Advanced Oxidation Technologies for the Treatment of
Contaminated Groundwater
(U.S.) Environmental Protection Agency, Cincinnati,
1994

-------
EPA/M0/A-94/00S
	NORMA M LEWIS
KIRANKUMAR TOPUDURTI
Advanced Oxidation Technologies
for the lYeatment of
Contaminated Groundwater
ABSTRACT
This paper presents information on two pilot-field applications of
advanced oxidation technologies for contaminated groundwater with
organics. The two UV/oxidation technologies were developed by
Ultrox International of Santa Ana, California and Peroxidation
Systems, Inc. of Tucson, Arizona. The ritrox technology was
demonstrated in 1989 with the U.S. Environmental Protection
Agency's Superfund Innovative Technology Evaluation (SITE) program
at the Lorertz Barrel and Drum (LB&D) site in San Jose, California.
Peroxidation Systems technology was applied at the Old O-Field site
located within the Aberdeen Proving Ground, in Maryland.
The Ultrox system was evaluated for its effectiveness in treating
the volatile organic compounds (VOCs) present in groundwater at the
LB&D site. Achievement of VOC removals were greater than 90
percent under best operating conditions at that time. Most VOCs
were removed through Chemical oxidation, however, for a few VOCs,
stripping also contributed toward removal. The treated groundwater
met the applicable discharge standards at 95 percent confidence
level for discharge into a local waterway. There were no harmful
air emissions to the atmosphere from the Ultrox system, which is
equipped with an off-gas treatment unit.
The Peroxidation Systems technology, achieved contaminant removal
efficiencies of about 96 percent, and the treated water met the
federal maximum contaminant levels for drinking water.
The information presented includes a description of the
technologies, factors affecting the technologies, and results from
the two pilot-scale studies of the UV/oxidation treatment system
applications.
REPRODUCED BY:
U.S. Department o* Commerce
National Technical Information Serve*
Springfield, Virginia 22161
Norma M Lewis. M A , US Environmental Protection Agency. Risk Reduction Engineering Laboratcy Oncinnat. Ohio USA
Mankumar Topudurti. Ph D PRC Env-ronmenta! Management Inc . Ch.cago. .nmo
-------
INTRODUCTION
The Superfund Innovative Technology Evaluation (SITE) program was
created in 1986 to provide information on alternative and
innovative technologies. The SITE program also generates reliable
performance and cost data for these technologies from each
demonstration, as well as a broader range of data on each process
from non-SITE activities. Therefore, technologies to destroy,
treat, detoxify, reduce mobility, volume or recycle hazardous waste
materials are being developed and demonstrated within the
Environmental Protection Agency (EPA).
Most conventional treatment processes, such as air stripping, steam
stripping, carbon adsorption, and biological treatment, are quite
effective in treating water contaminated with organics, but have
certain limitations. These limitations include transfer of
contaminants from one medium (water) to another (air or carbon)
when using stripping and adsorption. In addition biological
treatment processes generate sludge that may require further
treatment and disposal.
Most of these limitations can be eliminated by chemical oxidation
processes using ozone, hydrogen peroxide or some other conventional
oxidant. However, because of kinetic limitations chemical
oxidation by conventional oxidants has yet to become a competitive
treatment option. Several studies have shown that the kinetic
limitations could be overcome by using hydroxyl radicals to carry
out the oxidation reactions (1-3). The hydroxyl radicals are known
to be less selective in carrying out oxidation reactions and have
much higher rate constants compared to ozone, hydrogen peroxide, or
ultraviolet (UV) radiation.
Hydroxyl radicals are generated by the combined use of (1) UV
radiation and hydrogen peroxide, (2) UV radiation and ozone, or (3)
ozone and hydrogen peroxide. These processes are commonly referred
to as "advanced oxidation processes" or, when UV radiation is used
to generate hydroxyl radicals, "UV/oxidation technologies."
This paper briefly describes (l) the chemistry of UV/oxidation
technologies and factors affecting these technologies and (2) the
results from pilot-field scale operations of two UV/oxidation
systems. These technologies differ in design and application, and
therefore present unique features that demonstrate the efficiencies
of both processes under different conditions. It is noted that
Ultrox International has already been demonstrated under the EPA's
SITE program in 1989 and has successfully been in the market for
several years. Likewise, the Peroxidation Systems technology has
been available in the market for a number of years, the technology,
however, will be in the SITE demonstration program in 1992 at the
Lawrence Livermore National Laboratory, in Livermore, California.
407
-------
UV/OXIDATION PROCESS CHEMISTRY
The generation of hydroxyl radicals is the key principle of
UV/oxidation technologies through UV photolysis of hydrogen
peroxide and/or ozone. When UV radiation is used to photolyze
hydrogen peroxide or ozone, the UV radiation may also photalyze
some organic contaminants. A summary of the chemistry of
UV/oxidation technologies is given below. More information of the
chemical reactions that occur during these applications is
available elsewhere (4).
UV Photolysis of Hydrogen Peroxide
Generation of hydroxyl radicals by UV photolysis of hydrogen
peroxide may be described by the following equation:
H202 + hv - 2 OH*
Most commercial applications are using low-pressure mercury vapor
UV lamps to produce UV radiation. The maximum absorbance of uv
radiation b/ hydrogen peroxide occurs at about 220 nanometers (run).
However, the dominant emission wavelength of low-pressure mercury
vapor UV lamps is at about 254 nm. Also, the molar extinction
coefficient of hydrogen peroxide at 254 nm is low, 19.6 liters per
mole-centimeter (M* cm" ) . Because of the low molar extinction
coefficient, a high concentration of hydrogen peroxide is needed in
the medium to generate sufficient hydroxyl radicals.
UV Photolysis of Ozone
UV photolysis of ozone in water yields hydrogen peroxide, which in
turn reacts with UV radiation or ozone to form hydroxyl radicals as
shown below.
03 + hv + H20 - H202 + 02
H202 + hv - 2OH"
2 03 + H202 - 2 OH' + 3 02
Because the molar extinction coefficient of ozone is 3,300 M*1 cm"'
at 254 nm, the UV photolysis of ozone is not expected to have the
same limitation as that of hydrogen peroxide when low-pressure
mercury vapor UV lamps are used.
FACTORS INFLUENCING PERFORMANCE
Factors influencing the performance of a UV/oxidation technology
can be grouped into three categories: (1) waste characteristics,
(2) operating parameters, and (3) maintenance requirements.
Following is a brief discussion of these factors.
408
-------
Waste Characteristics
The type of contaminants to be treated influence, the removal
efficiencies of the UV/oxidation processes. For example, organics
with double bonds, such as trichloroethene (TCE), tetrachloroethene
(PCE), and vinyl chloride, and aromatic compounds, such as phenol,
toluene, benzene, and xylene, are easily removed because they are
readily oxidised. In systems that use ozone, organics without
double bonds and with high Henry's law constants, such as 1,1-
dichloroethane (1,1-DCA) and 1,1,1-trichloroethane (1,1,1-TCA), are
also removed. However, because they are difficult to oxidize their
removal is primarily due to stripping. Organics without double
bonds and with low Henry's law constants, such as diethylamine and
1,4-dioxane, would be difficult to remove because they are not
easily oxidized or stripped.
UV/oxidation technologies are intended for the destruction of
organic contaminants, other species that consume oxidants are
considered an additional load for the system. These species are
called scavengers and include anions such as bicarbonate,
carbonate, sulfide, nitrite, bromide, and cyanide. Metals present
in their reduced states, such as trivalent chromium, ferrous iron,
inanganous ion, and several others, are likely to be oxidized.
These reduced metals, in addition to acting as scavengers, cause
additional concerns. For example, trivalent chromium is oxidized
to more toxic hexavalent chromium, and ferrous iron and manganous
ions are oxidized to less soluble forms, which precipitate in the
reactor and can cause UV lamp scaling and suspended solids
formation. Nontarget organics (for example, humic compounds) could
also act as scavengers. Other parameters such as suspended solids
and oil and grease would reduce UV transmission, thereby decreasing
the treatment efficiency. For these reasons, pretreatment may be
required for proper functioning of UV/oxidation units depending on
the waste characteristics.
Operating Parameters
Operating parameters are those parameters that are varied during
the treatment process to achieve the desired treatment
efficiencies. Such parameters include hydraulic retention time,
ozone dose, hydrogen peroxide dose, UV lamp intensity, influent pH
level, and gas-to-liquid flow rate ratio.
In general, increasing the hydraulic retention time will increase
treatment efficiency up to a certain point. At this point, the
system tends to proceed toward equilibrium, and increasing the
hydraulic retention time no longer increases treatment efficiency.
The higher the dose of oxidants, the better the treatment rate.
However, the molar ratio of the oxidant doses must be considered.
For example, when treating water containing TCE and PCE, maximum
removals were observed when the molar ratio of ozone dose to
hydrogen peroxide dose was equal to two, and the removals were
significantly ]ass when the ratio was not equal to two. In this
409
-------
case, the expected stiochiometry for pure water agreed with the
molar ratio at which optimum removal was observed; however, several
factors may influence the molar ratio (1). These factors are
summarized as follows:
Hydrogen peroxide can act as a free radical scavenger itself,
thereby decreasing the hydroxyl radical concentration if it is
present in excess.
Ozone can react directly with hydroxyl radicals, consuming
both ozone and hydroxyl radicals.
Ozone and hydroxyl radicals may be consumed by scavengers
present in the water being treated.
Therefore, the optimum proportion of the oxidants for maximum
removals cannot be predetermined. Instead, the proportion needs to
be determined for the waste under consideration using pilot- or
bench-scale treatability tests.
In addition to photoly2ing hydrogen peroxide and ozone to generate
hydroxyl radicals, the UV radiation may also photolyze some organic
contaminants, such as PCE, aromatic halides, and pesticides,
increasing the contaminant removal.
If water has bicarbonate and carbonate alkalinity at a level
greater that 400 milligrams per liter (mg/L) as calcium carbonate,
lowering the pH to a range of 4 to 6 should improve the treatment
efficiency. Low pH decreases the concentration of these scavengers
by shifting the equilibrium toward carbonic acid. If the carbonate
and bicarbonate alkalinity is low, then a high pH should improve
the treatment efficiency. High pH favors hydroxyl radical
formation because of the reaction between ozone and the hydroxyl
ion.
The ozone gas flow rate can also influence treatment rate. In
practice, once the ozone dose is selected, several combinations of
ozone gas phase concentration and ozone gas flow rate can be
applied. According to Venosa and Opatken (5) , the ratio of gas
flow rate to liquid flow rate will dictate the hydraulic
characteristics of the reactor, as shown in Figure l. This figure
shows that, at low gas-to-liquid flow rate ratios, the mixing
regime in a reactor is close to that of a plug flow reactor (shown
as Curve A); whereas at high ratios, the reactor mixing regime is
close to that of a mixed reactor (shown as Curve C) . For reactions
with a positive reaction order, plug flow mixing characteristics
offer higher treatment rate than mixed reactor mixing
characteristics (6) . Since most reactions have a positive reaction
order, low gas-to-liquid flow rate ratios should be considered. In
addition to increasing the treatment rate, low gas-to-liquid flow
rate ratios reduce stripping of volatile organics.
410
-------
Maintenance Requirements
Regular maintenance by trained personnel is essential for a
successful treatment operation. The following components require
maintenance: (1) ozonation system, (2) UV lamp assembly, (3) ozone
decomposer unit, and (4) miscellaneous components. A brief summary
of the maintenance requirements for these components is presented
in Table 1.
PILOT-FIELD STUDY ONE
The Ultrox technology demonstration occurred in February, 1989 at
the i^rentz Barrel and Drum (LB&D) sit-> in San Jose, California,
through an agreement between EPA's Region TX, Ultrox International,
and the EPA's SITE program. The LB&D site was used primarily for
drum recycling operations from about 1947 to 1987. The drums
contained residual aqueous wastes, organic solvents, acids, metal
oxides, and oils. The preliminary site assessment report for the
LB&D site showed that the groundwater and soil were contaminated
with organics and metals.
The upper aquifer at the LB&D site was selected as the waste stream
for evaluation of the UV/oxidation technology. Samples from the
shallow aquifer were collected in December 1988 which indicated
that volatile organic compounds (VOCs) were present in the
groundwater. VOCs detected at high levels included TCE (280 to
920ji/L), vinyl chloride (51 to 146 mg/L), and 1,2-trans-
dichloroethylene (42 to 68 ng/L). The pH and alkalinity of the
groundwater were about 7.2 and 600 rag/L as CaC03 , respectively.
These measurements indicated that the bicarbonate ion (HC03) , which
acts as an oxidant scavenger, was present at high levels. Other
oxidant scavengers, such a bromide, cyanide, and sulfide were not
detected. Iron and manganese were present at low levels (less than
1 ing/L) . Detailed information is available (7) to describe the
parameters and design of the SITE demonstration.
Ultrox System
The Ultrox UV/oxidation treatment system uses UV radiation, ozone,
and hydrogen peroxide to oxidize organics in water. The major
components of the Ultrox system are the UV/oxidation reactor
module, air compressor/ozone generator module, hydrogen peroxide
feed system and the catalytic ozone decomposition (Decompozon) unit
(Figure 2) .
The UV/oxidation reactor used has a volume of 150 gallons and is 3
feet long by 1.5 feet wide by 5 feet high. The reactor is divided
by five vertical baffels into six chambers and contains 24 65-watt
UV lamps in quartz sheaths. The UV lamps are installed vertically
and are evenly distributed throughout the reactor (four lamps per
chamber). Each chamber also has one stainless steel sparger that
extends along the width of the reactor. These spargers uniformly
diffuse ozone gas from the base of the reactor into the water.
Hydrogen peroxide is added to the influent line to the reactor. An
411
-------
in-line static mixer is used to disperse the hydrogen peroxide into
the contaminated water in the influent feed line.
The Decompozon unit (Model 3014 FF) uses a nickel-based proprietary
catalyst to convert reactor off-gas ozone to oxygen. The
Decompozon unit can accommodate flows of up to 10 standard cubic
feet per minute and can destroy ozone concentrations in ranges of
1 to 20,000 ppm (by weight) to less than 0.1 ppm.
Testing Approach
The study was designed to evaluate the Ultrox system by adjusting
the levels of five operating parameters: (1) influent pH, (2)
hydraulic retention time, (3) ozone dose, (4) hydrogen peroxide
dose, and (5) UV radiation intensity. Eleven test runs were
performed to evaluate the Ultrox system under various operating
conditions. After these Runs, two additional runs were performed
to verify that the system's performance was reproducible. The
verification Runs (Runs 12 & 13) were at the best operating
conditions which were determined to be those of Run 9, pH 7.2;
hydraulic retention time 40 minutes; ozone dose 110 mg/L? hydrogen
peroxide dose 13 mg/L; and all UV lamps operating.
During the study, a preliminary estimate of Ultrox system's
performance in each run was obtained from the effluent
concentrations of three indicator VOCs. The vocs selected for this
purpose were TCE, 1,1-DCA, and 1,1,1-TCA. TCE was selected because
it is a major volatile contaminant at the site, and 1,1-DCA and
1,1,1-TCA were selected because they are relatively difficult to
oxidize. At the end of the study, data from all samples was used
to evaluate the system's effectiveness.
Results and Conclusions
Results of the Ultrox system are summarized to present the overall
effectiveness of the UV/oxidation technology in removing VOCs from
the groundwater at the LBSD site. The removal efficiencies and
concentration profiles of all VOCs are not presented in this paper,
but additional information can be obtained from the Technical
Evaluation Report and the Application Analysis Report published by
EPA (8).
Summary of Results for VOCs
Based on overnight analysis performed during the demonstration
(when two of the six replicates determined the average effluent
concentrations for each indicator VOC), Runs 8 and 9 showed that
the effluent met the discharge standard at either set. of
conditions. Since a lower hydrogen peroxide dose was used in Run
9, compared to Run 8, Run 9 was chosen as the preferred operating
run. However, based on a complete analysis of the four remaining
replicates for Run 9 performed after the demonstration, the mean
concentration of 1,1-DCA was found to be slightly higher J'
jig/L, the discharge standard for the VOC. Sinc^ ~
412
-------
preferred operating conditions during the demonstration, the
verification runs (12 and 13) were also performed at those
conditions.
Figure 3 shows that the total VOC removals were about 90 percent,
while removal efficiencies for TCE were about 98 percent and those
for 1,1-DCA and 1,1,1-TCA were about 60 and 85 percent,
respectively. Higher removal efficiencies for TCE than for 1,1-DCA
and 1,1,1-TCA support the rationale used in selecting the indicator
VOCs.
Figure 4 compares the 95 percent upper confidence limits (UCLs) for
the effluent VOCs with the National Pollutant Discharge Elimination
System (NPDES) limits. The UCLs were calculated using the one-
tailed Student's t-test. The effluent met the discharge limits for
all regulated VOCs at the 95 percent confidence level in Runs 12
and 13; in Run 9, the mean concentrations for 1,1-DCA and 1,2-DCA
exceeded the discharge limits.
The gas chromatography (GC) and GC/mass spectrometry analyses
performed for VCCs, semivolatile organics, polychlorinated
biphenyls, and pesticides did not indicate the formation of new
compounds in the treated water. Because VOCs made up less than 2
percent of the total organic carbon, the general claim that
UV/oxidation technologies convert VOCs to carbon dioxide and water
could not be verified.
Because the Ultrox system treated the groundwater by bubbling ozone
gas through it, some VOC removal could be attributed to stripping
in addition to oxidation. To determine the extent of stripping
within the treatment system, VOC samples were collected from the
reactor off-gas and emission rates for four VOCs were compared to
the VOC removal rates from groundwater. The results are summarized
in Table 2. Because the extent of stripping for any particular VOC
is expected to be proportional to the ratio of air flow rate to
water flow rate, this ratio is also presented in the table. The
ratio for Runs 1 to 5 is approximately 2; for Run 6 and Runs 8 to
13, it is about 4.5? and for Run 7, it is 1. If stripping
contributed to the total removal of the four VOCs, the extent of
stripping would be expected to be least in Run-7, and most in Runs
6 and 8 to 13. The data presented in the table follow this trend
for three of the four VOCs (except for the vinyl chloride in Runs
6,7, and 9). A quantitative correlation of the extent of stripping
cannot be made, because the operating conditions were different in
each run. For example, at a given air to water flow ratio, when
oxidant doses are varied, the extent of oxidation also varies.
Therefore, the extent of stripping will be indirectly affected.
Table 2 presents Henry's law constants for the four VOCs (9). 3y
comparing these constants for the VOCs, their volatility is
expected to increase from left to right as shown below:
1,1-DCA -• TCE - 1,1,1-TCA - vinyl chloride
413
-------
However, significant removal fractions for 1,1,1-TCA and 1,1-DCA
were observed to be due to stripping. Conversely, the extent of
stripping was low for vinyl chloride and TCE. This is because it
is easier to oxidize vinyl chloride and TCE than 1,1-DCA and 1,1,1-
TCA because of the double bonds between the carbon atoms in TCE and
vinyl chloride. Therefore, UV/oxidation processes using ozone,
stripping is a significant removal pathway for compounds that are
difficult to oxidize.
Performance of the Decompozon Unit
The ozone concentrations in the influent to and the effluent from
the Decompozon unit were analyzed in each run. Ozone destruction
efficiencies greater than 99.99 percent were achieved in Runs 1 to
10. The effluent ozone concentrations were low (less than 0.1 ppm)
for Runs 1 to 8, approximately 1 ppm in Runs 9 and 10, and greater
that 10 ppm in Runs 11, 12, and 13. The high ozone levels (greater
than 1 ppm) in the effluent are attributed to the malfunctioning
heater in the Decompozon unit. The temperature in the Decompozon
unit should have been 140* F for the unit to properly function,
whereas the temperature for Runs 11 to 13 was only about 80' F.
Although the primary function of the Decompozon unit is to remove
ozone, significant VOC removal also occurred when the unit
functioned as designed (Runs 1 to 8) . For example, the Decompozon
unit removed TCE, 1,1-DCA, 1,1,1-TCA and vinyl chloride (present in
the gas phase in the reactor at levels of approximately 0.1 to 0.5
ppm) to below detection levels.
PILOT-FIELD SCALE STUDY TWO
This pilot-field scale study was performed at the Old O-Field site
located in the Edgewood Area of the Aberdeen Proving Ground,
Maryland by the Peroxidation Systems, Inc. (PSI). The site was
used for disposal of the chemical-warfare agents, munitions,
contaminated equipment, and various other hazardous materials
during the 1940s and early 1950s. The disposal of these hazardous
materials contaminated several media at the Old O-Field site,
including groundwater, surface water, and sediments.
During the pilot-field study, groundwater samples were collected
which indicated that several organics, including VOCs, organosulfur
compounds, and explosives, were present at the sitt in
concentrations of 10 nq/L to 500 /ig/L. Iron and manganese were
present at levels of 120 mg/L and 2.5 mg/L, respectively (10).
Trace levels of arsenic were also observed in some locations which
were sampled before the pilot-field scale study.
Treatability studies were performed in April and May 1991 as part
of the remediation process. A total of 37,000 gallons of
groundwater from three wells was used to perform the treatability
studies. The contaminated groundwater was pumped to two holding
tanks and then treated by a metals precipitation system. The
414
-------
metals precipitation was performed at pH 11, primarily to remove
arsenic observed in samples collected before the pilot study began;
however, the pilot study samples showed no arsenic contamination.
Iron and manganese were removed to levels of 0.2- mg/L and 0.02
mg/L, respectively. The pH of the metals precipitation system
effluent was adjusted to 7 and then treated using two parallel
systems: (1) an air stripping system followed by carbon adsorption
for both liquid and vapor phase effluents from the air stripper and
(2) a UV/oxidation system. The air stripping/carbon adsorption
system was developed by Carbonair and the UV/oxidation system was
developed by Peroxidation Systems,Inc.
Peroxidation Systems Process
Figure 5 shows a schematic of the PS I UV/oxidation system. The
system used at the Old O-Field site consisted of two parallel
cartridge filters rated at 10 micrometers (/im) followed by the
UV/oxidation reactor. The UV/oxidation reactor had a volume of 80
gallons and was divided by three horizontal baffles into four
chambers. Each chamber contained one high intensity, broad band,
mercury-arc-type 15-kw UV lamp. A splitter was used so that
hydrogen peroxide could be added at multiple points, such as the
influent line and at several locations inside the reactor, making
hydrogen peroxide available for hydroxyl radical formation
throughout the reactor. The effluent from the reactor was passed
through an optional manganese-greensand filter to remove any
residual hydrogen peroxide, followed by a pH adjustment to raise
the pH to an acceptable level.
Testing Approach
Four tests were conducted at a flow rate of 15 gpm (hydraulic
retention time of about 5.3 minutes). In Tests 1,2, and 3,
hydrogen peroxide doses were 45 mg/L, 90 mg/L and 180 mg/L,
respectively, with the splitter in operation; and in Test 4,
hydrogen peroxide dose was 45 mg/L with the splitter not in
operation. When the splitter was used, the total hydrogen peroxide
dose was split into three equal parts, which were added at (1) the
influent line to the reactor, (2) the effluent line from the first
chamber, and (3) the affluent line from the second chamber. In
Test 4, when the splitter was not used, all hydrogen peroxide was
added at the influent line to the reactor. Treated and untreated
water samples were collected for (1) chemical analyses to estimate
removal efficiencies and compare them with federal maximum
contaminant levels (MCLs) for drinking water, and (2) bioassay to
evaluate whether the water was acutely toxic to fathead minnows,
daphnia magna, sheepshead minnows, and mysid shrimp.
415
-------
Results and Conclusions
A discussion of the optimum operating conditions of Test 3 is
presented here, a more detailed description on the PSI system
performance is available in an unpublished report by the U.S. Army
Corps of Engineers Toxic and Hazardous Materials Agency (10).
Also, Table 3 summarizes the influent and effluent contaminant
levels and the removal efficiencies of several contaminants. These
results show that for most compounds the effluent levels were below
detection levels, and the removal efficiencies for these compounds
were greater than 82 to 99 percent. The effluent levels of
chloroform and 1, 3,5-trinitrobenzene were 1.2 Mg/L and .53 Mg/L,
respectively. The removal efficiencies for these compounds were in
the range of 96 to 97 percent.
The treated effluent met the federal MCLs for all compounds. The
influent to and the effluent from the PSI system passed the
bioassay tests. The pH decreased by about one unit, indicating
that some of the oxidation byproducts were acidic. Although the
manganese-greensand filter was effective in removing residual
hydrogen peroxide, it increased the manganese levels in treated
water from about 0.02 mg/L to 1.4 mg/L, which is above the National
Secondary Drinking Water Standard (50 Mg/L)• Therefore, the use of
manganese-greensand filter is not recommended. Instead, other
methods should be considered to neutralize any residual hydrogen
peroxide in the treated water samples (for example, addition of
ascorbic acid, thiosulfate, or catalase-D) . If the residual
oxidant level is greater than 1 mg/L and is not neutralized, it
would continue to react with the contaminants in the sample bottles
until analysis could be performed. This continued reaction may
introduce a bias in the treatment system evaluation.
CONCLUSIONS ON THE UV/OXIDATION TECHNOLOGIES
The UV/oxidation technologies present an efficient and competitive
alternative treatment, especially for the removal of organics
present in water at low concentrations (less than about 100 mg/L).
For higher concentration levels of contaminants, these technologies
may prove cost-effective when used in combination with biological
or adsorption processes. The UV/oxidation technologies are often
preferred over adsorption or biological processes, because in the
UV/oxidation technologies (1) contaminants are destroyed rather
than transferred to some other medium and (2) no residuals
requiring further handling, such as sludge or spent carbon, are
generated. Due to the contaminants present it may be necessary to
implement pretreatment processes to minimize shut down or delays.
Operation and maintenance data are currently being documented, and
this information is instrumental in moving the technology to more
efficient design and application techniques.
416
-------
A greater understanding of the actual chemistry needs to be
researched further, especially to understand the byproducts of the
UV/oxidation of organics. Several technology developers claim that
byproducts are carbon dioxide, water, and halides, but little
published data are available to support these claims when
UV/oxidation is used to treat the contaminants present in
groundwater. Research is also being established in the area of
using the UV/oxidation technologies for the destruction of organics
in the air phase. Promising results are being generated for this
application.
REFERENCES
1.	Aieta, E.M., K.M. Reagan, J.S. Lang, L. McReynolds, J.W. Kang,
and W.H. Glaze, 1988. Advanced Oxidation Processes for
Treating Groundwater contaminated with TCE and PCE: Pilot-
Scale Evaluation. Journal of the American Water Works
Association. 5:64.
2.	Glaze, W.H., and J.W. Kang, 1988. Advanced Oxidation Processes
for Treating Groundwater Contaminated with TCE and PCE:	
Laboratory Studies. Journal of the American Waterworks
Association, 5:57.
3.	Weir, B.A., D.W. Sundstrom, and H.E. Klei, 1987. Destruction
of Benzene bv Ultraviolet Light Catalyzed Oxidation with
Hydrogen Peroxide. Hazardous Waste and Hazardous Materials,
4:2:165.
4.	Glaze, W.H., J.W. Kang, and D.H. Chapin, 1987. The Chemistry
of water Treatment Processes Involving Ozone. Hydrogen
Peroxide. and Ultraviolet Radiation. Ozone Science and
Engineering, 9:335.
5.	Venosa, A., and E.J. Opatken, 1979. Ozone Disinfection-State
of the Art. In: Proceedings. Preconference Workshop on
Wastewater Disinfection. Atlanta, GA, Water Pollution Control
Federation.
6.	Levenspiel, 0., 1972. Chemical Reaction Engineering. 2nd
Edition. John Wilev & Sons. Inc.. New York. NY.
7.	PRC Environmental Management, Inc., 1989 (Unpublished Report).
Demonstration Plan for the Ultrox International Ultraviolet
Radiation/Oxidation Process. U.S. EPA SITE program.
8.	EPA/540/A5-89/012,1990. Ultrox International Ultraviolet
Radiation/Oxidation Technology: Applications Analysis Report.
U.S. EPA, Office of Research and Development, Washington, DC.
9.	EPA 540/1-86/060, 1986. Superfund Public Health Evaluation
Manual. U.S. EPA, Office of Emergency and Remedial Response,
Washington, DC.
417
-------
TECHNICAL REPORT DATA
(Hum rt*d Iniuucitoni m the rtvtnt be/ore completing)
<*«#• w • .
1.MIPORTNO. 2.
EPA/6UG/A-94/00S
3 RECIPIENT S At.Lt
4. TITLE and subtitle
"Advanced Oxidation Technologies for the Treatment
of Contaminated Groundwater"

6. PERFORMING ORGANIZATION COOC
7. AUTHORIS)
Norma M. Lewis and Kirankumar Topudurti
8 PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
USEPA, Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
PRC Environmental Management, Inc.
Chicago, Illinois
10 PROGRAM ELEMENT NO.
11 CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND AOORESS
Risk Reduction Engineering Laboratory--Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERlOO COVEREO
Rnnlr renter
14. SPONSORING AGENCY COOE
EPA/600/14
is. supplementary notes Book Chapter in "Chemical Oxidation - Technologies for the
Norma M. Lewis, 513-569-7665 Ninaties", Volume 2, p:406-417
ii. AsSTttAdf
This paper presents information on two pilot-field applications of advanced
oxidation technologies for contaminated groundwater with organics. The two UV/
oxidation technologies'were developed by Ultrox International of Santa Ana,
California and Peroxidation Systems, Inc. of Tucson, Arizona. The Ultrox
technology was demonstrated in 1989 with the U.S. Environmental Protection
Agency's Superfund Innovative Technology Evaluation (SITE) program at the
Lorentz Barrel and Drum (LB&D) site in San Jose, California. Peroxidation
Systems technology was applied at the Old 0-Field site located within the
Aberdeen Proving Ground, irv Maryland.
The information presented includes a description of the technologies,
factors affecting the technologies, and results from the two pilot-scale
studies of the UV/oxidation treatment system applications.
(7. KEY WORDS AND OOCUM6NT ANALYSIS
a. descriptors
b.IDENTIFIERS/ OPEN ENDED TERMS
c. COSati FieldiGioup
Groundwater
Superfund
Hazardous Waste


ia. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS {This Rtporl)
UNCLASSIFIED
21 NO. OF pages
20 SECURITY Class iTIiii pagf!
UNCLASSIFIED
22. PRICE
EPA Fofdi 2220-1 (R«». 4-77) pwkyioui idition i obiolete
-------