United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-87/033  June 1987
 Project  Summary
Destruction of  Dioxin-
Contaminated Solids  and
Liquids by Mobile  Incineration
Helge Mortensen, Alan Sherman, William Troxler, Richard Miller, and
Charles Pfrommer
  The EPA Mobile Incineration System,
which consists of a kiln, secondary
combustion chamber, air  pollution
control unit, and separate continuous
stack gas analysis capabilities, was
rigorously tested in 1982-1983 using
PCB-contaminated liquids and other
chlorinated organic fluids. Destruction
and  removal  efficiencies of at least
99.9999% were consistently attained
at a heat release of 10 GJ/hr.
  As  a  result of these favorable
performance  data, a project  was
initiated to  evaluate the technical,
economic, and administrative feasibil-
ity of on-site  incineration of dioxin-
contaminated materials. During 1984,
the system was extensively modified
for field use and performance-tested
with a variety of uncontaminated soils
and  other solid  wastes  at the  EPA
facility in Edison, NJ.
  Based on the results of laboratory and
pilot plant studies conducted to estab-
lish optimum kiln conditions and the
available literature, the  EPA  system
was judged to be more than adequate
for detoxifying dioxin-contaminated
solids and liquids, and thus could be
expected to  accomplish a successful
dioxin  trial  burn.  Accordingly the
system was transported  in December
1984 to the  Denney Farm  site in
McDowell, Missouri, which had been
selected for the trial burn in the inter-
vening months.
  Destruction and removal efficiencies
exceeding 99.9999% were achieved
for 2,3,7,8-TCDD during a trial burn
on dioxin-contaminated  liquids and
solids conducted in April 1985. The kiln
ash  and  process  wastewater by-
products had no detectable dioxins and
were in accordance with  guidelines
identified  by EPA's Office of Solid
Waste.
  A field demonstration on a variety of
dioxin-contaminated  materials was
conducted between July  1985 and
February 1986. A total of 0.9 million
kg of solids and 81,600 kg of liquids
was successfully decontaminated dur-
ing that time. Operations  were sus-
pended on February 6, 1986 pending
Superfund reauthorization. When oper-
ations resume, the Field Demonstration
will be completed and a second trial
burn will  take place on  materials
designated by the Resource Conserva-
tion and Recovery Act and the Toxic
Substances Control Act.
  This Project Summary was deve-
loped by  EPA's Hazardous  Waste
Engineering Research Laboratory, Cin-
cinnati, OH, to announce key findings
of the research project that is fully
documented in a separate report of the
same title  (see Project Report ordering
information at back).

Introduction
  The continued discovery of abandoned
hazardous  waste sites by  Superfund
investigations, decreasing availability of
landfill  sites,  and  increasing public
opposition  to toxic and hazardous waste
transport have placed increasing pres-
sure on the U.S. Environmental Protec-
tion Agency (EPA) to find alternatives for

-------
treating  and  disposing  of  toxic  and
hazardous wastes. The treatment  and
disposal problem is particularly acute in
the case of the highly toxic dioxin isomer
2,3,7,8-tetrachlorodibenzodioxin
(2,3,7,8-TCDD). In recognition  of these
difficulties and as a result of preliminary
favorable tests of the technology, EPA's
"Dioxin Strategy" (November 28, 1983)
recommended  that  high-temperature
incineration of dioxin-contaminated
materials  be evaluated further by the
Office  of  Research  and Development
(ORD).
  The  EPA Mobile Incineration System
(MIS) was designed and constructed for
ORD to provide such treatment  at the
actual  site of dioxin contamination. The
system consists  of  a refractory-lined
rotary  kiln,  a  secondary combustion
chamber (SCC), and air pollution control
equipment mounted on three heavy-duty
semi-trailers. Monitoring equipment is
carried by a  fourth trailer. The  ability of
the MIS to destroy tetrachloromethane
(carbon tetrachloride), dichlorobenzene,
trichlorobenzenes, tetrachlorobenzenes,
and PCBs while complying with applica-
ble Federal and State regulations for the
emissions of HCI and particulate matter
was demonstrated during a Liquid Trial
Burn conducted  between  September
1982  and January  1983 at  the  EPA
facility in Edison, NJ.
  In March  1984, the ORD  Releases
Control Branch (RGB) of the  Hazardous
Waste Engineering Research Laboratory
(HWERL),  at the request of  the  EPA
Region VII, embarked on a field validation
project to evaluate the MIS for on-site
treatment and disposal of  toxic  and
hazardous  wastes,  particularly  soils
contaminated  with  2,3,7,8-TCDD.  The
purpose of this research was  to deter-
mine the economic  feasibility of the
technique and to establish: (1) test burn
protocols; (2) health and safety protocol;
(3) site specific, risk assessment protocol;
(4) an economic model for estimating the
cost of treatment per  unit of  material
processes; and (5)  national and  state
permit protocol.
Site Selection and Planning
  Agreements were  reached in April
1984 to operate the Mobile Incineration
System on the Denney Farm near McDo-
well, MO, where over 90 drums of dioxin-
contaminated wastes had been exca-
vated  and stored in a diked  shelter. A
second  covered  concrete basin  on the
site contained over 180  m3 of soil that
had  become  contaminated when  the
buried drums leaked.
  The Denney Farm was chosen in part
because the safe removal and destruc-
tion  of dioxin that contaminated soil,
liquids, drums, trash, and chemical solids
on the  site  would  demonstrate  the
versatility of the MIS. Further, the variety
of soil types available in the immediate
area  would demonstrate that  incinera-
tion   could  decontaminate  dioxin-
containing soil found  elsewhere in
Missouri.

Laboratory and Pilot Studies
  These studies, performed concurrently
with  the Federal and State permitting
processes, investigated whether  the
objective of decontaminating the soil to
less than 1 ppb dioxin was feasible given
the operating limits of the MIS, and to
develop recommended operating condi-
tions for  the  trial burn and field
demonstration.
  Soils from  two  Missouri sites  with
confirmed  dioxin  contamination were
selected for laboratory treatability test-
ing. The average 2,3,7,8-TCDD concen-
trations were relatively high  in both soils
(563 ppb in Denney Farm soils; 338 ppb
in Piazza Road  soils), enabling investi-
gation of the maximum treatability range.
In addition, soils from both sites had wide
ranges  of pH, conductivity, organic
matter  content,  and  particle  size
distributions.
  Three series of treatability tests were
conducted. The first determined  the
optimal kiln residence time and temper-
ature that would  produce the target
treatment effectiveness of 1 ppb or  less
of 2,3,7,8-TCDD residue.  The  second
series characterized the effect of soil type
(Piazza or Denney), initial soil mositure
content, and gas phase composition on
treatability under  fixed residence time
and  temperature conditions. The  third
series included additional treatment
conditions to fill in data gaps and also
several  special tests in which 5-cm
"cubes"  (to  simulate  clay lumps) of
Piazza Road  soil  were tested under
various conditions.
  A  linear regression analysis  of  the
treatability data for  Denney farm  soil
allowed prediction  of the final 2,3,7,8-
TCDD concentrations at different time-
temperature  conditions as shown in
Figure 1  and Table  1.  There  was  no
significant correlation between  soil
treatability and either  moisture or atmos-
phere; soil type had  a  relatively minor
influence.
  In the third series of tests, a substant
lag in achieving the target test tempe
ature within a cube core was attribut
largely to the drying process. The eva
oration rate of the initial 20% moistu
content from the cube was dependent i
the heat and mass transfer character!
tics of the cube and  the  external g
temperature, which in the MIS kiln wou
be higher than 500°C.
  The results from the laboratory testii
demonstrated that the  clean-up criterii
of 1 ppb could be achieved at reasonab
kiln operating conditions and provid
part of the information needed to proj€
the specific  kiln  residence time  ai
temperature for various feed rates a
feed conditions for the MIS.

Modifications to the MIS
  Several  changes  were  made in t
original MIS  design, including genei
modifications  affecting the  refractoi
the burner controls; the stack  g
monitoring system; the electrical syste
and the design, specification, procur
ment, installation,  and shakedown  of
solids feed system.
  Further  design modifications,  inclu
ing a wet electrostatic precipitator, a
planned.

Site Preparation and
Community Relations
  After selection of the Denney Farm 1
the MIS  demonstration, detailed eni
neering and  design  were started
satisfy operating and permitting requir
ments. The actual incinerator site  w
determined by the physical dimensio
of the solids feed handling system a
the location of the contaminated mat
rials in a  prefabricated metal buildir
Further site  contamination was  pi
vented by  maintaining the incinerator
an uncontaminated area  and the fe
system in a contaminated area with
connecting sealed conveyor system.
  After several changes in the origir
design and hot and cold shakedown tes
to ensure reliable operation, the MIS w
transported to and set  up on the Denn
Farm site in mid-December 1984.

Field Shakedown  and Trial
Burn
  Final preparations, component chec!
and  on-site personnel safety traini
were completed by early January 19E
The incineration system was then start
up with fuel oil to check its performan
after transport  from  New  Jerse

-------
   900
   800
   700.
   600-
   5OO-
   400-
   300.
                                 Residual Concentration of 2,3,7.8- TCDD
                                0.1 ppb
                                ~-    -
                                 Ijipb
                                 10 ppb
    0          10         20         30        40         SO         60         70
                                                     Time Required (Minutest

Figure 1.    Effect of time and temperature on removal of 2.3.7,8- TCDD from Denney Farm soil.
                                                                                            80
                                                                                                       90
                                                                          100
Mechanical  and weather-related prob-
lems delayed the start of the Trial Burn.
Dioxin-contaminated liquids and solids
were fed to  the incinerator for the first
time at the end of February. Several more
minor problems were encountered and
corrected, and by April 1985, four dioxin
trial burn runs had been completed.

Trial Burn Plan
  The trial burn program  was originally
designed to consist of  three  tests.
However, due to  the operational prob-
lems noted above. Test 1, a burn using
a PCB matrix mixed with 5 wt % hex-
achloroethane to achieve a  5% PCB
concentration and liquid tetrachlorome-
thane, was postponed to the end of the
field demonstration. The two tests that
were performed are described below.

Test 2
  2,3,7,8-TCDD  (dioxin)-contaminated
soil and dioxin-containing waste liquids
(primarily trichlorophenol  in a solvent
mix of methylene chloride and butanol)
were fed to the rotary kiln to demonstrate
the ability of the MIS to destroy dioxin
with a destruction and removal efficiency
(DRE) of 99.9999%.

Test3
  Bromine-contaminated   industrial
chemical sludge was fed to the rotary kiln
to demonstrate the ability of the MIS to
control  bromine/hydrogen bromide
emissions while incinerating bromine-
contaminated wastes.
  The  Quality Assurance  Project Plan
(QAPP) for the Trial Burn ensured that
the trial burn data were technically sound
and  acceptable  to  regulatory offices.
Standard analytical protocols were used
whenever possible,  but the  determina-
tion  of 2,3,7,8-TCDD  in  incinerator
emissions and  by-products required
state-of-the-art analyses to demonstrate
the required DRE.  The samples were
analyzed in two  laboratories to provide
independent verification of test results.
  The  incinerator operating conditions
during  the Dioxin Trial Burn were essen-
                                                                             tially the  same  as those during the
                                                                             previous Liquid Trial Burn successfully
                                                                             conducted in New Jersey. Waste liquids
                                                                             and solids were  fed to the rotary kiln.
                                                                             The solids were  retained in the rotary
                                                                             kiln, which operated at a gas exit
                                                                             temperature of 845-955°C, for approx-
                                                                             imately 30 minutes before being dis-
                                                                             charged at 750°C into drums. The gases
                                                                             from the combustion of wastes flowed
                                                                             into the SCC where they were heated
                                                                             to 1150-1230°C. The combustion gases
                                                                             in  the  SCC were  mixed  with  excess
                                                                             oxygen (air) to a control level of 4-7%
                                                                             02 and were retained for 2.4-3 sec. The
                                                                             relatively long retention time was due to
                                                                             the operation of the incineration system
                                                                             at  low gas flow  rates  to  minimize
                                                                             particulate carryover from  the kiln into
                                                                             the  SCC,  not  to   a  DRE-related
                                                                             requirement.
                                                                              The combustion  gases then passed
                                                                             through three stages  of air pollution
                                                                             control equipment  to  cool, filter, and
                                                                             remove acid  gases (by-products from
                                                                             waste combustion) and particulate mat-
                                                                             ter (from  the solid wastes processed).

-------
Table 1.   Summary of Time-Temperature Effect on Removal of 2,3,7.8-TCDD
Nominal
Test
Temperature
PC)
429
430
429
428
429
475
478
477
479
550
550
554
616
616
.. . 616
803
808
803
Time at
Test
Temperature*
(minj
0
15
30
90
90
0
15
30
30
0
0
15
0
15
30
30
30
90
Soil
Type"
A
A
A
A
B
A
A
A
A
A
A
A
A
A
A
A
B
A
Residual
2,3.7,8-TCDD
Concentration
(ppo)
377
60
30.8
10.2
2.86
67
8.4
3.7
3.37/3.30"
24
27.5
0.16
0.2
NO (0.08)
NO (0.06)
NO (0.02)
NO (0.04)
ND (0.08)
'This time begins when the target test temperature is reached; therefore, zero time is actually
 six to nine minutes after start of heat-up.
"A: Denney Farm Soil; B: Reference Soil.
Analytical duplicate; separate aliquots of treated soil were analyzed.
Other process by-product streams (kiln
ash, CHEAP mat, and purge water) were
collected and  analyzed  in  accordance
with delisting  guidelines and the Trial
Burn Plan. The incinerator performance
during the trial burn was actually better
than reported since the actual emissions
were lower than what is measurable by
current sampling and analytical technol-
ogy. No 2,3,7,8-TCDD was  detected in
the stack, using state-of-the-art high
resolution mass spectrometry. The low-
est  ORE  for   2,3,7,8-TCDD  was
99.999973%;   the  best   DRE   of
99.99999%  during  the  trial burn
occurred in Test 2, Run 4, which had the
greatest analytical sensitivity. The
results of Test 3 were also satisfactory
in that  no bromine or chlorine was
detected in the stack gas.
  In summary, the Trial Burn  obtained
data, which, when combined with data
from the Liquid Trial Burn in Edison, NJ,
verified that (1) dioxins and other hazard-
ous organic liquid and solid materials are
destroyed by incineration in the EPA MIS
to a residual ash concentration of less
than 1 ppb,  (2) by-product ash, CHEAP
media, and water met delisting stand-
ards,  and  (3) the resulting  stack emis-
sions do not pose an unacceptable health
or  safety risk to  the surrounding
communities.

Field  Demonstration
  The  objective of the field demonstra-
tion is to  determine the rates at which
various types  of  dioxin-contaminated
liquids and solids can be fed into the
system and decontaminated. In addition,
the  demonstration will result in the
cleanup  of the majority of  dioxin-
contaminated material in southwestern
Missouri.  As of February 6, 1986,  more
than 900,000 kg of  solids  and 81,600
kg of liquids have been incinerated.


Recommendations
  The  following recommendations are
the  result of  the experience  gained
during  the design, operation, and main-
tenance of the MIS Trial Burn and  Field
Demonstration in Missouri.  The recom-
mendations center around the need to
prepare for the following circumstances:

• extreme weather conditions

• mechanical failures
• road bed failure due to inadequate site
   preparation

• unavailability of spare parts

  In addition, it is recommended that the
permitting process be started as soon at
the site is chosen. The delisting protoco
required  analytical tests on  every tank
of wastewater  and  on relatively small
quantities of treated soil to confirm that
the extracts of the material  met RCRA
Extraction Procedure (EP) toxicity criteria
before discharge  or disposal. In retro-
spect, securing a  National Pollution
Discharge Elimination System (NPDES]
permit may have allowed more efficient
site preparations and decreased analyt-
ical  costs (due to. longer  analytical
turnaround time)  since NPDES permits
normally require less frequent sampling
and reporting of analytical results.

Economic Analysis
  The unit cost ($/Mg) of waste material
processed by a  mobile incinerator
strongly  depends on  its capacity, on-
stream factor, and the associated cost.
The capacity of the MIS results from a
complex  interaction between the phys-
ical size of its components; the operating
conditions necessary to meet the DRE
and delisting requirements;  and the
waste characteristics.  The EPA  MIS
utilizes the largest size components that
can be accommodated on semi-trailers.
Operating conditions are selected con-
servatively to process a variety of waste
materials and meet the DRE and delisting
requirements. For the specific  compo-
nent sizes and operating conditions, the
capacity  of the MIS was about 450 kg/
hr for 8.8 MJ/kg heating value  lagoon
sludge and about 900 kg/hr for a low
heating value soil.
  The MIS,  being a prototype system,
experienced a low (45-55%) on-stream
factor during its operation at Missouri.
Based on the lessons learned from its
operation, a similar mobile system can
be expected  to  demonstrate  an on-
stream factor of 70%. The costs of an
MIS can be grouped into three general
categories: capital costs,  including all
costs that can be  amortized  over the
service life of the system; mobilization/
demobilization costs that are associated
with the  startup and shutdown at a given
site and  can be amortized while the unit
is located (and operated) at a given site;
and operating and maintenance costs.
   Capital costs include  the costs for
design and fabrication; development of

-------
operating procedures; providing operator
training; initial startup and shakedown;
application costs for a permit;  and trial
burn costs for general performance data.
Mobilization/demobilization   costs
include the costs for site preparation and
logistics; transportation  and  system
setup; on-site system  checkout; site-
specific testing for proof-of-performance;
and  decontamination  and demobiliza-
tion.  A review of  the  operating  costs
during the 116  days  that El  was the
operating contractor (October 3, 1985 to
February 6, 1986,  less the Christmas
holiday shutdown  of 10 days) showed
that  the MIS operating  is  field-labor
intensive.
  A  simplified economic analysis  was
performed  for an  incineration system
designed and fabricated utilizing all the
EPA MIS data and drawings. The actual
MIS  associated cost factors were  used.
The  capital cost  of $5.1  million  and
mobilization/demobilization costs of
$0.9 million were used. Site-specific cost
factors such as costs for site preparation
and logistics and system transportation
were  not included.
  An  on-stream  factor  of  70% was
assumed for this  analysis.  Operating
costs extrapolated from the actual field
operation of 116 days were $4.2 million/
yr. Based on a 15-year system life with
equipment  relocation assumed every 2
years, the unit cost of the incineration
system was calculated to vary  between
$750/Mg for low  heat content soil to
$1500/Mg for lagoon sludge.
  The planned  modifications to the
system will double its capacity, increase
the on-stream factor to 80%, increase the
capital costs by about 20%, and increrase
the mobilization/demobilization  and
operating costs by about 10%. Therefore,
the unit  costs for the modified system
is expected to be approximately $3607
Mg for low heat content soil and $7207
Mg for lagoon sludge.
Future Use of the MIS
  Further use of the  EPA MIS after the
field demonstration at Denney Farm will
be at the direction of the EPA Office of
Solid Waste and Emergency Response.
  The intention of future operations of
the MIS is to encourage commercializa-
tion of  on-site cleanup technologies
rather than to use the system strictly for
cleanup activities. The  private sector is
likely to build  improved,  more  reliable,
larger capacity, lower-cost systems of at
least equivalent performance for use in
routine cleanup operations.
  The  full report was  submitted in
fulfillment  of Contract  Number 68-03-
3255  by Enviresponse,  Inc., under the
sponsorship of the U.S. Environmental
Protection Agency. The report covers a
period from February 1984 to February
1986 and work was completed as of April
15, 1987.
  H. Mortensen and A. Sherman are with Enviresponse, Inc.. Livingston  NJ
    07039; W.  Troxler, R. Miller, and C.  Pfrommer are with IT Corp.. Knoxville
    TN 37923.
  Frank Freestone is the EPA Project Officer (see below).
  The complete report, entitled "Destruction of Dioxin-Contaminated Solids and
    Liquids by Mobile Incineration." (Order No. PB 87-188 512/AS; Cost: $18.95
    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:
          Re/eases Control Branch
          Hazardous Waste Engineering Research Laboratory—Cincinnati
          U.S. Environmental Protection Agency
          Edison, NJ 08837

-------
United States                       Center for Environmental Research                                         BULK RATE
Environmental Protection              Information                                                      POSTAGE & FEES Pf
Agency                           Cincinnati OH 45268                                                       EPA
                                                                                                   PERMIT No G-35


Official Business
Penalty for Private Use $300

EPA/600/S2-87/033
                                      0000329   PS
                                          cC
                                      CHICAGO               IL   $0604

-------