EPA/600/A-96/033
ES96-35
Destruction of Halogenated VOCs Using Premixed Radiant Burner
David F. Bartz and Bruce N. Marshall
Alzeta Corporation
2343 Calle del Mundo
Santa Clara, CA 95054
Kevin Bruce and Anthony Lombardo
Acurex Environmental Corporation
4915 Prospectus Drive
Research Triangle Park, NC 27709
C.W. Lee
Air Pollution Prevention and Control Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
INTRODUCTION
Alzeta Corporation has developed a natural-gas-fircd thermal oxidizer to provide emission
control for industrial exhaust streams where stringent emission limits of volatile organic compounds
(VOCs) are required. Measurements made with assistance from the U.S. Environmental Protection
Agency (EPA) at Research Triangle Park, North Carolina, show destruction efficiencies (DEs) between
99.9766 and 99.9999 percent for eight common halogenated compounds. This thermal oxidizer
technology utilizes an inward-fired premixed radiant burner that operates at high levels of excess air
(typically 80 to 100 percent) to achieve nitrogen oxide (NO*) and carbon monoxide (CO) emissions
below 10 ppmv, corrected to 3 percent oxygen (02). A brief discussion of pertinent regulations and
emissions of concern is presented. The evaluation test program is presented with results. A description
of three similarly designed commercial thermal oxidizers is presented including emission test results.
Regulations and Current Health Risk Assessments
The need for ultraclean burning thermal oxidation technology is driven by air emission
regulations aimed at reducing toxic or hazardous air pollutants. On the list of 190 hazardous air
pollutants proposed to be regulated by the EPA (1) are many halogenated VOCs, including carbon
tetrachloride, trichloroethylene, and methylene chloride.
Halogenated compounds present health hazards and are more difficult to fully oxidize relative to
non-halogenated compounds (2). Further, combustion of these compounds may result in the formation
of products of incomplete combustion (PICs) that may be more toxic than the original compounds;
dioxins and furans, for example (3). A permitted thermal oxidizer must meet the DE limit set for the
given industrial process (typically 95 to 99 percent of the untreated VOCs must be oxidized). If a
hazardous air pollutant is present, a screening for health risk is also required. Table 1 lists several
compounds and emission trigger limits used by the San Francisco Bay Area Air Quality Management
District (BAAQMD). For reference, the BAAQMD trigger limit for carbon tetrachloride is 2.1 kg/yr
-------
ES96-35
(4.6 Ib/yr). This corresponds to an emission level of less than 0.022 ppmv in a 28,322 L/min (1,000
scfrn) air stream flowing continuously.
Future Title III Proposals
Future air toxic regulations are expected from the EPA as part of Title III, Section 112 of the
Clean Air Act. Maximum Available Control Technology (MACT) standards for emission limits are
scheduled to be set for various source categories, including chemical manufacturing facilities, ethylene
oxide sterilization facilities, gasoline distribution stations, and dry cleaners.
Dioxins, Furans, and Formation Mechanisms
Of particular concern is control of polychlorinated dibenzodioxins and dibenzofurans
(PCDDs/PCDFs). These compounds are also potentially formed from oxidation of gas-phase
halogenated VOCs (4). Table 1 shows a BAAQMD-established health risk trigger level of 5.4 x 10"7
kg/yr (1.2 x 10*6 lb/yr) for these compounds.
There are 75 PCDD congeners and 135 PCDF congeners, all referred to genetically herein as
"dioxins," the most toxic of which is 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD). This
compound is used as a toxicity standard. The emission of these compounds from an oxidation process
may result from any of the following: 1) the presence of dioxin in a feed stream that is not fully
oxidized; 2) the formation of the dioxin from PICs; and 3) "de novo" formation from dioxin constituents
catalyzed from solid surface contact. PICs that are dioxin precursors are typically chlorinated aromatic
hydrocarbons. At temperatures above 1,000°C (about 1,800°F) the likelihood for formation and survival
of dioxins is very low, with destruction efficiencies over 99.99 percent. De novo formation occurs
typically in low temperature regions of between 230 to 400°C (450 to 750°F) with the following
components: oxygen-, hydrogen-, carbon-, and chlorine-containing compounds with particulate or metal
wall surfaces to promote catalytic reactions (3,4,5,6).
OXIDIZER TESTING AND COMMERCIAL APPLICATIONS
Evaluation Test Program
Objective. Recognizing the difficulty in oxidizing chlorinated chemicals and the potential for PIC
formation, and with encouragement from the South Coast Air Quality Management District (SCAQMD)
and funding from the Gas Research Institute (GRI), AJzeta provided Acurex Environmental Corporation
(Acurex) with an inward-fired thermal oxidizer system for VOC destruction testing. The testing was
performed by Acurex under the guidance of EPA's Control Technology Center (CTC) at Research
Triangle Park, North Carolina. CTC is set up jointly by EPA's Office of Air Quality Planning and
Standards (OAQPS) and Office of Research and Development (ORD) to assist state and local
environmental agencies to solve specific air pollution problems.
Interest in testing Alzeta's inward-fired burner technology stemmed from SCAQMD's
continuing support of emission abatement technologies that can simultaneously reduce nitrogen oxides
(NOJ and VOC emissions, without generating other harmful PICs. Thermal treatment technologies
capable of achieving this goal offer industries unable to eliminate halogenated VOCs from their
chemical processes (such as electronics and pharmaceuticals) with an another emission control option.
Burner Description. The thermal oxidizer sent for testing (Figure 1) was the twin of a commercial
system used for destroying BTEX (benzene, toluene, ethylene, and xylene) compounds at soil
remediation sites. Operating at a control temperature of approximately 871°C (1600°F), this particular
burner can process up to 2,832 L/min (100 scfin) of gaseous process media. The burner is
approximately 25.4 cm (10 in.) in diameter and 61 cm (24 in.) long. In this application the inward-fired
burner typically operates at a surface firing rate of approximately 157 kW/m2 (50,000 Btu/hr-ft2) which
corresponds to a firing rate of 73.3 kW (250,000 Btu/hr) for this burner.
-------
ES96-35
With the inward-fired radiant burner geometry, the burner is formed on the interior surface of an
annular air-fuel plenum (Figure 2) so that the radiant surface fires inwardly upon itself. Premixed air,
fuel, and VOCs enter the plenum, flow through the ceramic fiber burner matrix, and combust flamelessly
at the burner's surface. Due to gas-phase radiation, the ceramic fiber surface becomes incandescent and
reaches approximately 982°C (1,800°F). Surface-to-surface radiation exchange results in near-adiabatic
operation, resulting in stable combustion at high excess air levels (to greater than 100 percent). High
excess air operation reduces flame temperatures, resulting in low NOx emissions. Since the reactants are
well mixed before combusting in the oxygen-rich environment, CO and hydrocarbon (HC) emissions arc
typically low also.
Test Program. VOC destruction tests were performed using the inward-fired thermal oxidizer. With
assistance and guidance from CTC for the program, a test matrix was prepared using eight VOCs (Table
2). Criteria for VOC selection included common use in industry (xylene, methylethyl ketone, and
selected halogenated VOCs) and potential for formation of PICs, such as PCDDs/PCDFs. In addition to
determining the VOC destruction efficiencies, criteria pollutant emissions of NOx (NO plus N02), CO,
and HCs were measured. It was decided to repeat the chlorobenzene test three times to verify
reproducibility of the burner system performance.
Sampling Methodology. In order to achieve the specified input concentrations, VOCs were mixed with
the combustion air upstream of the burner in two ways, depending on their standard states. Gaseous
VOCs were metered into the combustion air through rotameters to result in concentrations specified in
the test matrix. Rotameters were calibrated for the specific gaseous VOC using a soap film bubble test
calibrator. Liquid VOCs were introduced by flowing a controlled portion of combustion air through an
impinger system filled with the compound prior to mixing with the remainder of the combustion air.
Calculations showed that saturation vapor pressures of the liquid-state compounds in air at room
temperature were sufficient to yield concentrations of approximately 100 to 500 ppmv. Continuous
monitoring of the VOC-laden combustion air using a flame ionization detector (FID) hydrocarbon
analyzer ensured minimal fluctuations during testing. The actual amount of each liquid-state VOC
expended was determined by mass loss measurements.
Stack emissions of 02, carbon dioxide (COj), NOx, CO, and HCs were measured using
continuous emission monitors (CEMs). HC measurements were based on piopane. Gas temperatures
were measured using thermocouples connected to system's controller displays. Stack concentrations of
target VOCs and volatile PICs were determined by Volatile Organics Sampling Train (VOST, EPA
Method 0030) (7) and Tcdlar bag methods (EPA Draft Method 0040) (8), as appropriate. Sampling and
analyzing for volatile PICs were performed for one of the three chlorobenzene tests to investigate the
potential for forming PCDD and PCDF compounds during combustion. PCDD and PCDF emissions
were determined by EPA Method 23 (9).
Prior to testing the specific compounds, field and combustion blanks were obtained as a matter of
course for quality assurance. Samples from each destruction test were analyzed for primary organic
hazardous constituents (POHCs) and PICs remaining after combustion.
Results. Destruction efficiency results for each test were presented in terms of nanograms of analyle
found on the VOST tubes or in the Tedlar bags, and in terms of the concentration in the stack associated
with the measured mass expended during the particular test.
Field blank results showed trace amounts of acetone, methylene chloride, and toluene, similar to
results from daily laboratory blanks. Combustion blanks showed trace amounts of acetone and toluene,
similar to the field blanks. Unexpected trace amounts of chlorinated PICs and some fluorinated and
brominated compounds were also found in some POHC test results. Though they were not found in the
combustion blanks, they may be due to VOST tube contamination.
-------
ES96-35
Results in Table 3 show destruction efficiencies (DEs) for all POHC test conditions. DEs
reported are not corrected for combustion blanks. Specifically:
2-Butanone CMEKV Two runs of three tests each were reported. VOST tube breakage rendered Run 2
invalid. The average DEs were 99.9998 and 99.9988 percent for Runs 1 and 3, respectively. Traces of
acetone, methylene chloride, and toluene, similar to the combustion blanks, were found in both runs.
Traces of chlorinated, fluorinated, and brominated compounds were present. The origin of the halogens
is not known, but is possibly from the natural gas.
Methylene Chloride. One run of three tests was reported. The average DE was 99.9766 percent. PIC
levels increased slightly, relative to MEK. results. Carbon tetrachloride appeared as a PIC at
approximately 0.2 ppb, probably indicating some incomplete breakdown of methylene chloride and
partial rearrangement of available chlorine.
O-Xvlene. Two runs using o-xylene at two concentrations, 36 and 61 ppm^ were conducted. Again,
VOST tube breakage invalidated Test 1 of Run 2. DEs for the 61-ppniy tests were higher than for the
36-ppmv tests, perhaps due to the higher inlet concentration. Average DEs for Runs 1 and 2,
respectively, were 99.9880 and 99.9996 percent.
Chlorobenzene. Three runs using chlorobenzene were conducted. Inlet concentrations were 50 ppmv for
Run 1 and 90 ppmv for Runs 2 and 3. DEs for the 90 ppmv tests again were higher than for the 50 ppmv
tests. Average DEs for the three runs, respectively, were 99.9982,99.9998, and 99.9996 percent.
Sampling and analysis of the stack gas from one of the tests indicated the presence of trace
PCDDs/PCDFs. This finding is discussed below.
Trichloroethvlene fTCE"). Two runs were conducted with TCE, one at approximately 100 ppm,, and the
other at 500 ppmv. The average DE for both runs was 99.9994 percent. The high DE indicates effective
treatment at high inlet concentrations. No significant PICs were noticed.
Dichloroethvlene fDCEl One run was conducted with DCE. Exhaust PIC concentrations were higher
than in the TCE test, but DEs were still high at an average of 99.9956 percent
Dichlorodifluoroethane. Two runs were conducted with dichlorodifluoroethane (Refrigerant-12). This
was introduced directly from a gas cylinder through a calibrated rotameter. Inlet concentrations were
100 and 1,000 ppmv for the two runs. The volatility of this compound required sampling with Tedlar
bags. DEs for these tests were calculated using a Practical Quantitation Limit (PQL) of 0.2 ng on the
column for the 1L injections. The average DE for both runs was 99.9999 percent No significant
concentrations of PICs were detected.
Hexafluoroethane fC^FgV Two runs were conducted with hexafluoroethane. This was introduced
directly from a gas cylinder through a calibrated rotameter. As with the Refrigerant-12 tests, inlet
concentrations were 100 and l,000v ppm for the two runs. DEs for these tests were calculated using a
PQL of 0.2 ng on the column for the 1L injections. Also, as with Refrigerant-12, no POHC was detected
in the Tedlar bag samples. The average DE for these runs was 99.9999 percent No PICs were detected.
Criteria Pollutants. Criteria pollutant emissions (as measured at approximately 10 percent 02, wet) are
shown in Table 4. In the combustion blanks, the NOx (NO plus NO2) concentration was 2.48 ppmv and
CO was 0.15 ppmv- NOx concentrations from POHC tests ranged from detectable to 2.50 ppm,,. CO
concentrations were similarly low, showing consistently negative values up to 1.87 ppmv. These results
demonstrate the performance of the burner as a low-NOx technology. No unburned HCs were detected.
4
-------
ES96-35
Actually, HC output consistently yielded negative values and are superseded by the VOST and Tedlar
bag organic data.
PICs. PCDDs/PCDFs were detected in one of the chlorobenzene tests. Analysis for these PICs was
performed by EPA Method 23 (9). Total PCDDs/PCDFs are shown in Table 5. While present, the
concentrations (7.01 ng/Nm3 corrected to 7 percent 02, total PCDDs/PCDFs) are well below the value
regulated for hazardous waste incineration in the U.S. (30 ng/Nm3, corrected to 7 percent O2). This
result is unexpected given the environment, because destruction is favored over formation at these
temperatures (6). The burner operates at approximately 1,000°C (approximately 1,800°F) and sampling
occurred at 850°C (approximately 1,560°F). Analysis of a chlorobenzene feed sample showed no
PCDD/PCDF contamination, suggesting either equipment contamination or formation in the sampling
train. The dioxin formation temperature window is between about 225 and 600°C (437 and 1,112°F).
EPA Method 23 protocol specifies a sampling probe temperature of 110°C (230°F). The sampling probe
in this case was not water-cooled, suggesting that sufficient residence time was available within the
formation temperature window. However, insufficient time was available to conduct repeat tests to
provide additional insight.
Quality Assurance. All recovery spikes for VOST measurement were within the limits of the method.
Inward-Fired Thermal Oxidizer Commercial Installations
Three commercial installations based on the inward-fired burner technology are described below.
Two installations involve semiconductor-related applications and one involves soil remediation.
NEC Electronics, Roseville, CA. In August 1994, Alzeta installed an abatement system (Figure 3) to
remove VOCs from the process air. The VOCs were primarily 1,2-dichlorobenzene, isopropyl alcohol,
phenol, and methylcthylbcnzene. The cumulative VOC mass flow was approximately 9.1 kg/hr (20
Ib/hr) in a process air stream flow rate of 28,316 L/min (10,000 cfm). The process flow diagram is
shown in Figure 4. The system uses an organic concentrator to remove VOCs from the process air. The
concentrator adsorbent is regenerated by counterflowing 24,000 L/min (850 cfm) heated air at
approximately 176°C (350°F) through the adsorbent matrix. A regeneration heater provides the heated
air stream. Desorbate concentration is typically ten times that of the process air which results in lower
fuel usage by the thermal oxidizer. The VOC-laden desorbate air forms the combustion air for the
thermal oxidizer which fires at approximately 733 kW (2.5 million Btu/hr). The user required
continuous operation with only brief annual facility-wide shutdowns for maintenance.
Test Results. Independent thermal oxidizer emission tests were required to determine compliance of
thermal oxidizer stack emissions with the local air pollution control district's regulations. With the
thermal oxidizer on-line, three separate 40-minute tests were conducted. Sampling was performed with
an on-line FID using California Air Resources Board (CARB) Method 1-100 (10). Inlet hydrocarbon
concentration was monitored using BAAQMD Method ST-7 (11). Concurrently, Tedlar bags and
Summa canisters were taken and analyzed using EPA Method TO-14 (12).
Destruction efficiencies for the combined concentrator and thermal oxidizer system were
calculated in two ways: first, based on FID measurements of total non-methane hydrocarbons, measured
as methane; and second, based on the average molecular weight of all VOCs using speciated results from
batch sample analyses. Table 6 shows the DEs for the three tests based on total non-methanc
hydrocarbons (TNMHCs). The average DE for the three 40-minute tests was 98.38 percent, with the
lowest being 98.14 percent. VOC-weighted results using spcciatcd data showed an average DE of 98.53
percent, with the lowest at 98.255 percent. An additional test was performed to determine thermal
oxidizer destruction efficiency alone. This turned out to be 99.2 percent. Portable emission analysis
results showed NOx and CO emissions of 8.9 and 12.6 ppmv, respectively.
-------
ES96-35
ECI Semiconductor, Santa Clara, CA. A 14,000 L/min (500 cfm) inward-fired thermal oxidizer
(Figure 5) was installed in mid-1995 to destroy xylene, contained in the process air which vents from
positive and negative photoresist applicators. The thermal oxidizer operates continuously.
Test Results. A compliance control efficiency test was performed as a permit requirement of the
BAAQMD. The process air inlet was continuously monitored during triplicate 30-minute tests for
TNMIICs and C02, using the BAAQMD combustion procedure, Method ST-7. Simultaneously, the
outlet was monitored for C02, 02, and TNMHCs using a FID. Inlet 02, CO, and methane were checked
and determined to be ambient since no combustion sources were related to the processes. Stack
moisture content and volumetric flow rate were measured. The DE averaged greater than 99.46 percent.
Carbon monoxide concentrations were between 15.1 and 19.2 ppmv. Analysis for NOx was not
performed.
Soil Remediation Unit. In March 1991, Alzeta supplied to an environmental remediation contractor the
twin of the unit used for the EPA tests described above. The thermal oxidizer was designed to destroy
BTEX emissions from a vacuum extraction system in the vicinity of an underground gasoline tank. The
extracted ventilation air flow was 2,832 L/min (100 scfm), with initial hydrocarbon concentrations of
1,000 ppmv. Table 7 shows results of hydrocarbon analyses performed at the inlet and exhaust streams.
Though no halogenated compounds were present, the removal of BTEX vapors to the detection limits of
0.05 ppmv was helpful in passing health-risk screening criteria.
CONCLUSIONS
Evaluation Test Program
The inward-fired radiant burner demonstrated effective treatment technology for halogenated
VOCs. The burner technology appears to offer an environment resulting in high destruction
efficiencies. Analysis of criteria pollutants using a CEM shows low emissions of NOx, CO, and HCs.
The unexpected finding of trace amounts (7.01 ng/Nm3, corrected to 1% 02) of PCDD/PCDFs
during the chlorobenzene test seems anomalous, given the thermal environment. However, further work
is required to determine the validity of the detection. PIC results showed small (0.2 ppb), but
measurable, quantities of chlorinated hydrocarbons in tests with methylene chloride and
dichlorobcnzene. Analysis of the VOST tubes used for testing is suggested. These results certainly
warrant further investigation to determine whether formation is due to the combustion process or
sampling train.
Destruction Efficiency
Based on the evaluation test program results and source tests on the commercial systems
described above, the premixed inward-fired radiant thermal oxidizer technology provides DEs of the
target compounds tested typically above 99.99 percent, including chlorinated and fluorinated
hydrocarbons. Commercial installations using this technology also show DEs above 99.9 percent,
consistent with evaluation test program results.
Suitable Applications
The inward-fired premixed radiant burner technology appears suitable for applications requiring
high destruction efficiencies and low criteria pollutant emissions potentially allowing for simplified
permitting and higher process throughput. In particular, halogenated-VOC processes would be
appropriate users, including pharmaceutical production, metal degreasing, and dry cleaning.
-------
ACKNOWLEDGMENTS
ES96-35
The authors would like to thank the South Coast Air Quality Management District for their
recommendation and support of the EPA test program. Alzeta Corporation greatly appreciates funding
made available by the Gas Research Institute for this program. A special note of thanks is given to
professional staff members at the U.S. EPA and Acurex Environmental Corporation, Research Triangle
Park, for their advice, guidance, and support for the test program.
REFERENCES
1. Public Law 101-549, "Clean Air Act Amendments of 1990," Title III, November 15,1990.
2. F. Amodio and A. Festa, "Thermal Destruction of Hazardous Chlorinated Hydrocarbons,"
International Flame Research Foundation, 1992.
3. P. Acharya, S.G. DeCicco, and R.G. Novak, "Factors that Can Influence and Control the Emission of
Dioxins and Furans from Hazardous Waste Incinerators," Paper 91-48, Western States Section
Combustion Institute Fall Meeting, UCLA, October 1991.
4. J.G. Griffith and D.M. Pitts, "Controlling Dioxin Emissions," Pollution Engineering.. November
1995, pp.50-52.
5. B. Dellinger, W. Ruby, D. Hall, and S. Mazer, "Laboratory Investigation of the High Temperature
Formation and Destruction of PCDFs," Electric Power Research Institute Conference. Palo Alto,
California, December 1984.
6. W. M. Shaub and W. Tsang, "Dioxin Formation in Incinerators," Environ. Sci. Technol.. 7(12): 721-
730 (1983).
7. EPA SW 846 Method 0030, "Volatile Organic Sampling Train," Test Methods for Evaluating Solid
Wastes, Volume II, EPA SW 846 (NTIS PB88-239223), Environmental Protection Agency, Office of
Solid Waste, Washington, DC, September 1986.
8. EPA SW 846 Draft Method 0040, "Volatile Organic Sampling Train." Test Methods for Evaluating
Solid Wastes. Volume II, EPA SW 846, Environmental Protection Agency, Office of Solid Waste,
Washington, DC, January 1995.
9. EPA Method 23, "Determination of Polychlorinated Dibenzo-p-dioxins and Polychlorinated
Dibenzofurans from Stationary Sources," Code of Federal Regulations. Title 40, Part 60, Appx. A, U.S.
Government Printing Office, Washington, DC (1991).
10. CARB Method 1-100, "California Air Resources Board Stationary Source Test Methods," Vol. I,
Method 100, June 1983.
11. BAAQMD Method ST-7, "Bay Area Air Quality Management District Manual of Procedures -
Source Test Policies "and Procedures," Vol. IV, December 1994.
7
-------
ES96-35
12. W. Winberry, N. Murphy, and R. Riggan, "Compendium of Methods for Determination of Toxic
Organic Compounds in Ambient Air," EPA-600/4-89-017 (NTIS PB90-127374), EPA Method TO-14,
June 1988.
8
-------
ES96-35
Table 1. Toxic Air Contaminant Trigger Levels for Bay Area Air Quality Management District
Compound
CAS Number
Trigger Level (lb/year)
Benzene
71432
6.70E+00
Carbon tetrachloride
56235
4.60E+00
Chlorinated dibenzodioxins and dibenzofurans
1746016**
1.20E-06
(TCDD equivalent)
Chlorobenzene
108907
1.35E+04
Chlorofluorocarbons
1.353+05
Dibromo-3-chloropropane, 1,2-(DBCP)
96128
9.703-02
Hexachlorobenzene
118741
3.90E-01
Methyl chloroform (1,1,1-TCA)
71556
6.18E+04
Methylene chloride
75092
1.90E+02
Perchloroethylene (tetrachloroethylene)
127184
3.30E+01
Trichlorethylene
79016
9.70E+01
Xylenes
1330207**
5.79E+04
• 1 lb = 0.45 kg
** This is a chemical compound group. If a CAS number is listed, it represents only a single chemical within the chemical
class (for metallic compounds, the CAS number of the elemental form is listed; for other compounds, the CAS number
of a predominant compound in the group is given).
Table 2. Matrix for Primary Organic Hazardous Constituents Used for Destruction Efficiency Tests
Test Condition Target POHCs POHCs PCDD/PCDF
Concentration (VOST) (Tedlar Bag) Test
2-Butanone (MEK)
100 ppm
3 x (2 tests)
0
0
Methylene chloride
100 ppm
3
0
0
(Dichloromethane)
30 and 60 ppm
O-Xylene
3 x (2 tests)
0
0
Chlorobenzene
100 ppm
9
0
1
Trichloroethylene
100 and 500 ppm
3 x (2 tests)
0
0
Dichloroethylene
100 ppm
3
0
0
Dichlorodifluoroethane
100 and 1,000 ppm
0
3 x (2 tests)
0
Hexafluoroethane
100 and 1,000 ppm
0
3 x (2 tests)
0
Field blank
N/A
3
1
1
Laboratory blank
N/A
2
N/A
0
Matrix spikes
N/A
2
1
0
N/A - Not Applicable
9
-------
ES96-35
Table 3. Destruction Efficiency Results
POHC or Condition
Average Run
Molecular
Average
Minimum
DRE
POHC
Weight
Destruction
Detection
Detection
Concentration
Efficiency
Limit"
(PPniv)
(g/molc)
(%)
(ng/liter)
(%)
2-Butanone (MEK)
99
72
99.9998
0.5
99.999843
2-Butanone (MEK)
92
72
99.9988
0.5
99.999831
Methylene chloride
97
85
99.9766
0.5
99.999864
(Dichlororaethane)
O-Xylene
36
106
99.9880
0.5
99.999706
O-Xylene
61
106
99.9996
0.5
99.999827
Chlorobenzene
55
12.5
99.9982
0.5
99.998371
Chlorobenzene
88
112.5
99.9998
0.5
99.999887
Chlorobenzene
92
112.5
99.9996
0.5
99.999892
Trichloroethylene
117
131
99.9994
0.5
99.999927
Trichloroethylene
542
131
99.9994
0.5
99.999984
Dichloroethylene
59
97
99.9956
0.5
99.999804
Dichlorodifluoroethane
100
102
99.9999
0.2
99.999956
Dichlorodifluoroethane "
1,000
102
99.9999
0.2
99.9999956
Hcxafluoroethane
100
138
99.9999
0 2
99.999967
Hexafluoroethane""
1,000
138
99.9999
02
99.9999967
Combustion blank
0
N/A
N/A
0.5
N/A
* Not corrected for combustion blanks.
DRE D.L. (%) - {[POHCil
-------
ES96-35
Tabic 5. PCDD/PCDF Results from Chlorobcnzene Test
Congener Lab Blank Sample Field Blank
(£|) (n§) (ng)
Total TCDD (0.007) 0.85 1.1
Total PeCDD 0.02 0.78 (0.01)
Total HxCDD 0.07 1.0 0.03
Total HpCDD 0.03 7 0.07
Total TCDF (0.006) 4.3 (0.008)
Total PeCDF 0.05 3.3 0.03
Total HxCDF 0.09 3.3 0.14
Total HpCDF 0.06 2.5 0.14
( ) Detection limit
Table 6. Destruction Efficiencies for 10,000-CFM EDGE PLUS* QR System
Condition
Average
Test Location
System Inlet
Concentrator Outlet
Oxidizer Outlet
Oxidizer temperature, °C (°F)
N.A.
N.A.
788°C (1,450)
Flowrate, L/min (dscfin)
278,345 (9,828)
293,639 (10,368)
30,446 (1,075)
Stack temperature
18°C(65)
26°C (79)
35°C (95)
H20, %
1.2
0.9
7.7
o2,%
20.9
20.9
9.7
C02, ppmv or (%)
395
(0.04)
(6.5)
CH4, ppmv
1.1
N.M.
N.M.
THC, ppmv
N.A.
20.0
2.1
TOC, ppmv
68 6
N.A.
NA
TNMHC, ppmv
274
3.9
1.1
TNMHC, lbs/hr (kg/hr) as
6.71 (3.02)
0.102(0.046)
0.003 (0.0014)
CH,
TNMHC D.E., %
98.38 (combined)
99.2 (oxidizer only)
N.A Not Applicable
N.M. Not Measured
11
-------
ES96-35
Table 7. Emission Measurements for 100-CFM EDGE QR Thermal Oxidizer for Soil Remediation
Location Inlet Concentration (ppniy) Exhaust Concentration (ppmv)
Compound
Benzene 21.3 <0.05
Toluene 73.8 <0.05
Ethylbenzene 9.5 <0.05
M&P Xylene 36.7 <0.05
O-Xylene 15.9 <0.05
Total Petroleum Hydrocarbons 457 <0.05
12
-------
ES96
Figure 1. 1OO-CFM Inward-Fired Thermal Oxidizer for VOC Destruction Testing at EPA
13
-------
ES96-35
INWARD-FIRED
POROUS BURNER
CLEAN EXHAUST
AIR + VOCs
NATURAL GAS
Figure 2. Prcmixed Inward-Fired Surface Combustion Radiant Burner
14
-------
Figure 3. 10,000-CFM EDGE PLUS4" QR Thermal Oxidizer
15
-------
ES96-35
. I EXHAUST I
PROCESS
FAN 1
SERVICE BY-PASS
PROCESS
MR
BLOWER
ISOLATION
DAMPERS
ISOLATION
DAMPERS
ISOLATION
DAMPERS
PROCESS
PAN 2
DRAFT-
BLOWER
AIR FILTER
•' (1WC)
•MO'F
DESORBATE
AIR STREAM
REOENAIR
HEATER .
ZEOUTE
CONCENTRATING
WHEEL 1W
. | NATURAL OAS |
QUENCH WATER
1M*r
.EXHAUST
TO SCRUBBER
NATURAL GAS
QUENCH CHAMBER
DESORBATE
FAN
EDOE on
THERMAL
OXIDIZER
TO DRAIN
Figure 4. Process Diagram for 10,000-CFM EDGE PLUS* QR
16
-------
ES96-35
Figure 5. 500-CFM EDGE QR Thermal Oxidizer
-------
T^rr-T^ • nn, TECHNICAL REPORT OATA
NRMRL-RTP- P~091 (PteasereadlmWuctionx on the reverie before completing)
—
1. REPORT NO. 2.
EPA/600/A-96/033
3. RE<
4.TITLE ANO SUBTITLE
Destruction of Halogenated VOCs Using Premixed
Radiant Burner
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7.authoh(s) Bartz, B. N. Marshall (Alzeta); K. Bruce,
A.Lombardo (Acurex); and C.W.Lee (EPA)
8. PERFORMING ORGANIZATION REPORT NO.
ES96-35
0. PERFORMING OROANIZATION NAME AND ADDRESS
Alzeta Corporation Acurex Environmental Corp.
2343 Calle del Mundo 4915 Prospectus Dr.
Santa Clara, CA Research Triangle Park, NC
95054 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
NA (Inhouse)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Published paper:
14. SPONSORING AGENCY CODE
EPA/60.0/13
IB. SUPPLEMENTARY NOTES AppCD pro;ject Qfficer ^ aw Mail Drop 65i 919/541-7663.
Presented at AWMA Specialty Conference, Emerging Solution to VOC and Air Toxics
Control. Clearwater. FT,. 2/2R-3 /I /9fi.
is.abstract paper describes the destruction of halogenated volatile organic com-
pounds (VOCs) using a premixed radiant burner. Alzeta Corporation has developed a
natural-gas-fired thermal oxidizer to provide emission control for industrial exhaust
streams where stringent emission limits of VOCs are required. Measurements, made
with assistance of the U.S. EPA at Research Triangle Park. NC, show destruction
efficiencies (DEs) between 99.9766 and 99-999% for eight common halogenated com-
pounds. This thermal oxidizer technology utilizes an inward-fired premixed radiant
burner that operates at high levels of excess air (typically 80 to 100%) to achieve ni-
trogen oxide (NOx) and carbon monoxide (CO) emissions below 10 ppmv. corrected to
3% oxygen (02). A brief discussion of pertinent regulations and emissions of concern
is presented. The EPA test program is presented with results. Three similarly de-
signed commercial thermal oxidizers are described, including emission test results.
17. KEY WORDS AND DOCUMENT ANALYSIS
I. DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Halohydrocarbons
Burners
Natural Gas
Organic Compounds
Emission
Pollution Control
Stationary Sources
Halogenated Volatile Or-
ganic Compounds
Radiant Burners
13 B
07C
13 A
21D
4G
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thh page)
Unclassified
22. PRICE
CPA Form 2220*1 (9*73)
------- |