United States
Environmental Protection
Agency
Solid Waste and
Emergency Response
(5102W)
EPA 542-N-94-002
March 1994
SEPA Ground Water Currents
Developments in innovative ground water treatment
Hydrodynamic Cavitation Oxidation Destroys Organics
By Richard Eilers, Risk Reduction Engineering Laboratory
The CAV-OX® technology
destroys organic contaminants
(including chlorinated hydro-
carbons) in water. The process
uses hydrogen peroxide, hydro-
dynamic cavitation and ultra-
violet (UV) radiation to
photolyze and oxidize organic
compounds present in water at
parts per million to nondetect-
able levels. Ideally, the end
products of the process are wa-
ter, carbon dioxide, halides
and-in some cases-organic
acids. The CAV-OX® technol-
ogy was evaluated at a SITE
(Superfund Innovative Tech-
nology Evaluation) demonstra-
tion at Edwards Air Force Base
Site 16 in California. Ground
Water at Site 16 is contami-
nated with volatile organic
compounds (VOCs), primarily
trichloroethene (TCE) and
BTEX compounds (benzene,
toluene, ethylbenzene and
xylenes).
Almost 8,500 gallons of
contaminated ground water
were treated during a two-week
period. Initial contaminant
concentrations were 1,475 to
2,000 parts per billion (ppb)
TCE, 240 to 500 ppb benzene,
8 to 11 ppb toluene and up
to 100 ppb xylene. The
CAV-OX® systems achieved
removal efficiencies of up to
>99.9% for TCE and BTEX
compounds.
The major components of
the CAV-OX® system are the
cavitation chamber, UV
reactor and control panel
unit. Prior to entry into the
cavitation chamber, ground
water was pumped from three
monitoring wells into a 7,500
gallon equalization tank. A
bladder tank was used as the
equalization tank to minimize
variability in influent charac-
teristics. From the equaliza-
tion tank, the water was
transferred to an influent
holding tank, where hydro-
gen peroxide was added. The
water was then pumped to
the cavitation chamber.
Cavitation occurs when a
liquid undergoes a dynamic
pressure reduction while un-
der constant temperature.
The hydrodynamic cavitation
is induced through the shape
of the cavitation chamber,
which causes pressure varia-
tions in a flowing liquid. A
pressure reduction causes gas
bubbles to suddenly develop,
grow and then collapse. This
cavitation decomposes water
into extremely reactive hy-
drogen atoms and hydroxyl
radicals, which recombine to
form hydrogen peroxide and
molecular hydrogen, which
help oxidize the organic com-
pounds. Flow can be recycled
through the cavitation cham-
ber to control the hydraulic
retention time before it is
transferred to the UV reactor.
The UV reactor houses
low-pressure mercury-vapor
lamps that generate UV
radiation, which further oxi-
dize the organic compounds.
Each lamp is housed in a
UV-transmissive quartz tube,
which is mounted entirely
within the UV reactor.
Hydroxyl and hydroperoxyl
radicals are produced by direct
photolysis of hydrogen perox-
ide at UV wavelengths.
During the SITE demonstra-
tion, no scaling of the quartz
tubes was observed. Treated
ground water was stored in an
effluent storage tank prior to
disposal.
Magnum Water Technolo-
gy manufactures both low-en-
ergy and high-energy UV
systems, both of which were
evaluated during the SITE
demonstration. The low-ener-
gy CAV-OX® I system con-
tains six 60-watt lamps per
reactor. The high-energy
CAV-OX® II system contains
two UV reactors with one UV
lamp each and can operate at
2.5, 5, 7.5 or 10 kilowatts
(kW). Flow capacity is esti-
mated to be less than 3 gal-
lons per minute (gpm) for the
low-energy system and less
than 5 gpm for the high-ener-
gy system. Three configura-
tions of the CAV-OX® tech-
nology were demonstrated
during the SITE evaluation:
the CAV-OX® 1 system oper-
ating at 360 watts and the
CAV-OX® II system operating
at both 5 kW and 10 kW. The
demonstration consisted of 15
runs for each configuration of
the CAV-OX® technology.
The high-energy system was
first operated with the UV re-
actor at 10 kW and then at 5
kW. Ground water samples
were collected before and after
treatment during each run to
determine the technology's ef-
fectiveness in removing VOCs
from ground water. The prin-
cipal operating parameters—
hydrogen peroxide dose, pH
and flow rate-were varied
during the demonstration to
evaluate the technology's per-
formance under different
conditions.
For more information, con-
tact Richard Eilers at EPA's
Risk Reduction Engineering
Laboratory at 511-569-7809.
An "Applications Analysis
Report and a Technology
Evaluation Report" will be
available in the summer of
1994.
This Month in Currents
UV Oxidation
Biosparging Update
Surfactant Research
Ground Water Models
Recycled/Recyclable
Printed with Soy/Canola ink on paper that contains at least 50% recycled fiber
-------
RESEARCH UPDATE
Biosparging Documented in Fuel Remediation Study
By Don Kampbell, Robert S. Kerr Environmental Research Laboratory
E PA's Robert S. Kerr Envi-
ronmental Research Laborato-
ry (RSKERL), through a
three-year research study, has
documented subsurface aera-
tion (biosparging) remedia-
tion of an aviation gasoline
spill at the U.S. Coast Guard
Air Station site in Traverse
City, Michigan. This case
study has shown that fuel vol-
atilization by aeration and va-
dose zone biodegradation of
vapors is a convenient way to
remove dissolved hydrocar-
bons from ground water in sit-
uations where large amounts
of spilled fuel have moved
downward through a porous
vadose zone and formed a
plume in the aquifer. Sparge
aeration can cleanse the water
of fuel hydrocarbons to meet
ground water quality stan-
dards. However, sparge cleans-
ing of the plume water is a
short-term solution unless
there is further remediation of
the aquifer. This study found
that complete remediation of
contaminants was prevented
by fuel globules trapped in
capillary pores of sand gran-
ules that protected them from
the sparge aeration. These oily
globules can recharge and
maintain the contaminant
plume once sparging ceases.
At the Traverse City site,
about 36,000 gallons of gaso-
line had spilled in 1969 as a
result of a flange failure of an
underground transfer line.
During the next 20 years a
plume 1,200 feet long down
gradient was formed. The wa-
ter table is at a depth of ap-
proximately 15 feet with an
oily phase smear of almost
five feet, due to fluctuations
in water depth resulting from
climatic changes. During the
study period about one-third
of the smear zone was in the
vadose zone and two-thirds
were at or below the water ta-
ble. Both the aquifer and the
vadose zone were composed of
relatively uniform beach sand.
Prior to the field-scale
study, an eight-month bio-
venting pilot-scale demonstra-
tion was conducted. At its
completion in 1991, the sys-
tem's performance showed that
99% of the fuel hydrocarbons
in the vadose zone were re-
moved, with only minimal
surface emissions. Upon com-
pletion of the pilot study, aera-
tion wells were installed in the
same plot to a depth of
10 feet below the water table.
The rate of aeration was the
same as for the pilot-scale
(SEE BIOSPARGING, PAGE 3)
Surfactant Flushing Research
to Remove Organic Liquids from Aquifers
By Linda M. Abriola and Kurt D. Pennell, University of Michigan
Organic liquids, such as gas-
oline and industrial solvents,
are a major source of ground
water contamination through-
out the United States.
Through the Great Lakes/
Mid-Atlantic Hazardous Sub-
stances Research Center, re-
searchers at the University of
Michigan have combined de-
tailed laboratory experiments
with the development of
mathematical models to in-
vestigate the potential useful-
ness of surfactant flushing as
an aquifer-remediation
technology. The specific ob-
jectives of this research were
to: (1) screen and select sur-
factants that will enhance the
solubility of organic liquids in
water; (2) measure the solubil-
ity of dodecane and tetrachlo-
roethylene (PCE) in aqueous
surfactant solutions; (3) quan-
tify the ability of selected sur-
factants to recover entrapped
dodecane from soil columns;
and (4) develop and evaluate
numerical models capable of
predicting surfactant-en-
hanced solubilization and
mobilization of organic liq-
uids in ground water systems.
First, commercially avail-
able surfactants were screened
based on their toxicity, bio-
degradability, molecular
structure and potential to sol-
ubilize organic compounds.
The screening process led to
the selection of three nonion-
ic surfactants for experimen-
tal testing with two organic
liquids, dodecane and PCE, as
model compounds. The re-
searchers found that adding
these surfactants to water
increased the aqueous solubili-
ty of PCE and dodecane by
200 times and one million
times, respectively. The large
enhancement in solubility re-
sults from the incorporation or
partitioning of organic com-
pounds within surfactant mi-
celles (colloidal-size clusters).
Surfactant molecules aggregate
to form micelles above a spe-
cific concentration, the criti-
cal micelle concentration
(CMC). The micelles possess a
(SEE SURFACTANT FLUSHING,
PAGE 4)
Ground Water Currents
-------
NEW FOR THE BOOKSHELF
Compilation of Ground-Water Models
EPA'S Robert S. Kerr Envi-
ronmental Research Laborato-
ry has published a report,
"Compilation of Ground-Wa-
ter Models" (Document No.
EPA/600/R-93/118). This re-
port is a review of ground wa-
ter models and is based on
information gathered by the
International Ground Water
Modelling Center (IGWMC)
under a research and technol-
ogy transfer cooperative agree-
ment with the EPA. The
IGWMC was established as an
international clearinghouse
and technology transfer center
for ground water modelling.
Ground water modelling, as
a computer-based methodolo-
gy for mathematical analysis,
is a tool for investigating and
managing the mechanisms
and controls of ground water
systems. Models are playing
an important role in the de-
termination of the physical
and economical effects of
proposed ground water protec-
tion policy alternatives and
thus the protection of human
and ecological health. Com-
puter models are important
tools in the screening of alter-
native remediation technolo-
gies and strategies in cleaning
up ground water systems pol-
luted in the (recent) past, in
the sound design of ground
water resource development
schemes for water supply and
for other land use modifica-
tions affecting ground water
systems.
The model selection pro-
cess for appropriate computer
codes is a vital step to con-
ducting these investigative
and management alternatives
for ground water systems. To
be able to select a computer
code appropriate for the type
of analysis to be performed,
ground water modelers need
to have an overview of avail-
able computer codes and their
characteristics. These model-
ling codes are used for the
evaluation of policies, actions
and designs that may affect
such systems. This report pre-
sents the methodology used by
the IGWMC to classify, eval-
uate and manage descriptive
information regarding ground
water modelling codes for the
purpose of model selection.
Furthermore, the report pro-
vides an overview of available
ground water modelling codes
and their major characteris-
tics. A section is included that
defines ground water model-
ling, presents the classifica-
tion approach taken by the
IGWMC and discusses differ-
ent types of models and the
mathematical approaches in-
voked for developing the
models. Separate sections dis-
cuss and review the different
categories of ground water
models: flow models, transport
models, chemical reaction
models, stochastic models,
models for fractured rock and
ground water management
models.
The appendices include a
listing and description from
the IGWMC Model Annota-
tion Search and Retrieval Sys-
tem (MARS) of selected
models from each category.
Currently this MARS data-
base is installed on a micro-
computer operating under
MS-DOS. Detailed informa-
tion on the reviewed models is
presented in a series of tables,
preceded by an introduction
on model classification and
principal characteristics of the
described models.
The report can be ordered
from EPA's Center for Envi-
ronmental Research Informa-
tion at 513-569-7562. Please
refer to Document No. EPA/
600/R-93/118 when ordering.
Biosparging,
from page 2
bioventing—an air flow pat-
tern upward that enabled the
air to remain below ground for
approximately 24 hours.
Plume water initially contain-
ing several hundred micro-
grams per liter (ug/1) of BTEX
compounds (benzene, toluene,
ethylbenzene and xylenes) was
cleansed to <1 ug/1.
After one year of operation,
and again after two years, rep-
licate vertical profile core
samples were collected from
the sparged plot and from an
adjacent non-sparged control
in the plume. Considerable
variations between replicated
profiles for fuel carbon con-
centrations were detected.
Averaged values for total fuel
carbon of replicates showed
that non-sparged control sam-
ples decreased by 10% while
sparged replicates showed a
42% decrease. Most of the
sparged decrease occurred dur-
ing the first year. The ability
of the system to completely
eliminate contaminants was
restricted because of fuel glob-
ules trapped in capillary pores
of sand granules which pro-
tected them from the sparge
aeration.
For more information, call
Don Kampbell at RSKERL at
405-436-8564. A history of
the first year of work has al-
ready been published; the ref-
erence is: Kampbell, D. H.,
C. J. Gnffm and F. A. Blaha,
"Comparison of Bioventing
and Air Sparging for In-Situ
Bioremediation of Fuels,"
Proceedings of Symposium on
Bioremediation of Hazardous
Wastes: Research, Develop-
ment, and Field Evaluations,
Dallas, Texas, 1993, pp. 61-65
(Document No. EPA/600/
R-93/054) and can he ordered
from EPA's Center for Envi-
ronmental Research Informa-
tion at 513-569-7562. A
publication on the full study
is anticipated for the Fall of
1994.
Ground Water Currents
-------
Surfactant Flushing,
from page 2
lipophilic (attracted to or sol-
uble in oils) core surrounded
by a hydrophilic (attracted to
or soluble in water) mantle.
When the concentration of
surfactant exceeds the CMC,
organic compounds dissolve
within the lipophilic core of
surfactant micelles.
The most promising surfac-
tant tested, polyoxyethylene
(20) sorbitan monooleate
(trade name Tween 80 or
Witconol 2722), was used in
the soil-column experiments.
This is a food-grade surfactant
commonly used in dietary
supplements, flavoring agents,
whipped toppings and short-
enings. Dodecane was used as
the model organic compound.
Prior to surfactant flushing,
dodecane was entrapped in
water-saturated soil columns
packed with a uniform sand.
After the introduction of a
4% surfactant solution, the
concentration of dodecane
exiting the column increased
dramatically. Removal of
10% of the residual dodecane
required 0.7 liters of surfac-
rant solution. Comparable re-
covery of dodecane without
surfactant would have re-
quired approximately 130,000
liters of water.
Although high, the con-
centrations of dodecane mea-
sured in the column effluent
were seven times less than
those measured in batch ex-
periments. These results im-
ply that the equilibrium
solubility of dodecane was not
reached within the soil col-
umn. Subsequent column
experiments conducted at
several flow rates confirmed
the existence of rate-limited,
rather than instantaneous,
solubilization of residual
dodecane. Numerical models
were then developed which
coupled surfactant transport
with the solubilization of re-
sidual organic liquids. The
models were used to interpret
laboratory experiments, eval-
uate alternative remediation
strategies and investigate the
factors which influence the
solubilization and mobiliza-
tion of organic liquids at the
field scale. Using these mod-
els, HSRC researchers ex-
plored optimal surfactant
technologies, based on
the amount of flushing time
and amount of surfactant so-
lution required to remove re-
sidual dodecane from soil
columns.
This research demonstrates
the ability of surfactants to
enhance the solubility or or-
ganic liquids and to promote
recovery of entrapped organic
liquids from soil columns.
Model simulations were
shown to be valuable tools in
Interpreting data and evaluat-
ing alternative pumping strat-
egies. The results of these
projects provide a basis for
further development of surfac-
tant flushing as an aquifer re-
mediation technology.
Ongoing research efforts focus
on processes influencing the
solubilization and mobiliza-
tion of PCE entrapped within
several aquifer materials.
For more information, con-
tact Linda Abriola at the
University of Michigan
(313-764-9406).
To order additional copies of Ground Water Currents, or to be included on the permanent mailing list, send a fax request to the National
Center for Environmental Publications and Information (NCEPI) at 513-891-6685, or send a mail request to NCEPI, 11029 Kenwood Road,
Building 5, Cincinnati, OH 45242-0419. Please refer to the document number on the cover of the issue if available.
Ground Water Currents welcomes readers' comments and contributions. Address correspondence to:
Managing Editor, Ground Water Currents (5102W). U.S. Environmental Protection Agency,
401 M Street S.W.. Washington, DC 20460.
United States
Environmental Protection Agency
National Center for Environmental
Publications and Information
P.O. Box 42419
Cincinnati, OH 45242-0419
Official Business
Penalty for Private Use $300
BULK RATE
Postage and Fees
Paid
EPA
Permit No. G-35
EPA 542-N-94-002
------- |