&EPA
United States
Environmental Protection
Agency
EPA/600/R-19/078 | June 2019
www.epa.gov/homeland-security-research
Design, Testing, and Deployment
of a Mobile Emergency Water
T reatment System
Office of Research and Development
Homeland Security Research Program
-------
This page left intentionally blank
-------
EPA/600/R-19/078
June 2019
Design, Testing, and Deployment of a
Mobile Emergency Water Treatment
System
by
James A. Goodrich and John S. Hall
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Mark Hogg and Kurtis T. Daniels
Water Step
Louisville, KY 40208
Gregory C. Meiners and Suzan M. Witt
Aptim Federal Services, LLC
Cincinnati, OH 45204
Cooperative Research and Development Agreement
CRADA #865-15, Contract EP-C-12-014
U.S. Environmental Protection Agency
Homeland Security Research Program
Cincinnati, Ohio 45268
-------
Disclaimer
The U.S. Environmental Protection Agency (EPA) through its Office of Research and
Development funded and managed the research described herein under Cooperative Research
and Development Agreement # 865-15 with WaterStep and contract EP-C-12-014 with Aptim.
It has been subjected to the Agency's review and has been approved for publication. Note that
approval does not signify that the contents necessarily reflect the views of the Agency. Any
mention of trade names, products, or services does not imply an endorsement by the U.S.
Government or EPA. The EPA does not endorse any commercial products, services, or
enterprises.
The contractor role did not include establishing Agency policy.
11
-------
Contents
DISCLAIMER II
LIST OF FIGURES IV
LIST OF TABLES IV
ABBREVIATIONS V
ACKNOWLEDGEMENTS VI
EXECUTIVE SUMMARY VII
1.0 INTRODUCTION 9
1.1 BACKGROUND 9
1.2 Project Objectives 10
1.3 Water Quantity and Quality Scenarios 11
1.3.1 Other Design Considerations 12
2.0 EVALUATION OF THE BASELINE WATER TREATMENT SYSTEM 12
2.1 2nd Generation WOW Cart Description 12
2.2 Preliminary Testing at the Water Security Test Bed 15
2.2.1 On-site Chlorinator Lagoon Water Testing 17
2.2.2 Bladder Tank Chlorinator Testing 20
2.2.2.1 Analysis of Test Results 21
3.0 PUERTO RICO DEPLOYMENT 24
3.1 Background 24
3.2 Achievements 25
3.3 Strong Communication 26
3.4 Success Highlights 26
4.0 FULL-SCALE DEPLOYMENT 28
4.1 DESCRIPTION OF THE FINAL VERSION OF THE WOW CART 28
4.2 Secondary Wastewater Challenge 31
4.2.1 Analysis of Test Results 34
4.3 WSTB Microbial and Diesel Fuel Challenge 37
4.3.1 Analysis of Test Results 40
5.0 CONCLUSIONS 42
6.0 REFERENCES 42
7.0 APPENDICES -USEPA T&E FACILITY CONTRACT TECHNICAL STANDARD OPERATING PROCEDURE
44
ill
-------
List of Figures
Figure 1. WOW Cart Proposed Schematic 13
Figure 2. 2nd generation WOW cart prototype (front) 14
Figure 3. 2nd generation WOW cart prototype with media filters installed (back) 15
Figure 4. Schematic overview of Water Security Test Bed 16
Figure 5. Aerial view of the Water Security Test Bed 16
Figure 6. Water Security Test Bed lagoon 17
Figure 7. WaterStep chlorine generator components 18
Figure 8. WOW cart setup at the lagoon 19
Figure 9. WaterStep chlorinator system with bladder tanks (dark blue) 21
Figure 10. Free chlorine concentration (orange) and Bacillus globigii spores (blue line) density over time
in the WaterStep bladder tank 22
Figure 11. The log reduction in spores during the WaterStep experiment plotted against the Ct value
(disinfectant concentration multiplied by time) 23
Figure 12. Municipalities that received a disaster kit within weeks of Hurricane Maria (red) 26
Figure 13. Disaster kit installation 27
Figure 14. 3rd generation of the WOW cart following Puerto Rico deployment 28
Figure 15. SingleHole Recirculation Manifold 30
Figure 16. Electrical outlets and phone charging station 30
Figure 17. WOW cart secondary wastewater challenge set-up with empty 1,250-gallon bladder tank at
T&E Facility 31
Figure 18. Secondary wastewater effluent holding tank and WOW cart at T&E Facility 32
Figure 19. Rotameter showing flow through the WOW cart 33
Figure 20. Bladder tank (660 gallons -1/2 full) 34
Figure 21. WOW cart inlet (left) and outlet (right) samples 35
Figure 22. WOW cart delivered on-site at the Water Security Test Bed 38
Figure 23. WOW cart deployed on-site 39
List of Tables
Table 1. WaterStep Technology-specific Consideration and Observations* 24
Table 2. Chlorine Concentrations at the WOW Cart Outlet (single pass) 35
Table 3. WOW Cart Summary of Escherichia coli and Total Coliform Disinfection Results.... 36
Table 4. Microbial Results from the Lagoon 40
Table 5. Diesel Fuel Removal Rates 41
iv
-------
Abbreviations
BG Bacillus globigii
BSL Biosafety Level
BWS bulk water sample
°C degrees Celsius
cfu colony forming units
CRADA Cooperative Research and Development Agreement
E. coli Escherichia coli
EPA U.S. Environmental Protection Agency
FTTA Federal Technology Transfer Act of 1986
ft foot
GAC granular activated carbon
gpm gallons per minute
hr hour
INL Idaho National Laboratory
L Liter
m meter
mg/L milligram per liter
ml milliliter
min minute
MPN most probable number
NPDES National Pollutant Discharge Elimination System
NTU Nephelometric Turbidity Units
PEX cross-linked polyethylene
MOP Miscellaneous Operating Procedure
T&E Test and Evaluation
TPH total petroleum hydrocarbons
WOW Water On Wheels
WSTB Water Security Test Bed
v
-------
Acknowledgements
Contributions from the following individuals to the field work described in this report are
acknowledged: Gary Lubbers, Dave Elstun, and John Brannon of Aptim Federal Services, LLC.;
Bill Parker, Dr. Joe Jacobi, Larry Friebert, Dr. Cindy Figueroa, Lynn Smith, and Dustin Alton
Strupp of EDGE Technologies; and Stephen Reese of the Idaho National Laboratory.
vi
-------
Executive Summary
The U.S. Environmental Protection Agency's (EPA) Homeland Security Research Program
partnered with Edge Outreach Technologies, LLC (DBA WaterStep) to develop and deploy a
mobile emergency water treatment system utilizing the Federal Technology Transfer Act
(FTTA) of 1986, that enabled the Government to enter into a Cooperative Research and
Development Agreement (CRADA) and to negotiate licenses for patented inventions. The
purpose of this study was to design, build, evaluate, and deploy a mobile emergency water
treatment system capable of treating a wide variety of contaminated water following a natural or
man-made disaster. Most emergency water treatment systems are very large and expensive
tractor-trailer mounted systems. They can be complicated to operate and maintain (very high
pressures and concentrated wastes) given their use of reverse osmosis water treatment
technology. Water may be contaminated with chemical, biological or radionuclide contaminants.
Therefore, an emergency water treatment system must be designed and built so the treatment
train can be configured on-site to treat a broad spectrum of contaminants without utilizing other
unnecessary and costly unit processes and without producing large amounts of contaminated
wastes. Bottled water is typically the first responder's choice when responding to an incident.
However, excessive dependence on bottled water creates a large solid waste disposal problem
and, often times, large vehicles transporting bottled water are unable to get to affected locations
because of road debris and damage. Bottled water in large or extended recovery situations
cannot be used for cooking, bathing and sanitation purposes. However, it could be used in
conjunction with an inexpensive and versatile mobile emergency water treatment system
providing water for other non-drinking water applications. Not all the water being treated needs
to be drinking water quality. In some cases, contaminated stormwater or wash water from
building decontamination activities need only to be treated to levels safe for disposal to the
wastewater treatment plants or to the environment. For longer-term mitigation efforts, large
volumes of contaminated wash water can be produced and needs to be safely transported and
disposed of in a hazardous waste facility. Mobile treatment of the contaminated water can
significantly reduce the volume of water to be transported and reduce the liability and cost of
transporting and disposing of a hazardous waste.
The mobile treatment technology system in this study is referred to as the Water-On-Wheels
(WOW) Cart. The mobile system originally consisted of a pre-filter, an on-site chlorine
generator, and a pump attached to a dolly or frame with wheels. The frame also provided space
to store accessory equipment and to transport two empty 1,250-gallon bladder tanks used to store
treated water. Building upon the original device, this study designed, built, challenged and
deployed an inexpensive mobile water treatment system with expanded water treatment and
power supply capabilities. The system integrated the pre-filtration step with additional media
filtration (e.g., granular activated carbon) and on-site chlorine gas generation with options for
UV LED and/or ultrafiltration membranes, which were all stored and transported on a wheeled,
powder-coated steel frame This study also added multiple power supply options that can be
operated from the electrical grid (1 lOv AC), a duel-fuel generator, and peripherals with a 12v
DC deep cell marine battery (with solar recharge). There are also additional electrical outlets
and USB ports for phones, computers, etc. The WOW Cart can now also produce liquid bleach
for sanitation purposes.
vii
-------
A prototype mobile system (Version 2.0) was challenged with Bacillus globigii spores (a non-
pathogenic surrogate for anthrax spores) from a dirty lagoon at the EPA Water Security Test Bed
located near Idaho Falls, Idaho. The mobile system was easily deployed and able to produce a
large amount of chlorine, but it could not overcome the large chlorine demand from the lagoon.
Thus, it was determined that the WOW Cart should be operated in batch mode utilizing a bladder
tank to overcome chlorine demand of the dirty water by providing controlled contact time. This
setup then demonstrated greater than 7 log reduction of the anthrax surrogate. Shortly after this
successful testing, Hurricane Maria slammed Puerto Rico. The non-profit organization
WaterStep was able to deploy over 100 disaster kits (pre-filter and chlorine generator) to
municipal governments and to other non-profit organizations providing access to safe drinking
water to approximately 225,000 people daily.
Learning from both the field challenge and Hurricane Maria experience, the final version of the
WOW Cart was fabricated. It was challenged with secondary wastewater at the EPA Test and
Evaluation Facility located in Cincinnati, Ohio and subsequently successfully tested again at the
Water Security Test Bed against lagoon water contaminated with diesel fuel and Escherichia
coli. The WOW Cart successfully removed 4 to 6 Logs of E. coli and Total Coliforms
respectively to non-detection levels from the contaminated lagoon simultaneously with diesel
fuel components. Diesel fuel components were removed to below detection levels as well, thus
making the water safe to drink. Given that in most cases, microbial water quality is of the
utmost importance given the shorter duration of use, disinfection by-products and long-term
health effects were not a focus of this research. Nor was an extensive evaluation of a number of
chemical contaminants, particulates, viruses, metals, or pathogens undertaken given their widely
known removal characteristics in the commercial and research literature, especially since GAC is
the likely first choice for most emergency responders dealing with an unknown quantity and type
of chemical contamination. In the event, the WOW Cart would be used for long-term
community drinking water supply, regulatory considerations for that particular community and
nation would need to be considered. The primary purpose of this report to document and
evaluate the ease of deployment, operation, and general efficacy of a mobile water treatment
system for emergency use, both for drinking and non-drinking water purposes. This report
describes the results of those evaluations and provides details on the WOW Cart design.
-------
1.0 Introduction
1.1 Background
The U.S. Environmental Protection Agency's (EPA) Homeland Security Research Program
partnered with Edge Outreach Technologies, LLC (DBA WaterStep) to develop and deploy a
mobile emergency water treatment system utilizing a Federal Technology Transfer Act (FTTA)
Cooperative Research and Development Agreement (CRADA). The CRADA was originally
signed in August 2015 with a subsequent modification in January 2017. The goal of this
CRADA was to provide potable water in areas without a safe traditional water supply and in
emergency response situations such as after a man-made or natural disaster. This mobile water
treatment system incorporated innovative on-site chlorine generation for disinfection, multiple
filtration steps, media adsorption, multiple alternative power supply options, and distribution
technologies. This mobile treatment technology system is referred to as the Water-On-Wheels
(WOW) Cart.
There are a variety of scenarios that can result in compromised or untreated water entering a
drinking water distribution system, wastewater, and/or stormwater collection systems, such as:
• Large or multiple pipe breaks
• Loss of power and pressure for days, weeks, months due to floods, hurricanes, tornadoes,
earthquakes
• Terrorist or disgruntled employees directly introducing contaminants into a system
In a drinking water distribution system where a boil water advisory has been issued, a mobile
emergency water treatment system can be quickly deployed. The system can provide an interim
potable water supply to critical institutions such as hospitals, nursing homes, and prisons where
populations cannot be easily relocated. The mobile system can even be stored on-site at such
institutions as part of their emergency preparedness plan. Under natural disaster scenarios, the
mobile emergency system can be an interim solution for days, weeks, or months. Developing
countries lacking a reliable water supply or intermittent power could also utilize such a low-cost,
easy to operate water treatment system as a permanent solution. For example, in Puerto Rico the
WOW Cart has become a permanent solution in some locations following Hurricane Maria.
Following a large-scale natural disaster, untreated wastewater and stormwater can be discharged
directly into the environment where they will mix with chemical contamination from road and
building surfaces. These waters and wastes require treatment to prevent excess contamination
from spreading further into the environment. The mobile emergency water treatment system can
provide localized mitigation at overflow points in the wastewater and stormwater infrastructure.
In some cases, contaminated stormwater or wash water from building decontamination activities
need only to be treated to levels safe for disposal to the wastewater treatment plants or to the
environment. For longer-term mitigation efforts, large volumes of contaminated wash water can
be produced and needs to be safely transported and disposed of in a hazardous waste facility.
Mobile treatment of the contaminated water can significantly reduce the volume of water to be
transported and reduce the liability and cost of transporting and disposing of a hazardous waste.
9
-------
In 2012, the downtown area of Louisville, Kentucky had six water main breaks in a short amount
of time, which put a burden on The Louisville Water Company and the city's Emergency
Management Agency. After those issues were resolved, WaterStep was contacted by Louisville's
Emergency Management Agency about a problem they faced during the event that affected the
downtown jail (2000 people) and juvenile detention center (600 people). The Emergency
Management Agency was only hours away from being forced to move the residents out of the
jail to a hotel because of the lack of water during the emergency. Obviously, moving the
population of the jail to another facility posed a large logistical issue. The Louisville Office of
Emergency Director wanted to discuss ideas WaterStep had as a result of its history working in
the developing world and in disaster relief. Over the years, WaterStep had designed simple,
affordable, efficient and sustainable equipment for people in developing countries to provide
their own safe water. The City of Louisville was provided three carts that consisted of a pre-
filter, on-site chlorine generator, and pump attached to a dolly or frame with wheels. These units
could serve approximately 2,000 people per day. The frame also provided space to store
accessory equipment and to transport two empty 1,250-gallon bladder tanks that could be used to
store treated water. Those carts began the concept of the current WOW cart.
1.2 Project Objectives
The objective of this CRADA was to design, develop, and deploy a turnkey robust water
treatment system, capable of being transported, set-up and operated for the treatment, storage,
and discharge of water contaminated by intentional acts, industrial accidents, natural disasters,
by alternative untreated drinking water sources, or when traditional water supplies are
unavailable. Any untreated runoff that enters the surrounding environment could spread the
contaminant outside the containment field, risking further public health and environmental
consequences. In situations where runoff water is contaminated from precipitation or by wash
water from cleaning contaminated roads, parking lots, or buildings, the untreated water is
typically collected and shipped offsite. For extremely large volumes of contaminated water, this
is very expensive. Onsite treatment of this water would reduce costs and waste volumes to be
shipped. The water treatment system is intended to address these emergencies and meet these
needs locally.
A robust system would most likely employ multiple treatment unit processes capable of treating
a broad suite, and broad concentrations, of contaminants ranging from volatile and non-volatile
organics, hydrophobic/hydrophilic (particulates) contaminants, pathogens, viruses, parasites, and
metals representative of untreated source water, wastewater, storm water, or contaminated wash
water. In addition to the mobile system being easy to set up and inexpensive, it must also be
easily configured (plug and play) to most effectively treat contaminants for any given
contamination event. The ideal system would also include real-time monitoring and
communication capabilities.
Special considerations to be evaluated were:
• Energy minimization and use of alternative renewable energy sources
• Packaging for rapid deployment, set-up and take-down
• Economies of scale for manufacture and operation
10
-------
• Real-time optimization for water quality vs. costs
• Easy set-up and operation for laymen
• Inline and batch operation
• Self-contained such that supplies, and materials needed to operate for an extended period
are included
1.3 Water Quantity and Quality Scenarios
The mobile system must be able to treat a variety of water treatment scenarios. Design
requirements were quite varied, depending on the quality of the untreated water and the ultimate
end use of the treated water. The mobile system must be able to acquire and use water from the
following sources:
• Open sources:
o Rivers
o Creeks
o Springs and Seeps
o Streams
o Lakes
o Rain water catchment
• Well water
• Contaminated wash water from wide-area decontamination events
• Municipal water that has been compromised
o Hydrants
o Distribution network pipes
• Tanker trucks
• Barges
The treatment goal will also be varied depending on the water's ultimate end use. The mobile
system must be able to treat water of sufficient quantity and quality for the following end uses:
~~~ Quantity
• Point of Use - Individuals (approximately 5 gallons per person per day)
• Small Batch - Families (25 gallons/day/family of five)
• Large Quantity - Community Size (Assume up to 50,000 gallons per day)
~~~ Quality
• Human Consumption
• Discharge to receiving wastewater treatment plant
• Discharge to stormwater drain or combined sewer system such that discharge criteria
is met
• Discharge to permitted National Pollutant Discharge Elimination System (NPDES)
outfalls or other non-permitted outfalls as the situation requires.
11
-------
• Discharge to receiving water bodies
1.3.1 Other Design Considerations
The ability to use multiple energy sources as well as ease of operation and maintenance are also
critical. The system must be able to be stored for long periods without sacrificing performance
and without requiring long start-up procedures.
Energy Consumption design considerations are:
• Worst-case scenario utilizing a hand pump and 12-volt DC battery operated system
• Best-case scenario is generator provided service or utility provided AC
• Need to design the system for the worst-case scenario (total blackout situation)
• For long-term disasters such as those following a hurricane, tornadoes, or tsunamis, the
mobile system should be capable of being operated off the electrical grid
Operation and Maintenance design considerations:
• Little to minimal assembly
• Little to minimal maintenance
• Portable
• Easily transportable to developing and rural areas
o Meet airline size and weight requirements
• Low purchase and operational costs amenable to small and/or rural communities lacking
technical, managerial, and financial resources
• No hazardous materials (i.e., adhere to airline restrictions)
• Can be easily modified to whatever water systems are currently being used locally
• Minimal electronics (harsh environmental conditions, no maintenance available locally)
• Ability to add components as needed in the field (i.e., each system can be customized
based on what is needed, such as adding different types of filters)
• Long-term operation (several months)
• Minimal, if any reliance on reagents, consumables, and calibration standards, which
might be inaccessible
• Allow for recirculation of water through the chlorination unit in case multiple passes are
needed to reach the desired level of disinfection.
2.0 Evaluation of the Baseline Water Treatment System
2.1 2Pd Generation WOW Cart Description
Version 1.0 proved effective for providing disinfection, but it became apparent that the system
could be improved to make it more user-friendly. Version 2.0 of the WOW Cart included the
WaterStep M-100 Chlorine Generator and pre-filters, plus pre-piped PVC manifolds and valves,
12
-------
a frame-mounted jet pump, media adsorptive filtration, alternative power supplies including a
12v DC deep cell marine battery, solar panel, two 500-gallon bladder tanks, quick connect
hoses, salt, and extra parts and supplies. Figure 1 conceptually describes the WOW Cart
treatment train. The solid arrow describes the full treatment train with the dotted arrow showing
alternative paths when less treatment is required.
MOBILE WATER TREATMENT SYSTEM
WATER SOURCE
CITY DISTRIBUTION SYSTEM
RIVER
SWIMMING POOL
WELL
STORAGE TANK
PRE-FILTRATION
(DISC, BAG,
CANISTER)
-
,x
r v
MEDIA FILTERS
CHLORNATOR
DISCHARGE TO ENVIRONMENT
OR POTABLE WATER
ADDITIONAL
TREATMENT
(UV, OZONE,
AND/OR
ULTRAFILTRATION.)
Full Treatment Train
> Alternate Paths
MOBILE -RAMEWORK
Figure 1. WOW Cart Proposed Schematic.
The built Version 2.0 of the WOW Cart (Figures 2 and 3) was assembled into a steel frame with
wheels to make deployment and use much easier.
13
-------
Steel Welded Frame
Chlorine Gas
Generator
Pre-Filters
Figure 2. 2nd generation WOW cart prototype (front).
14
-------
Additional Media
Filters
Diaphragm
Tank
Figure 3. 2nd generation WOW cart prototype with media filters installed (back).
2.2 Preliminary Testing at the Water Security Test Bed
Preliminary testing was conducted to determine the operability and performance of the new
WOW Cart. This experiment was designed to assess the ability of the portable disinfection unit
within this treatment train to treat a large volume of water containing Bacillus globigii spores, a
surrogate for anthrax contamination, as well evaluate the ease of operation and setup. This was
conducted at the National Homeland Security Research Center's Water Security Test Bed
(WSTB) located near Idaho Falls, Idaho at the Department of Energy's Idaho National
Laboratory (INL).
The WSTB consists primarily of an 8-inch (20 cm) diameter drinking water pipe oriented in the
shape of a small drinking water distribution system (US EPA. 2016b). The WSTB contains ports
for simulating water demand from service connections and a 15-foot (5 m) removable coupon
section designed to sample the pipe interior. Figure 4 schematically depicts the main features of
the WSTB.
15
-------
Drinking Water
from (
INL Pumphouse
Existing Fire Hydrant
Fire Hose
Start of WSTB j—
Injection Port -tXl—
Lagoon
—ixi
Parking
Area
IP1)
Not to Scale
End of
WSTB
15-ft Coupon Section
Drainage
Ditch
Bulk Water Sample Tap
(P2)
Legend
FM Flow Meter
PG Pressure Gauge
IP1 Instrument Panel 1 (Upstream, Cellular)
IP2 Instrument Panel 2 (Downstream, Radio)
00 Valve, Open
~4 Valve, Closed
[>4 Valve, Partly Open
^ Fire Hydrant
Flushing Hydrant
—] Blind Flange
jCj^j Pressure Reducing Valve
Check Valve/Backflow Preventer
—Service Connector (Closed)
Figure 4. Schematic overview of Water Security Test Bed.
Figure 5 shows the aerial view of the WSTB. The lower right corner shows the upstream and
system inlet; the upper left corner shows the lagoon.
Lagoon
Downstream Sensors
WSTB End
WSTB Start
Upstream Sensor and Injection
Figure 5. Aerial view of the Water Security Test Bed.
16
-------
The water from the WSTB system is discharged to a lagoon (Figure 6) which has a water storage
capacity of 28,000 gallons (105,980 L).
Figure 6. Water Security Test Bed lagoon.
Water from this lagoon was used for studies on disinfection technologies to determine their
ability to treat large volumes of biologically contaminated water. Water in the lagoon contained
dirt and sediment from the surrounding area, as well as algae. The dirt and algal growth created
disinfectant demand in the water and rendered the water "dirty." Bacillus globigii (BG) spores
were dumped into the lagoon in order to simulate contaminated wash water resulting from the
decontamination of a drinking water pipeline or building with a contamination goal of 103 to 10 '
cfu/100 ml.
The effectiveness of the treatment technology was evaluated by sampling the lagoon water
containing BG spores before it entered the WOW Cart. The concentration of BG spores in the
influent (or before treatment began) was then compared to the concentration in the effluent (after
treatment).
2.2.1 On-site Chlorinator Lagoon Water Testing
The WOW Cart was challenged to assess its disinfection capability. The self-contained device
was shipped in a pallet/skid for easy deployment. It was mounted on one locking, rolling storage
cart with the following components:
1. The WaterStep M-100 chlorinator (an onsite chlorine generator)
2. Pumps: circulating pump (12V DC), distribution pump (120V AC) and a hand pump
3. Electrical Components: connectors and cords for equipment needing a power supply
including a ground fault interrupter, one 12V DC, a deep cycle battery, a storage case, a
solar panel, and one 10/2/50 ampere automatic battery charger
4. Plumbing Components: tubing and quick-connect cam-lock fittings for all water
connections
17
-------
The system setup is depicted in Figure 7.
Chlorine gas (Venturi Tube)
Salt addition
Chlorine generator
Chlorinated water outlet
oha'h
Contaminated water
inlet
Figure 7. WaterStep chlorine generator components.
The chlorinator uses salt (sodium chloride) and the process of electrolysis using direct current
from a 12-volt battery to produce chlorine gas and sodium hydroxide. Table salt purchased from
a grocery store was used in this experiment. The system runs an electrical current between the
two electrodes, separated by a membrane, in a solution of sodium chloride. Electrolysis breaks
up the salt molecules and frees chlorine gas from the brine. The chlorine gas is used as the
disinfectant. A small amount of sodium hydroxide is generated, which can be reused for other
purposes at the response site as needed.
The chlorine gas is introduced into the water stream using a venturi tube connected to the
chlorine generator. A pressure pump (a shallow well pump with bladder tank and a pressure
switch) is used to draw water from the lagoon and to circulate it through the venturi using a
garden hose. As the water passes through the venturi, it creates a vacuum which draws the
chlorine gas out of the chlorine generator. As the water is mixed with the chlorine gas, it flows
through and returns to the source or a bladder tank for storage and disinfection contact time. This
process is typically continued until the free chlorine concentration in the finished water reaches
the desired level.
The WOW Cart has the capability to pump water into 10,000 gallon (37,850 L) portable
bladders, where the contaminated water is temporarily stored to provide contact time for
disinfection before treated water is disposed of. These bladders were not used during tests at the
Idaho National Laboratory. Instead, the WOW Cart was set-up to pump contaminated water
directly from the lagoon through the chlorinator and then recirculated back into the lagoon for
storage and contact time allowing for disinfection to occur. During planning of the water
18
-------
treatment experiments, the research team felt that pumping water from the lagoon directly into
the WaterStep unit (and bypassing the bladders) would be a more accurate representation of how
the unit might be deployed during an emergency water treatment scenario. They expected the
enclosed lagoon would provide the necessary contact time and storage.
Operationally, water was drawn from near the lagoon inlet (the presumed point of highest
contamination in the lagoon) into the WOW Cart. The chlorinated effluent from the WOW Cart
was pumped back into the far end of the lagoon while a portion of the untreated effluent water
was re-directed to another portion of the lagoon, away from the inlet near the WSTB piping, to
increase or promote mixing within the lagoon that was not mechanically mixed. Figure 8 shows
the operational setup of the experiment.
Chlorinated Water
Outlet
Inlet to WOW Cart
Recirculation
Water Outlet
Figure 8. WOW cart setup at the lagoon
The unit operated for 4 hours and 40 minutes. Throughout this period, samples from the
chlorinated water outlet were collected and analyzed for free chlorine using a swimming pool kit.
The numbers reported were consistently above 5 ppm (the kit can only report values up to 5
ppm). Field dilution was not performed because this was simply a check to determine if chlorine
was being generated by the system. Grab samples were collected from the lagoon to evaluate the
chlorine levels and were submitted offsite for analysis of BGto determine if disinfection was
being accomplished.
After the chlorine treatment, the lagoon sampling results indicated that each of the BG values
reported were greater than 1Q5 cfu/100 ml. Although the reported chlorine values produced by
19
-------
the on-site chlorinator were consistently above 5 mg/L, the field methodology of delivering the
chlorine disinfectant to the lagoon without the bladder tanks was ineffective for disinfection of
such a dirty water source. The highest free chlorine residual detected in the lagoon was 0.03
mg/L, but the highest total chlorine residual detected was 1.71 mg/L. This indicated that the free
chlorine being generated by the WOW Cart was being transformed into total (or combined)
chlorine once it entered the lagoon. The large exposed surface area of the lagoon, in combination
with shallow depth, and intense sunlight, may all have contributed to the rapid degradation of the
chlorine delivered to the lagoon. Another confounding factor was the high organic load from the
dusty lined lagoon. Thus, the research team concluded that temporary storage bladders would
need to be used in emergency situations to provide sufficient contact time, reduce surface area,
remove the adverse effects of sunlight on the disinfection process, and reduce the impact of the
organic load that could be found in the environment.
2.2.2 Bladder Tank Chlorinator Testing
Next, the experiment was designed to assess the ability of the WOW Cart to disinfect a large
volume of water containing BG spores utilizing a bladder tank rather than the lagoon (US EPA,
2016a).
As in the previous experiment, the lagoon contained dirt and sediment from the surrounding area.
Disinfection experiments with the WOW Cart chlorine generator were conducted by spiking a
vendor supplied 1,250-gallon (4,732 L) tank with BG spores (10 spores/100 ml), filled with
lagoon water, and then chlorinated. The system set-up only included the chlorine generator and
power supply and is depicted in Figure 9. The free chlorine flows into a bladder tank where it
could disinfect the contained water. The system was operated using a 12-volt DC battery on the
cart (as shown in the middle of Figure 9). The battery was placed on a trickle charger to maintain
full charge for operational stability during the testing. There is one contained volume of
contaminated water that is exposed to free chlorine, which can facilitate disinfection of the BG
spores over time. The bladder tank was manually agitated by pushing on its side to mix the spores.
Manual agitation took place approximately every 15 minutes throughout the experiments. Before
disinfection, the bladder tank was sampled to determine the initial spore density, and then the
chlorination started. Subsequent samples were considered as treated, or disinfected water samples.
The bulk water samples (BWSs) for BG concentrations were collected from the same sampling
port that served as both inlet/outlet of the system using the grab sampling technique in 100-ml
sterile sample bottles with a 10 mg sodium thiosulfate tablet. The BWS sampling port was
opened and the water was drained for 15 seconds prior to collection of the sample.
20
-------
Battery
Water Step
Chlorinator
Water Inlet
Bacillus globigii
injection inlet/outlet port
Figure 9. WaterStep chlorinator system with bladder tanks (dark blue).
2.2.2.1 Analysis of Test Results
Data analyses and results from the disinfection experiments are presented in the following
sections.
Figure 10 shows the increase in free chlorine concentration inside the bladder tank over the
course of the experiment, and the subsequent decrease in BG spores. No free chlorine was
detected in the water at the time the experiment began. During the first 60 minutes after the
chlorinator was turned on, the free chlorine concentration in the bladder tank increased slowly
due to the organic demand in the water (turbidity was measured as 11 to 13 Nephelometric
Turbidity Units (NTU). However, after the first hour, the demand was overcome and free
chlorine in the bladder tank increased at a faster rate. The chlorinator was turned off after 210
minutes. The free chlorine was around 12 mg/L free chlorine at that time. The subsequent free
chlorine samples reflect the decay due to demand and temperature in the bladder tank.
At the start of the experiment, BG spores were mixed in the bladder tank volume by pushing on
the outside of the bladder tank to slosh the water around and promote mixing. The first three
samples taken from the bladder tank show that the volume was well mixed. BG spore density
averaged 2.4x 1Q7 cfu/100 ml over the first three samples. Figure 10 shows that even as the free
chlorine concentration rose from 0.14 to 3.30 mg/L from 60 to 120 minutes, spore density
remained the same. This is due to a well-known phenomenon in the field of disinfection knowns
as a "lag phase" or "shoulder". Bacillus spores are well known to be resistant to inactivation via
21
-------
oxidative disinfectants, and their concentration will remain stable for a period time in the
presence of disinfectants before decreasing (AWW A, 1999; Rice et al., 2005). Once free
chlorine did inactivate the BG spores, approximately 7-log reduction was achieved after 300
minutes of contact time.
l.E+08
^ l.E+07
o
S l.E+06
D
S. l.E+05
« l.E+04
-------
0 500 1000 1500 2000
Ct (mg-min/L)
Figure 11. The log reduction in spores during the WaterStep experiment plotted against
the Ct value (disinfectant concentration multiplied by time).
Ct values have been compiled in the literature for disinfection of pathogenic and non-pathogenic
Bacillus spores. These Ct values were often collected in experiments focused on disinfection of
drinking water, which generally has less disinfectant demand than the lagoon water used in these
experiments. For example, a Ct of 106 mg-min/L was needed for a 3-log reduction of B.
cmthracis Ames at pH 7 and 25° C in the presence of 1 mg/L free chlorine. The 3-log reduction
Ct value for BG spores at similar conditions was 136 mg-min/L (US EPA, 2012). In the
WaterStep experiments with lagoon water, the 3-log reduction Ct was 707 mg-min/L at pH 7 and
temperature ranging from 20 to 25°C.
Some of the increase in the Ct values found in lagoon water comes from the fact that temperature
started lower than in the drinking water Ct experiments (15°C to 25°C), where temperature was
constant (25°C). Disinfectant concentration is generally fixed in lab Ct studies, where in this
experiment it had to increase from zero once the chlorinator was started. Furthermore,
disinfectant demand is much less of a factor in lab studies, unlike this field study where
disinfectant concentration had to build over time in the presence of an organic load. These
factors resulted in a Ct value that is approximately 5 to 6 times higher than those found for the
same or similar spores observed under drinking water treatment conditions.
In summary, the WOW Cart achieved 6.8 log removal in a 1,250-gallon (4,732 L) bladder tank
within 5 hours of the start of the experiment while achieving 12.2 mg/L free chlorine. This was a
small volume and appropriate under certain scenarios, but evaluation of larger volumes of water
under flow-through conditions also needed to be challenged and will discussed later in this
report.
Table 1 contains a summary of the WOW Cart technology-specific equipment observations
recorded during the treatment experiments and considerations for similar field deployments. The
terms Low, Medium and High are the opinions of the authors of this study and are based on their
experience operating the equipment in the field. The text in the table is meant to support these
23
-------
opinions, and they are specific to this piece of equipment. Other equipment operators may come
to different conclusions under different conditions.
Table 1. WaterStep Technology-specific Consideration and Observations*
Technology
Considerations
Rating and Comments
Market Availability
High. Commercially available off-the-shelf product from a non-profit
organization for producing drinking water in communities in developing
countries. Self-contained kit could be used in disaster zone to purify
water if there was no power available from the electrical grid. Available
from httD://watersteD.org/
Capital Cost
Medium (estimate $15,000). Includes storage bladders, pump, battery,
charger, solar cell, mounting/transportation rack, and salt-based chlorine
generator (chlorinator).
Shipment to Site
Medium. Needs to go on a truck or commercial transportation. Could be
transported in a smaller vehicle, if mounting and transportation rack are
not used. The unit weighs approximately 460 pounds and is built to meet
airline size and weight requirements.
Setup
Considerations
Medium. Need flat surface to spread out the bladder tanks. Need to
recirculate chlorinated water to provide contact time for disinfection.
Not a flow through system. Test kit (strips or colorimetric) required to
periodically check chlorine generation. After disinfection, if chlorine is
not consumed, the excess chlorine may need to be neutralized before
being discharged to the environment.
Operational
Considerations
Low. Simple to operate on a short-term basis. If extended contact period
is required greater than 3 hours, the salt solution needs to be replenished,
electrolytic cell must be drained, and, if not on 110-volt AC power, the
battery needs to be charged. Each unit can treat up to 10 GPM providing
drinking water to approximately 500 people daily. Consumables depend
on the useage, primarily consisting of only salt, filtration media, and
power source.
Maintenance and
Consumables
Low. Table salt is the only consumable. For optimal chlorine generation,
the electrolytic cell needs to be cleaned periodically. Gasoline or
propane for generator. Pumps, hoses and O-rings need to be checked
periodically for wear and cracking.
Result Summary
Under the tested conditions, a 7-log removal of Bacillus globigii was
observed in a batch type operation with 300-minutes of contact time.
* Mention of trade names or commercial products does not constitute endorsement or recommendation for use of a specific product
3.0 Puerto Rico Deployment
3.1 Background
24
-------
Just three weeks after Hurricane Maria made landfall in Puerto Rico Wednesday, September 20,
2017, WaterStep's team was on the ground training emergency workers and distributing kits with
components of the 2nd generation WOW Cart. The impact continues after over 100 kits were
deployed and hundreds of people trained in the proper use of the equipment. Though some kits
are still being used, many are now positioned and poised to be used during the next disaster.
Below is a summary of the work WaterStep accomplished in its disaster response.
3.2 Achievements
WaterStep received a generous donation from General Electric Appliance Park, a Louisville,
Kentucky foundation, as well as funding from many donors to respond in Puerto Rico. In
addition, WaterStep received a donation of the use of a DC-3 for the cost of fuel, the staff time of
the response team, and the initial shipment of disaster kits. These kits consisted of:
• Hand pump
• Pre-filters
• On-site chlorine gas generator
• BleachMaker*
• Pump
• Power supply
• Solar charger
• Single hole recirculation manifold
• Bladder tank
• Quick connect fittings and hoses
• Salt
* 1-litre containers of a 1% solution of liquid bleach produced concurrently with the water
treatment to be used for general cleaning and support of medical triage by emergency personnel.
Deployment onto the island occurred within three weeks after the hurricane. Coordination for
training and distribution of equipment was coordinated in conjunction with the National Puerto
Rican Leadership Council Education Fund. Training and equipment were first given to the most
affected municipalities. After receiving a grant from Unidos por Puerto Rico, WaterStep was
charged to make sure each municipality (78) had one disaster kit to be used by their emergency
management office. The municipalities identified in red in Figure 12 denotes those who had
received disaster kits by Spring, 2018. Thanks to the commitment of Doctoras Boricuas and the
Unidos por Puerto Rico grant, more than 100 bleach makers and chlorine generators were
delivered in total around the island including more than 20 to non-profit and medical
organizations. Hundreds of people from the government, private sector, doctors, non-profit
volunteers, and students have been trained on how to properly use each disaster kit. Each unit is
in the responsible hands of trained people in a government or non-profit organization who will
keep the equipment safe and operational until another emergency occurs. If all equipment is
working at one time, the WaterStep equipment placed in Puerto Rico has the potential to
generate 1 million gallons of safe water per day serving approximately 50,000 people.
25
-------
Culebra
Vieques
RESPONSE IN PUERTO RICO 2017-18
Figure 12. Municipalities that received a disaster kit within weeks of Hurricane Maria
(red).
3.3 Strong Communication
In Puerto Rico, WaterStep was careful to ensure that the government was involved in
deployment and training in the proper use of the equipment. This created better communi cati on
and a higher sense of confidence. It was noticed that emergency management teams shared their
experiences with each other and also shared best practices on the equipment's various uses
during the emergency.
There was some hesi tati on about drinking the water after just treating it with the chlorine gas
generator. There was some concern among the communities as to whether or not the Puerto Rico
Department of Health and the EPA had approved the use of the WaterStep system. The
successful testing of the WOW Cart at the EPA Water Security Test Bed was important to the
local governments' acceptance of the system. It is recommended that there should be additional
education and a water-safety educational campaign, so everyone knows about safe ways to treat
water. Everyone from government officials, business people, volunteers, citizens, elderly, to
even children must be educated in order to better respond to another emergency. Also,
coordinating work with other non-profit organizations is key.
3.4 Success Highlights
Figure 13 is an example deployment. A training event was conducted near San Juan for many of
the municipalities and emergency responders. One of the first disaster kits was given to the
municipality of Orocovis. The Orocovis emergency management team was trained and returned
with a logistics plan to disinfect and distribute water throughout the most affected areas in
26
-------
Orocovis. This included most of the municipality's population of 24,000 people. A few months
later, the Director of Emergency Response in charge of the Orocovis equipment reported how
effective the system had been in terms of being simple to use and providing drinking water. An
additional system was soon installed permanently at a baseball complex to treat water from a
nearby spring. Piping was installed to bring the water to the treatment system and then to be
stored in new tanks. The primary lesson learned was to be as inclusive and active as possible
communicating with government officials, emergency responders, and community leaders on the
use and operation of the technology.
In the municipality of Isabela, a unique approach was to use a flatbed truck to house the bladder
tank and fill it from local rivers and streams; the disinfection process was started in transit. Safe
water was transported and easily distributed to the community.
The disaster kit was shown to be a powerful tool to provide safe water during the initial weeks of
response after Hurricane Maria in Puerto Rico. However, due to the sustainable housing and
structure of the equipment, it is designed to be used again and again. With over 100 disaster units
on the Island of Puerto Rico, emergency workers and medical personnel are now prepared for
storms in the future. The link below provides information on the Puerto Rico deployment and
the FTTA utilization.
https://www.voutube.com/watch?v=Db9MlSiOJkk&feature=voutu.be
Figure 13. Disaster kit installation
27
-------
4.0 Full-Scale Deployment
4.1 Description of the Final Version of the WOW Cart
The WaterStep saltwater chlorine gas generator is at the heart of the Version 3.0 WOW Cart
treatment system. The unit is customizable with flow rates ranging up to 10 gallons per minute
(gpm) (37.85 L/min). In addition to chlorine-based disinfection, the WOW Cart first utilizes
100 micron and 25-micron disc pre-filters to remove particulates. The small media cartridges
were replaced with larger media tanks to prolong filter life. In many situations, media such as
granular activated carbon (GAC) is a likely choice given its ability to remove a broad spectrum
of chemical contaminants (Figure 14). Other types of media could be used for their ability to
remove radioactive or other types of inorganic contamination.
Chlorine Gas
Generator
L_
kJ
New Larger Tanks
for Media
New Frame
BleachMaker
Dual Fuel I
Generator
Deep Cycle
12V Battery
Larger Wheels
Figure 14. 3rd generation of the WOW cart following Puerto Rico deployment.
The WOW Cart is self-contained and self-supported; therefore, it does not require any additional
installation beyond connections to the raw water source and electric power. If necessary, the
28
-------
WOW Cart can also be powered by a generator that comes with the cart. Version 3.0 of the
WOW Cart utilizes a user-friendly duel-fuel gasoline/propane 3,500-watt generator. The chlorine
generator and a small recirculation pump can be powered by a deep cycle marine battery and
charged by a solar panel.
During the redesign, a few other issues were addressed. The frame of the cart was slightly
increased to accommodate the generator, the larger media tanks, and other possible treatment
devices such as UV LED and/or ultrafiltration membranes if required by the particular
emergency response incident. The new frame material is poly coated steel. The cart frame can
now be pre-cut and assembled without welding enabling size adjustments according to the
situational needs without waiting on frame design, welding, and powder coating. Larger wheels
were added to insure better mobility on different terrains and with the larger frame.
The use of schedule 80 solvent piping and valves was changed to PEX (cross-linked
polyethylene) piping and brass for durability. An additional section of PEX piping was inserted
into the frame to accommodate extra filters or other accessories such as UV disinfection or
additional filtration.
The extra room of the larger cart allowed for the installation of five 1-liter containers of the new
WaterStep BleachMakers. This enables the production of a 1% solution of liquid bleach
concurrently with the water treatment. The bleach solution can then be used for general cleaning
and support of medical triage by emergency personnel.
This newest version still fits on a standard skid and weighs less than 700 pounds (weight will
vary depending on tank bladder size.) A new SingleHole Manifold (Figure 15) for connections
from the WOW cart to the bladder tanks was developed. This manifold allowed for the reduction
in the amount of plumbing (hoses) and has proven to be much more user friendly to re-circulate
the stored water. Auxiliary 120v electrical outlets, and USB ports have also been added into the
system for on-site access to recharge phones, tablets, and lap-tops (Figure 16).
29
-------
Sample Hose
Figure 15. SingleHole Recirculation Manifold.
Figure 16. Electrical outlets and phone charging station.
30
-------
4.2 Secondary Wastewater Challenge
The Secondary Wastewater challenge evaluated the ability of 3rd generation WOW Cart to
disinfect a turbid non-chlorinated secondary effluent discharged from the Greater Cincinnati
Metropolitan Sewer District's Gest St. Wastewater Treatment Plant. Secondary wastewater was
selected to simulate a contaminated surface drinking water source or a combined stormwater and
sanitary sewer effluent. The challenge was based on free chlorine residual produced by the
chlorine gas generator and the subsequent inactivation of Escherichia coli and total coliforms in
the secondary effluent. (Total coliforms are a group of related bacteria that are common in the
environment [soil and vegetation] and are used as a general indicator of drinking water quality.)
This unit was evaluated at the EPA Test and Evaluation (T&E) Facility, located in Cincinnati,
Ohio. The unit and its associated bladder tank used in this evaluation is shown in Figure 17.
Figure 17. WOW cart secondary wastewater challenge set-up with empty 1,250-gallon
bladder tank at T&E Facility.
Non-chlorinated secondary effluent enters the T&E Facility directly from the Greater Cincinnati
Metropolitan Sewer District through an 8-inch PVC pipe. For this experiment, a portion of the
secondary effluent flow was directed into a 1,000-gallon stainless steel tank located on the floor
31
-------
of the T&E high bay (Figure 18). The WOW Cart was connected to the tank through a manifold.
The WOW Cart's optional onboard jet pump pulled the secondary effluent from the tank through
the disc pre-filters and then through the chlorine generator. The treated water was then pumped
into the 1,250-gallon bladder tank.
Figure 18. Secondary wastewater effluent holding tank and WOW cart at T&E Facility.
The test was performed on June 27, 2018. The secondary effluent made a single pass through
the WOW Cart's disc pre-filters and then through the chlorinator and was discharged into the
bladder tank. The water flow through the WOW Cart was approximately 6 gpm (22.71 liters per
minute) as shown on the rotameter in Figure 19.
32
-------
Figure 19. Rotameter showing flow through the WOW cart.
When the bladder tank was approximately half full (-660 gallons), as shown in Figure 20, the
secondary effluent flow from the tank was shut off to the WOW Cart and the pump lines were
reconfigured to allow the bladder contents to continuously recirculate through the WOW Cart's
disinfection system and back into the bladder tank. The manifold "mixer" bar inside the bladder
tank was used to mix and recirculate the water within the bladder tank. The recirculation and
chlorination of the secondary effluent continued for one hour. The water passed through the
chlorinator throughout the entire test period.
33
-------
Figure 20. Bladder tank (660 gallons - 1/2 full).
4.2.1 Analysis of Test Results
During the test, inlet and outlet samples from the WOW Cart were collected and analyzed fori?.
colt and total coliforms. The secondary effluent was the source of the bacteria microorganisms:
E. coli and total coliforms. The inlet samples were collected from the WOW Cart's lower
sample port located just before the water enters the cart's disinfection device, while the outlet
samples were collected from the port on the short side of the mixer manifold located at the
bladder inlet. Samples were collected as the bladder was being filled and during the recirculation
of the water through the bladder tank. Throughout the test, the inlet and outlet water were
analyzed for free and total chlorine. Results of these analyses are presented in Table 2. Figure
21 shows the oxidizing capabilities of the WOW Cart. The blue-green color of the inlet
secondary effluent sample (left) was stripped from the water and the resulting outlet sample
(right) was clear.
34
-------
Table 2. Chlorine Concentrations at the WOW Cart Outlet (single pass)
Free chlorine,
Total chlorine,
Sample Time
mg/L
mg/L
Startup - single pass through cart
0.06
0.12
20 minutes of single pass water
12
14.2
2 hours of single pass water
13.6
15.4
2.5 hours of single pass water
12.9
14.1
Start recirculation in bladder tank
30 minutes of recirculating water
23
27
45 minutes of recirculating water
25
31
1 hour of recirculating water
27
33
Note: It was not necessary to achieve these high chlorine levels to completely disinfect the E.
coli. It was just done to demonstrate the capability of the system
Figure 21. WOW cart inlet (left) and outlet (right) samples.
TheE. coli samples were analyzed at the T&E Facility Biosafety Level (BSL) 2 Laboratory
following the APTIiYl T&E MOP [Miscellaneous Operating Procedure] 310, Revision 2: "Total
Coliform and E, coli Analysis Using IDEXX Colilert-18." The 100 ml samples and/or diluted
samples were mixed with Colilert®-18 media powder (IDEXX Laboratories, Inc., Westbrook,
35
-------
Maine). When the media powder dissolved, the sample-media mixture was poured into an IDEXX Quanti-tray®/2000. After
incubating at 35°C for 24 hours, the trays are examined under UV light to count the number of fluorescent wells. The number of
fluorescent wells is cross-referenced with a most probable number (MPN) table to obtain the MPN of E. coli in the original sample.
The results from the E. coli and total coliform analyses of water samples collected from the WOW Cart are summarized in Table 3.
Individual inlet concentrations were compared to the corresponding outlet concentrations to compute the log reduction values shown in
Table 3. Log reduction values for E. coli and total coliforms from the initial outlet samples were not presented since the chlorination had
not started. The data show that E.coli and total coliforms were removed to below the level of detection by the WOW Cart during the first
15 minutes of the recirculation of the water through the bladder tank. Comparing inlet and outlet E. coli concentrations, the WOW Cart
produced log reductions up to 2.8 while operating in the single pass mode. Log reductions of 4 or greater (complete removal of E. coli
and total coliforms) were achieved while recirculating the contents of the bladder tank. As shown in Table 2, this is most likely due to
the higher chlorine concentrations present in the recirculated bladder tank water. Utilizing the recirculating manifold from the initial
start-up would have most likely reduced the time to complete inactivation of the E. coli and total coliforms.
Table 3. WOW Cart Summary of E. coli and Total Coliforms Disinfection Results
Sample
Condition
Total Elapsed
Time (min)
Inlet E. coli
(MPN/100 ml)
Inlet Total
Coliforms
(MPN/100 ml)
Outlet E. coli
(MPN/100 ml)
Outlet Total
Coliforms
(MPN/100
ml)
E. coli
Log
Reduction
Total Coliforms
Log Reductions
Single Pass through the WC
)W Cart
Started filling bladder with secondary wastewater and turned on chlorinator
Initial
2
1.07E+04
2.40E+05
1.50E+04
2.40E+05
NA
NA
330 gal in
bladder
42
2.00E+04
1.60E+05
6.40E+01
6.10E+02
2.4
2.6
660 gal in
bladder
143
1.40E+04
2.04E+05
2.70E+01
3.00E+02
2.8
2.9
Average inlet concentration
1.70E+04
2.13E+05
E. coli and Total Coliforms were removed below detection level within 15 minutes of recirculation
NA - Log reduction cannot be calculated at start-up of the test
36
-------
4.3 WS TB Microbial and Diesel Fuel Challenge
The objective of the test was to next evaluate the efficacy of the WOW Cart treatment train to
decontaminate a mixed water supply contaminated with bacteria and petroleum-based chemicals.
The scenario is reflective of a surface water contaminated by a barge spill or flood waters
contaminated with untreated sewage.
Figure 22 shows the WOW Cart being deployed on-site at the WSTB. The WOW Cart was
unpacked and wheeled to its location adjacent to the lagoon (Figure 23). For this experiment, the
lagoon was to be filled to -7,000 gallons of potable water from the Water Security Test Bed
pipeline. Prior to the WOW Cart being deployed, the main WSTB pipeline experienced a severe
joint failure flooding the area causing the lagoon to overflow, thus creating a much more realistic
emergency response scenario. Power was available on-site, so the duel fuel generator was not
necessary.
Non-pathogenic E. coli K-12 bacteria and diesel fuel were added to the lagoon. The
contaminants were allowed to mix and disperse overnight. The next day, the water was pumped
through the WOW Cart. The desired contaminant concentrations in the lagoon were -20 mg/L
diesel fuel and 105 MPN/100 ml E. coli. To achieve those concentrations throughout the lagoon
the contaminants were physically mixed by walking through the lagoon. Benzene, toluene,
ethylbenzene, and xylene (BTEX) make up 0.5-1.2% of typical diesel fuel. The effluent from
the WOW Cart was collected in a 2,000-gallon bladder. Water samples from the lagoon were
compared to water samples taken from the bladder.
37
-------
aHmaBat ¦'•*!&&/¦
Figure 22. WOW cart delivered on-site at the Water Security Test Bed.
38
-------
39
-------
4.3.1 Analysis of Test Results
Table 4 describes the E. coli and total coliform reduction and/or inactivation of microbial contaminants by the WOW Cart being
operated in flow through mode. In order to provide increasing Ct for disinfection, bladder tank outlet samples were collected at set
intervals to show an increasing disinfection rate. Effluent samples were taken from the bladder tank over time providing Contact
Time. After about 45 minutes of operation, the WOW Cart effluent showed a reduction of around 1 log of both microbial
contaminants. Following another 45 minutes of treatment and contact time (90 minutes total) both the WOW Cart effluent and
bladder tank contents showed reductions of E. coli and total coliforms of 6 log and 4 logs respectively.
Table 4. Microbial Results from the Lagoon
WOW Cart Inlet (Lagoon Source Water)
Treated WOW Cart Outlet to
Recirculated Treated Bladder
Bladder Tank
Tank
Sample Time
Total Coliforms
E. coli
Total Coliforms
E. coli
Total Coliforms
E. coli
(MPN/100 ml)
(MPN/100 ml)
(MPN/100 ml)
(MPNt 100 ml)
(MPN/100 ml)
(MPN/100 ml)
15:15
>2.4E+06
1.5E+04
2.1E+05
8.4E+03
1.7E+01
2.0E+00
16:00
>2.4E+06
3.5E+03
1.0E+00
1.0E+00
1.0E+00
ND
16:45
>2.4E+06
5.6E+04
1.0E+00
ND
1.0E+00
ND
17:30
2.4E+06
1.5E+05
1.2E+05
1.5E+04
2.0E+01
5.1E+01
18:15
1.3E+06
1.1E+04
ND
ND
2.0E+01
ND
19:00
1.4E+06
1.3E+03
ND
ND
ND
ND
E. coli and Diesel fuel added to the lagoon at 14:25 and
4:27 respectively
ND - non-detect (<1 MPN/lOOmL)
40
-------
Table 5 describes the influent and effluent levels of the diesel fuel components. Because of the extremely high turbidity in the lagoon
(> 100 NTU), the GAC filter media tanks became clogged and were removed from the WOW Cart after 3 hours of operation prior to
sampling at 17:30 hours. Results indicate that diesel range organics (DRO) CIO - C20, oil range organics (ORO) C20 - C34, gasoline
range organics (GRO) C6 - C12, and Total Petroleum Hydrocarbons (TPH) were removed through the first two hours of operation
prior to the GAC being removed as shown by the BWS-4-1 sample. One sample taken at 14:00 hours indicates that DRO and TPH
may have started to breakthrough given the GAC clogging prior to the removal of the GAC media.
Table 5. Diesel Fuel Removal Rates
WOW Cart Inlet (Lagoon Source Water)
Treated WOW Cart Outlet to Bladder Tank
Sample
Time
DRO
(mg/L)
ORO
(mg/L)
GRO
(mg/L)
TPH
(mg/L)
DRO
(mg/L)
ORO (mg/L)
GRO
(mg/L)
TPH
(mg/L)
15:15
6.500
1.300
0.110J
7.910
U
U
U
U
16:00
U
0.120
u
0.120
0.110
U
U
0.110
16:45
U
0.120
0.200J
0.320
u
U
u
u
17:30
0.150
0.140
0.170J
0.460
0.140
0.120
0.120J
0.380
18:15
0.170
0.170
0.140J
0.480
0.250
0.120
u
0.370
19:00
0.190
0.140
0.120J
0.450
0.140
0.110J
0.120J
0.370
WOW Cart Inlet (Lagoon Source Water)
Treated WOW Cart Outlet to Bladder Tank
Sample
Time
Benzene
(ug/L)
Ethylbenzene
(ug/L)
Toluene
(ug/L)
Total
Xylene
(ug/L)
Benzene
(ug/L)
Ethylbenzene
(ug/L)
Toluene
(ug/L)
TPH
(ug/L)
15:15
U
U
U
U
U
U
U
U
16:00
U
U
U
U
U
U
U
U
16:45
u
u
u
1.200J
u
u
u
u
17:30
u
u
u
1.200J
u
u
u
u
18:15
u
u
u
u
u
u
u
u
19:00
u
u
u
u
u
u
u
u
E. coli and diesel added to the lagoon at 14:25 and 14:27 respectively
U = Non-detect value
J = Estimated value
41
-------
5.0 Conclusions
The WOW Cart has shown that it is able to treat natural waters highly contaminated with
Bacillus globigii (an anthrax surrogate), E. coll., and total coliforms. Simultaneously, it was also
able to treat diesel-fuel contaminated water. Based upon the field evaluations and Hurricane
Maria response, the deployment, training, and operation of the WOW Cart was determined to be
easy and cost-effective. Capital cost of the WOW Cart has been estimated to range between
$15,000 and $25,000 depending on quantity fabricated.
• Bacillus globigii showed a 7-log reduction with a free chlorine residual around 10 mg/L
when operated in a batch mode over a few hours in a 1,250-gallon bladder tank.
• Following Hurricane Maria in Puerto Rico, over 100 disaster kits (pre-filtration and
chlorination) were deployed, providing microbially safe drinking water to tens of
thousands of people.
• The 3rd generation WOW Cart disinfected 4 and 5-log levels of E.coli and total
coliforms respectively, found in secondary wastewater while exhibiting high levels of
free and total chlorine.
• The WOW Cart successfully removed similar levels as above of E. coli and total
coliforms from a contaminated lagoon while simultaneously removing diesel fuel
components to below detection levels. Because of the extreme high turbidity (> 100
NTU), the granular activated carbon media treating the diesel fuel clogged after
approximately two hours of operation. This suggest that if a better source water cannot
be utilized during an actual emergency response, an additional pre-filtration step may be
necessary to reduce excess media replacement.
• Additional research is ongoing to evaluate the integration of additional pre-filtration
technologies (e.g. electro-coagulation), multi-media filtration, UV-C LED, ozone, and
ultrafiltration membranes to increase the WOW Cart's ability to treat an even broader
suite of contaminants and increase unit process longevity (e.g. GAC) and decrease
O&M replacement costs.
6.0 References
American Water Works Association (AWW A). 1999. Water Quality and Treatment: A
Handbook of Community Water Supplies, 5th ed. McGraw-Hill, New York.
Rice, E.W., Adcock, N.J., Sivaganeson, M. and Rose, L.J. 2005. Inactivation of spores of
Bacillus anthracis Sterne, Bacillus cereus, and Bacillus thuringiensis subsp. israelensis
by chlorination. Applied and Environmental Microbiology, 71(9):5587-5589.
42
-------
U.S. Environmental Protection Agency (US EPA). 2012. Technical Brief: Inactivation of
Bacterial Bioterrorism Agents in Water: Summary of Seven Studies. US EPA,
Washington, DC. EPA/600/R-12/521.
U.S. Environmental Protection Agency (US EPA). 2016a. Testing large volume water treatment
and crude oil decontamination using the Water Security Test Bed at the Idaho National
Laboratory. U.S. EPA, Washington, DC, EPA/600/R-16/126. Accessed on May 21, 2019
at:
https://cfpub.epa.gov/si/si public record report.cfm?Lab=NHSRC&dirEntryId=327000
U.S. Environmental Protection Agency. 2016b. EPA Water Security Test Bed Experiments at the
Idaho National Laboratory. U.S. EPA, Washington, D.C., EPA/600/R-15/146._Accessed
on May 21, 2019 at:
https://cfpub.epa.gov/si/si public record report.cfm?Lab=NHSRC&dirEntryId=322581
43
-------
7.0 Appendices -USEPA T&E Facility Contract Technical
Standard Operating Procedure
Appendix A: Total Coliform and E. coli Analysis Using
IDEXX Colilert® 18 Method
MOP
Coliforr
Pff
310 Total
n and E coli
Appendix B: Free Chlorine Analysis by HACH®
Method 8021 and Total Chlorine Analysis by HACH®
Method 8167
N.N-diethyl-p-phenylene-diamine (DPD) Colorimetric
Method (0.02 to 2.00 mg/L C12)
504 Fr
and Tot£
J™ ]
i w- J
ee Chlorine
1 Chlorine An
44
-------
vvEPA
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |