United States
                   Environmental Protection
                   Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
                   Research and Development
EPA/600/S2-86/121 May 1987
£EPA         Project  Summary
                   Pilot-Scale  Incineration  Test
                   Burn  of  TCDD-Contaminated
                   Trichlorophenol  Production
                   Waste
                   R. W. Ross, II, T. H. Backhouse, R. H. Vocque, J. W. Lee, and
                   L. R. Waterland
                     A series of three tests to evaluate the
                   incinerability of the toluene stillbot-
                   toms waste from trichlorophenol pro-
                   duction previously generated by the
                   Vertac Chemical Company were per-
                   formed in the Combustion Research Fa-
                   cility (CRF) rotary kiln incineration sys-
                   tem. This waste contained 37 ppm
                   2,3,7,8-TCDD as its principal organic
                   hazardous constituent (POHC). Flue gas
                   2,3,7,8-TCDD levels were less than de-
                   tectable at all locations sampled. Corre-
                   sponding incinerator destruction and
                   removal efficiencies (DREs) were
                   greater than 99.9997 percent, based on
                   individual sampling train analyses. By
                   analyzing combined extracts from four
                   simultaneous sampling trains, it was
                   concluded that 2,3,7,8-TCDD DRE was
                   indeed greater than 99.9999 percent.
                   These results suggest that incineration
                   of the Vertac waste  is capable of
                   achieving the required DRE and should
                   be considered a treatment option for
                   this waste.
                     This Project Summary was devel-
                   oped by ERA'S Hazardous Waste Engi-
                   neering Research Laboratory, Cincin-
                   nati, OH, to announce key findings of
                   the research project that is fully docu-
                   mented in a separate report of the same
                   title (see Project Report ordering infor-
                   mation at back).

                   Introduction
                     A primary function of the U.S. Envi-
                   ronmental Protection Agency's (EPA)
                   Combustion Research Facility (CRF) is
                   to perform incineration testing of trou-
blesome hazardous wastes to support
decisions regarding  whether incinera-
tion is a proper waste treatment and dis-
posal option. Waste contaminated with
2,3,7,8-tetrachlorodibenzo-p-dioxin
(2,3,7,8-TCDD or dioxin) is one class of
such wastes.
  An example of a well-documented,
highly dioxin-contaminated waste is the
toluene stillbottoms from trichlorophe-
nol production previously generated
and currently being stored, pending a
decision regarding appropriate treat-
ment and disposal, at the Vertac Chemi-
cal Company in Jacksonville, Arkansas.
The generator is currently considering
onsite incineration in a mobile incinera-
tor system for disposal of this waste and
wishes to have a permit for a trial burn.
The primary objective  of the tests re-
ported herein was to evaluate the incin-
erability of the Vertac toluene stillbot-
toms waste  by determining whether
99.9999 percent DRE could be achieved
as required by current regulations. Re-
sults of these incineration tests could in
turn be used to support any subsequent
permit decision.
  All tests were performed in the CRF
rotary kiln incineration system. The test
program consisted of a total of four trial
burns performed in  September 1985.
These trial burns consisted of the fol-
lowing:

  • A blank burn with the incinerator
    fired with auxiliary fuel (propane)
    to establish  background  emission
    levels of pollutants of concern

-------
  • A  miniburn  of short duration
    (4 hours) with waste fired at nomi-
    nally 17 kg/hr (38 Ib/hr) to demon-
    strate the ability to feed and incin-
    erate the waste  and to gain
    experience with the sampling pro-
    tocols specified
  • Two full waste test burns of nomi-
    nally 10-hour duration with the
    waste fired at about 10 and 18 kg/hr
    (22 and 39 Ib/hr) to specifically ad-
    dress the test  objectives

Facility Description, Waste
Characterization, and System
Operation
  The rotary kiln incineration system at
the CRF consists of a rotary kiln primary
combustion chamber,  a fired after-
burner, and a primary air pollution con-
trol system consisting of a  venturi
scrubber, wetted  elbow, and packed
tower scrubber. In addition, a backup air
pollution control system (APCD)  con-
sisting of a  carbon-bed absorber and a
HEPA filter is in  place. The primary
APCD might be considered reflective of
what might exist in an actual commer-
cial or industrial incinerator. The backup
system is in place to ensure that organic
pollutant and paniculate emissions to
the atmosphere are negligible. A sche-
matic of the system is given in Figure 1.
  For these tests, waste was introduced
at the feed face through the front face
lance with  a diaphragm-type pump,
while auxiliary fuel (propane) was fired
through a burner located at the transfer
duct end of the kiln. The afterburner was
also fired with  auxiliary fuel.
  Table 1 summarizes the nominal in-
cinerator operating conditions for each
of the  tests performed. For all four,
propane was fired in the kiln  and the
afterburner to maintain the kiln at about
1800°F and the afterburner at about
2000°F. This corresponded to  heat in-
puts of about 260 to 350 kW (0.9 to
1.2 x 106 Btu/hr) in the  kiln  and about
470 to 560 kW (1.6 to 1.9  x 106 Btu/hr) in
the afterburner for all four tests.
  The amount  of waste fed into the kiln
was monitored  by recording the
changes in the waste container weight
reading. The feedrates fluctuated over a
wide range with mean rates of 17 kg/hr
(38 Ib/hr) on  September 9; 10 kg/hr
(22 Ib/hr) on September 20; and 18 kg/hr
(39 Ib/hr) on September 21, 1985.
  System residence times were calcu-
lated based on volumetric flowrate
measurements using a helium tracer
system and the assumption that the kiln
and afterburner chamber temperatures
were isothermal and equal to the single
point  measurement noted in Table 1.
The calculated residence time in the kiln
main chamber was 5.7 sec for the blank
burn,  5.3 sec for the miniburn, and 4.9
and 6.0 sec for the two still bottoms
waste full burns. Residence times in the
afterburner were 2.1,  1.9, 1.8, and 2.3
sec for the respective  tests.
  The generic composition of the tol-
uene stillbottoms waste based on previ-
ous data developed by the Vertac
Chemical Company and the physical
characteristics of the  waste based on
analysis performed on a sample of the
waste at the CRF are  given in Table 2.
The waste was also analyzed for organic
and trace element priority pollutants.
Results of these analyses are given in
Table 3. Specific analyses for the Princi-
pal Organic Hazardous Constituent
(POHC) in  the waste,  2,3,7,8-TCDD,
showed that it contained 37 ppm of this
compound.
  Significant problems  were experi-
enced in attaining and maintaining
waste feed throughout  the test pro-
gram. Specific problems included con-
tinued feed lance clogging, which was
caused by carbon buildup (coking of the
waste material), in the lance, pump
check valve seal failure, and the ability
to pump waste. The feed lance clogging
problem was solved by cofeeding water
with the waste so that when the lance
clogged it would heat and vaporize the
water, thereby, clearing the clog. Feed
line clogging and check valve seal stick-
ing were temporarily solved by cleaning
all feed line components with solvent
(toluene). However, in  retrospect the
choice of a diaphragm pump for this
waste was inappropriate. The waste
was waxy and very  viscous at room
temperature. Only at about 95°C (200°F)
would it flow sufficiently to be consid-
ered pumpable. Hot water heating coils
were  immersed in the waste for these
tests, however, pumping problems per-
sisted. Perhaps a pump of another de-
sign, such as a  progressive cavity
pump, would have provided better serv-
ice.

Sampling and Analysis
Protocol
  The combustion gas generated dur-
ing each test was monitored at various
locations in the system for CO, C02, 02,
NOX, total hydrocarbon (THC), HCI, par-
ticulate, and trace semivolatile organic
compounds, most  importantly  2,3,7,8-
TCDD. Additionally, grab samples we
obtained of the waste feed, the scrubb
system  blowdown liquid, and the a
collected in the ash pit during the tes
Ambient air sampling, both in the hi)
bay incinerator room and in the outsit
vicinity of the CRF, was also performs
Figure 2 summarizes the sampling loc
tions and types of samples obtained.
  Waste samples were analyzed f
2,3,7,8-TCDD  by dilution, cleanup, ar
high resolution gas chromatograph
low  resolution  mass spectromet
(HRGC/LRMS); for the halogenatt
volatile organic priority pollutants by c
lution, purge and trap  GC/electron ca
ture detector (ECD); for the semivolati
organic  priority pollutants  by dilutio
cleanup, and HRGC/LRMS in accor
ance with Method 8270; and for the pi
ority pollutant trace elements by ac
digestion and atomic  absorption tecl
niques.
  Kiln ash and blowdown water san
pies were analyzed for polychlorinate
dibenzo-p-dioxins (PCDDs) and pol'
chlorinated dibenzofurans (PCDFs) i
chlorine substitution 4 through 8,  ar
for 2,3,7,8-TCDD by benzene extractioi
extract concentration and cleanup, an
HRGC/HRMS.
  Blowdown water samples were als
analyzed for  the  halogenated volatil
organic priority pollutants by purge an
trap GC/ECD; for the  semivolatile o
ganic priority  pollutants by benzene e:
traction, extract  concentration, an
HRGC/LRMS;  and for the priority polli
tant trace elements by atomic absorj
tion.
  Modified Method 5 (MM5) train san
pies were benzene extracted, extracl
for all train components combined, cor
centrated, and subjected to extrac
cleanup procedures.  These extracl
were then analyzed for 2,3,7,8-TCDD b
HRGC/LRMS.  In addition, extracts fc
the four simultaneous MM5 trains ope
ated  downstream of the scrubber sy;
tem for the two  full-burn tests wer
combined and analyzed for 2,3,7,{
TCDD. The samples from this area of th
system, the "virtual stack" or "E-duct1
are very important since these data wi
be used  to design future systems.

Test Results
  Levels of 02, C02, CO, and NOX in th
flue gas at the afterburner exit and i
the stack for  the four  tests performe
are summarized in Table 4. As showr
all tests were  performed at high exces
air; flue  gas O2 was in  the 10 to 17 pei

-------
                                          Venturi
                                        Inlet Duct
 Propane
     Transfer
      Duct
 Propane
                                                                             Cyclone     Packed Tower
                                                                             Separator     Scrubber
Burner
No. 2
         Burner  (TAsh
         No. 1    II Bin
                                                        Retire ulation    0   .   ...
                                                            Pump       *eclrculat,on
                                                                                                          Sanitary
                                                                                                           Sewer

                                                                                                        -» Chemical
                                                                                                           Sewer
 Figure  1.    Simplified rotary kiln system schematic.
                                                                           Slowdown   Blowdown
                                                                              Tank        Tank
                                                                             No. 1        No. 2
Table 1.    Incinerator System Operating Conditions

Kiln operation
Propane heat input, kW
(106 Btu/hr)
Waste feedrate, kg/hr
llb/hr)
Waste heat input, kW
(106 Btu/hr)
Exit gas temperature, °C
<°f)
Background burn
(9/4/85
0950 to 1940)
Range Average

260 to 290 280
(0.9 to 1.0) (0.95)
0
0
0
0
950
(1750)
Miniburn
(9/9/85
0945 to 1520)
Range

180 to 290
(0.6 to 1.0)
8 to 23
(18 to 50)
37 to 101
(0. 13 to 0.35)

Average

220
(0.72)
17
(38)
77
(0.26)
990
(1820)
First full burn
(9/20/85
1030 to 2230)
Range

320 to 440
(1.1 to 1.5)
Oto25
(0 to 56)
Oto 114
(0 to 0.39)

Average

384
(1.3)
10
(22)
44
(0.15)
980
(1800)
Second full burn
(9/21/85
1735 to 2330)
Range

290 to 350
(1.0 to 1.2)
2.7 to 40
(6 to 87)
12 to 180
(0.04 to 0.60)

Average

330
(1.1)
18
(39)
80
(0.27)
990
(1810)
Nominal residence time sec8

Afterburner operation

Propane heat input, kW
<106 Btu/hr)
                    470 to 500
                    (1.6 to 1.7)
                                     5.7
  480
(1.64)
440 to 500
(1.5 to 1.7)
                            5.3
 470
(1.6)
500 to 590
(1.7 to 2.0)
                                                                                      4.9
 560
(1.9)
500 to 560
(1.7 to 1.9)
                                                                                                               6.0
 520
(1.8)

-------
 Table 1.    (continued)

Exit gas temperature, °C
m
Background burn
(9/4/85
0950 to 1940)
Range Average
1120
(2040)
Miniburn
(9/9/85
0945 to 1520)
Range Average
1120
(2050)
First full burn
(9/20/85
1030 to 2230)
Range Average
1110
(2030)
Second full burn
(9/21/85
1735 to 2330)
Range Averagi
111
(203C
Nominal residence time seca

APCD operation

System water makeup rate, Umin
(gpm)

System blowdown rate, Umin
(gpm)

Scrubber liquor pH
  12 to 17
(3.2 to 4.4)

 6.8 to 9.1
(1.8 to 2.4)

 8.4 to 8.5
                2.1
  15
(4.0)

 8.2
(2.2)

 8.5
  10 to 19
(2.7 to 4.9)

 4.9 to 8.3
(1.3 to 2.2)

 8.0 to 8.7
                        1.9
  15
(3.9)

 7.2
 1.9
  9.5 to 30
(2.5 to 7.8)

  3.4 to 13
(0.9 to 3.4)
 8.2   8.0 to 8.6
                                                           1.8
 21
(5.5)

 8.7
(2.3)

 8.2
  6.8 to 23
(1.8 to 6.0)

  4.5 to 11
(1.2 to 2.8)

 6.8 to 8.8
                                                                   2.
(4.2

 8.
(2.3

 8..
Venturi pressure drop, kPa
(in. WC)
Venturi exit gas temperature, °C
<°F)
Packed tower pressure drop, kPa
(in. WC)
Packed tower exit gas temper-
ature, °C (°F)
10.7 to 10.9
(43 to 44)

0.35 to 0.50
(1.4 to 2.0)

10.7
(43)
79
(174)
0.44
(1.8)
78
(172)
8.2 to 10.0
(33 to 40)

0.30 to 0.50
(1.2 to 2.0)

9.5
(38)
80
(176)
0.40
1.6
78
(173)
7.5 to 10.0
(30 to 40)

0.47 to 1.3
(1.9 to 5.2)

8.6
(35)
81
(178)
1.0
(4.1)
80
(176)
7.0 to 9.7
(28 to 39)

0.35 to 1.3
(1.4 to 5.4)

8.
(34
8.
(18C
0.8.
(3.4
8
(176
'Residence times based on volumetric flows calculated using the helium tracer system data.
 Table 2.   Stillbottoms Waste Generic
          Composition and Physical
          Characteristics

Compound
Methanol
Toluene
Dichlorobenzenes
Trichlorobenzenes
2,4,5-trichloroanisole
Na-trichlorophenol
Dichloromethoxybenzene
2,4,5-T, Na salt

Parameter
Bulk density, g/ml
Loss on drying, percent
Ash
Heating value, MJ/kg
(Btu/lb)
Concentration
(percent)
1
8
1.5
1.5
56
7
16
7
Value
(Percent)
1.37
13.2
5.1
16.11
(6945)
 cent range in the afterburner exit and in
 the 13 to 17 percent range in the stack.
 CO emissions were  always low, <10
 ppm, as were NOX levels, <30 ppm. The
 missing data noted in Table 4 illustrate
 another problem experienced during
 the tests, that of continuous monitor
         failure. Although the CRF has two com-
         plete monitoring systems (02, CO2, CO,
         and NOX), during these tests only one of
         each set of instruments was working
         and even some of these working instru-
         ments would periodically fail.
           Table 4 also noted the HCI emission
         rates for those tests for which these
         rates were measured. For the miniburn,
         the HCI emission rate was 0.45 kg/hr as
         measured by the continuous HCI moni-
         tor at the E-duct; for the September 20
         full burn the HCI emission rate was 0.25
         kg/hr as measured both by the HCI mon-
         itor and by the MM5 trains operated at
         the E-duct.  Both of these are less than
         the CRF permit level of 0.5 kg/hr.
           Table 5 summarizes the participate
         and 2,3,7,8-TCDD emission levels meas-
         ured at various locations in the inciner-
         ator system for each of the tests. As
         shown in the table and as expected,
         paniculate  levels at both the kiln and
         afterburner exit were quite low during
         the background burn with propane fuel
         alone.  Flue gas paniculate levels for the
         tests with waste feed were highly vari-
         able and ranged from several to several
         hundred mg/dscm. Paniculate levels at
         the virtual stack were indicated to be as
                                  high as 340 mg/dscm for one test. Dui
                                  to technical factors and since the  0
                                  monitor at this location was not operal
                                  ing properly at the time of the test, accu
                                  rate figures, corrected to 7 percent 0;
                                  cannot be  derived from the raw date
                                  Since results are uncertain, further wor
                                  on measurement of particulate emis
                                  sions from dioxin-contaminated wast
                                  is needed.
                                    Particulate emissions at the stack syj
                                  tern (after the High Efficiency Particulat
                                  Filter)  were  also  indicated  as bein
                                  much higher than anticipated. The va
                                  ues are 5 to 15 times higher than desig
                                  values and are therefore suspect. If the
                                  are as  high as indicated, further test in
                                  is needed to  determine the nature an
                                  source of the particulate.
                                     Flue gas levels of 2,3,7,8-TCDD wer
                                  less than method detection limits at a
                                  locations for all tests. These levels coi
                                  respond to the DRE values noted  i
                                  Table  5 for the tests with waste feec
                                  Two sets of DRE values are noted in th
                                  table for the E-duct and stack location;
                                  These correspond  to two differer
                                  measures  of flue gas flowrate. One c
                                  these was  based on a helium tracer ir
                                  jection system; the other was based o

-------




Rotary
Kiln


I

Afterburner




Venturi
Scrubber/
Wetted
Elbow


Packed
Tower
Scrubber


I

Carbon
Bed
Filter



HEPA
Filter





(1) (2) (3) (4) (5) (61 (7> (8)
Test Sample No. of
Parameter Points Methodology Duration Samples
C02

02


CO

/vo,

HCI

THC


Semivolatile
organics
including
2,3.7.8-TCDD

Waste

Kiln ash
Scrubber
blowdown
3.4.8

3.4.8


3.4.8

3,4,8

6

6,7


3,4.6,8




 1

2
/VO//?

Zirconium oxide
sensor

NDIR

Chemiluminescence

Color/metric

Flame ionization
detector

MM5
Grab

Grab


Grab
Continuous

Continuous


Continuous

Continuous

Continuous

Continuous"
10 hours
(7 dscm)
(250 dscfj
                                                8 composites after
                                                entire burn series

                                                8 composites after
                                                entire burn series
2 at test pt 3
2 at test pt 4
4 at test pt 6
2 at test pt 8

     3

     8
'The two test points were monitored by one instrument on a 5-min time-sharing basis.

 Figure  2.    Summary of the general sampling protocol.
Table 3.   Composition of the Stillbottoms Waste


              Component
                    Concentration3
                      (ppm, wt)
                  Detection limit
                     (ppm, wt)
Volatile organic priority pollutants

Methylene chloride
1,1-dichloroethylene
1,1-dichloroethane
t- 1,2-dichloroethylene
Chloroform
1,2-dichloroethane
1,1,1-trichloroethane
Carbon tetrachloride
Bromochloromethane
1,2-dichloropropylene
t- 1,3-dichloropropylene
Trichloroethylene
Benzene
1,1,2-trichloroethane
Bromoform
Tetrachloroethylene +  tetrachloroethane
Chlorobenzene
Toluene
                          277
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                       159,000
                        42
                        42
                        70
                        84
                        42
                        42
                        70
                        42
                        84
                        42
                        42
                       na
                        42
                        70
                        84
                        42
           the MM5 train velocity measurements.
           The data in Table 5 suggest that 2,3,7,8-
           TCDD ORE was  generally greater than
           99.9997 percent in the virtual stack
           which would likely correspond to the
           stack of an actual  hazardous waste in-
           cinerator.
             Method detection limits did not allow
           unambiguously establishing  that
           greater than 99.9999 percent ORE was
           achieved either at the virtual stack or at
           the system stack. Therefore,  the ex-
           tracts from the four MM5 trains oper-
           ated  at  the virtual stack for the two
           full-burn tests were combined and rean-
           alyzed in an attempt to achieve better
           detection limits.  Calculated  2,3,7,8-
           TCDD levels for the virtual stack and
           corresponding 2,3,7,8-TCDD DREs
           based on the combined extract analyses
           for the second full burn  (9/21/86) are
           given in Table 6.  Data for the second full

-------
Table 3.   {continued}
             Component
                    Concentration3
                      (ppm, wt)
                        Detection limit
                          {ppm, wt)
Semivolatile organic priority pollutants

1,2-dichlorobenzene
1,2,4-trichlorobenzene
All other base/neutral semivolatile
  priority pollutants
1,2-dichlorophenol
4-chloro-3-methylphenol
2,4,6-trichlorophenol
2,4-dinitrophenol
4-nitrophenol
2-methyl-4,6-dinitrophenol
Pentachlorophenol
All other acid semivolatile priority
  pollutants

Trace Elements
                         2,690
                         3,410
                         NO

                           159
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
aND denotes not detected at the detection limit noted.
Table 4.    Emission Monitor and HCI Emission Rate Data
    Parameters
Background
   burn
  9/4/85
Minibum
 9/9/86
  First
full burn
9/20/85
Afterburner exit:

  O2 (percent)
  CO2 (percent)
  CO (ppm)
  NOX (ppm)

E-ducf.

  HCI (kg/hr)
    Continuous
    Analyzer
    MM5 train

Stack:

  O2 (percent)
  CO2 (percent)
  CO (ppm)
  NOX (ppm)
     75
     8

     30
    12
    7
   10
    4
     17
      7
   20
                     0.45
   18
    3

   30
10





0.25

0.25
                             500
                             600
                             300
                             300
                             500
                             500
                             500
                             100
Antimony, Sb
Arsenic, As
Beryllium, Be
Cadmium, Cd
Chromium, Cr
Copper, Cu
Lead, Pb
Mercury, Hg
Nickel, Ni
Selenium, Se
Silver, Ag
Thallium, Tl
Zinc, Zn
ND
ND
ND
ND
ND
ND
4
ND
ND
ND
ND
ND
ND
2
1
3
1
1
1
1
1
1
1
1
1
10
          Second
         full burn
          9/21/85
             17
             4
                20
                13
                 8
— Denotes monitor not operating or measurement not made.
burn  clearly show that greater the
99.9999 percent DRE was achieved. Tf
extracts for the first  full burn wei
spiked  with an order of magnituc
higher level of recovery standard the
deemed appropriate by the offsite labi
ratory that originally analyzed the inc
vidual train extracts. Consequent!
method detection limits correspondir
to ng/dscm of flue gas were not any be
ter  than the individual train analys
data summarized in Table 5.
  The kiln ash and the scrubber blo\
down water from this entire test serii
were  analyzed for PCDDs and PCDF
chlorine substitution 4 through 8. Tl
kiln ash samples were devoid of PCOE
and PCDF to detection limits rangir
from  3 to 40 ppt as shown in Table
Similarly, the data in Table 8 show th
scrubber blowdown samples were d
void  of all PCDDs and PCDFs exce
octa-CDDs which were present at 0.(
ppt. This is not surprising since oct
CDDs are relatively common in enviro
mental samples.
  The scrubber blowdown was also a
alyzed for the organic and trace eleme
priority pollutants. Results are summ
riz'ed  in Table 9. As shown, no organ
priority  pollutant was present  in  tt
blowdown at levels greater than 10 pp
In addition, none of the trace elemen
were  present at concentrations whi<
would cause the blowdown water to t
considered EP (Extraction Procedur
toxic. Based on all analytical data, tr
blowdown would  not  be  considered
hazardous waste.

Conclusions
  A series of incineration experimen
was performed with the Vertac Chenr
cal Company's toluene  stillbottorr
waste from trichlorophenol productio
This  waste is one of the more we
known of the  dioxin-contaminate
wastes presently in existence. Sample
of the waste tested in this study coi
tained an  average of  37  ppm  2,3,7;
TCDD (37 u,g/g).
  Three incineration tests were  pe
formed  during  September  1985. A
were  performed in the CRF rotary  kil
incineration system with waste feedral
nominally 20 kg/hr.
  With  regard to  the  principal  objei
tives of these tests, the following can b
concluded:
  • 2,3,7,8-TCDD DRE based  on  th
    combined extracts from the foi
    MM5 trains  at the virtual  stack wa
    greater than 99.9999 percent fc

-------
Table 5. Paniculate and 2,3,7,8-TCDD Emissions and 2,3,7,8-TCDD ORE
Paniculate emissions
Date and Test location
Background burn 9/4/85
Kiln exit: Train 1
Train 2
Afterburner Train 1
exit: Train 2
E-duct: Top train
Bottom train
Left train
Right train
Stack: East train
South train
Miniburn 9/9/85
Kiln exit: Train 1
Train 2
Afterburner Train 1
exit: Train 2
Full burn 9/20/85
Kiln exit: Train 1
Train 2
Average
Afterburner Train 1
exit: Train 2
E-duct: Top train
Bottom train
Left train
Right train
Average
Stack: East train
South train
Average
Full burn 9/21/85
Kiln exit: Train 1
Train 2
Afterburner Train 1
exit: Train 2
Average
E-duct: Top train
Bottom train
Left train
Right train
Stack: East train
South train
Average
(mg/dscm
as measured)

0.28
0.50
0.83
3.4
__b
	 b
—b
	 b
2.3
0.4

463
175
585
2,030

2.71
1.46
2.09
371
266
564
291
251
343
38.4
56.0
47.2

753
1,940
154
199
176
<0.1S
2.3
10.8
<0.13
72.6
208
140
(mg/dscm 2,3,7,8-TCDD
corrected to emissions
7 percent O2a) (ng/dscm)

<0.95
<0.36
1.9 <1.7
7.9 <0.43
<0.18
<0.24
<0.31
<0.31
8.1 <0.21
1.4 <0. 10

<0.74
<1.4
<13
<11

<0.55
<0.17
<3.8
<1.3
<0.43
<0.51
<2.5
<0.55
<2.3
<1.5

<6.4
<21
<3.9
<2.8
<1.9
<2.0
<2.2
<0.76
132 < 1.5
378 < 1.6
255
2,3,7,8-TCDD ORE (percent)
Based on
helium tracer






>99.99995
>99.99991
>99.9984
>99.9986

>99.99992
>99.99998
>99.99903
>99.99967
>99.99982
>99.99979
>99.99896
>99.99977
>99.99907
>99.99945

>99.99964
>99.99988
<99.99956
>99.99969
>99.99969
<99.99967
>99.99864
>99.99988
>99.99975
09.99973
Based
on flue gas
velocity











>99.99987
>99.99985
>99.99925
>99.99983
>99.99916
>99.99951



>99.99974
O9.99973
>99.99970
>99.999897
>99.99974
>99.99973
aO->meter not functioning properly. Figures, when given, are estimates.
bNo filters were used in the E-duct trains for this test.
°Evidence of a significant sampling train leak was discovered after sampling was complete. Data were considered invalid for paniculate
 measurements.

-------
Table 6.  2,3,7,8-TCDD Emissions and ORE Based on Combined E-Duct Train Extracts

                                                    2,3,7,8-TCDD ORE
Test date
Full bum 9/21/86
2,3,7,8-TCDD
emissions
(ng/dscm)
<0.066
Based on
helium tracer
>99.999989
Based on
flue gas velocity
>99.999991
Table 7.  Levels of PCDD and PCDF in Kiln Ash Samples

                                       Concentration" (ppt)b
Analyte
2,3,7,8-TCDD
TCDDs-CDD
Penta-CDDs
Hexa-CDDs
Hepta-CDDs
Octa-CDDs
2,3,7,8-TCDF
TCDFs
Penta-CDFs
Hexa-CDFs
Hepta-CDFs
Octa-CDFs
Sample 1
(13)
(13)
(28)
(7.4)
(14)
(44)
(7.0)
(7.0)
(11)
(4.6)
(12)
(33)
Sample 2
(10)
(10)
(6.5)
(4.5)
(6.3)
(18)
(16)
(16J
(2.7)
(2.8)
(6.2)
(21)
Sample 3
(5.6)
(5.6)
(4.2)
(5.0)
(3.4)
(15)
(7.1)
(7.1)
(1.7)
(3.1)
(3.4)
(15)
Sample 4
(28)
(28)
(16)
(37)
(8.4)
235
(10)
(10)
(6.4)
(7.4)
(8.4)
(40)
'Numbers in parentheses denote analyte not detected to the detection limit noted in parenthe-
 ses.
b1 ppt = 1 pg/g.
    one test. For the  other tests
    method detection limits preventec
    quantitating that  better thar
    99.9998 percent ORE was achieved
  • Accurate determination of particu
    late emissions at the virtual stacl
    and system stack was not achieved
    Further research is needed to ob
    tain data on the amount, nature anc
    source of paniculate  emission:
    from these sources.
  • HCI emissions in the virtual stacl
    ranged from 0.2 to 0.45 kg/hr
    These  were below the CRF Part E
    limit of 0.5 kg/hr.
  The above conclusions  suggest  thai
incineration should be considered a vi
able disposal  method for this stillbot
toms waste, that is if appropriate safe
guards are employed. The data in this
study confirm that an incinerator oper
ating under  proper conditions  car
achieve greater than 99.9999 percen
ORE for 2,3,7,8-TCDD  with HCI emis
sions below the  regulatory limit.
Table 8. Priority Pollutant Composition of the Scrubber Slowdown Water

                                       Concentration' (ppt)b
Analyte
2,3,7,8-TCDD
TCDDs-CDD
Penta-CDDs
Hexa-CDDs
Hepta-CDDs
Octa-CDDs
2,3,7,8-TCDF
TCDFs
Penta-CDFs
Hexa-CDFs
Hepta-CDFs
Octa-CDFs
Sample 1
(0.005)
(0.06)
(0.04)
(0.03)
(0.02)
0.07
(0.02)
(0.1)
(0.02)
(0.02)
(0.1)
(0.1)
Sample 1
duplicate
(0.02)
(0.08)
(0.05)
(0.04)
(0.04)
0.04
(0.01)
(0.07)
(0.02)
(0.06)
(0.02)
(0.08)
Sample 2
(0.02)
(0.09)
(0.04)
(0.04)
(0.03)
0.07
(0.02)
(0.1)
(0.07)
(0.06)
(0.03)
(0.06)
Sample 2
duplicate
(0.04)
(0.09)
(0.02)
(0.03)
(0.05)
0.07
(0.02)
(0.06)
(0.03)
(0.06)
(0.03)
(0.04)
'Numbers in parentheses denote analyte not detected to the detection limit noted in parenthe-
 ses.
b1 ppt = 1 pg/ml.

-------
Table 9.  Priority Pollutant Composition of the Scrubber Slowdown Water

                                                          Detection
Component
Volatile organic priority pollutants
Methylene chloride
1, 1-dichloroethylene
1, 1-dichloroethane
t- 1,2-dichloroethylene
Chloroform
1,2-dichloroethane
1, 1, 1-trichloroethane
Carbon tetrachloride
Bromodichloromethane
1,2-dichloropropylene
t- 1,3-dichloropropylene
Trichloroethylene
Benzene
1, 1,2-trichloroethane
Bromoform
Tetrachloroethylene + tetrachlorethane
Chlorobenzene
Concentration
(ppb, wt)a

ND
ND
6.1
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
limit
(ppb, wt)

12.6
2.6
2.6
2.6
4.2
5.0
2.6
2.6
4.2
2.6
5.0
2.6
2.6
6.8
4.2
5.0
12.6
Semivolatile organic priority pollutants

All base/neutral semivolatile priority             ND
  pollutants
  4-chloro-3-methylphenol                      ND
  2,4,6-trichloropehenol                        ND
  2,4-dinitrophenol                             ND
  4-nitrophenol                                ND
  2-methyl-4,6-dinitrophenol                    ND
  Pentachlorophenol                           ND
  All other acid semivolatile priority             ND
    pollutants
50
30
30
50
50
50
10

Trace elements
Antimony, Ab
Arsenic, As
Beryllium, Be
Cadmium, Cd
Chromium, Cr
Copper, Cu
Lead, Pb
Mercury, Hg
Nickel, Ni
Selenium, Se
Silver, Ag
Thallium, Tl
Zinc, Zn
(ppm, wt)

ND
ND
ND
ND
ND
4"
1»
ND
1»
ND
ND
ND
ND
(ppm, wt)

2
1
3
1
1
1
1
1
1
1
1
1
10
EP toxicity
concentration
limit (ppm)

—
5
—
7
5
—
5
0.2
—
1
5
—
—
aND denotes not detected; duplicate samples analyzed.
bFound in only one sample; not detected in the others.

-------
R. W. Ross, II. T. H. Backhouse, R. H. Vocque, J. W. Lee, and L R. Water/and
  are with Acurex Corporation, Jefferson, AR 72079.
R. A. Carries is the EPA Project Officer (see below).
The  complete report,  entitled "Pilot-Scale Incineration Test Burn of TCDD-
  Contaminated TrichlorophenolProduction Waste,"(OrderNo. PB87-145835/
  AS; Cost: $13.95, 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:
        Hazardous Waste Engineering Research Laboratory
        U.S. Environmental Protection Agency
        Cincinnati, OH 45268
                                10

-------

-------
  TJ =;•
^ S
_» 0)
  ?• ®

M^"
2 *
  c
  CO
                TJ
                 '
              Da
              1!
              -. o

              O3
              I

              4k
              01
              M
              O>
              00
                S
            TJ   CO
            m   H
            =imm£
            9   S5i


            »   5

-------