EPA/540/2-89/017
SUPERFUNDTREATABILITY
CLEARINGHOUSE
Document Reference:
U.S. EPA. "Preliminary Report on Treatment/Detoxification Alternatives for PCBs and
Chlorinated Organics." U.S. EPA ORD, HWERL. Cincinnati, Ohio. 31 pp.
September 1985.
EPA LIBRARY NUMBER:
Superfund Treatabillty Clearinghouse - EUZD
-------
SUPERFUND TREATABILITT CLEARINGHOUSE ABSTRACT
Treatment Process: Physical/Chemical - Dechlorination
Media: Soil/Generic
Document Reference: U.S. EPA. "Preliminary Report on
Treatment/Detoxification Alternatives for PCBs and
Chlorinated Organics." U.S. EPA ORD, HWERL.
Cincinnati, Ohio. 31 pp. September 1985.
Document Type: EPA ORD Report
Contact: Charles Rogers
U.S. EPA, ORD-HWERL
26 W. St. Clair Street
Cincinnati, OH 45268
513-569-7757
Site Name: Manufactured Waste (Non-NPL)
Location of Test: Buffalo, NY
BACKGROUND; The EPA Hazardous Vaste Engineering Research Laboratory
(HWERL) report summarizes the development of systems to dechlorinate
polychlorinated biphenyls (PCBs), chlorinated dibenzo-p-dioxins (PCDDs) and
chlorinated dibenzofurans (PCDFs) using a series of reagents prepared from
alkali metals and polyethylene glycols (KPEG).
OPERATIONAL INFORMATION; The data for this document are pilot-scale data
for the KPEG-350 slurry process and bench-scale data with various reagents
for the slurry.
The pilot-scale slurry process was tested on a Buffalo, NY PCB
contaminated site on July 15-20, 1985. The slurry reactor was a 55-gallon
metal drum equipped with a lid, electric heating tape and a rocking
mechanism that mixed reagent into soil. The original PCB concentration in
soil ranged from 22-66 ppm. Approximately 150 Ibs. of soil were added to
the reactor along with 50 Ibs. of reagent. The treatment time ranged from
2-2.5 hours at temperatures of 75°-100°C. PCBs were reduced from 22-66 ppm
to less than 1 ppm after 2.5 hours of reaction with more than 902 of the
reagent recovered for reuse.
The bench-scale data included several of the tests conducted on the
effects of radio-frequency (RF) heating on the in-situ process. The
document reports that RF heating of the soil was effective.
PERFORMANCE; The report indicates PCBs and dioxin concentrations can be
reduced to less than 1 ppm and 1 ppb respectively by the slurry process.
The document concludes that the in-situ process under ambient conditions is
not as effective as the slurry process in the destruction of PCB- or
PCDD-contaminated soils. It should also be noted that the document does
not report any analysis on transformation products. This needs to be
addressed, because when chemically altering PCBs, it is necessary to know
what the transformation products are and their potential toxicities.
3/89-13 Document Number: EUZD
NOTE: Quality assurance of data may not be appropriate for all uses.
-------
Costs of the process are estimated at $100 to $300/ton with the in-situ
cost being higher due to reagent loss. The document reports on some
methodology, procedures, and QA/QC protocols and indicates gas chromato-
graph/mass spectroscopy as the primary method of analysis. Laboratory
QA/QC is not discussed in detail.
CONTAMINANTS;
Analytical data is provided in the treatability study report. The
breakdown of the contaminants by treatability group is:
Treatability Group CAS Number Contaminant
W02-Dioxins/Furans/PCBs 1336-36-3 Total PCBs
3/89-13 Document Number: EUZD
NOTE: Quality assurance of data nay not be appropriate for all uses.
-------
PRELIMINARY REPORT ON TREATMENT/DETOXIFICATION
ALTERNATIVES FOR PCBs AND CHLORINATED ORGANICS
by
Charles J. Rogers, Donald Ly Wilson, Alfred Kernel
Chemical & Biological Detoxification Branch
Alternative Technologies Division
Hazardous Waste Engineering Research Laboratory
Cincinnati, Ohio 45268
HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER .
.This document is intended for internal Agency use only. Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use.
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial
products and practices frequently carry with them the increased generation of
solid and hazardous wastes. These materials, if improperly dealt with, can
threaten both public health and the environment. Abandoned waste sites and
accidental releases of toxic and hazardous substances to the environment also
have important environmental and public health implications. The Hazardous
Waste Engineering Research Laboratory assists in providing an authoritative
and defensible engineering basis for assessing and solving these problems.
i
Its products support the policies, programs and regulations of the Environ-
mental Protection Agency, the permitting and other responsibilities of State
and local governments and the needs of both large and small businesses in
handling, their wastes responsibly and economically.
This report describes the application of alkali metal polyethylene
glycolate complexes to the destruction of PCBs in the environment. The ob-
jective is to develop information useful to the U. S. Environmental Protection
Agency for the in-situ destruction of chlorinated organics in contaminated soils
For further information, please contact the Alternative Technologies
Division of the Hazardous Waste Engineering Research Laboratory.
David G. Stephan, Director
Hazardous Waste Engineering Research Laboratory
-------
PRELIMINARY REPORT ON TREATMENT/DETOXIFICATION ALTERNATIVES
FOR PCBs AND CHLORINATED ORGANICS
Executive Summary
Polychlorinated biphenyls (PCBs), chlorinated dibenzo-p-dioxins (PCDDs),
and chlorinated dibenzofurans (PCDFs) are three series of related compounds
that gained notoriety due to their high toxicity and persistence in the environ-
ment. Increasingly, in the last few years, PCBs, PCDDs and PCDFs have been
identified as major contaminants in soil, sediments, and sludge as well as
effluents from incineration processes throughout the country. The accumulation
of PCB, PCDDs and other toxic halogenated compounds in the environment and
living systems is a serious problem that has been well documented.
Although a .great amount of work has been done by many research groups in
the area of direct chemical decomposition, relatively little effort has-been
directed toward in-situ or on-site decomposition of halogenated organics.
In 1979, EPA scientists identified criteria for a chemical treatment
system to decontaminate chlorinated pollutants in the open environment.
Such a system should be capable of:
0 Reaction under ambient conditions
0 Cleaving carbon-halogen bonds
0 Reacting rapidly
0 Complete
0 Exothermic
0 Self-sustaining
0 Yield non-toxic by-products
iv
-------
Scientists at the Franklin Research Institute (FRI) began an EPA
sponsored study in 1979 to devise such a system that would lend to the
dehalpgenation of PCBs. A series of reagents prepared from alkali metals
and polyethylene glycols (KPEG) have been developed, in the interim, by
FRI, U.S. EPA, Galson Research Corporation, and General Electric Company to
dehalogenate PCBs and other halo-organics.
Early Experiments Using Laboratory Contaminated Soils
Laboratory experiments using simulated dry soil substrates spiked with
PCBs confirmed that PCBs are dechlorinated .by the KPEG Reagents under mild,
ambient conditions. Subsequent experiments using actual soils spiked with
PCBs clearly show, however, that water in soil greatly reduces the rate of
dechlorination and the effectiveness of the reagents on PCBs in real soils.
These adverse effects were greater than expected from the preliminary
results obtained in previous studies. In addition, it was found that the
most effective reagent formulation for the dechlorination of PCBs in con-
trolled one-phase reactions and model substrates was not the most effective
reagent for PCBs in soil, specifically because of its extreme sensitivity to
water. The reagents described in the full report are less sensitive to
deactivation with water.
The next phase of experiments show that soil freshly contaminated with
1000 ppm of Aroclor 1260 can be decontaminated to below 50 ppm PCBs within
28 days with a direct application of Potassium Polyethylene Glycol (KPEG)
(average molecular weight, 305-1). Particularly encouraging is the fact that
this reagent can be used on soil containing some'water and organics, and that
continues to react after several days in open air.
-------
On-Site Treatment of PCB Contaminated Soil
The first completed on-site experiment in the application of reagent to
contaminated soil was carried out by FRI and by Galson Research Corporation,
Syracuse, New York in August 1983. This experiment produced inconclusive
results in contrast to those obtained under laboratory controlled conditions.
The only known major problem which has adversely affected the results of
this initial field study involves the inhibiting effect of water (early
morning rain) on the chemical dehalogenation rate.
Results using PCB contaminated soil from a Philadelphia, PA. site showed
that significant dechlorination is achieved by simply air drying the soil at
room temperature prior to the application of the reagent. Without this pre-
treatment, insignificant decontaminatron was observed for the relatively wet
soil, even when KPEG 350-1 was used.
Recent laboratory studies conducted by the Hazardous Waste Engineering
Research Laboratory (HWERL) has established that both polyethylene glycol
(M.W.350-400) and the KPEGs prepared from the glycols are hygroscopic, and
can absorb within 13 weeks, 50% of their weight in moisture. High humidity
contributes to KPEG deactivation.
Heated Treatment Processes
In laboratory studies, it was determined that a heated KPEG reagent is not
adversely affected by moisture compared to its use under ambient temperatures.
Two different processes, namely, a heated slurry process and a radio frequency
(RF) heated in-situ process are being tested to decontaminate soils, sediments,
and sludges without concern for levels of moisture.
-------
In the laboratory-scaled slurry process, 250 g. of PCDD contaminated
soil (2000 ppm) is placed in a one liter reactor along with 250 g. KPEG
reagent. The contents of the reactor are heated to 75° C with mixing for 1-2
hours. At the end of the reaction time, the reactor contents are allowed to
cool, soil is allowed to settle out and the reagent is decanted and saved
for reuse. PCDDs in the slurry process are reduced from 2000 ppb to below
one ppb.
The slurry process was tested on a Buffalo, N. Y. PCS contaminated site
on July 15-20, 1985. The slurry reactor was a 55-gallon metal drum equiped
with a lid, electric heating tape and a rocking mechanism that allowed for
mixing of reagent into soil. The original PCB concentration in soil ranged
I
from 28-66 ppm. Approximately 150 Ibs-. of soil was added to the reactor along
with 50 Ibs. of reagent. The treatment time ranged from 2-2.5 hours at
temperatures of. 75°-100°C. PCBs were reduced from 28-66 ppm to less than 1 ppm
after 2.5 hours of reaction with more than 90% of the reagent recovered for
reuse.
The heated in-situ studies have shown that soil containing 10-30% moisture
can be heated to 140°C by a RF heating process. The heating mechanism is similar
to that of a microwave oven, except that the frequency is different and the scale
of operation is much larger. The energy penetrates the target volume and is
absorbed by the molecules present in the volume of PCB/PCDD contaminated soil.
-------
A treatment system of RF heating (60°-140°c) in conjunction with KPEG reagent
is being developed to reduce the adverse influence of soil moisture, and to
enhance the possibility of in-situ decontamination of landfills or large volumes
of soil contaminated with PCBs or PCDDs. •
Conclusions
The results of these studies clearly indicate that KPEG reagents can- sig-
nificantly reduce the PCBs and other toxic halo-organics in contaminated soils
under low moisture conditions. There are several areas in which improvements
must be achieved in order to develop a viable and economical process for
chemically treating PCB contaminated soils in the open environment. It has
been established that the KPEG reagents are extremely hygroscopic, capable of
absorbing moisture from the surrounding environment, resulting in their deacti-
vation. Heating the-reagents reverses their deactivation by moisture and en-
hances the rate of dehalogenation. Two technologies, a low temperature (70-140°)
heated slurry process and a RF in-situ process have been demonstrated to destroy
PCBs and TCDDs in the presence of moisture. Preliminary costs for the slurry
process are on the order of $200/ton of soil. An estimated cost for RF in-situ
heated process is $300/ton of soil.
Recommendations
Both methods, the heated slurry and the RF in-situ, should be scaled-
up from laboratory data to field verification units.
These processes should be tested on PCB and PCDD contaminated
sites.
VTM
-------
' Studies on environmental fate and toxicity of products
should be completed.
" Studies should be conducted with field scale units to
determine the technical/economical feasibility of each
process.
-------
ABSTRACT
. In recent years, PCBs, PCDDs, and other hazardous chemicals have been
discovered as contaminents at sites throughout the country. Conventional
waste management methods cannot be used, in many cases, for in-situ or on-
site treatment of contaminated soils. '
The EPA has initiated a research program to identify chemical methods
for decomposing halogenated pollutants in the environment. The overall
objective of the program is to identify, validate, and demonstrate effec-
tive and economical chemical processes for removal/destruction of chlorinated
pollutants in soil, sludges and sediments. This report summarizes research
progress on chemical methods development for the detoxification/destruction of
halogenated pollutants in the environments.
-------
CONTENTS
Disclaimer ii
Fo rewo rd i i i
Executive Summary iv
Abstract x
Tab! es xi i
1 Introduction 1
2 Concl usi ons 4
3 Recommendati ons 6
4 Experimental Procedures 7
Preparation and Nomenclature of the APEG Reagents 8
Soil Treatment Methods 9
Sampl e Preparation Methods 10
Analytical Method Quantitation of PCBs 12
5 Laboratory Studies .' 14
Treatment of PCB-Spiked Soils ! 14
Analysis and Treatment of PCB Contaminated Soil
from the Bengart & Memel Site 15
Results of Laboratory Studies 19
6 Development of Heated KPEG Treatment Methods 20
SI urry Process 20
General Slurry Test Procedure 20
Slurry Process Kinetics 21
Recovery of Reagent from Slurry Samples 23
7 Di scussi on 25
xi
-------
TABLES
Number Page
1 Treatment of Aroclor 1260 Spiked Soil with
NaPEG 400 16
2 Three Consecutive Daily Treatments of Aroclor
1260 Spiked Soil with KPEG 350-1 17
3 Results of PCB Analyses of Bengart & Memel Soil 18
4 Results of Laboratory Treatment of Bengart & Memel
Soil with Two KPEG Reagents 19
5 Slurry Process Results Using KPEG 22
6 In-Situ KPEG Data 24
xii
-------
SECTION 1
INTRODUCTION '
The enormous variety and amount of toxic halogenated organic materials
which have pervaded our environment during the past fifty years have left us
with major, crucial problems of disposal. These problems were clearly de-
lineated in a recent U.S. EPA publication, "State-of-the-Art Report: Pesticide
Disposal Research." The accumulation of polychlorinated biphenyls (PCBs)
and polychlorinated dibenzodioxins (PCDDs, "dioxins") in soil, sediment, and
living tissue is a serious problem that has received considerable public atten-
tion in recent years. Although a great amount of work has been done by many groups
the area of direct chemical decomposition of these and other halogenated
organics, relatively little effort has been directed toward in-situ chemical
detoxification.
The "clean up" of a contaminated site, usually involves landfilling, and
is not really a permanent detoxification but rather a transfer of a toxic spill
from one region to another. Landfilled toxic materials are still in the
environment and may persist there for many years. The Hazardous Waste
Engineering Research Laboratory (HWERL) past and current research in the area
of chemical detoxification is addressed to the well-known chemical stability
of chlorinated aromatics since these comprise the bulk of the problematic
halogenated waste types.
During the summer of 1978 a new chemical reagent was synthesized and used
to effectively dechlorinate PCB oils at the Franklin Research Center (FRC).
Since that time a series of reagents have been prepared from potassium hydroxide
-------
and polyethylene glycolates (KPEG) which, with heating, produce rapid dehalogena'
tion of halo-organic compounds of all types—in open air systems.
The apparent reduced sensitivity of the KPEG System to moisture immediately
suggested its potential for use on soils contaminated with halo-organics. In
August of 1979, the U. S. EPA provided grant support to FRC to investigate the
chemistry underlying the dehalogenation process, concentrating on dechlorination
of PCBs.
Additional EPA grant assistance was awarded in 1982 for the detailed
investigation of the effects of variable reaction parameters on the rate and
extent of chemical decontamination of substrate. This research focused
almost exclusively on the direct chemical treatment of PCB-contaminated soil.
The continued laboratory investigation-was aimed at identifying treatment
conditions necessary for the most efficient decontamination in a direct field
application. During the study an optimum reagent composition was selected
based on chemical reactivity considerations. A system which exhibits maximum
reactivity toward PCBs and other halogenated organics coupled with minimum
sensitivity to reagent-deactivating side reactions was sought.
Recent laboratory studies conducted in HWERL has established that KPEG
reagents are extremely hygroscopic and are capable of absorbing up to 50% of
their weight in moisture from out of the surrounding environment within thirteen
weeks. The discovery of the hygroscopic property of KPEG and that the deactiva-
ti on influence of moisture can be eliminated by heat is expected to lead to the
-------
development of new effective methods for in-situ or on-site decontamination
of PCB contaminated soils. This report summarizes the research performed and
the results obtained.
-------
SECTION 2
CONCLUSIONS
The feasibility studies have established that the KPEG reagents can
significantly reduce the concentration of PCBs and PCDDs in contaminated soils
under controlled laboratory conditions. There are, however, several areas in
which improvement and optimization must be achieved in order to develop a
viable and economical process for chemically treating PCB and PCDD contaminated
soils in the field. Preliminary results indicate that the slurry process (as
described in Section 6) can reduce the levels of PCDDs in soil to less than 1
ppb in a two hours and that up to 99% PCDDs in spiked soil may be destroyed by
heating soil in the presence of KPEG at a temperature of 70° to 160°C.
Developmental research on the slurry process is being conducted under
Contract Number 68-13-3219 with Gal son Research Corporation and sponsored by
the U. S. Environmental Protection Agency and witlrassistance from The Air
Force Engineering and Services. Field verification of the process is on-going
and a final report is expected in March 1986.
Radio frequency heating to enhance in-situ KPEG decontamination of soil is \
presently being investigated with IIT Research Institute. A report on this
development will be available in June 1986.
Results of early experiments using actual soils spiked with PCBs clearly
showed that water, particularly water in soil, greatly reduces the rate of
reaction.
-------
However, heating the reagents prevents their deactivation by moisture
and increases the dehalogenation rate. The two technologies under development,
a low temperature slurry process and an RF in-situ process have been demon-
strated to destroy PCBs and PCDDs in the presence of moisture.
From these studies, it is estimated that the cost of decontamination
utilizing the slurry process will be on the order of $100-$200/ton of soil.
An estimated cost for the RF in-situ heated process is approximately $300/ton
of soil.
-------
SECTION 3
RECOMMENDATIONS .
Additional areas in which investigations should continue are outlined
below:
0 Both methods, the heated slurry and the RF in-situ, should be
scaled-up from laboratory data to field verification units.
8 These processes should be tested on PCBs and PCDDs contaminated
sites.
0 Studies on environmental fate and toxicity of products should
be completed.
0 Studies should be conducted with field scale units to determine
the technical/economic feasibility of each process.
-------
SECTION 4
EXPERIMENTAL PROCEDURES
KPEG reagents for certain laboratory experiments were prepared in FRC's
Philadelphia (PA) laboratory. Reagents of various compositions are described
in this section.
Soils treated in the laboratory were prepared and/or pretreated as de-
scribed in the following. After a predetermined amount of time, the soil was
analyzed for unreacted PCBs. Soil extraction methods employed in this study
are described.
Gas chromatography employing packed columns with electron-capture detection
(GC-ECD) was used for the quantitative determination of PCBs in soils. Aroclors
(PCB mixtures) were used as standards, Treated and untreated soils were analyzed
for their PCB mixtures. The concentrations of PCBs were expressed a parts-per-
million (ppm) (w/w) based on the weight of extracted, dried soil.
The gas chromatograph was a Hewlett Packard Model 3380A equipped with a
Model 5700A integrator, a Ni63 electron capture detector and a 6' by 1/4" OD
silanized glass column of 3% OVI on 80-100 mesh Supelcoport. The operating
conditions were as follows: carrier gas 90% Argon - 10% Methane at 33 ml/min;
temperatures, column 195°C isothermal; injector, 250°C; detector, 300°C.
Carbowax polyethylene glycol 400 (PEG 400) was obtained from the Quaker
City Chemical Company and poly(ethylene glycol monomethyl ether) 350 (PEG 350)
-------
was obtained from the Aldrich Chemical Company. Other chemicals used to
manufacture the reagent or in sample preparation were obtained from various
suppliers.
All analyses were run in duplicate. In most cases the individual experi-
ments were performed in duplicate.
PREPARATION AND NOMENCLATURE OF THE KPEG REAGENTS
All formulations of reagent are referred to as "KPEG reagents". The pre-
paration of the specific reagents is as follows:
KPEG 350: Poly(ethylene glycol monomethyl ether), average molecular
weight 350 (PEGM 350) was reacted with an equimolar amount of potassium metal
under nitrogen. The temperature of the reaction was maintained slightly above
the melting point of potassium. The greenish amber, relatively nonviscous
liquid reagent was cooled to room temperature and stored under nitrogen until
needed.
The reagent is composed primarily of the potassium salt of PEGM 350 in
which the potassium ion is complexed by the polyglycol chain.
KPEG 350-1: KOH pellets were dissolved in an equimolar amount of PEGM 350
under nitrogen at approximately 100°C. The amber, viscous reagent was stored
under nitrogen until needed. The reagent contains primarily a KOH/PEGM
complex in equilibrium with the potassium salt of PEGM, which is the reactive
species. The reagent contains some precipitate (mainly KOH) at room temperature,
-------
NaPEG 400: PEG 400 was reacted with molten sodium metal at approximately
100°C under nitrogen. As with all reactions involving alkali metal reactants,
the metal was added slowly, in small portions. When 1.1 moles of sodium was
consumed per mole of PEG 400, the reaction slowed considerably and was termi-
nated. The viscous amber liquid was cooled and stored in the same manner as
the other reagents. This reagent consists primarily of mono- and disodium
salts of PEG 400 in which the sodium ion is complexed by the PEG chain.
SOIL TREATMENT METHODS
Laboratory Treatment Method for PCB-Spiked Soils
Approximately 10 kilograms of a sandy soil was obtained from a non-
contaminated site in New Jersey. A 500 gram portion of this soil was placed
in a one-liter container. PCB 1260 (500 milligrams) dissolved in 25 ml of
hexane) was added to the container. The ingredients were mechanically mixed
for 24 hours. Portions (50 gram each) of the spiked soil (1000 ppm PCB) were
extracted and analyzed to assure homogeneity and to quantify PCB recovery.
The reagent was warmed to approximately 100°C before being added to 500 g
samples of contaminated soils. The treated soil samples were mixed thoroughly
for one minute. In all cases, the reactions were conducted entirely open to
the atmosphere. Each APEG-treated soil sample was kept in an open jar so that
the depth of soil was approximately 3 inches.
Laboratory Treatment Procedure for Buffalo, NY Soil - Preliminary Indoor Tests
Soil samples from test area I (0-6 inches below the surface), and test
area II (0-6 inches and 6-12 inches) were combined. Approximately equal weights
-------
of each of the four samples were mixed thoroughly and divided into two equal
portions. To one portion was added 10 weight percent of KPEG-350. The reagent
was added and mixed thoroughly in the soil in an open air indoor environment.
The treated soil was let stand at room temperature for 37 days, after which time
two aliquots of each sample were analyzed for residual PCBs. The other portion
of "combined soil" was similarly treated with KPEG 350-1, a reagent more like
that used in the field tests in that it was prepared with KOH.
Standardized Laboratory Treatment Procedure for PCB Contaminated Soil
Soil was air dried for 3 days in a hood prior to use, or wet-soil was used
directly. One hundred grams of soil was weighed into a 4 oz. screw-cap jar,
the soil was equilibrated in an oven at the appropriate reaction temperature.
The reagent was added by inserting a syringe filled with reagent into the soil;
the soil and reagent were then mixed thoroughly for 1-5 minutes. The jars were
returned to the oven, then analyzed at the appropriate time. Reaction tempera-
tures, reagent-to-soil ratios and methods of mixing were varied in the experimental
program.
SAMPLE PREPARATION METHODS
Modified Soxhlet Extraction of PCBs from Soil
Ten gram aliquots of soil were placed in a round bottom flask. In
chemically treated soil samples the reaction was quenched prior to extraction
by adding 2 g of water to the soil aliquot. This prevents the heat from the
10
-------
condensing solvent from accelerating dechlorination reactions during the ex-
traction step. The aliquots were then extracted'with 1:1 (v/v) acetone/hexane
in a Soxhlet extractor. The workup was as follows:
a. 150 ml of solvent was used for Soxhlet extraction.
b. Extract was added to 250 ml of deionized water in a 500 ml
separatory funnel to which an additional 75 ml of acetone
was added (to prevent emu!sification).
c. The organic layer was separated. Four 30 ml portions of
hexane were then used to extract the aqueous layer.
d. The organic portions were combined and placed in a 200 ml
volumetric flask. Hexane was added to fill to volume.
e. Appropriate dilutions were made to provide a hexane solu-
tion with PCB concentration in the range of 1 mg/1.
f. 10 ml of the solution was stirred vigorously with 5 ml of
concentrated H2S04 for 3 to 5 minutes.
g. 5 ul of the H2S04-treated solution was injected into the
gas chromatograph. PCB concentration was calculated by
the method described below.
h. The extracted soil was dried and weighed.
Shaker Extraction of PCBs from Soil
Ten gram aliquots of soil were removed from the reaction vessel and placed
in a 4 oz. jar with Teflon«-lined screw cap lid. Fifty ml solvent (1:1 mixture
11
-------
acetone:hexane) was added, then the jar was agitated for three (3) hours on a
wrist action shaker. The soil was allowed to settle and the liquid was decanted
into a 250 ml Erlenmeyer flask. The soil was rinsed three times with 25 ml
solvent and all extracts were combined. The soil was weighed after it had
dried.
The combined extracts were added to 250 ml deionized F^O in a 500 ml
separatory funnel to which was added an additional 75 ml acetone. The organic
layer was separated and four 30 ml portions of hexane were used to extract the
aqueous layer. The organic portions were combined and placed in a 200 ml
volumetric flask; hexane was added to fill to volume. After appropriate dilu-
tions, 10 ml of the solution was stirred vigorously with 5 ml concentrated
H2S04. The hexane solution was analyzed for PCBs by the GC-ECD as previously
described.
ANALYTICAL METHOD QUANTITATION OF PCBs
Unreacted PCBs extracted from soil samples were quantitated by gas
chromatography with electron capture detection (GC-ECD). Aroclors 1242, 1254
and 1260, 1 mg/liter in hexane, were used as standards. Five ul of standards,
and samples diluted with hexane to approximately 1 mg/liter PCBs, were injected
for each analysis. The method of Webb and McCall* was used to quantitate PCBs
by individual peak areas using predetermined response factors for each peak.
*Webb, R. G. and McCall, A. C., J. Chromatog. Sci. II, 366 (1973)
12
-------
When sample chromatograms closely resembled thd.se of one of the above Aroclors,
an approximation was made by comparing the total (summed) peak area of the sample
to that of the standard.
The concentration of PCBs was calculated as follows:
(MS)(V*)(R*)
c (ppm) = — — ;
(v*)(W*)(RS)
where
Ms = mass of standard injected, in nanograms
Vx = total volume of sample times dilution factors, in milliliters
Rx, Rs = chromatpgram response of sample and standard peaks, respec-
tively, in peak area units
vx = volume of sample solution injected, in microliters
Wx = weight of the sample matrix (extracted, dried soil), in grams
13
-------
SECTION 5
LABORATORY STUDIES
Two series of laboratory tests were conducted. Some tests evaluate the
effect of KPEG treatment on PCB spiked soils. Other tests examined the effects
of adding these reagents to soil obtained from the Bengart & Memel site (Buffalo,
N. Y.).
TREATMENT OF PCB-SPIKED SOILS
This phase of testing was directed primarily toward evaluating APEG
reagents under controlled laboratory conditions. In these laboratory"experi-
ments, the reagent was preheated and applied to the contaminated soil in an
indoor environment at room temperature. Multiple applications were employed
and the reagent was heated prior to application to the soil.
A sandy soil was obtained from a non-contaminated site in New Jersey.
The soil was spiked to a level of approximately 1000 ppm with Aroclor 1260.
The spiked soil was extracted to assure quantitative recovery. The acetone/
hexane Soxhlet extraction procedure described in Section 4 was used. The
reagent was added to the soil as described in Section 4 for PCB-spiked soils.
After predetermined reaction time, the soil was analyzed for PCBs, and/or
additional reagent was added.
Ten gram aliquots of the soil were removed periodically and analyzed for
PCBs using the procedure described above for untreated soils. Additional
portions of reagent were applied to the soil in most of these experiments at
various time intervals up to twenty-eight (28). The results are presented in
14
-------
Tables 1 and 2. Analyses of two aliquots of the same reaction were performed
in most cases. The most successful of these experiments were those of Table 2
when initial concentrations of PCB of approximately 1,000 ppm were reduced to a
below 50 ppm within twenty-eight (28) days.
ANALYSIS AND TREATMENT OF PCB-CONTAMINATED SOIL FROM
THE BENGART & MEMEL SITE
Analysis of Soil Samples from Various Locations at the Bengart & Memel Site
Ten soil samples (a heavy, wet clay) from the Bengart & Memel Site,
Buffalo, New York were received at FRC in July 1983. Soil samples weighing
approximately 50 to 100 grams each were obtained from various depths at the
two test plots and the "control" area. Two 10-gram aliquots of each soil
sample were extracted by the acetone/hexane Soxhlet extraction procedure'de-
scribed in Section 4 and analyzed for PCBs by 6C-ECD. In each case the
chromatogram of the extracted PCBs resembled that of Aroclor 1260, but quantita-
tion was done by comparing individual peaks to 3 Aroclor standards. The total
concentrations of PCBs are expressed as ppm based on the weight of the extracted,
dried soil as shown in Table 3.
15
-------
TABLE 1. Treatment of Aroclor 1260-Spiked Soil With NaPEG 400
Day
0
0
0
0
0
1
1
5
5
5
6
6
11
13
13
Description
Unspiked Soil (Blank)
Unspiked Soil (Blank)
Spiked Soilb
Spiked Soil .
50 g Reagent Appl ied
Soil Analyzed
Soil Analyzed
Soil Analyzed
Soil Analyzed
51 g Reagent Applied
Soil Analyzed
Soil Analyzed
100 g Reagent Applied
Soil Analyzed
Soil Analyzed
[PCB]a, ppm
0.64
0.58
1349
1238
515
492
410
535
399
408
399
426
Analyzed
as Aroclor9
1260
1260
1260
1260
—
1260
.1260
1260
1260
—
1254
1254
--
1254
1254
a PCBs were quantified by comparing total chromatogram peak area to
that of the listed Aroclor standard as per the method of Webb and
McCall listed under Analytical Method Ouantitation of PCBs in
Section 4.
b Soil (1000 g) spiked with approximately 1000 ppm Aroclor 1260.
16
-------
TABLE 2. Three Consecutive Daily Treatments of Aroclor 1260-
Spiked Soil with KPEG 350-1
Day
0
0
1
1
2
3
3
7
9
9
14
14
28
28
Description
Spiked Soilb
51 g Reagent Applied
Soil Analyzed
Soil Analyzed
56 g Reagent Applied
Soil Analyzed
Soil Analyzed
100 g Reagent Applied'
Soil Analyzed
Soil Analyzed
Soil Analyzed
Soil Analyzed
Soil Analyzed
Soil Analyzed
[PCB]a, ppm
951
-
722
713
-
326
320
-
124
110
110
67
40
48
Analyzed
as Aroclor3
1260
—
1260
1260
—
-1254
1254
—
1254
1254
1254
1254
1254
1254
a See footnote in Table 1.
b See footnote in Table 1.
17
-------
TABLE 3. Results of PCB Analysis of Bengart & Memel Soil
Soil
0"
6"
12"
18"
Depth
- 6"
- 12"
- 18"
- 24"
Total [PCBs], ppm
127.0
138.0
109.0
127.0
30.0
21.2
39.3
25.6
*C 125a
12.6
14.4
8.7
12.1
C 280a
19.0
22.1
12.3
17.0
-
a Concentration (ppm) of PCBs in each chromatogram peak, or set of peaks
as calculated by the Webb & McCall method. The concentration of peaks
117 plus 125, which should be used for comparison, is roughly 25% higher
than that of peak 125 alone.
Laboratory Treatment of PCB-Contaminated Soil From The Bengart & Memel Site
A series of laboratory experiments were conducted using PCB-contaminated
soil from the Bengart & Memel site. The soil samples were obtained, prepared
and treated as described in Section 5. Duplicate 10-gram aliquots were analyzed
after 37 days for residual PCBs by the Soxhlet extraction method described in
Section 4. Quantisation was done by comparing individual peaks of the sample
chromatogram to Aroclor standards (as previously mentioned). The results are
shown in Table 4.
18
-------
TABLE 4. Results of Laboratory Treatment of Bengart X
with Two KPEG Reagents
Description Total [PCS]
Control (Sample from Approx 101
0-6" depth)
Soil 4 KPEG-350, 33.6
37 days 52.3
Soil & KPEG 350-1, 114
37 days 94
a Concentration (ppm) of PCBs in each chromatogram peak, or set
calculated by the Webb & McCall method (Section 4).
RESULTS OF LABORATORY STUDIES
The results of the data presented in Tables 1 and 2 clear!
significant drop in Aroclor 1260 levels when the KPEG 350-1 rea
(90-95% reduction) as compared to the NaPEG 400 reagent (60-707
The NaPEG 400 reagent was the original reagent suggested for us
however, after these results the KPEG 350 series of reagents we
for continuation of these studies. This led to a comparison o
and KPEG 350-1 formulations presented in Table 4. From these
determined that the KPEG 350 reagent was more effective than K
ambient conditions. Preparation and nomenclature of these rea
sented in Section 4.
These ambient temperature studies led to elevated temper;
i.e., the slurry process as described in Section 6. The resl-
elevated temperature studies were very promising with respect
levels in soils to less than PPB levels in short (2-4 hour) t
Future studies will focus on both heated in-situ and heated s
19
-------
SECTION 6
DEVELOPMENT OF HEATED KPEG TREATMENT METHODS
The results from laboratory studies (Tables 1-4) has clearly demonstrated
the inability of KPEG under simulated in-situ conditions to dehalogenate PCBs
to acceptable levels in soil. However, the requirements necessary for KPEG to
dehalogenate PCBs, TCDDs and other halo-organic contaminants in soil, recently
have been well defined in laboratory studies. To affect dehalogenation in
soils it requires intimate contact of KPEG with pollutants, heat to enhance
rates and control of moisture levels since KPEG is extremely hygroscopic and is
deactivated in high moisture environments. Two processes, namely a slurry and
heated in-situ, are under development to address treatment requirements and
different types of treatment sites.
SLURRY PROCESS
The slurry process development test series focused on two areas: reaction
kinetics as a function of temperature and reagent recovery. These areas will
be discussed separately.
GENERAL SLURRY TEST PROCEDURE
For laboratory slurry reactions, 250 g of soil were placed in a 1 liter
round bottom flask. The samples and spikes were spiked with 1,2,3,4-TCDD to a
nominal level of 2000 ng/g. Each portion of soil was spiked individually. The
samples were slurried with 250 g of reagent including KOH. Spikes and blanks
were slurried with the same amount of glycol and dimethyl sulfoxide (DMSO, a
20
-------
catalyst/co-solvent). The soils were heated for two hours with agitation.
At the end of the reaction, the soils were allowed to cool with agitation.
The cooled soil was allowed to settle and the reagent was decanted and saved
for analysis. The soil was washed with three separate volumes (250 ml) of
deionized water. These water washes were also saved for analysis. The washed
soil was quenched with 50 ml of 50% sulfuric acid and kept in the reaction
flask or transferred to a clean sample jar pending analysis.
SLURRY PROCESS KINETICS
The results of the slurry process testing are listed in Table 5.
21
-------
Table 5. Slurry Process Results Using KPEG
Sample 1
85012201
85012401
8501 2903
85012902
85012901
84111920
841 1 1 921
841 1 1 923
84111924
84111925
84111926
84111927
85012809
85012810
85031 501
85031502
85031503
85031801
85031 802
85031 901
85041901
85041 902
Type5
Sample
Sample
Sample
Control
Blank
Sample
Sample
Spike
Conrtol
Blank
Blank
Blank
Sample
Control
Sample
Samp! e
Control
Samp! e
Control
Blank
Sample
Sample
Temp
(max°C)
260
260
260
260
260
180
180
180
180
180
180
180
150.
150
70
70
70
70
70
70
70
25
Time
(hr)
4
4
. 4
4
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
0.5
2
added
2200
2200
2200
2200
0
2100
2100
2100
2100
0
0
0
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
_ppb TCDD
soil after treatment
< 1
< 1
< 1
120
< 1
<1
<1
1.4
. 2.4"
<1
<1
<1
<1
82
<1
<1
40
<1
43
<1
15
36
a Type: Sample - TCDD, PEG/DMSO, KOH
Control - TCDD, PEG/DMSO, no KOH
Blank - No TCDD, PEG/DMSO, no KOH
These data indicate that the slurry process is capable of extracting TCDD
levels to < 1 ppb under a variety of conditions, and is relatively insensitive
to changes in reaction time and temperature.
22
-------
RECOVERY OF REAGENT FROM SLURRY SAMPLES
The slurry process uses a very large ratio'of reagent to soil, generally
around 1:1 by weight, in order to rapidly bring about contact between reagent
and dioxin. This compares to a reagent:soil ratio of about 1:5 for the in-situ
process. In order to operate the slurry process in an economical fashion, it
is necessary to recover the reagent for-re-use. This reagent recovery can be
achieved by decanting the reagent from soil or by washing the reagent out of
the soil with 100 ml of water, followed by distillation of the reagent/water mix-
ture. Washing of the soil to recover reagent has been successful as shown below.
% Recovery % Recovery
Sample » of DMSO of KPEGa
601 94 - 92
602 98 96
603 93 95
801 69 92
802 73 82
803 68 96
Average 82 92
a Note that KOH does not distill over.
Tests on other soils from a PCB test series gave DMSO recoveries of
94-99%, indicating that much higher recoveries than shown here may be possible.
GENERAL IN-SITU TEST PROCEDURE
The samples of soil decontaminated by in-situ methods were treated in
one liter amber glass jars with Teflon«-lined screw caps. Reagent was added at
a level of 20% by weight; 50 g of reagent for 250 g of soil. Heated in-situ
soils were kept open at 70°C in a water bath for the various times, and then
23
-------
quenched with 50 ml of 50% sulfuric acid and sealed pending analysis. The
nine ambient temperature in-situ soils were kept- sealed at room temperature
for seven days and then quenched with acid and resealed pending analysis.
Each set of treated soil samples had an associated set of blanks (soil
with PEG/DMSO but without addition of hydroxide or dioxin) and controls. Con-
trol soils were mixed with 1,2,3,4-TCDD and an incomplete reagent (no hydroxide).
Controls, blanks and samples were extracted and analyzed as a set.
Soil samples were mixed with a 1:1:1 solution of potassium hydroxide (KOH)/
polyethylene glycol 400 (PEG)/dimethyl sulfoxide (DMSO) and held for seven (7)
days at 25°C. The results of these tests are listed in Table 6.
The ambient temperature seven (7) day samples showed a 24-76% reduction
by extraction in TCDD level compared to >99.95% reduction of TCDD in the heated
slurry process.
Table 6. In-Situ KPEG Data
Sample #
841 11 91 0
841 11 911
841 11 91 2
841 11 91 3
841 11 91 4
841 1 1 91 5
841 11 91 6
841 11917
84111918
Type
Sample
Sample
Sample
Control
Control
Control
Blank
Blank
Blank
Time
(days)
7
7
7
7
7
7
7
7
7
ppb TCDD
Initial
2100
2100
2100
2100
2100
2100
0
0
0
in soil
Final
500
980
1600
540
2500
1800
<1
<1
<1
% Reduction
76
53
24
74
0
14
24
-------
SECTION 7
DISCUSSION .
Gas chromatographic mass spectrometric analysis have shown that KPEG
extracts and destroys PCBs and PCDDs in all samples tested. The PCDDs in
control samples containing polyethylene glycol was also extracted from soil
but was not destroyed due to the absence of potassium. The slurry process has
been tested on 2000 ppb TCOD and 1000 ppm Aroclor 1260 contaminated soils and
was determined to destroy the TCDD to less than one ppb and the PCB 1260 to
less than one ppm.
As shown in Tables 1, 2, 4, and corroborated in Table 6, the in-situ
process under ambient conditions is not as effective as the slurry process
t
in the destruction of PCB or PCDD contaminated soils. Future research plans
call for the de-emphasis of ambient condition studies and a re-emphasis on
the heated slurry and RF heated in-situ method for decontaminating PCB and
PCDD contaminated materials.
25
------- |