&EPA
                             Untied states
                             Environmental  Protection
                             Agency
                             Office of
                             Research and Development
                             Cincinnati,  OH 45268
EPA/540/R-94/527a
March  1995
SITE Technology Capsule
 IITRI  Radio  Frequency
 Heating Technology
Abstract

    Radio frequency heating (RFH) technologies use electro-
magnetic energy in the radio frequency (RF) band to heat soil
insitu, thereby potentially enhancing the performance  Of Stan-
dard soil vapor extraction(SVE) technologies.  Contaminants
are remove d from in situ soils and transferred to collection or
treatment facilities.

    The Illinois Institute of Technology  Research Institute
(IITRI) RFH process was evaluated under the U. S. Environ-
mental  Protection Agency (EPA)   Superfund InnovatlveTech-
nology Evaluation (SITE) Program at a site containing various
organic contaminants in a heterogeneous soil matrix. Due to
changes in the original design and operational problems ex-
perienced during the demonstration, the treatment area was
evaluated as two separate zones:  the "revised' design treat-
ment zone and the  "heated' zone. The revised design treat-
ment zone reflects both changes made to the design of the
RFH  system and operational  problems associatsd with shal-
low groundwater at the test site. The heated zone consists of
the area that achieved the target temperature of 150°C.

    Concentration changes were  calculated  from  paired pre-
and  post-demonstration soil samples; these concentration
changes were evaluated for statistical significance. Conclu-
sions have been drawn based only on data thai were statisti-
cally significant at greater than or equal to the 90 percent
confidence level.

    Within the revised design treatment zonethe estimated
mean concentration  decrease for Total Recoverable Petro-
leum  Hydrocarbons  (TRPH) was 60 percent. Estimated mean
concentration  decreases for twosemivolatile  organic com-
pounds (SVOCs), pyrene and bis(2-ethylhexyl)phthalate were
87 and  48 percent  respectively. There were statistically  sig-
nifiiant increases in the concentrations of four volatile organic
compounds  (VOCs);  the estimated mean concentration in-
creases were  457 percent for 2-hexanone; 263 percent for 4-
methyl-2-pentanone;  1,073 percent for acetone; and 683
percent for methyl ethyl ketone.
                              Outside  of the revised design treatment zone, only TRPH
                          showed a statistically significant change with an estimated 88
                          percent increase in the mean concentration.

                              Within the heated zone the estimated mean concentra-
                          tion decrease was 95 percent for TRPH.

                              Outside the heated zone, the estimated mean concentra-
                          tion decrease was  37 percent for  bls(2-ethylhexyl)
                          phthalate; estimated mean concentration  increases were 423
                          percent for 2-hexanone: 249 percent for 4-methyl - 2-pentanone;
                          1,347 percen t for acetone:  and 1,049 percent for methyl ethyl
                          ketone.

                              Several possible reasons exist for changes In concentra-
                          tion observed. They include  inward contaminant migration,
                          low extraction rates,  widely varying soil temperatures, low
                          pretreatment contaminant concentrations in the soil, and the
                          potential degradation of TRPH  and SVOCs.

                              The estimated cost to treat 10,152 tons of contaminated
                          soil based on a scaleup of the revised design treatment zone
                          is $619 per ton; the estimated cost to treat 8,640 tons  of
                          contaminated soil based on IITRI's theoretical system design
                          is $340 per ton.

                              The IITRI RFH technology was evaluated based on the
                          nine criteria used for decision making in the Superfund feasi-
                          bility study (FS) process. Results of the evaluation ars sum-
                          marized in Table 1. This evaluation was based on information
                          from the SITE demonstration conducted at Kelly Air Force
                          Base (AFB).

                          Introduction

                              In 1980, the U.S. Congress passed  the Comprehensive
                          Environmental Response, Compensation, and Liability Act
                          (CERCLA), also known as Superfund, committed to protect-
                          ing human  health and the environment  from uncontrolled
                          hazardous waste sites. CERCLA was amended by the Super-
                          fund Amendment and Reauthorization Act (SARA) in 1986.
                                       SUPERFUND  INNOVATIVE
                                       TECHNOLOGY EVALUATION
                                                                                    Printed on Recycled Paper

-------
 Table 1. Evaluation Criteria for the  IITRI RFH Technology1
Evaluation Criteria
                                                    Performance
Overall Protection of Human Health and the
Environment
Compliance with Federal ARARs1
Long-form Effectiveness and Performance
Reduction of Toxicity, Mobility, or Volume
through Treatment
Short-term Effectiveness
 Implementability
Cost3


State Acceptance

Community acceptance
Site-specific treatability studies will be needed to verify #10 levels of
contaminant removal achievable
Requires measures to protect workers during installation and treatment
Additional contaminants may form at high temperatures if not properly
designed or operated

Vapor collection and treatment are needed to ensure compliance with air
quality standards
Construction and operation of an onsite vapor treatment unit may require
compliance witfi location-specific ARARs
RF generator must be operated in accordance with National Institute of
Occupational Safety and Health (NIOSH) and Federal Communication
Commission (FCC) requirements

As with all SVE-based systems, the contamination source may not be
adequately removed
Involves some residuals treatment (vapor stream)

Potentially concentrates contaminants, reducing waste volume
Potentially reduces contaminant mobility, although downward mobility of
contaminants during treatment has not been quantified
May form new contaminants, thereby potentially reducing or increasing
toxicity, if not properly designed or operated

Presents minimal short-term risks to workers and community from air
release during treatment
No excavation is required, although drilling will disturb the soil to some
extent

RF generator must be operated in accordance with NIOSH and FCC
requirements
Pilot-scale tests have been completed at two other sites to evaluate the
technology; no full-scale applications to date
Because of operational problems experienced during the SITE
demonstration, consistent soil heating was not observed

Cost evaluation based on revised design treatment zone is $619 per ton;
cost evaluation based on IfTRI's theoretical system design is $340 per ton

No excavation is required, which should improve state acceptance

No excavation Is required, which should improve community acceptance
Potential health effects of RF fields may be an issue
   Based on the results of the SITE demonstration at Kelly AFB
   ARARs = Applicable or Relevant and Appropriate Requirements
   Actual cost of a remediation technology is site-specific and dependent on the target cleanup level, contaminant concentrations,
   soil characteristics, and volume of soil. Cost data presented in this table are based on the treatment of 10,152 tons of soil
   (scale-up based on the revised design treatment zone) and 8,640 tons of soil (based on IITRI's theoretical system design).
These amendments emphasize the  achievement of long-term
effectiveness and  permanence of  remedies at Superfund sites.
SARA mandates implementing permanent solutions and using
alternative treatment technologies or resource recovery tech-
nologies, to the maximum extent  possible, to clean up hazard-
ous waste sites.

    State and federal agencies,  as well as  private  parties, are
now exploring a growing number of  innovative technologies for
treating hazardous wastes. The sites on the National Priorities
List total more than  1,200 and comprise a broad spectrum of
physical, chemical, andenvironmental conditions requiring vary-
ing types  of remediation. The   EPA has  focused on  policy,
technical,  and informational issues related to exploring  and
           applying  new remediation technolo gy  applicable to Super-
           fund sites. One such  initiative is  EPA's SITE Program,  which
           was established  to   accelerate  development,  demonstration,
           and use of innovative  technologies for site cleanups. EPA SITE
           Technology Capsules summarize  the  latest information  avail-
           able on selected innovative  treatment and  site remediation
           technologies and related issues.  These capsules are designed
           to help EPA remedial  project  managers, EPA  on-scene coordi-
           nators, contractors,  and  other site cleanup managers  under-
           stand  the type  of  data  and  site characteristics  needed  to
           effectively evaluate a  technology's applicability for cleaning up
           Superfund sites.  Additional  details regarding technology  dem-
           onstrations are presented in  the  Innovative Technology  Evalu-
           ation  Reports.

-------
    This capsule provides information on thellTRI in situ RFH
technology. This technology was  developed to  improve the
removal ofVOCs and SVOCs from  the  soil  using  standard
SVE technologies. The IITRI  RFH  process was evaluated un-
der the SITE Program from January through August 1993 at
Kelly  AFB in  San Antonio, Texas.  This demonstration was
performed  In  conjunction with  a technology evaluation per-
formed by the U.S. Air Force (USAF). Information in this cap-
sule emphasizes specific  site characteristics and results of the
SITE  field demonstration  at  Kelty AFB. The  capsule presents
the following  information:
        Technology  Description
    •   Technology Applicability
    •   Technology  Limitations
    •   Process Residuals
        Site  Requirements
    •   Performance  Data
    •   Technology  Status
    •   Sources of Further Information
        References

Technology Description

    RFH  technologies use RF energy to heat soil in situ,
thereby potentially  enhancing  the  performance of standard
SVE technologies. The RF  energy causes dielectric heating of
the soil, which  is a faster and more  efficient mechanism  for
heating solids than is convective  heating.  Some conductive
heating also occurs  in the soil.

    In IITRI's  proprietary system, an RF  generator supplies
energy to  exciter electrodes, which are copper pipes installed
in vertical  boreholes. Copper balls welded onto the ends of the
exciter pipes help distribute  the energy, which tends to concen-
trate at these points. As the soil is heated  due to the dissipa-
tion of the RF  energy, contaminants and moisture  in the soil
are vaporized and pulled toward ground electrodes, which also
serve es vapor extraction wells. The  vaporized water may act
as a steam sweep  to further enhance the removal  of organic
contaminants.  A standard SVE  system provides a vacuum to
the ground  electrodes and transfers the vapors to collection or
treatment  facilities.  Contaminants are treated using standard
vapor treatment techniques. After soil treatment is complete,
the soil is  allowed to cool. The SVE system may be operated
during  part or all of  this cooling period. The exact numbers of
exciter and ground  electrodes,  electrode configurations,  vapor
collection or treatment techniques, and other design  details are
site-specific.

    Figure 1 is a schematic diagram  of the IITRI  RFH system
used  for  the  SITE  technology  demonstration  at Kelly AFB;
Figure 2 is a cross-section of IITRI's  RFH system. A 40-kW RF
generator  served as the energy source for  the system. Energy
was supplied  to the exciter electrodes for approximately  9
weeks  via coaxial  cables.  Reflected  energy was  measured,
and the electrical characteristics of the transmitted RF energy
were  adjusted  as   necessary.   Exciter electrodes were con-
structed of 2.5 and  4-inch (nominal diameter) copper pipe and
were  installed  in 10-inch boreholes  to  a  depth of 19.5  feet
below  the surface.  The boreholes were backfilled around the
electrodes using a material  similar to the surrounding soil. Four
exciter electrodes spaced 2.5 feet apart were installed in a row.
    Two rows of eight ground  electrodes each were  installed
parallel  to and on either side of the exciter electrode row.  The
ground electrodes were fabricated from 2-inch-diameter alumi-
num pipes. The electrode configuration was designed to  direct
the flow of RF energy through  the soil and contain the energy
within the treatment zone. With  the exception of the four corner
electrodes  that were not perforated,  the ground  electrodes
were perforated  on the side facing the treatment zone to permit
the collection of vapors from the soil. They were perforated in  8
uniform  pattern over the full length of the electrodes. Each
perforated  ground electrode was  connected  to a  manifold,
which led to the vapor treatment system. Two additional perfo-
rated vapor extraction  pipes were installed parallel to the ground
surface  to prevent buildup of vapors below the vapor barrier.

    As shown  in Figure 1, the insulated vapor barrier extended
several  feet over  the surface of the ground end exciter elec-
trodes to  prevent heat loss,  contaminant emission,   and air
infiltration. A sheet of expanded aluminum covered the same
area as the vapor barrier and was designed  to contain  RF
energy.  An  arched  aluminum   shield  that  covered only  the
electrodes in the treatment zone was also designed  to  ocntain
RF energy. To complete the  RF ocntainment system,  the
arched  RF shield was electrically connected  to  the ground
electrodes by aluminum wire mesh.

    The system was designed to  heat the soil evenly to a
temperature of approximately 150°C. The soil heating began at
the top  of the treatment zone near the exciter  electrodes,  and
the heated zone expanded outward and downward  over time.
The original area to be treated  measured  10 feet wide  by 17.5
feet long by 29 feet deep. Due to shallow groundwater and
operational problems  encountered  during the  demonstration,
the treatment zone was revised to 10 feet wide by 14.1  feet
long by  23.3 feet deep. The  RFH syaem  was evaluated using
this '"revised"  design treatment zone.  The  RFH system was
also evaluated using  the "heated" zone, which is the area  that
achieved the target temperature of  150°C. This zone measures
5.7 feet wide by  10.3 feet long by 20.0 feet deep. Tempera-
tures wtihin and outside  both treatment zones were monitored
at various depths throughout treatment. Contaminant removals
inside and  outside each of these zones were evaluated sepa-
rately; results of these evaluations are  presented in the Perfor-
mance Data Section.

    The vapor extraction system was operated throughout the
9-week  heating period and for approximately 2 months during
the cooldown  period. The vapor extraction system was  oper-
ated at a suction pressure of at least 7 inches of water  column
while heat was being  applied to the  soil. Vapors were extracted
from the soil at a rate of approximately  120 standard cubic  feet
per minute (scfm).  Condensate that formed in the system  was
collected; uncondensed vapors  were burned in a propane flare.

Technology Applicability

    The heat  provided  by the RFH  process increases  the
vapor pressure of contaminants In  the soil, thereby potentially
improving the effectiveness of SVE. RFH may make it possible
to  remove SVOCs that would  not  normally  be removed by
standard SVE technologies. RFH may  also speed the  removal
of VOCs, which can  be removed by standard  SVE technolo-
gies. Contaminants that can potentially be removed  using RFH
include  a wide variety of organics such as halogenated  and

-------
           60 Hz
         AC Power
         Distribution!
                                                              Vapor Barrier Perimeter


                                                                  Gratmd ElectrodBS
                                                          A1  A2
                                                                   A3
                                                                           A5   A6  A7
                                                                      1&5
                                                                  Exciter Electrodes
                                                            '1  TW2  31   82

                                                               XXX
                                                              TW3 TW4  TW6
       Ground Eec-trodes
                                                           C1   C2  C3
                                                                       C4

                                                                       TW7
                                                                            C5  C6  C7   C8
                                   Vapor
                                 Collection
                                  Manifold
                                                    Stage 1
                                                    Matching
                                                    Network
                                                     Teirperature
                                                     Measurement
                                                      Probe Lines
Figure 1. Schematic diagram otltTRt's RFH system.
nonhalogenated solvents and straight-chain and polycyclic aro-
matic  hydrocarbons found in gasoline, jet fuel, and diesel fuel.
Inorganics, metals,  and other nonvolatile contaminants will not
normally  be treated by SVE or RFH technologies.

    A previous study (not sponsored by the SITE program)
conducted at Volk AFB indicated that IITRI's  RFH system
effectively removed VOCs and SVOCs from a small treatment
zone  containing homogenous, sandy soils. The soil in the Kelly
AFB treatment zone  was a heterogeneous mixture of sand,
clay, and gravel. Due to the operational  problems encountered
during the Kelly AFB study, it cannot be determined how the
heterogeneous soil matrix affected RFH system effectiveness.
However, soils containing large amounts of silt, clay, and
humic  substances tend to adsorb organic contaminants more
tightly. Soils containing a large fraction of clay may also have
insufficient air permeability for adequate contaminant removal
by SVE or RFH technologies. Site-specific, in situ treatability
studies will  need to be conducted to confirm the  applicability of
this technology.

Technology  Limitations

    IITRI claims its technobgy is not ready for commercializa-
tion. Considerable development  and optimization of the pro-
cess would be required before a full-scale system would be
ready for field use. For example, IITRI must demonstrate the
system's ability  to  treat  an entire site  and to use  an RF
generator that can  supply more than 40 kW of RF energy. The
IITRI  RFH technology cannot be used  as  a stand-alone tech-
nology.  Vaporized  contaminants  and steam  must be  collected
for reuse or treatment. In some cases, residual contaminants
may  remain in the soil  after treatment

    This technology currently  appears to be limited  to unsatur-
ated soils, although  groundwater  pumping  may be used to
lower the water table and  extend the treatment zones.  RFH is
further limited to soils contaminated wirh VOCs and SVOCs.
Nonvolatile organics,  metals,  and inorganics will not normally
be removed by RFH or SVE technologies.

Process Residuals

    The IITRI RFH process generates one process waste
stream  that contains vaporized oontaminants  and steam di-
luted in extraction air. This waste stream can be treated by any
of a  number  of standard vapor treatment technologies includ-
ing vapor-phase carbon, condensation, or incineration. During
the SITE  demonstration at Kelly AFB, a  propane  flare  was
used to treat uncondensed contaminants in the vapor waste
stream.

    Steam and contaminants in the vapor stream  that con-
densed in the vapor collection  system were collected prior to
the flare. These condensed residuals were handled along with
other site wastes at Kelly AFB.  When groundwater is pumped
to lower the water table, the groundwater must also be handled
as a liquid residual.

    Soms soil contaminants may remain after treatment. Sig
nificant quantities of several organics  remained In the soil after
treatment was completed at Kelly AFB, although  remediation of

-------
                          Plastic weattier cove* •
                         RF Shisld-
                       {corrugated
                        aluminum)
Protective teflon tubing
(for fiber-optic temperature
measurements)
                                        Coaxial cable (supplies
                                        RF power to exciter
                                        electrodes)
               Ground electrode _
               (2* I.D. aluminum'
                          pipe)
                  PwforatiorB for,
                  vapor collection"*
                     (face inward
                   toward energy
                        source)


"" r



:

• i

\ 1

; <

j I

	 »
NJ
ri
r






Exciter electrode (2-5" V
or 4" nominal diameter .
copper pipe} 	 '







!



;

;

f
]i
1
rj
C ' •
c • :
j
C ;
C :| -Balb" {threaded
i; .5* or 4" I.D. ooppar ^ ^ •
^;: sphwes} :"
:: i :
\















E





































^
3 ;'

; a

3

i

3 :

3 •

3 :

i>
J :
3 :
D

3

i 3 I

                                      , Fiber-optic probe
                                       {removed after each
                                       measurement)
                                                Three-layer
                                                vapor barrier
                                                and aluminum
                                                lattice shield
                                            -I" connector
                                             (tor routing rising
                                             vapor horizontally
                                             through vapor
                                             oolleetiori manifold)
                                    A3
           B1
                                                                            TW4
C3
Figure 2.  Cross-section of IITRI's RFH system.

-------
the site was  not an  objective  of the demonstration.  Further
treatment will  be required to fully remediate these soils.

Site Requirements

    Onsite assembly of the full-scale system will take several
weeks,  including drilling time.  It is expected that medium to
large sites will be divided into several sections  that will be
treated consecutively.  The soil must be allowed to cool  before
final soil samples can be collected  to confirm  cleanup of each
section.  Soil cooldown will take  up to several months if  portions
of the soil  reach 1,000°C,  as they did during  the SITE  demon-
stration.  However, with proper  disign and operation, soil  heat-
ing may be more uniform,  resulting in lower soil temperatures
and faster cooldown. After treatment is completed, system
demobilization may take up to a week.  Access roads are
needed for equipment transport.  Approximately  4,600 ft2 of
level ground  are needed to accommodate  the trailer-mounted
RF generators, controllers, and other support  equipment.  A
bermed  area IS needed for decontamination of the drill rig.
Area is also needed for storage of condensed vapors, if appli-
cable, and the selected vapor treatment system.

    Remediation using the RFH process will  require that cer-
tain utilities be available at the site. Water must be  available for
steam-cleaning and  other equipment  and personnel decon-
tamination  acivities. Electrical power must  also  be available.
The RF generator requires 480-volt, 3-phase power, and most
of the minor system components require 240-volt,  single-phase
power.  If carbon adsorption is used to treat vapors, com-
pressed air may be required  for system  control,  and steam or
hot air will be needed if the carbon is regenerated  onsite.
Natural gas or propane will  be required if a flare is used to
control vapors.

    The ground electrodes, expanded aluminum sheet,  arched
aluminum shield, and aluminum wire mesh minimize RF en-
ergy emissions. Monitoring  should  be conducted  to ensure that
the RF  field outside  of the treatment zone does not exceed
NIOSH or FCC requirements. These regulations were report-
edly met during the SITE  demonstration  and other previous
field studies.  The system  is relatively quiet,  and  only during
installation will dust and vapors be  a potential problem.  There-
fore, the RFH technology  can  be applied  near residential ar-
eas.

Performance Data
    The system was designed to heat the soil evenly to a
temperature of approximately 150°C.  Soil temperatures were
monitored by the vendor throughout the 61 day treatment pe-
riod during which RF energy was applied to the soil. Before
treatment began, the soil  temperature throughout the area was
approximately 20°C. During the SITE demonstration, the RF
energy applied to the exciter electrodes progressed gradually
from the  surface to the deepest point of each exciter electrode.
The soil temperature near the ground electrodes  increased
gradually as RF energy flowed from the exciter electrodes to
the ground electrodes. Near the end of the demonstration all
exciter electrode temperatures varied  widely;  maximum tem-
peratures near the exciter electrodes exceeded 1,3000°C.  Tem-
peratures near the  ground electrodes did not exceed  112°C
near the surface, and did not exceed 52°C below 24 feet.
Based on temperature measurements,  it appears that a system
malfunction could have resulted in incomplete heating of the
revised design treatment zone.  IITRI believes this malfunction
was a result of electrical problems due  to shallow groundwater.
    Groundwater levels may have been higher than originally
expected by the vendor. Due to the  presence of shallow ground-
water,  IITRI shortened the exciter electrodes to avoid potential
system  operational problems.  Groundwater levels were  moni-
tored infrequently,  so insufficient data are available to deter-
mine exact water  levels during the demonstration.  However,
exciter electrodes removed after the demonstration had melted.
Since copper melts at approximately 1,100°C, these electrodes
provide evidence of a system malfunction that prevented full
utilization of RF power for soil heating. Based on IITRI's soil
temperature logs,  it appears  that the system malfunction  oc-
curred during the last weeks of the heating period.

    Contaminant removal during the demonstration was evalu-
ated  by measuring contaminant concentrations in the soil be-
fore  and after treatment. Soil samples were collected  using
split  spoons with stainless steel sleeves before and after the
soil was treated. The samples were collected as matched
pairs; each post-treatment sample was collected as near as
possible to its corresponding pretreatment  sample. Within  the
revised design treatment zone,  26 matched pairs  of soil samples
were collected; 9 matched pairs  of soil samples were collected
outside this zone. Within the heated zone, 6 matched pairs of
soil samples were  collected; 31  matched pairs of soil samples
were collected  outside  this zone. Only complete matched pairs
were used in the evaluation of the data. A complete matched
pair  was defined as a matched pair in which  both pre- and
post-treatment samples were collected and analyzed, and in
which the compound being evaluated was measured at a
concentration greater than the  detection limit in at least one of
the two samples. The  number of complete  matched pairs was
limited by the low pretreatment concentrations of many con-
taminants  in the soil. This made t difficult to determine any
change after treatment because pretreatment concentrations
were often below analytical quantitation limits.

     For each  contaminant, the population distribution of the
concentration  data was evaluated. Most contaminants were
log-normally distributed,  and  as a result, concentration data
were log transformed. The ratio of post-treatment concentra-
tion  to  pretreatment concentration was calculated  for each
complete matched pair.  The  resulting set of ratios for each
contaminant of interest was evaluated using a t-test to deter-
mine whether the contaminant concentration exhibited a statis-
tically significant change  between  the pre- and  post-treatment
sampling events. The  mean ratio  of post-treatment  concentra-
tion to  pretreatment concentration was also calculated for each
contaminant. The  mean  ratio was  then converted to either the
mean percent decrease or the mean percent increase, as
appropriate. Conclusions have been drawn based only on
changes that were statistically significant at the  90 percent
confidence level or greater.

     Because of the high concentrations of TRPH expected at
the site, TRPH was designated as an indicator (i.e., critical)
contaminant.  Method 418.1 [I] was used to determine TRPH
concentrations in the soil samples following extraction with
Freon according to Method 9071 [2].

     Soil samples were extracted  by Method 3540 [2] and
analyzed  for SVOCs using Method 8270 [2]. Five SVOCs were
designated as indicator (i.e., critical) compounds  because their
concentrations  were expected  to be significant. These included
2-methylnaphthalene;   naphthalene;  1,2-dichlorobenzene;   1,3-
dichlorobenzene;  and 1,4-dichlorobenzene.  Because a prelimi-
nary statistical evaluation did not indicate any significant changes
in the  concentrations of these compounds,  none of them were
included in the final statistical evaluation and no conclusions

-------
can be drawn. Two other (i.e., non-critical) SVOCs were statis-
tically evaluated  because  they  exhibited  concentration  de-
cre ases.

    Method 824 0 [2] was used to determine VOC  concentra-
tions in the soil samples. Benzene, toluene, ethylbenzene, total
xylene. and chlorobenzene were designa  d as indicator (i.e.,
critical )VOCs  because of their expected  high  concentrations.
A preliminary  statistical evaluation  conducted for  these  five
compounds indicated  that  only  chlorobenzene  exhibited  a
change  in concentration, and was, therefore, the only indicator
VOC  included in the  final  evaluation. However  in the  final
statistical  evaluation, chlorobenzene  did  not exhibit a  statisti-
cally significant change at the 90  percent  confidence level.
Four othe rVOCs were  statistically  evaluated  because  they
exhibite d concentration increases.

 Revised Design  Treatment Zone Results

    The following results were  observed within the revised
design treatment  zone:

  * There was a statistically significant decrease in TRPH  con-
    centration at the 95 percent confidence level: the estimated
    decrease in the mean concentration was 60 percent.

  • No conclusions can be drawn  regarding the five indicator
    SVOCs.

    There were statiscally significant decreases in the  concen-
    trations  of  two  other SVOCs, pyrene  and bis(2-.
    ethylhexyl)phthalate, at the 97.5 percent   confidence level;
    estimated decreases in the mean concentrations were 87
    and 48 percent .respectively.

    The decreases in TRPH and SVOCs are likely due to the
performance of the RFH system, which may have resulted in
the volatilization of contaminants, allowing them to be collected
by  the  SVE system. These decreases may also have been
caused  by the degradation of these compounds from the el-
evated temperatures of the RFH system.  Decreases from out-
ward migration are unlikely, since the configuration  of the  SVE
system would  limit this type of migration.

    For th eVOCs within the revised  design treatment zone,
the following  results were  observed:

  * There wss no statistically significant change in the concen-
    tration o fchlorobenzene at the 90 percent confidence level.
    No  conclusions can  be drawn  regarding the other four
    indicator VOCs.

  - There were statistically  significant increases in  the concen-
    trations of four other VOCs (all ketones) at the 99 percent
    confidence level; estimated increases in the mean concen-
    trationswere  457 percent for 2-hexanone, 263 percent for
    methyl-2.pentanone, 1,073 percent for acetone, and 683
    percent for methyl  ethyl ketone.

    The ketones  may have been formed by the degradation  of
TRPH  and SVOCs  near the exciter electrodes,,  where soil
temperatures were  highest. A  possible degradation  pathway
may be the pyrolytic conversion of TRPH to unsaturated hydro-
carbons. In the presence of sufficient oxygen and a catalyst
(e.g.,  silica in the  soil), the RF energy  may  convert these
hydrocarbons  into ketones. The increase  in ketones may  also
have  been caused  by inward migration.  Possible  sources  of
ketones are the groundwater, which was not sampled,  and the
soil beyond the sampled  area, although these sourc es  cannot
be verified.  However,  there are insufficient  data to confirm or
disprove  either of these  hypotheses.  Statistically significant
changes  in TRPH,  SVOC,  and VOC  concentrations in the
revised design treatment zone are listed in Table 2.

    Outside of the revised design treatment zone, only TRPH
showed a  statistically significant change at the 95  percent
confidence  level, with  an  estimated  88 percent mean concen-
tration  increas e. Based on the configuration of the SVE sys-
tem, this increase may have  been due  to  inward migration;
however, it is not likely  this  increase was due  to outward
migration from the revised design treatment zone.

Heated Zone  Results
    The  following results were observed  within the heated
zone:
  • There was a statistically significant decrease in TRPH con-
    centration at the 97.5 percent confidence level  the estimated
    decrease in the mean concentration was 95 percent.

  • No conclusions can  be drawn regarding the five indicator
    SVOCs.

  • No conclusions can  be drawn regarding the five indicator
    VOCS.

    The TRPH decrease may be the result of the SVE system
pulling the contaminants out of the heated zone into the vacuum
wells. As in the  revised design treatment zone, this  decrease
may also have  been the result of the  degradation  of these
compounds caused by the  elevated temperatures of the RFH
system.

    There  was also a statistically significant decrease in the
concentration of bis(2-ethylhexyl)phthalate at the 90 percent
confidence  level outside the heated zone;  the  estimated de-
crease in the  mean  concentration was 37  percent.  As inside
the two zones,  this decrease  may  be caused by  the SVE
system pulling the contaminant  into the vacuum  wells or by
degradation due  to  the elevated temperatures of  the RFH
system.

    There  were  statistically significant  increases  at the  99
percent confidence level in the concentrations of four other
VOCs  (all  ketones)  outside the heated zone.  The  estimated
mean increases for these four  ketones were 423 percent for 2-
hexanone,   249 percent for 4-methyl-2-pentanone,  1,347 per-
cent for acetone, and 1,049 percent for methyl ethyl ketone. As
previously explained, these  ketones may have been formed by
the degradation of TRPH and SVOCs or inward migration.

    Several possible reasons exist for changes in concentra-
tion observed. They include inward contaminant migration, low
extraction rates, low soil temperatures achieved in some areas,
and low pretreatment contaminant concentrations in  the soil.

    Two-dimensional modeling  of gas flow rates was used to
qualitatively evaluate inward migration and treatment zone  ex-
traction rates.  The results of this modeling  indicate inward gas
flows.  Due  to  inefficiencies  in the design of the SVE system,
gas flows between the  outer edge of the impermeable cap and
the extraction  wells are five times  greater than those between
the two  rows of extraction wells.  As a  result, contaminant

-------
migration  into the treatment zone is possible,  especially near
the outer edges,  and contaminant removal from the treatment
zone may have been relatively slow as compared to inward
contaminant migration.

    Concentrations of TRPH and specificVOCs and SVOCs In
the gas stream were monitored by  a USAF  subcontractor and
were not part of the SITE demonstration. However, the appro-
priateness of the  methods used and the  quality of the data  are
unknown.  The results appear to qualitatively  indicate  removals
of TRPH and certain VOCs and SVOCs.  Because of limitations
of the sampling  and analytical methods, the  quantity of con-
taminants removed  cannot be estimated .

Technology  Status

    Information Is currently available   from  IITRI on two field
studies (not  sponsored by the SITE program) conducted at
VolkAFB  and Rocky Mpuntain Arsenal  [3][4]. IITRI  is perform-
ing a larger test at Sandi a   National  Laboratory. Results of  the
Kelly AFB demonstration are documented in  this capsule.

    IITRI  claims itstechnology  is not ready for commercializa-
tion.  Considerable development and optimization  of the pro-
cess would be required before a 1
ready for field use.
ll-scale system  would be
    A 300-kW full-scale syste m has been proposed. The cost
of a full-seal etreatment system , based on the result s obtained
from the revised deslg n treatment zone,  is  estimated   to be
$619 per ton for a site containing 10,192 tons of contaminated
soil. The cost of a full-scale system treatment system, based
on IITRI's theoretical system design, Is $340 per ton for a site
containing 9,640 tons of contaminated soil.  Costs  associated
with  analyses, site preparation, permitting, effluent treatment
and  disposal,  and  residuals and waste shipping are considered
site-specific coststhat will be assumed by the site owner or
responsible party.  As a result, these costs are not  included in
the per ton  cost estimates.

Disclaimer

    Although the   technology  conclusions presented in  this
report may  not change, the data have not been reviewed by
EPA Risk Reductio n Engineering Laboratory Quality Assur-
ance personnel at this time.

-------
Sources of Further Information

EPA Contact:
Laurel Staley
U.S.  Environmental Protection Agency
Risk Reduction Engineering Laboratory
26 West Martin Luther King Drive
Cincinnati. OH 45268
Telephone No.: 513-569-7863
Fax No.: 513-569-7620

Technology Developer:
Harsh Dev
NT Research Institute
10 West 35th Street
Chicago, IL 60616-3799
Telephone No.: 312-567-4257

Kelly AFB Project Engineer for Site S-1 RFH
Field  Demonstration:
Victoria Wark
SA-ALC/EMRO
305 Tinker Drive, Suite 2, Building 305
Kelly AFB, TX 78241-5915
Telephone No.: 210-925-1812
Brown and Root Environmental Project
Manager:
Clifton Blanchard
Brown and Root Environmental
800 Oak Ridge Turnpike, Suite A600
Oak Ridge, TN 37830
Telephone No.:  615-483-9900

References

1.   U.S. Environmental Protection Agency, EPA Methods for
    Chemical Analysis of Water and Wastes. 1983.
2.   U.S. Environmental Protection Agency, Test Methods for
    Evaluating Solid Waste (SW-846): Third Edition, Novem-
    ber, 1986, and Final Update I, September,  1990.

3.   Program Manager for Rocky  Mountain Arsenal, Final
    Rocky Mountain Arsenal Task 012 In Situ Radio Fre-
    quency/Vapor Extraction Pilot Test Plan. Document Con-
    trol Number 53300-01-12-AACB.
4.   Illinois Institute of Technology, In Situ Soil  Decontamina-
    tion by Radio-Frequency Heating Field Test, September
    1989.

-------