EPA/600/A-98/046
                                                    March  1998
  PERFORMANCE CHARACTERISTICS OF A MULTI-STAGE
     PROCESS FOR EX-SITU TREATMENT OF SOLIDS OR
 LIQUIDS CONTAMINATED WITH CHLORINATED ORGANICS
                           By
     Michael LTaylor, George L. Huffman, Thomas R. Clark,
   E. Radha JCrishnan, William E. Gallagher and John A. Prohaska
                 Paper to be presented at the
   Sixth International FZK/TNO Conference on Contaminated Soil
                     {"Con-Soil'98"}
       to be held in Edinburgh, Scotland on May 17-21, 1998
                       March 1998
CM Technologies, Inc./U.S. EPA/IT Corporation/Prohaska Consulting
                     Cincinnati, Ohio
                         U.S.A.

-------
 PERFORMANCE CHARACTERISTICS OF A MULTI-STAGE PROCESS FOR
 EX-SITU TREATMENT  OF SOLIDS OR  LIQUIDS  CONTAMINATED  WITH
 CHLORINATED ORGANICS

 Michael L. Taylor1, George L. Huffmair , Thomas R, Clark3, E. Radha Krishnan3, William
 E. Gallagher, and John A, Prohaska4

 1 CM Technologies, Inc., 10561 Cranwood  Ct., Cincinnati, OH 45240, USA
 2 USEPA, 26 W. Martin Luther King Dr., Cincinnati, OH 45268, USA
 3 IT Corporation, 11499 Chester Road, Cincinnati. OH 45246, USA
 4 Prohaska Consulting, 7411 Ridge Meadow Ct., West Chester, OH 45069, USA

 Key words: chemical treatment, contaminated soil, dioxins, gas emissions, organic chlorinated
 compounds, PCB, PCDD/Fs, thermodesorption

 SUMMARY
 The  US. Environmental Protection  Agency has sought to develop soil decontamination
 technologies which are effective, but which do not involve incineration. This paper presents
 the results of a study in which a steam-assisted, thermodesorption system utilized on Guam was
 characterized in terms of its effectiveness for removing PCBs and PCDD/Fs from soil. Data
 are also presented which are indicative of the performance of the air pollution control system
 which is directly coupled to the thermodesorber. The potential for formation of PCDD/Fs
during thermodesorption is discussed.

 KURZFASSUNG
Die US-Behoerde fuer Umweltschutz  hat nach einer Entwicklung einer wirkungsvollen
Technologic  zur  Bodenaufbereitung  gesucht, ohne  eine  Verbrennung  des Bodens
herbeizufuehren. Der vorliegende Artikel praesendert die Ergebmsse einer Studie, in der ein
dampfunterstuetztes Thermodesorbuonssystem, angewendet auf Guam, auf seine Effektifitaet
untersucht wurdc, PCBs und PCDD/Fs aus dem Erdrcich zu entfernen. Es werden auch Daten
vorgestellt, die fuer die L«istung des an den Thermodesorber, unmirtclbar angeschlossenen
Luftverschmutzungs-Kontrollsystems kennzeichnend sind. Die Moeglichkeit einer Bildung von
PCDD/Fs waehrend der Thermosorbtion wird diskutiert.

RE'SUMET
L'Agence Americaine de Protection de renvironnement (US EPA) cherche a developper des
technologies efficaces de decontamination des sols sans incineration. Cet article presente les
resultats d'une etude dans laquelle un systeme de thermodesorption assiste par vapeur utilise
a Guam est evalue en terme de d'enlevement (removal) des PCBs et  PCDD/Fs contenus dans
le sol Des donnees sont aussi presentees et indiquent la performance du systeme de controle
de la pollution de I'air. Ce systeme est directement couple a celui de thermodesorption. La
possibilite de la formation de PCDD/Fs durant la thermodesorption est discutee.

-------
 INTRODUCTION
 The United States Environmental Protection Agency's (USEPA's) National Risk Management
 Research Laboratory (NRMRL) has been involved in developing chemical dechlorination
 technology for nearly 20 years. Initial efforts focused on developing a process which utilized
 a  reagent consisting of an  alkali metal salt of a polyethylene glycol.  This process was
 successfully applied to treat 1-ton batches of soil which was contaminated with several hundred
 parts-per-million of polychlorinated biphenyls (PCBs)'.  Subsequently, NRMRL personnel
 developed and patented another chemical dechlorination process termed Base Catalyzed
 Decomposition (BCD)"4. For treating solids (e.g. soil, sand) the BCD process involves two
 steps. In the first stage of the process, soil [to which sodium bicarbonate (NaHCO,) has been
 added in  concentrations ranging from  2% to 12% on a  weight:weight basis] is fed  into a
 thermodesorption unit. The soil is heated to 340°C. During heating, the NaHCO, may react
 with  halogenated organics  such as chlorophenols and PCBs to  form partially or totally
 dechlorinated organic reaction products, plus inorganic chlorides such as sodium chloride. The
 NaHCO,  may  also assist  in breaking up agglomerated soil particles, thus  facilitating
 volatilization of contaminants.  The volatilized contaminants and entrained  fine particles
 emanating from the thermodesorption treatment must be captured and subsequently treated in
 the Second Stage of the BCD Process. In the Second Stage of the BCD process, the collected
 chlorinated organics are treated using a mixed reagent which includes a base, a hydrogen donor
 and a catalyst.

 In April 1996, the US Navy began utilizing a technology - the Base Catalyzed Decomposition
 Process (BCDP) -  on Guam for treating soil contaminated with polychlorinated biphenyls
 (PCBs) as well as polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans
 (PCDD/Fs). This technology is similar to the BCD process developed by the USEPA. The
 BCDP technology entails heating soil (to which 1% NaHCO,  has been added) to 300°C in a
 rotary calciner while injecting steam. The volatilized organics and fine particulates produced
 during the 1-hour thermodesorption are swept into an Air Pollution Control System (APCS)
 which removes particulates  and  organics  from the flue gas stream prior to venting to the
 atmosphere. The treated solids from the thermodesorber are returned to the site or deposited
 in a landfill. The particulates and liquids collected in the APCS must be subjected to further
 treatment since these materials typically are contaminated  with halogenated organic residues
 including PCDD/Fs. The Second Stage of the BCD process is not presently used in Guam.

 It is generally  recognized that thermodesorption can remove many types of organics from soil
and similar solids (sand, gravel, sediments) and the effectiveness of the BCDP technology for
removing PCBs from Guam soil has been reported".  PCDD/Fs are often found in association
with PCBs and these compounds,  especially the 2,3,7,8-substituted congeners, have been
 identified as being particularly toxic. PCDDs and more likely PCDFs can be generated during
the manufacture of PCBs; in addition,  PCDD/Fs can be  formed  from PCBs  under certain
conditions7'8.   The BCDP system operating on Guam presented an opportunity to not only
assess the overall performance of the BCDP system but to obtain data regarding the fate and
 formation of PCDD/Fs during implementation of the BCDP system. With the cooperation  of
the US Navy,  a plan was formulated for sampling the various process streams associated with
the BCDP unit followed by subsequent analyses of these samples to obtain a more complete
understanding of the chemistry of soil-bound PCBs and PCDD/Fs during BCDP  treatment.

-------
 The study had two objectives:
 1.   Determine the efficiency of the  thermodesorption process for removing PCBs and
     PCDD/Fs from Guam soil

 2.   Concurrent with Objective  1, determine the location in the APCS where PCBs and
     PCDD/Fs are being removed.

 In the sections which follow, the BCDP Demonstration Unit located in Guam will be described
 and  the results of the sampling and analysis program will be discussed.

 DESCRIPTION OF THE BCDP TECHNOLOGY
 The  BCDP Demonstration Unit3-6 includes the Rotary Kiln Reactor (RKR) which is directly
 coupled with the Air Pollution Control System (APCS). The BCDP unit comprises several
 modules which are coupled in series. These modules are described below.

 Rotary Kiln Reactor System.- The reactor is a rotary calciner consisting of two concentric
 carbon steel shells. Soil containing 1% sodium bicarbonate is fed into the calciner's inner shell
 at the rate of 900 to 1800 kg per hour. The inside shell rotates and is heated externally to
 315°C to 485°C.  Steam, at a flow rate of 48 kg/hr., is injected countercurrently to the soil
 flow. Volatilized organics and entrained particulates are swept into the APCS by the flow of
 steam and the slight negative pressure created by the in-line blower positioned just prior to the
 carbon column in the APCS.

 Multiclone (MC).- This unit removes particles by centrifugal force and is electrically heated
 to minimize condensation.

 Wet  Electrostatic Precipitator  (WESP).-  Additional steam, at a flow rate of 265  kg/hr, is
 injected to prevent formation of an explosive mixture.  A charged grid imparts electrostatic
 energy to suspended particles to promote  particle agglomeration and precipitation.

 Primary Condenser (PC).- This is a shell-and-tube condenser which cools the gases from
 100°C to 30°C. The  flow of gases entering the condenser is 3.68 cubic meters per minute and,
due to condensation,  the gas flow leaving the condenser is 0.85 cubic meters per minute.

 High Efficiency Mist Eliminator (HEME).- This module consists of a tightly woven fiberglass
pad about 7.6 cm thick. The HEME is designed to remove nearly 100% of the particles >3
micrometers and 99% of the <3 micrometer particles.

Carbon Column.- After the HEME, the gas passes through an induced draft fan.  The exhaust
 is routed through a carbon column to remove residual organic vapors, prior to exhausting to
the atmosphere.

 FIELD SAMPLING AND LABORATORY ANALYSES
The sampling and analysis program was designed to provide data to be used in assessing the
performance of the RKR to remove PCBs and PCDD/Fs from the Guam soil, and to determine
the location  in the APCS where these compounds are being removed.  The various process
streams which were  sampled and the analyses which were performed are discussed in  the
paragraphs which follow.

-------
 Feed Soil (RKRF), Treated Soil (RKRP), Breach Soil, and Cyclone Fines.- The concentrations
 of PCBs, PCDD/Fs in the feed soil, treated soil and cyclone fines were quantitated using gas
 chromatographie/mass spectrometric analytical methods [EPA Method 680 for PCBs" and EPA
 Method 8290 for PCDD/Fs'0]. The concentrations of these compounds in the treated soil were
 compared to the  concentrations of these  compounds in  the untreated  soil  to calculate
 percentage removals achieved during each of three 6-hour tests. During each test, samples of
 the treated and untreated soil were taken each hour, then composited to obtain a single sample
 representing the soil before and after treatment during the six-hour test. Similarly, PCBs and
 PCDD/Fs were quantitated in the cyclone fines. The breach soil is a mixture of feed soil which
 drops out of the front end of the kiln and cyclone fines which are collected and ultimately
 recycled back through the process.  The concentration of PCBs in the feed soil and in the
 cyclone fines were quantitated and the ratio of soil  to cyclone fines was calculated and utilized
 to determine the concentrations of PCDD/Fs in the breach soil.

 WESP Blow-down, WESP Makeup Water, and Primary Condensate.- Each of these streams
 was sampled hourly during each 6-hour run and the samples were then composited to obtain
 one sample for each stream.  The composited liquid samples were analyzed by EPA Method
 680 for PCBs and EPA Method 8290 for PCDD/Fs.  It was necessary to analyze the WESP
 makeup water because this is recycled process water and contains the analytes of interest.
These data, in conjunction with the other data, provided an indication of the performance of
the WESP in removing PCBs and PCDD/Fs from the gas stream.

Gas  Sample Before HEME and Gas Sample After HEME.- The  gas streams  entering the
 HEME and exiting the HEME were simultaneously sampled and the samples subsequently
analyzed using the methodology described in USEPA Method 23".   Gas sampling was
performed throughout the duration of each of the three six-hour runs. PCBs and PCDD/Fs
were quantitated using gas chromatographic/mass spectrometric methods. The resulting data
were used to calculate the efficiency of the HEME for removing the target compounds from
the gas stream.

Rationale for Not Assessing the Performance of the Carbon Column.- The efficiency of the
carbon column in the APCS was not assessed due to the fact that stack sampling data were
available and funding for the present project was limited.

RESULTS AND DISCUSSION
PCBs and PCDD/Fs in Soil Samples Before and After Thermodesorption
Table 1 lists the concentrations of PCBs, PCDDs, and PCDFs found in feed soil (untreated
soil) and treated soil for each of the  three tests.  It should be noted that the table  includes the
limits of detection for those cases where no PCBs were found and includes the number of
isomers in those cases where PCBs were detected. In some cases, the values in the table are
flagged due to the fact that high concentrations proved difficult to quantitate accurately.

For PCDDs, comparatively low levels of PCDDs are  present in the soil  before treatment
(values range from 3 ppb to 7 ppb) and concentrations were below detection  limits after
treatment (refer to RKRP values).  Detection limits, however, were not much lower than the
range of concentrations before treatment.

-------
Table 1. Concentrations of PCBs, PCDDs, and
Dale Sample
Collected In
Guam
2/25/97
2/25/97
2/26/97
2/26/97
2/27/97
2/27/97
Dale Sample
Collected In
Guam
2/25/97
2/25/97
2/26/97
2/26/97
2/27/97
2/27/97
Dale Sample
Collected in
Guam
2/25/97
2/25/97
2/26/97
2/26/97
2/27/97
2/27/97

Sample
Number
RKRP-I
RKRF-U
RKKP-2
RKRF-2I
RKKP-3
RKRF-3X
Sample
Number
RKRP-I
RKRF-I
RKRP-2
RKRF2
RKRP-3
RKRF3
Sample
Number
RKRP-U
RKRF-I
RKKP-2»
RKRF-2
RKRP-3*
RKKF3

MONO
RO'
ND[20)
ND[IO)
ND120.0]
ND[1.0]
ND[20)
Dl
10(6)'
ND(20]
2(3)
ND[200]
30(3)
ND[20]
PCDFs in Dnlrealt.l
(RKRF) and Treated (RKRP) Soil Samples from Base Catalyzed Decomposition
PCB Concentration In pph (nanogran
TRI TETRA
40(10) 160(18)
670(4) 2600(8)
9(7) 40(13)
80(1) 3400(8)
20(7) 110(18)
80(1) 3200(7)
PENTA
480(22)
21,500(15)
240(19)
61.200(16)
780(24)
58.700(15)
IIEXA
1300(23)
94,000(17)
1400(23)
283,000(20) y
3600(24) y
273,000(21) y
us per gram)
HEPTA
460(15)
51,700(15)
770(18)
164.000(19)
1900(18)
158.000(18)

Cuncent ratio
OCTA NONA DKCA n (ppb)
400C>) ND(l.O)' ND(20| 24»0
9800(9) 280(1) NI)[I(X)| 180550
110(9) 2(1) NI)[IO| 2573
.30.100(9) 990(2) Nl>(90| 54.).000
270(9) 7(X2) Nl)|IO| 6700
30.000(9) 820(1) NI)[90| "124.000
PCDDs Concentration In ppb (nanograms per gram)
TETRA
ND[1.5]
0049(1)
ND[I 5]
ND [0 02]
ND [36]
ND|002]
PENTA
NO [18]
RO
ND(23]
ND (0 03]
ND[4.3]
NDI.003]
HEXA
ND[I 8]
0 44 (4)
ND[1.9]
030(4)
ND[40|
019(3)
HEPTA
ND|28|
1 42(2)
ND[34]
081
(2)
ND[5.2)
074(2)
OCTA
ND [40]
4 51
ND[47|
248
ND[57]
227
PCDFs Concentration In ppb (nanograms per gram)
TETRA
186(3)
32.4(14)
29 9 (4)
167(13)
104(5)
964(16)
PENTA
ROB
21.1 (11)
21.6(2)
129(11)
854(5)
9 13(9)
HEXA
0.83 (1)B
6.76 (7)
4.33(2)8
556(8)
22.9(3)
959(9)
HEPTA
ND[I 7]
2.03 (4)
ND(2 I]
706(4)
ND[3 3]
507(3)
OCTA
Nl)(35]
1.63
ND[4 11
705
NO [4 9]
50
Total PCDI)
Conccnlralion (|>pb)
NO
642
ND
.3 58
ND
.3 ><>
Total PCI)F
Concentration (ppb)
194
63.9
5.V8
493
212
384
1 RO = peak(s) delected bul ratio out (indicating ratios of either the M'/M'1 peaks or the M':/MH peaks fall outside acceptable limits) D = analyle found in blank.
2 ( ) = No. of isomers observed
1 [ 1 - detection linul
* iirpoilcftl djla for OK- maniple mil Ifi mciy aincclcit

-------
 The concentrations of PCDFs are significantly higher than that of PCDDs in all samples. For
 example, the PCDFs in the reactor feed (RKRF) ranged from 38 to 64 ppb whereas the PCDDs
 in the same samples  ranged from 3 ppb to 7 ppb.  Therefore, the ratio of PCDF to PCDD is
 approximately 10:1  in the RKRF samples. This is as expected since the Guam soil was
 contaminated  with discarded transformer oil, and PCBs are more likely to be converted to
 PCDFs than PCDDs  during heating of transformer oil.

 Concentrations of PCDFs in the treated soil deserve special comment. On February 25, the
 first day of this testing program, the levels of PCDFs in the treated soil were found to be 30%
 of the concentration of PCDFs in the before-treatment soil. However, in the second test, PCDF
 concentrations in the treated soil were approximately equal to the levels in the untreated soil.
 In the third test, PCDF concentrations in the treated soil were 5.5 times the level of PCDF in
 the untreated soil. Also, the ratio of concentrations of the tetra-octa congener groups changed
 during treatment. The concentrations  of hepta-octa chlorinated PCDFs decreased  and  the
 concentrations of tetra-, penta-, and hexa-congeners generally increases after treatment. These
 results suggest that PCDFs are formed during treatment using stage 1 of the BCD process, i.e.,
 the thermal treatment stage.  This result is not surprising considering the reports in  the
 literature7'8 which indicate PCBs can give rise to PCDFs especially when air is present, at or
 near 250°C and up to around 650°C. Air leakage was a  problem during these test  runs due to
 worn knife gate valves.

Additional Data
Although a large amount of additional data were obtained, it cannot be included in this paper
and will be published elsewhere. Some of the data are summarized in the following table. The
International Toxicity Equivalent Values for 2,3,7,8-substituted PCDD/Fs published in 198912
were used to calculate Toxicity Equivalent (TEQ) values for feed soil, treated soil, and air
samples collected before and after the HEME (and upstream of the final carbon column). These
values are shown in Table 2.

Table 2. Toxicity Equivalent Values Calculated on the Basis of PCDD/F Concentrations
Date Sample
Collected in Guam
25 Feb 97
25 Feb 97
25 Feb 97
26 Feb 97
26 Feb 97
26 Feb 97
27 Feb 97
27 Feb 97
27 Feb 97
Sample Number
RKRF-1
RKRP-1
HEME Outlet
RKRF-2
RKRP-2
HEME Outlet
RKRF-3
RKRP-3
HEME Outlet
Sample Type
Feed soil
Treated soil
BCDP Exhaust
Feed soil
Treated soil
BCDP Exhaust
Feed soil
Treated soil
BCDP Exhaust
TEQ
3.15ng/g
1.01 ng/g
12.9 ng/m3
1.75 ng/g
2.02 ng/g
1.36 ng/m3
1.50 ng/g
15.3 ng/g
1.72 ng/m3

-------
 The increase in TEQ values for treated soil is due to the increased concentrations of residual
 PCDFs,

 The overall performance of the thermodesorption unit and the APCS are discussed in the
 section which follows.

 Treatment Efficiency of the BCDP Thermodesorption Unit
 The performance of the thermodesorption unit of the BCDP during each of the six-hour test
 runs was characterized by determining the percentage of removal by BCDP treatment of the
 PCBs, PCDDs, and PCDFs originally present in the Guam soil.

 To determine the treatment efficiency, the following formula was developed:

        Efficiency of Treatment (Ib/hr basis), % = 100 x (Target input in feed
        soil - target returned) -f Target input

 where:
        target = total PCBs, PCDDs, PCDFs, or total 2,3,7,8-substituted PCDD/Fs
        target returned =  amount of target in treated soil + amount of target in breach soil

 The BCDP treatment efficiency for PCBs averaged 96 percent for the three six-hour test runs.
 This means that 96 percent of PCBs initially present in the soil were removed  by BCDP
 treatment.

 In the case of PCDDs, however, the available data could not be utilized to calculate treatment
 efficiencies. This was due to the fact that no PCDDs were detected in the samples of treated
 soil and the limits of detection for the PCDDs were higher than the concentration of PCDDs
 found in samples of the feed soil.

 The treatment efficiency values calculated for PCDFs and for total 2,3,7,8-substituted isomers
 were lower than the values for PCBs and these values decreased from run-to-run to the extent
 that the third run demonstrated that PCDFs  were being formed instead of removed (as
evidenced by negative treatment efficiency values). The concentration of PCDFs in the treated
 soil (product) was over five times higher than the level in the feed for the third test run. Since
 the ftiran isomers contributed the most to the total 2,3,7,8-PCDD/F values, the total PCDF and
 total 2,3,7,8-PCDD/F results follow a similar pattern throughout the data.

 Control Efficiency of Air Pollution Control System (prior to the Carbon Column)
Table 3 summarizes the control efficiency of the APCS based on the test results. The control
efficiency values shown in this table indicate the fraction of the target compounds entering the
 unit (total PCBs, PCDDs, PCDFs, or total 2,3,7,8-substituted PCDD/Fs) which were removed
 by that unit of the APCS expressed as a percentage.  For example, a 4 percent control
efficiency value for total PCBs on 2/25/97 (see top panel of Table 3) indicates that the multi-
cyclone removed 4 percent of the total PCBs emitted by the reactor and entering the multi-
cyclone during the 6-hour test run.  Similarly, on that same day, the WESP removed 24 percent
 of the total PCBs which were not captured by the multi-cyclone and therefore entered into the

-------
Table 3. Summary of Control Efficiency by Pollutant
Total PCBs
Date
2/25/97
2/26/97
2/27/97
Average
Total PCDD
Date
2/25/97
2/26/97
2/27/97
Average
Total PCDF
Date
2/25/97
2/26/97
2/27/97
Average
Total 2,3,7,8
PCDD/F
Date
2/25/97
2126191
2/27/97
Average
Unit Air Pollution Control Efficiency %
Multi-
Cyclone
4%
6%
3%
4%
WESP
24%
74%
84%
61%
Primary
Condenser
99.9%
99.9%
99.9%
99.9%
HEME
72%
85%
78%
78%
Unit Air Pollution Control Efficiency %
Multi-
Cyclone
25%
34%
17%
25%
WESP
65%
99%
99%
88%
Primary
Condenser
98%
60%
65%
74%
HEME
73%
97%
93%
88%
Unit Air Pollution Control Efficiency %
Multi-
Cyclone
10%
59%
24%
31%
WESP
33%
95%
99%
76%
Primary
Condenser
98%
81%
79%
86%
HEME
74%
98%
97%
90%
Unit Air Pollution Control Efficiency %
Multi-
Cyclone
17%
62%
28%
35%
WESP
39%
97%
99%
78%
Primary
Condenser
98%
81%
79%
86%
HEME
76%
98%
97%
91%
Overall
Removal
99.98%
99.995%
99.996%
99.99%
Overall
Removal
99.7%
99,97%
99.95%
99.9%
Overall
Removal
99.7%
99.98%
99.98%
99.9%
Overall
Removal
99.7%
99.99%
99.99%
99.9%

-------
WESP. The total target input to the WESP also included the quantities of targets present in
the makeup water. The removal efficiencies shown for each unit of the APCS indicate that
except for the multi-cyclone, each unit removes a significant percentage of the target species,
with the APCS exhibiting excellent overall control efficiencies.  In calculating overall control
efficiencies, uncontrolled target emissions from the reactor were compared with HEME outlet
emissions.

As stated earlier in this report, the carbon column was not included in this study. Terres et al
reported3'6 that the dioxin/furan stack emissions (expressed as Toxicity Equivalents) were 0.18
ng per cubic meter.

SUMMARY/CONCLUSIONS
The BCD? technology was found to effectively remove PCBs and achieve the clean level of
2 ppm per PCB congener group using the system operating on Guam. The data for PCDD/Fs
indicate that PCDFs apparently form under certain conditions.  Although this study did  not
provide definitive  data  indicating the cause of formation  of PCDD/Fs in the  third test,
published reports indicate that oxygen levels may play a role.

The  APCS  was  found to perform very well and PCDD/Fs  were effectively lowered  to
concentrations which would readily be removed  by carbon column technology.  The stack
emissions were found to fall well within proposed U.S. EPA guidelines.

Clearly there is a  need to treat the solids and liquids removed from the exhaust gas stream by
the APCS.  This latter  treatment can probably  be accomplished in one of several ways
including the patented Second Stage of the BCD process, incineration, landfilling, or other
physical/chemical processes.

-------
                                   References
 1,  Wentz, J., Taylor, M.L., Gallagher, W. Chan, D,B. and Rogers, C.J., "Results of a Field
    Scale Test of an APEG Reactor for Chemically Degrading PCBs in Contaminated Soil,"
    presented at the  82nd Annual Air and Waste Management Association Annual Meeting
    and Exhibition, Anaheim, CA, June 1989.

2.  Rogers, C.J., Kernel, A. and Sparks, H, "Method for the Base-Catalyzed Decomposition
    of Halogenated and Non-Halogenated Organic Compounds in a Contaminated Medium"
    U.S. Patent  No. 5,064,526, Nov. 12, 1991.

3.  Rogers, C.J., Kernel, A. and Sparks, H, "Method for the Decomposition of Halogenated
    Compounds in a Contaminated Medium" U.S. Patent No. 5039350, August 13, 1991.

4.  Rogers, C.J., Kornel, A. and Sparks, H, "Method for the Destruction  of Halogenated
    Organic Compounds in a Contaminated Medium" U.S. Patent No. 5019135, May 28,
    1991.

5.  Terres, S. et al,  "Base Catalyzed Decomposition Process Proven  on Guam" Pollution
    Engineering, 1997, pp 22-25.
6.  Terres, S. And Gallagher, W.E. "The Base Catalyzed Decomposition Process, A Proven
    Remediation Technology," Proceedings of the  1997 International  Conference  on
    Incineration and Thermal Treatment Technologies, Oakland, CA. pp 805-809.

7.  Morita, M., Nakagawa, J. and Rappe, C, 1978, "Polychlorinated Dibenzofuran (PCDF)
    Formation from PCB Mixture by Heat and Oxygen," Bull. Environ. Contain, Toxicol. 13.
    665-70.

8.  Scholz, M., Stieglitz, L., Will, R. And Zwick, G., 1997, "The Formation of PCB on Fly
    Ash and Conversion to PCDD/PCDF," Proceedings of the 17th International Symposium
    on Chlorinated Dioxins and Related Compounds," Volume 31, pp 538-541.

9.  Modified Method 680 for Special  Analytical Services Test,  Identification  and
    Measurement of Pesticides and PCBs by Gas Chromatography/Mass Spectrometry," Ann
    Alford-Stevens and James  Eickelberger, Physical  and  Chemical Methods Branch,
    Environmental Monitoring and Support Laboratory, Office of Research and Development,
    US EPA, Cincinnati, OH 45268, November 1985.

10.  EPA Method 8290, Test Methods for Evaluating Solid Wastes, SW-846, Revision 2,
     December 1996.

11.  EPA  Method 23,  "Determination  of  Polychlorinated Dibenzo-p-Dioxins  and
     Polychlorinated Dibenzofurans from Stationary Sources," 40 CFR 60, Appendix A,
     February 1991.

12.  USEPA. Interim Procedures for Estimating Risks Associated with Exposures to Mixtures
     of Chlorinated Dibenzo-p-dioxins and dibenzofurans (CDDs and CDFs) and the 1989
     Update. Risk Assessment Forum, Washington, DC; EPA/625/3-89/016, 1989.

                                       10

-------
                                    TECHNICAL REPORT DATA
                             (rlctst rtwl Imtnictiotii on tkt rtvtne Ixfore compJc
 1. REPORT NO.
  EPA/600/A-98/046
 4. TITLE AND SUBTITLE
  Performance Characteristics  of a Multi-Stage Process
  for  Ex-Situ Treatment of  Solids and Liquids
  Contaminated with Chlorinated Organics
              3.
              S. REPORT DATE
               March 1998
             «. PERFORMING ORGANIZATION CODE
 7. AUTHORIS)
 M.L.Taylor,  G.L. Huffman,  et  al
              I. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND AOORESS
 CM Technologies,  Inc./IT Corporation
 Cincinnati,  OH 45246
              10. PROGRAM ELEMENT NO.
              1. CONTRACT/GRANT NO.
                                                              68-C5-0039(WA# 2-12)
 13. SPONSORING AGENCY NAME AND AOORESS
 U.S. EPA
 National Risk Management Research Laboratory
 Cincinnati, OH  45268
              13. TYPE OF REPORT AND PERIOD COVERED
                                    1Q98)—
              14.

              EPA/600/14
IS. SUPPLEMENTARY NOTES
 For comments, questions or additional  information,  contact EPA's  George L, Huffman
 on 513/569-7431.
IS. ABSTRACT                                           "~———___——_____________

The  US Environmental  Protection Agency has  sought to develop soil  decontamination
technologies which are  effective, but which do  not involve incineration.   This paper
presents the results  of a    study in which  a  steam-assisted, thermodesorption system
utilized on Guam was  characterized in terms of  its effectiveness for  removing PCBs
and  PCDD/Fs fromsoil.   Data  are also presented  which are  indicative  of.the performance
of the air  pollution  control system which is  directly coupled to the  thermodesorber.
The  potential for formation  of PCDD/Fs during thermodesorption is  discussed
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                                               b.lOiNTIFIERS/OPEN ENDED TEAMS
                           C COSATI FteM/Gioup
 hemical treatment,  contaminated soil,
 ioxins, gas emissions,  organic chlorinated
 ompounds PCS, PCDD/Fs,  thermodesorption
 8. DISTRIBUTION STATEMENT

 Release to the Public
19. SECURITY CLASS (This Report!
  UNCLASSIFIED
21. NO. OF PAGES
   10
                                               20. SECURITY CLASS

                                               UNCLASSIFIED
                                                                          22. PRICE
IPA ?*• 2230-1 (R«*. 4-77)   Previous COITION is OBIOLITI
                                              11

-------