989
> 99.9967
> 99.9970
> 99.9977
> 99.956
> 99.9967
> 99.99922
> 99.99968
> 99.99972
> 99.99976
> 99.99970
> 99.9868
> 99.9898
> 99.9904
> 99.985
> 99.987
> 99.9974
> 99.99896
> 99.99909
> 99.99922
> 99.99907
> 99.9944
> 99.9957
> 99.9955
> 99. 93 7
> 99. 9926
> 99.9983
> 99.99932
> 99.99947
> 99.99944
> 99. 99929
70@
99.9957
> 99.9926
> 99.9922
> 99.9865
> indicates that the DREs are based on the analytical method detection limit
a = air, o = oxygen
* the second set of DREs. are from duplicate samples
@ tests 7 and 8 were done at the request of Region 9 and involve spiking Stringfellow soil with the two POHCs listed.
Table 3. Particulate Emission Summary
Particulate
Concentration (mg/dscm
at 7 percent O2)a
Scrubber
Discharge
Test No. Flue Gas Stack Gas
1 (12-9-87)
2 (12-11-87)
3 (12-17-87)
4(1-14-88)
5 (1-20-88)
6 (1-21-88)
7(1-27-88)
8 (1-29-88)
b
. b
21
26
' b
b
27
38
8
9
99
59
63
21
37
38
aMeasured particulate concentration
directly corrected to 7 percent O2 using
flue gas O2 level. RCRA standard is 180
mg/dscm corrected to 7 percent O2- This
does not provide a direct comparison for
tests with O2 enhancement (Tests 3, 4, 5,
6, and 7).
bDenotes measurements not performed.
Particulate levels increased at the stack
partly because of particulate entrainment
downstream of the scrubber discharge.
capacity must be provided if throughput
is to be increased with this kind of waste.
During this demonstration water was
used. This was sufficient for the
24.16MJ/kg (10,400 Btu/lb) waste treated
during the demonstration. In some
situations, however, water injection may
not provide sufficient heat absorbing
capacity. In these cases throughput
increases may be difficult to achieve.
Third, levels of NOX produced by the
Pyretron were elevated over those that
occurred without oxygen enhancement.
The high flame temperatures that result
when the Pyretron is used with oxygen
enhancement are responsible for this.
Air-only operation resulted in average
NOX levels of 92 ppm. Use of the
Pyretron with oxygen enhancement
resulted in average NOX levels of 1073
ppm. Appendix C of the Technology
Evaluation Report contains all of the NOX
data obtained during the demonstration.
The Applications Analysis Report
discusses the implications of the
Pyretron's high NOX levels.
Conclusions and
Recommendations
Based on the results and experience
obtained from this SITE demonstration,
the following conclusions and
recommendations; can be made
concerning the operation and
performance of the American Combustion
Pyretron oxygen-enhanced burner.
1. Overall, the Pyretron may be useful
in increasing the efficiency of
incinerators that are treating many of
the wastes that are found at Superfund
sites.
2. With respect to the first claim made
about the Pyretron, we were unable to
conclusively determine whether the
Pyretron system with oxygen
enhancement was able to reduce the
magnitude of transient emissions
produced when waste is batch charged
to a rotary kiln. Part of the reason for
this is that the waste feed was not
uniformly contaminated with high levels
of organic waste. Because of this,
variations in the levels of transient
emissions observed could not be solely
attributed to the action of the Pyretron.
Further, there was not a clear
difference in the frequency or level of
transient emissions produced by the
Pyretron with oxygen enhancement
over conventional incineration.
3. As for the second claim made about
the Pyretron, the demonstration clearly
showed that thorough waste decon-
tamination can be obtained at
throughput rates double those
achievable without oxygen enhance-
ment provided that sufficient heat
absorption capacity is provided when
high heating value wastes are treated.
4. As for the third claim made for the
Pyretron, the results of the
demonstration indicate that the
incremental cost of operating an
incinerator equipped with the Pyretron
will vary depending on the size and
-------
Table 4. Utility Costs Incurred During the Pyretron Site Demonstration
Mode
Air
Total Feed
kg db)
1180(2596)
Total Cost
Oxygen $
HIGH
LOW
ACTUAL
.--
--
—
Total Cost
Propane $
HIGH
LOW
ACTUAL
6000
3000
4008
Water
Injection
Cost $*
HIGH
LOW
ACTUAL ,
-
;
--
Total Utility
Cost$
HIGH
LOW
ACTUAL
6000
3000
4008
Unit Cost
$lkg ($/lb)
HIGH
LOW
ACTUAL
5.08 (2.31)
2.54(1.15)
3.39 (1.54)
02
1820 (4004)
3870
3250
3560 +
4000
2000
2672
6.12
6.12 i
6.72
7876
5256
6238
4.32 (1.97)
2.89 (1.31)
3.43 (1.56)
* only needed for high heating value wastes
* average value
location of the application as well as on
the magnitude of the throughput
increases achievable and is predom-
inantly influenced by the costs of
oxygen and fuel.
5. The NOX levels observed during the
demonstration may limit the
applicability of the Pyretron in
situations requiring stringent control of
these emissions. Further development
of the Pyretron may alleviate this
problem.
The EPA \ Project Manager, Laurel J. Staley, is with the Risk Reduction
Engineering Laboratory, Cincinnati, OH 45268 (see below).
The complete report, entitled Technology Evaluation Report, Site Program
Demonstration Test: The American Combustion Pyretron Thermal Destruction
System \at the U.S. EPA's Combustion Research Facility," (Order No. PB89-
167894/AS; Cost $28.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
A related report, which discusses application and costs, is under development.
The EPA Project Manager can be contacted at:
Risk Reduction Engineering Laboratory
i U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
EPA/540/S5-89/008
------- |