Technical BRIEF

INNOVATIVE RESEARCH FOR A SUSTAINABLE FUTURE

Treatment of Chemical, Biological, and Radiological (GBR)
Contaminants in Storm water Runoff

Results from testing different media in a pilot-scale stormwater drainage system inlet
Introduction

Testing treatment technologies prior to spills of
hazardous or infectious materials is critical so that
the impact of such disasters can be quickly
minimized, and environmental resources and
human health preserved, A better understanding
of distributed stormwater treatment options for
homeland security contaminants is critical to
protect against immediate discharge to nearby
waterways, spatial spread of contamination,
sorption onto pipe material that can cause
persistent releases, and/or protection of
wastewater treatment plant unit processes. The
use of media-based catch basin inserts in separate
or combined sewer networks for removing
conventional contaminants such as solids, metals,
nutrients, and oil and grease from stormwater
runoff has been previously studied [1-6]. The
products demonstrated mixed success depending
on factors such as hydraulic retention time,
installation geometry, and stormwater quality.
However, there is a lack of data focused on
homeland security contaminants of concern (i.e.,
chemical, biological, and radiological (CBR)
contaminants). This study investigated the
feasibility to reduce CBR contaminant
concentrations entering the conveyance system by
installing treatment media in a pilot-scale inlet.

Capability

To conduct this research, a pilot-scale storm inlet
was built and characterized at EPA's Test and
Evaluation (T&E) Facility in Cincinnati, OH (Figure
1). The system is comprised of a feed tank, flow
distributor arm, flow table, piping, and inlet. The
polyethylene feed tank (VT0850-54, Ace Roto-
Mold, Den Hartog Industries Inc., Hospers, IA) has

U.S. Environmental Protection Agency's Homeland
Security Research Program (HSRP) develops
scientific products based on research and
technology evaluations. Our products and expertise
are widely used in preventing, preparing for, and
recovering from public health and environmental
emergencies that arise from terrorist attacks or
natural disasters. Our research and products
address biological, radiological, chemical, and oil
contaminants that could affect indoor areas,
outdoor areas, and water infrastructure. The HSRP
provides these products, technical assistance, and
expertise to support EPA's roles and responsibilities
under the National Response Framework, statutory
requirements, and Presidential Directives.

a capacity of 850 gallons. The distributor is
constructed from 3 in. diameter polyvinyl chloride
(PVC) pipe and sends water onto a 6.5-foot-long
flow table constructed from PVC sheets, which in
turn directs water to a box with a discharge pipe
located 7 in. from its' bottom. The on-grade box is
a commercially available high-density
polyethylene insert (2400 BLKIT, NDS, Woodland
Hills, CA) with dimensions of 2 ft. wide x 2 ft. long
x 2 ft. deep and comes with a grate divided into
four sections with 13 rows of 1-in. x 5-in. slots. The
system can generate flows from approximately 1-
50 gpm. For this study, its' interception capacity
was 100% as the bypass was kept closed. Flows
were measured using electromagnetic flowmeters
(#335-379, Toshiba America Inc., Irvine, CA). Note,
this design is considered a "simple" inlet since
there is no solids storage under the effluent pipe
once the media is inserted such as in a
conventional catch basin design, but it does
replicate the upper portion of a catch basin where
an insert would be installed.

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028

2025


-------
78"

4.5"
o

* ~

11.5"
*

47"

A

11.5"

2.5"

* * It

1.5"|
Bypass

Drain

6"

Flowmeter

Effluent
Sample

Oistri

butor

(1/I6"x42" long)

To Drain ~

Figure 1. Pilot-scale stormwater inlet top-view schematic and pictures

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028

2025


-------
Experimental Methods

To-date, switchgrass (Big Switch, BEG Group LLC,
Cambridge, OH), a coal-based granular activated
carbon (GAC) (FILTRASORB 400, Calgon Carbon
Corporation, Pittsburgh, PA), and a coconut-based
GAC (OLS Plus 12x30, Calgon Carbon Corporation,
Pittsburgh, PA) have been tested. These media
were selected due to their high availability and
low cost. Prior to testing the media for treatment
efficacy, hydraulic performance tests were
conducted that determined a 5-gpm flowrate
avoided flooding the table for all media types,
although switchgrass could accommodate higher
flowrates. Treatment efficacy tests were then
conducted by filling the inlet with 1-3 mesh bags
containing the media according to Figure 2. The
placement of the bags was selected to minimize
short-circuiting of the water and/or expedite
contaminant breakthrough testing. Simulated
stormwater was used for the testing according to
the recipe in Table 1.

Figure 2. Media Configuration for A) Coal &

Coconut GAC 800-gallon tests, B)
Switchgrass, and C) Coal & Coconut GAC
breakthrough tests

For these tests, malathion (ORTHO Max
Malathion, The Scotts Company LLC, Marysville,
OH) was used as a simulant for the nerve agent
Venomous Agent X (VX), Bacillus globigii (Bg) for

the bioagent Bacillus anthracis, and stable (non-
radioactive) cesium chloride (CsCI) for the
radionuclide Cs-137. Table 2 summarizes the
testing matrix. Contaminants were quantified
using EPA method 622 for malathion, ICP-MS for
cesium, and membrane filtration and culture for
Bg. The CsCI, Bg, and 0.2-0.3 mg/L malathion
experiments were conducted using 800 gallons of
water and the media configurations in Figure
2A/2B. The average weight of media used was 83
lbs (37.6 kg), for GAC and 16 lbs. for switchgrass.
Higher concentration malathion-GAC experiments
were conducted to discern contaminant
breakthrough. A cinder block was added to the
inlet (Figure 2C) so less GAC could be used
(approximately 28 lbs.) without short-circuiting.

Table 1. Simulated Stormwater Recipe

Chemical

Amount
mg/L

Source

Product
Code

Magnesium

Sulfate
Heptahydrate

15

Fisher
Chemical

M63-3

Potassium
Chloride

125

Fisher
Bioreagent
s

BP366-1

Potassium
Nitrate

90

Sigma
Aldrich

P8394-2.5KB

Sodium Nitrate

90

Fisher
Chemical

S342-3

Calcium
Chloride
Dihydrate

47

Fisher
Chemical

C79-3

Sodium Chloride

302.6

Fisher
Chemical

S271-10

Copper (II)
sulfate
pentahydrate

0.16

Acros
Organics

197722500

Zinc Sulfate
Heptahydrate

0.41

Fisher
Chemical

Z76-500

Soil Humic Acid

1 mg/L

Total Organic
Carbon

GS Plant
Food

Not
Applicable

Table 2. Testing Matrix

Media

Contaminant

Average
Influent

Replicates

Switchgrass

CsCI

3.2 mg/L

2

Coal GAC

Bg

4.9 x 105
CFU/100 mL

2

Coal GAC

CsCI

3.0 mg/L

2

Coal GAC

Malathion

0.2 mg/L

4

Coal GAC

Malathion

13 mg/L

3

Coconut GAC

Malathion

0.3

3

Coconut GAC

Malathion

7.2 mg/L

1

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028

2025


-------
Results

Out of the combinations tested, malathion removal was the most successful (Figure 3). The media were
ineffective at removing Bg spores (Figure 3A) and cesium (Figure 3B) during 800-gallon experiments and
therefore not investigated further. Across all sampled effluent points (100, 200, 400, 600, and 800 gallons),
the coal GAC averaged 86% removal of malathion and the coconut GAC averaged 91% removal of
malathion. This variation was within the range of standard deviation for both media types (Figure 3C).

800 gallons of Treated Water

Coal GAC

Coal GAC	Switchgrass

Media

Coal GAC

Coconut GAC

Figure 3. Average treatment efficacy by media type for A) Bg spores, B) cesium, and C)

malathion

Data from 800-gallon testing indicated steady removal of malathion in which breakthrough had not yet
been achieved (Figure 4A). Further testing was performed to determine how long the media would last.
During these experiments with 55 times higher malathion concentrations, 3 times less media, and 6 times
more water, breakthrough was achieved for both media types (Figure 4B).

Timeseries Samp ing Data

200

400	600

Cumulative Volume (gal)

Media ~ Coal GAC Coconut GAC

800

1000	2000	3000	4000

Cumulative Volume (gal)

Media ¦+¦ Coal GAC Coconut GAC

5000

Figure 4. Treatment efficacy at sampled timepoints for GAC media and malathion-
contaminated simulated stormwater A) 800-gallon tests, B) breakthrough tests (n = 1

Coconut GAC and n =3 Coal GAC)

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028


-------
To estimate an approximate sorption capacity, the
breakthrough data was averaged on each
cumulative volume data point in Figure 4B. An
average percent reduction was applied to the
averaged influent malathion concentration to
estimate the total malathion removed between
sample points. This resulted in an estimated
sorption capacity of 7.46 mg malathion / g of
carbon. Freundlich isotherm coefficients for
malathion from the literature[7] of k = 1.6 and 1/n
of 0.60 were also applied to the average influent
malathion data and resulted in an estimated
theoretical sorption capacity of 6.54 mg malathion
/ g of carbon for the breakthrough experiments.
The difference is reasonable considering the
averaging of the data, granularity of sample
points, and different carbons in the experiments
vs. literature study. The collected data indicate
that for preliminary design of a GAC inlet capture
system, the use of the Freundlich isotherm is a
reasonable starting point.

Discussion

Due to a limited testing budget and historical
projects showing low removal for some test
combinations only a limited number of
combinations of contaminant-media pairings were
tested. Switchgrass and GAC were selected as the
test media because of their availability and low
cost. Not unexpectedly, their large grain size and
lack of ion exchange capacity led to poor
treatment efficacy of stormwater contaminated
with Bg and CsCI. Further evaluation of media for
catch basin inserts is needed to document
treatment options for biological and radiological
contaminants under storm drain hydraulics. On
the other hand, GAC has a proven history of
successfully treating pesticides [8, 9] and
malathion, an organo-phosphate pesticide,
performed well under the pilot-scale stormwater
inlet hydraulic conditions. Still, the feasibility of
inlet treatment with GAC requires further site-
specific analysis to determine its suitability. For
example, determining a media use rate is crucial
for a utility or emergency responder to understand
implementation logistics. A use rate is dependent
on the amount of malathion spilled, the surface

that is contaminated, and the climate (e.g., when
precipitation occurs and temperature).

To put the malathion data in perspective, consider
a 1-acre asphalt parking lot with a 5% slope in
Research Triangle Park, North Carolina that
experienced a nerve agent attack of 100 mg/ft2.
Assuming a 5% washoff rate and using the
experimental sorption capacity of 7.45 mg/g, a
minimum of 65.5 lbs. of GAC would be needed in
the inlet to treat the stormwater runoff. This is
congruent with what was feasible in pilot-scale
experiments. Next, a hydraulic understanding of
the capture efficiency of the inlet + media is
necessary to evaluate how much water would be
treated and if surface flooding would occur. For
the RTP parking lot, a total runoff volume of 1.06 x
106gallons is estimated using EPA's Storm Water
Management Model (SWMM) and 2024
precipitation data. Further, peak runoff was
estimated at 801 gpm (from a 1.77-inch event)
and an average runoff of 23 gpm. This indicates
that many storm events would have a portion of
water bypass the carbon treatment at the current
5 gpm testing conditions and that surface flooding
might occur. As such, holding tanks or a different
configuration of the media would need to be
considered. Performing modeling to estimate the
exceedances and bypasses at the specific location
is a key component to developing a disaster
management protocol involving the deployment
of media in stormwater inlets and helps answer if
this technology is suitable in a specific location.
This work demonstrates the stormwater inlet
capability at EPA's T&E facility that is available to
test novel media and/or flow regimes to continue
providing data to assist responders in preparing
for disaster.

Contact Information

Technical Contacts

Anne Mikelonis, mikelonis.anne@epa.gov
Jeff Szabo, szabo.ieff@epa.gov
Jim Goodrich, goodrich.iames@epa.gov
Josh Steenbock, steenbock.ioshua@epa.gov

General Feedback
CESER@epa.gov

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028

2025


-------
Disclaimer: This document has been reviewed in accordance with U.S. Environmental Protection Agency,
Office of Research and Development, and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.

References

1.	Pitt, R. and R. Field. An evaluation of storm drainage inlet devices for stormwater quality treatment.
in Water Environment Federation Technical Exposition and Conference. 1998. Orlando, FL.

2.	Morgan, R.A., et al. An evaluation of the urban stormwater pollutant removal efficiency of catch
basin inserts. Water environment research, 2005. 77(5): p. 500-510.

3.	Kostarelos, K., et al. Field study of catch basin inserts for the removal of pollutants from urban
runoff. Water resources management, 2011. 25(4): p. 1205-1217.

4.	Basham, D., et al. Design and construction of full-scale testing apparatus for evaluating performance
of catch basin inserts. Journal of Sustainable Water in the Built Environment, 2019. 5(1): p.
04018013.

5.	Alam, M.Z., et al. Characterising stormwater gross pollutants captured in catch basin inserts. Science
of The Total Environment, 2017. 586: p. 76-86.

6.	Alam, M.Z., et al. Improving stormwater quality at source using catch basin inserts. Journal of
Environmental Management, 2018. 228: p. 393-404.

7.	Sharma, S., H. Rathore, and S. Ahmed. Studies on removal of malathion from water by means of
activated charcoal. Ecotoxicology and environmental safety, 1987. 14(1): p. 22-29.

8.	Jusoh, H.H.W., et al. Granular activated carbon optimization for enhanced environmental disaster
resilience and malathion removal from agricultural effluent, in E3S web of conferences, 2023. EDP
Sciences.

9.	Kearns, J., E. Dickenson, and D. Knappe. Enabling organic micropollutant removal from water by full-
scale biochar and activated carbon adsorbers using predictions from bench-scale column data.
Environmental Engineering Science, 2020. 37(7): p. 459-471.

U.S. Environmental Protection Agency - Office of Research and Development EPA/600/S-25/-028

2025


-------