United States
               [Environmental Protection
               Agency	
Air and Energy Engineering
Research Laboratory
Research Triangle Park. NC 27711
               Research and Development
EPA/600/SR-93/103   August 1993
EPA       Project  Summary

               Characterization  of  the Organic
               Emissions from  the Thermal
               Destruction  of  CFCs
               Jeffrey V. Ryan
                  As a result of the Montreal Protocol,
               an International accord Implemented to
               reduce the production and use of strato-
               spheric  ozone  depleting substances,
               considerable quantities of chlorofluo-
               rocarbons (CFCs) and halons may be
               accumulated and ultimately require dis-
               posal or destruction. Incineration  Is a
               potential destruction technology; how-
               ever, little Is known of the combustion
               emission characteristics from CFC In-
               cineration. A study has been performed
               that characterizes the organic emis-
               sions resulting from the pilot-scale In-
               cineration of trichlorofluoromethane
               (CFC-11) and dlchlorodlfluoromethane
               (CFC-12) under varied feed rates. A 293
               kW  (1 million  Btu/h)  Incinerator  was
               made available to the EPA for these
               tests. The emissions characterizations
               focused on determining the destruc-
               tion efficiencies (DEs) and major prod-
               ucts of  incomplete combustion (PICs)
               for each CFC evaluated. Sampling was
               performed to screen  for volatile  and
               semlvolatlle organic emission products
               Including chlorinated aliphatic*, chtoroben-
               zenes, chlorophenols, polychlorlnated
               dbenzodfoxins  and furans (PCDDs and
               PCDFs), and  polyaromatlc  hydrocar-
               bons (PAHs). Results Indicate that five
               nines (99.999%) DE can be achieved at
               a CFC-11 feed  rate as high as 69% by
               mass. The  formation of volatile  and
               semlvolatlle organic PICs was minimal.
               "Less than" concentrations are  pre-
               sented for target analytes not detected.
               Total PCDD/PCDF emission concentra-
               tions did not exceed 140 ng/Nm9. The
               Injection of water Into the combustion
               zone may Improve the thermal destruc-
               tion process.
   This Project Summary was developed
 by EPA'e Air and Energy Engineering
 Research  Laboratory,  Research Tri-
 angle Park, NC, to announce key find-
 ings of the research project that Is fully
 documented In a separate report of the
 same title (see Project Report ordering
 Information at back).

 Introduction
   Halogenated  hydrocarbons such as
 chlorofluorocarbons (CFCs) have  been
 implicated in stratospheric  ozone deple-
 tion. International accords are in place to
 phase out the production and/or use of
 these ozone-depleting substances (ODSs)
 before the end  of the century. Although
 some of these ODSs will be recycled, it
 will probably be  necessary to destroy sub-
 stantial quantities of some CFCs  to re-
 duce current inventories. A United Na-
 tions Environment  Programme (UNEP)
 technical advisory committee was formed
 in 1991 to evaluate the most appropriate
 ODS destruction technologies.  Incinera-
 tion was identified as a potentially  viable
 CFC destruction technology. However, the
 combustion emissions from  CFC incinera-
 tion have not  been well characterized.
 Characterizing  products of incomplete
 combustion  (PICs)  and determining  de-
 struction efficiencies (DEs) are required to
 fully evaluate the viability of incineration
 as a CFC destruction technology.
   Relatively little information is  available
 regarding CFC incineration, particularly in
 the  area of  PIC characterizations. Data
 have been collected indicating that vari-
 ous CFCs have been destroyed effectively
 by full-scale  incineration. However, infor-
 mation regarding PICs is virtually nonex-
 istent. EPA's Air and Energy Engineering
                                                                 Printed on Recycled Paper

-------
Research Laboratory (AEERL) initiated a
program to evaluate the viability of CFC
incineration, including characterization of
PICs. As part of this program, a bench-
scale study was performed thai characterized
the emissions from tricNorofluoiQmethane
(CFC-11) and dichlorodifluoromethane
(CFC-12) incineration. An emission sample
was collected to screen for pdychlorinated
dbenzo-p-dtoxins (PCDDs) and polychlo-
rinated  dibenzofurans  (PCDFs). The
screen  revealed that substantial  quanti-
ties (23.8 ng/Nm» total  PCDD/PCOF) of
PCDD/PCDF material were present in the
incinerator emissions. Full-scale CFC  in-
cineration emissions data were needed to
substantiate or refute this finding.
  Through an agreement with  EPA, a  pi-
lot-scale incinerator was made available
to evaluate the incineration of CFC-11 and
CFC-12  under conditions representative
of full-scale incineration facilities.  Under
this agreement, T-Thermal, Inc., provided
the equipment and labor support  to pre-
pare and operate the facility for the CFC
incineration tests.  Acurex Environmental
Corporation directed these tests including
coordination of sampling  and analytical
efforts.

Objectives
  The primary objective of this study was
to characterize organic emissions result-
ing from CFC incineration using a repre-
sentative incineration facility.  Particular
emphasis was placed on characterizing
PICs. A secondary objective was to con-
firm or  refute the presence of PCDDs/
PCDFs  in the incineration emissions  at
concentrations similar to those observed
during the previous  AEERL-sponsored
CFC  incineration  study. Should  similar
PCDD/PCDF emission concentrations be
observed, the screening  of  incinerator
emissions for volatile and semivolatile or-
ganic PICs may provide insight into po-
tential PCDD/PCDF formation precursors
or intermediates.

Experimental
  The incineration tests were  performed
at T-Thermal,  Inc.'s Conshohocken, PA,
test facility. The test  materials (CFC-11
and CFC-12) were incinerated at varied
feed conditions.  Four tests  were per-
formed.  Table 1 presents the target CFC
feeds for each test. Note the inclusion of
a combustion blank (no CFC incineration)
as a test condition.
  Emissions samples were collected for
volatile and semivolatile organics and sub-
sequently analyzed to determine DEs and
screen for PICs. Emissions were sampled
downstream of all pollution control devices.
Scrubber liquor samples were collected to
 Table 1. Target CFC Feed Conditions


 Test          Test Conditions

 ~1    No.2htetonfy

 2    3% (by mass) CFC-12/balance
         No. 2 fuel oil
 3    3% (by mass) CFC-11/balance
         No. 2 fuel oil
 4    50% (by mass) CFC-11/balance
          No. 2 fuel OH
also screen for semivolatile organic PICs.
T-Thermal collected and provided the data
pertaining to test facility operation includ-
ing, waste and air flows, critical tempera-
tures,  continuous  emission  monitoring
(CEM) data, and CFC feed rates.
  The T-Thermal pilot-scale test facility is
a down-fired,  turbulent flame incinerator
nominally rated at 293 kW (1  million Btu/
h). A diagram of the test facility  is pre-
sented in Figure 1. The incinerator con-
sists of a T-Thermal LV-1.3 high intensity
vortex  burner  mounted tangentially near
the top of the  vertical, refractory-lined  in-
cineration chamber. The No. 2 fuel oil and
the CFC waste stream  are introduced
through the side mounted burner, while
cooling water  is introduced through the
sixially mounted top injector. The cooling
water is  injected into the flame  region to
maintain  a consistent incineration tempera-
ture of 1,093°C (2,000°F).
  Hot gases leaving the incinerator pass
through a water-washed downcomer and
bubble up  through a pH-controlled water
bath. The saturated gas leaves the quench
tank at approximately 88°C (190°F) and
enters a liquid separating tower prior to
passing to the venturi scrubber for par-
tiiculate removal. An alkaline solution (po-
tassium hydroxide) is added to the quench
tank to neutralize add gases resulting from
CFC destruction. Gases leaving the ven-
turi scrubber pass through another liquid
disentrainment tower prior to exhausting
to a packed column scrubber.  The pH-
controlled packed column  scrubber neu-
tralizes any remaining acid gases. No  in-
duced  draft fan is used on the incinera-
tion/pollution control system. Instead, the
gases  are directed through the system
under positive  pressure. All sampling and
monitoring  locations were  at the  exit of
the packed tower.

Results
  The  incinerator  operational data are
summarized in Table 2. The data con-
tained  in this table  were provided  by  T-
Thermal. The  summary data presented
are based on the average  of the mea-
surements taken over each test period.
The CFC feed rates are presented as a
percentage of the sum of fuel and waste
mass flows. The CFC feed rates obtained
were in accordance with those established
in the original test matrix. However, the
obtained CFC feed rate for the high CFC
feed rate test condition (68.9%) was actu-
ally significantly greater than the  targeted
level (50%). The 50%  feed rate  was ex-
pected to  represent the maximum feed
rate likely to be employed.
   Five nines (99.999%) DE was achieved
for the CFC-12 low feed  rate (2.3%) and
the CFC-11 high  feed rate (68.9%). Only
three nines DE was achieved for the CFC-
11 low feed rate  (2.9%).  The DE for this
test condition may have been affected by
a CFC-11 sampling contaminant  CFC-11
was presented in CFC-12 and No. 2 oil
baseline  test  samples as  well  as field
blanks.
  The addition of water to the combustion
chamber to control  burner  temperature
may also enhance the thermal destruction
of CFCs. The addition of  water would re-
sult in an increase in hydroxyl (OH) radi-
cals. The OH radicals provide a bimolecu-
lar destruction mechanism in addition to
unimolecular bond rupture decomposition.
  Very few volatile PICs  were present in
the baseline, low feed  rate CFC-12, and
low feed rate CFC-11  test conditions. Many
of the PICs present were at or near practi-
cal quantifiable levels.  The CFC-11 high
feed rate test condition did reveal several
PICs in substantial concentrations. Chlo-
roform was present  at a relatively high
concentration (1,500-1,600 ng/Nm3). Car-
bon tetrachbride  was also evidenced but
at a much lower concentration (170  jig/
Mm3). Unexpectedly, the most prevalent PICs
were brominated. Bromodtchloromethane,
dbromochloromethane, and titxornomethane
(bfomoform) were the three most prevalent
PICs present. The presence of these PICs
is surprising because no source  of bro-
mine was identified in  the fuel and/or waste
feed.
  PCDDs and/or PCDFs were detected in
all samples  collected.  However, the
masses of PCDD/PCDF material present
in most of the test samples were at  or
near levels present  in  field  blanks. The
data are more than sufficient to provide a
quantitative comparison with PCDD/PCDF
emission  concentrations observed during
the previous AEERL bench-scale study.
Total  PCDD/PCDF emissions measured
in this study  were a factor  of 100 less
than those observed earlier. It is difficult
to determine if PCDD/PCDF concentra-
tion is a function of CFC feed rate. Figure
2 graphically depicts total  PCDD/PCDF

-------
                                      Water
                                                                                                 Sampling ^^^
                                                                                                 Locations  x""- p*
                                                                   Water Separators
               Packed Tower Scrubber
Flgun 1.  T-Thermal incineration facility.
Table 2.   T-Thermal Incinerator Operational Data
                                                         Testl
Test 2
Test3
                                                                                                                       Test 4
Primary Combustion Air Flow-kg/h (Ib/h)
Secondary Combustion Air Flow-kg/h (Ib/h)
Purge Air Flow-kg/h (Ib/h)
Cooling Water Flow-kg/h (Ib/h)
No. 2 Fuel Oil Ftow-kgfr (IMi)
CFC-12 Flow-kg/h (Ib/h)
CFC-11 Flow-kg/h (Ib/h)
Total Fuel/CFC Flow-kg/h (Ib/h)
% CFC of Total Flow
% Excess Air
Firing Rate, kW(MMBtufr)
Incinerator Temperature-0 C (°F)
Oxygen (% dry)
Carbon Dioxide (% dry)
Carbon Monoxide (ppmv dry)
Nitrogen Oxides (ppmv dry)
285.7(629.3)
100.9 (222.3)
8.6 (19.0)
NA
16.2 (35.6)
0.0
0.0
16.2 (35.6)
0.0
25.1
198 (0.676)
1,070 (1,958)
15.9
7.0
0
65
293.7(647.0)
106.4 (234.3)
9.2 (20.3)
14.7(32.4)
19.1 (42.0)
0.5 (1.0)
0.0
19.5 (43.0)
2.3
6.2
234(0.797)
1,092 (1,998)
8.9
10.7
a
43
300.5(661.8)
106.9 (235.5)
9.8 (21.5)
9.1 (20.1)
18.5 (40.8)
0.0
0.5 (1.2)
19.1 (42.0)
2.9
12.0
227(0.776)
1,092 (1,998)
9.0
9.3
10
149
295.3 (650.5)
110.8(244.0)
9.1 (20.0)
8.7(19.2)
18.8(41.4)
0.0
41.6(91.7)
60.4(133.1)
68.9
-6.1
246 (0.839)
1,121 (2,049)
7.3
10.6
15
50
NA = Not available.

-------
emissions for each test, ft appears that
the high CFC-11  feed  rate condition re-
sulted in slightly increased PCDD/PCDF
emissions. PCDD/PCDF analysis of scrub-
ber water samples yielded  inconclusive
data.  The inconsistent results observed
were likely attributable to a sampling or
analytical contaminant
  Method 23 and scrubber water samples
were  screened for sem (volatile  organic
PICs considered to be  PCDD/PCDF pre-
cursors. Essentially  all  of  the target
analytes were not detected. Based on in-
strumentation detection levels, "less than"
emission concentration levels  are  pre-
sented.   No  semivolatile organic target
analytes were detected in the  scrubber
water samples.

Summary and Conclusions
  This study effectively characterizes the
organic  emissions resulting from the pilot-
scale incineration of CFCs. CFC-12 and
CFC-11  were thermally destroyed at feed
rates  representative of full-scale thermal
destruction facilities (2.3 and 2.9%, re-
spectively).  A  high CFC-11 feed  rate
condition (68.9%) was also evaluated.
Greater  than five nines DE (99.999%) was
observed for the CFC-12 and high CFC-
11 test  conditions. Only three nines DE
(99.9%) was observed for the low CFC-11
feed rate test condition.
  The presence of volatile and semivolatile
organic  PICs was screened for. The PIC
screens included target analytes  such as
chlorinated  aliphatics, chlorobenzenes,
chtorophenols, PAHs, and PCDDs/PCDFs.
Essentially no target PICs were found in
the low CFC feed  rate test conditions. For
the high CFC-11 feed  rate test condition,
PIC screens indicated that several volatile
organic  target PICs as well  as several
non-target volatile organic PICs were indeed
present  Chloroform, bromodichloromethane,
dibromochloromethane, and  bromoform
were emitted in substantial concentrations
(1,500-2,300 ug/Nm3).  Carbon tetrachto-
ride was also emitted, but at a lower con-
centration (170 |ig/Nm3). The presence of
brominated PICs was particularly surpris-
ing, as no source of bromine was readily
identifiable; the CFC-11 and fuel  oil used
      1
                No. 2 Fuel Oil    CFC-12 Low    CFC-11 Low     CFC-11 High
                       Train A
Train B
Average
  Figure 2.  Total PCDD/PCDF emissions lor each test condition.
during testing were analyzed specifically
for trace bromine. Prior tests on the incin-
eration test facility were suspected as a
residual bromine source.
   Essentially all semivolatile organic PICs
target analytes were  not detected. This
finding is significant in that chlorobenzenes
and  PAHs,  PICs identified in a bench-
scale CFC incineration  study, were  not
detected. A bench-scale study found the
formation of these PICs to  be a function
of the halogen/hydrogen ratio.
   The PCDD/PCDF emission concentra-
tions  measured  (2-140  ng/Nm3) were a
factor of 100 less than those reported in
another AEERL-sponsored CFC incinera-
tion study, indicating that the formation of
PCDDs/PCDFs from the incineration of
CFCs may not be as large  a concern as
was  initially  suspected,  ft does appear,
however,  that increased  PCDD/PCDF
 emissions were realized at the high CFC-
 11 feed rate test condition.  The mecha-
 nisms involved in PCDD/PCDF formation
 from CFC incineration are not fully under-
 stood.
   The injection of water into the combus-
 tion zone to control incinerator tempera-
 ture  may have  several  added benefits.
 The injection of water may enhance CFC
 destruction  efficiency.  Water injection
 would lead to an  increase in hydroxyl radi-
 cal population, thereby providing a bimo-
 tecular destruction mechanism in addition
 to unimolecular thermal bond rupture. The
 injection of water may also minimize the
 formation of PICs. The injected water also
 provides  an  additional source of hydro-
 gen. Hydrogen  is  involved  in reactions
 that scavenge halogen free-radicals, po-
 tentially reducing PIC formation.
                                                                        £u.S. GOVERNMENT PRINTING OFFICE: 1993 - 750-071/80041

-------

-------
   Jeffrey V. Ryan is with Acurex Environmental Corp.. Research Triangle Park, NC
     27709.
   C.W.Leeis the EPA Project Officer (see below)
   The complete report, entitled 'Characterization of the Organic Emissions from the
     Thermal Destruction of CFCs," (Order No. PB93-205 557/AS; Cost: $36.50
     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:
           Air and Energy Engineering Research Laboratory
           U.S. Environmental Protection Agency
           Research Triangle Park, NC 27711
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/600/SR-93/103

-------