EPA/600/R-18/017 | February 2018
www.epa.gov/research
United States
Environmental Protection
Agency
&EPA
Persistence and Decontamination
of Radioactive Cesium-137 in a
Model Drinking Water System
Office of Research and Development
Western Ecology Division
-------
EPA/600/R-18/017
February 2018
Persistence and Decontamination of Radioactive
Cesium-137 in a Model Drinking Water System
by
Jeffrey Szabo
U.S. Environmental Protection Agency
Office of Research and Development
Homeland Security Research Program
Cincinnati, OH 45268
Ryan James
Battelle
Columbus, OH 43201
-------
DISCLAIMER
The U.S. Environmental Protection Agency (EPA) through its Office of Research and
Development funded and managed the research described herein under contract EP-C-15-010
with Battelle. Reene Watt served as the U.S. Environmental Protection Agency Contracting
Officer's Representative. 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
-------
TABLE OF CONTENTS
DISCLAIMER ii
EXECUTIVE SUMMARY vi
1.0 INTRODUCTION 1
2.0 SUMMARY OF PIPE DECONTAMINATION EXPERIMENTAL DESIGN 1
2.1 Experimental Reactor System 2
2.2 Persistence Evaluation Approach 3
2.3 Decontamination Approaches 5
2.3.1 Ammonium Chloride Chemical Cleaning Agent Evaluation 6
2.3.2 Potassium Chloride Chemical Cleaning Agent Evaluation 6
3.0 QUALITY ASSURANCE/QUALITY CONTROL 7
3.1 Experimental Controls 7
3.2 Measurement Methods 8
3.2.1 Liquid Scintillation Counting 8
3.2.2 Sodium Iodide Spectroscopy 8
3.2.3 Enumeration of Biofilm Growth 8
3.3 Quality Control 9
3.4 Audits 10
3.4.1 Technical Systems Audit 10
3.4.2 Data Quality Audit 10
4.0 RESULTS 11
4.1 Evaluation of Contaminant Persistence 11
4.2 Chemical Cleaning Agent Evaluations 14
4.2.1 Ammonium Chloride 14
4.2.2 Potassium Chloride 17
4.3 Biofilm Measurements 20
4.4 Water Quality Measurements 21
4.5 Summary and Conclusions 22
REFERENCES 24
iii
-------
LIST OF FIGURES
Figure 1. Annular reactor 2
Figure 2. Persistence evaluation for clean water exposure - average percent persistence (%P) of
Cs-137 by coupon type (Cu, PVC, concrete-lined [Con]); Sodium iodide spectroscopy (Nal)
method 12
Figure 3. Persistence evaluation for clean water exposure - percent persistence (%P) of Cs-137
of liquid scintillation counting (LSC) and sodium iodide spectroscopy (Nal) results on copper,
PVC, and concrete-lined (Con) coupons individually 13
Figure 4. NH4CI chemical cleaning agent evaluation - average (Ave) percent persistence (%P) of
Cs-137 by coupon type (Cu, PVC, concrete-lined [Con]); sodium iodide spectroscopy (Nal)
method 15
Figure 5. Copper coupons with green and blue precipitate from the reaction of NH4CI with the
copper coupons. Left photo: One day of exposure to NH4CI. Right photo: Four days of exposure
to NH4CI 15
Figure 6. Ammonium chloride chemical cleaning agent evaluation - percent persistence (%P) on
of Cs-137 on copper, PVC, and concrete-lined (Con) coupons; sodium iodide spectroscopy (Nal)
and liquid scintillation counting (LSC) methods 16
Figure 7. KC1 chemical cleaning agent evaluation - average (Ave) percent persistence (%P) of
Cs-137 by coupon type (Cu, PVC, and concrete-lined [Con]); sodium iodide spectroscopy (Nal)
method 18
Figure 8. Potassium chloride chemical cleaning agent evaluation - percent persistence (%P) on
of Cs-137 on copper, PVC, and concrete-lined (Con) coupons; sodium iodide spectroscopy (Nal)
and liquid scintillation counting (LSC) methods 20
Figure 9. Testing coupons with biofilm growth: concrete-lined (left), PVC (middle), and copper
(right) 21
LIST OF TABLES
Table 1. Experimental Steps of Persistence Evaluation 4
Table 2. Evaluation of Chemical Cleaning Agent as Decontamination Approach 6
Table 3. Experimental Controls 8
Table 4. Data Quality Objectives for Sodium Iodide and Liquid Scintillation Analyses 10
Table 5. Initial Cs-137 Activity Levels for the Persistence Evaluation 12
Table 6. Initial Cs-137 Activity Levels for the NH4C1 Cleaning 14
Table 7. Initial Cs-137 Activity Levels before Tap Water Introduction 17
Table 8. Initial Cs-137 Levels for the KC1 Cleaning 18
Table 9. Colony Forming Units Grown on Coupon Surfaces 21
iv
-------
LIST OF ABBREVIATIONS/ACRONYMS
AR annular reactor
ASTM ASTM International
AWWA American Water Works Association
BSTC BioSurface Technologies Corporation
cm centimeter
CFU colony forming units
CCA chemical cleaning agent
Cs-137 cesium-137
DI deionized water
DPM disintegrations per minute
°C degrees Celsius
EPA U.S. Environmental Protection Agency
ft/s foot/second
HPC heterotrophic plate counts
in. inch
KC1 Potassium chloride
L liter
LSC liquid scintillation counting
|iCi microcuries
|iL microliter
mg milligram
min minute
mm millimeter
mL milliliter
molar M
NaISS sodium iodide spectroscopy
NH4CI Ammonium chloride
NHSRC National Homeland Security Research Center
%P percent persistence
PE persistence evaluation
PVC polyvinyl chloride
QA quality assurance
QC quality control
rpm revolutions per minute
s second
S Siemen
TTEP Testing and Evaluation Program
UL Underwriters Laboratories
v
-------
EXECUTIVE SUMMARY
The objective of this study was to evaluate the persistence of radioactive cesium-137 (Cs-
137), on concrete-lined, copper, and polyvinyl chloride (PVC) pipe, and possible
decontamination approaches. During this study, conditions within operational drinking water
pipes were simulated using annular reactors (ARs) (i.e., ring-shaped reactors). The ARs consist
of a glass outer cylinder and a rotating polycarbonate inner cylinder with 20 flush mounted
rectangular slides (referred to as coupons) that are made of the aforementioned pipe materials
with biofilm. Shear was applied to the coupon surfaces by setting the reactors' inner cylinder
rotation to 100 revolutions per minute, which produces shear forces similar to those produced
from 1 ft/sec flow in a 6-inch pipe.
The study was initiated with an evaluation of Cs-137 persistence. Pipe material
coupons with biofilm were contaminated with Cs-137 for 24 hours and then exposed to fresh
tap water in annular reactors operating at 100 rpm for up to five days. Exposure to fresh tap
was meant to simulate flushing of water pipes in a real distribution system. The results of the
persistence evaluation showed that exposure to clean drinking water reduced the adhered Cs-
137 levels by 75% for the copper pipe coupons, and 91% and 93% for the PVC and concrete
pipe coupons, respectively.
After exposure to fresh tap water, a decontamination evaluation with two separate
chemical cleaning agents was performed to remove the remaining Cs-137 adhered to the
coupons. The pipe materials were exposed to either a solution containing 1 molar (M)
ammonium chloride (NH4CI) or a 1 M potassium chloride (KC1) solution, as cleaning agents.
The reactors were operated at 100 rpm for up to six days after application of the cleaning
agent. Decontamination with NH4CI removed 89% of the remaining Cs-137 activity from
the copper coupons. However, there was variability in the decontamination results on the
copper coupons, possibly due to a precipitate that formed during decontamination. The
precipitate was likely corrosion of the copper piping material that resulted from exposure to
NH4CI. The NH4CI cleaning treatment on PVC and concrete piping materials removed 91%
and 85%) of the remaining adhered Cs-137, respectively. Decontamination with the KC1
solution removed 50% of the Cs-137 activity remaining on the copper coupons, and 90% of
the Cs-137 activity remaining on both the PVC and concrete coupons.
vi
-------
It should be noted that when Cs-137 activity removal from both exposure to clean water
(simulated flushing) and decontamination with the cleaning agents are added together, 99% of
the activity was removed from PVC and concrete, regardless of which agent was applied or the
duration of clean water exposure after contamination. Total Cs-137 activity removal from
copper ranged from 97% removal with 5 days of clean water exposure and NH4CI treatment, to
89%) removal with one day of clean water exposure and KC1 treatment. These results suggest that
KC1 may be a better decontamination chemical compared to NH4CI since the decontamination
performance is comparable, and KC1 does not corrode the copper pipe. However, a noted earlier,
exposure to clean water (simulated flushing) alone reduced Cs-137 levels by 75% for the copper
pipe coupons, and 91% and 93% for the PVC and concrete pipe coupons. Since most of the
adhered Cs-137 was removed via clean water exposure, flushing alone may be a sufficient
decontamination method.
vii
-------
1.0 INTRODUCTION
Contamination of drinking water distribution systems can occur following an intentional
contamination incident, an industrial accident, a massive main break or a natural disaster. Natural
and man-made incidents are further exacerbated by the declining integrity of the aging water
infrastructure in the United States. Decontamination of drinking water distribution systems is
critical for the effective and rapid return of the system to operation, and for restoring the safety
of the water for drinking and other applications. The ability to reliably and cost effectively
decontaminate miles of distribution system pipes and premise plumbing will be a critical
capability that utilities will need to ensure public safety following a contamination incident.
Research is needed to better understand the adherence and persistence of selected
contaminants on pipe walls, and to identify effective decontamination methods. In recent years,
research studies have been conducted to determine the adsorption of chemical, biological, and
surrogate (non-radioactive) radiological contaminants to various drinking water pipe materials,
and to test various methods to destroy, reduce or remove adsorbed contaminants 1'2. However,
data on the persistence of radioactive species on common distribution system and premise
plumbing pipe material are lacking. The following report attempts to address this data gap by
examining the persistence of radioactive Cs-137 on copper, polyvinyl chloride (PVC) and
cement-mortar, which are materials commonly used in home plumbing (copper and PVC) and
drinking water distribution pipes (cement-mortar lined iron). Decontamination of Cs-137 is also
examined using flushing with clean water, followed by pipe cleaning with 1 M ammonium
chloride (NH4CI) or 1 M potassium chloride (KC1).
2.0 SUMMARY OF PIPE DECONTAMINATION EXPERIMENTAL DESIGN
One of the most important features in the experimental protocol is the use of an annular
reactor (AR) (i.e., ring-shaped reactor) as the device that simulates flow through drinking water
pipe, which is represented by coupons (excised samples) of various materials. Shown in Figure
1, the AR simulates pipe flow with a variable speed motor that drives an inner rotating cylinder.
The rotation of this cylinder provides surface shear between pipe surface coupons and water
within the AR.
-------
Figure 1. Annular reactor.
ARs are commercially available, providing ease of repeatability across laboratories, as opposed
to requiring the fabrication of a specialized setup.
The contaminant used during testing was radioactive Cs-137 (Cs-137), applied in
separate experiments on three simulated pipe surfaces (concrete-lined, polyvinyl chloride [PVC],
and copper).
2.1 Experimental Reactor System
For the persistence and decontamination experiments, the conditions within operational
drinking water pipes were simulated in ARs (Model 1320 LS Standard Laboratory AR,
BioSurface Technologies Corporation, Bozeman, MT). The ARs consist of a glass outer
cylinder and a rotating polycarbonate inner cylinder with 20 flush mounted rectangular coupons
that can be manufactured from materials such as concrete, PVC, and copper and obtained from
BioSurface Technologies Corporation (BSTC). For the concrete-lined pipe, the specifications of
the concrete used for the coupons meet the requirements of the CI 50-07 ASTM Standard
Specification for Portland Cement3. The thickness of the concrete was approximately 1.3
millimeters (mm), which is slightly less than 1.6 mm as specified in American Water Works
Association (AWWA) C104-03 Standard for Cement-Mortar Lining for Ductile-Iron Pipe and
Fittings for Water4. For the PVC pipe, the coupons carry an Underwriters Laboratories (UL) 94
2
-------
V-0 rating and conform to the D-1784 ASTM standard specification for Rigid PVC Compounds
and Chlorinated PVC Compounds5. The copper coupons are made of solid copper.
These coupons, which have surfaces that are 14 mm x 148 mm, simulated the inner
surface of drinking water pipes. Shear stress was applied to the coupon surfaces by setting the
inner cylinder rotation to 100 revolutions per minute (rpm), which produces shear similar to 30.5
centimeter (cm)/second (s) (1 foot [ft]/s) flow in a 15.2 cm (6 inch [in.]) pipe6.
Prior to any persistence or decontamination experiments, a biofilm was grown on the
surface of the coupons. Biofilm was formed by submerging 21 coupons (for each AR used) of
the applicable surface material in a 20 L reservoir containing recirculating dechlorinated tap
water (at approximately 200-300 milliliter [mL]/ minute [min]) and 1 gram of yeast extract for
seven days. Biofilm was formed from the native microorganisms in the dechlorinated tap water.
The coupons were kept in the dark during the biofilm formation period. Following the 7-day
biofilm formation period, growth was measured (procedure described in Section 3.2.3) using
heterotrophic plate counts (HPC) to obtain the concentration of colony forming units (CFU)/mL
from one of the coupons.
2.2 Persistence Evaluation Approach
The persistence evaluation (PE) approach for concrete-lined, PVC, and copper surface
coupons is summarized in Table 1. For each of the three experiments (one for each pipe
material), biofilm was grown on 21 coupons. One coupon of each surface type was used for
HPC analysis to confirm biofilm growth. At the conclusion of the biofilm formation period,
background measurements were taken. Two coupons with biofilm growth were measured (non-
destructive^) directly using an EG&G Ortec Maestro Sodium Iodide Spectroscopy System
(NaISS), and one coupon was measured (destructively) using liquid scintillation counting (LSC).
Once counting was complete, the two coupons measured by NaISS and the remaining coupons
were placed in the ARs. The reactors were contaminated with a bulk solution of Cs-137 at a
concentration of 100 microcuries per liter (|iCi/L) for the ARs containing PVC and copper, and
10 |iCi/L in the AR containing the concrete-lined coupons. The ARs were set to continuously
rotate at 100 rpm for 24 hours in contact with the Cs-137.
The initial level of Cs-137 used in the ARs was determined from small scale beaker
experiments with pipe material coupons prior to testing. This was to ensure there would be
3
-------
enough Cs-137 signal for the analytical measurements on all three pipe materials, but also to
minimize Cs-137 dose. The results of these initial experiments showed that the concrete-lined
coupons absorbed the Cs-137 contamination much more readily than PVC and copper, as
evidenced by gamma counts that were several orders of magnitude higher than the other two
surface types. Therefore, during the experiments, the concrete contamination solution was
prepared and used at a concentration 10 times lower than the contamination solution used for the
PVC and copper pipe materials.
Table 1. Experimental Steps of Persistence Evaluation (PE)
PE Step
Description
Net
Coupons
Consumed
PE 1
Develop a biofilm on 21 coupons. Confirm the biofilm with
heterotrophic plate count (HPC) on one coupon, leaving 20 for use in
experiments.
1
PE 2
Remove two coupons from the AR to serve as uncontaminated blanks
(NaISS) and one coupon for LSC analysis. Measure the background
activity on these coupons and return the NAISS coupons to the AR.
1
PE 3
Add Cs-137 contamination solution to AR, sample the bulk
contamination solution; wait 24 hrs with AR rotation at 100 rpm.
0
PE 4
Following the 24 hr exposure to the Cs-137 solution, sample the solution
and remove 4 coupons. Measure two coupon using LSC and the other
two using NaISS. Return the NAISS coupons to the AR (these two
coupons were measured by NaISS each day and returned to AR).
2
PE 5
Replace the contamination solution with fresh drinking water and allow
coupons to reside in 100 rpm AR for 24 hrs; sample the water and
remove one or two coupons and measure the activity with a LSC.
Remove NaISS coupons, measure activity and return to AR. This step
was repeated four additional times.
10
PE 6
Calculate percent persistence (%P) for PE coupons by comparing to
contaminated control coupons.
0
Following 24 hrs of exposure to Cs-137, the contamination solution was sampled and
drained. The same two coupons from each AR previously measured by NaISS were measured
again to establish the contamination activity remaining on the coupons. Being a nondestructive
method, the same two coupons were measured by NaISS each day and returned to the ARs. Two
contamination coupons were also removed from the ARs and their activity measured by LSC.
The ARs were then filled with deionized (DI) water and drained to ensure the contamination
solution was removed from the ARs. Then they were filled with fresh drinking water and
4
-------
rotation of the AR at 100 rpm was restarted (Day 1). After 24 hrs of exposure to drinking water,
the two NalSS-designated coupons were removed from the AR and their activity measured by
NalSS. Again, two coupons were removed for LSC analysis. This procedure was repeated daily
for five days; however, because of the limited number of coupons available for duplicate
samples, only one coupon was removed for LSC analysis during Days 3 to 5. The activity of the
water was also measured at every daily exchange. These measurements were accomplished by
transferring 1 mL aliquots of water from the AR into a vial and measuring the activity using the
LSC.
Following the activity measurements of the coupons removed from the AR, activity from
the surface was compared with the activity from the contaminated coupons measured after
contamination with Cs-137. This comparison was made by calculating the percent persistence
(%P) of the contaminant on the coupons as described by the following equation.
A pe
o/oP = -f± X 100
Ac
Where Ape is the activity of the contaminant recovered from the PE coupon surface and Ac is the
average activity of the contaminant originally measured from the surfaces of the contaminated
control coupons. Note that any background activity was subtracted from Ape and Ac.
2.3 Decontamination Approaches
With Cs-137 persisting on the surface of the pipe materials after drinking water exposure,
decontamination experiments were conducted. The ARs were spiked with either ammonium or
potassium chloride to remove the residual Cs-137. The chemical cleaning agent evaluation was
performed as shown in Table 2, and used the coupons remaining in the AR following the PE.
5
-------
Table 2. Evaluation of Chemical Cleaning Agent (CCA) as Decontamination Approach
Step
Description
Net
Coupons
Consumed
CCA 1
Drain the AR and add the CCA solution. Wait 24 hrs with AR rotation
at 100 rpm.
0
CCA 2
Following 24 hrs of exposure to the CCA solution, sample the CCA
solution, remove one or two coupons and measure their activity with the
LSC. Remove two NaISS coupons, measure, and return them to the AR.
Replace the CCA solution with fresh CCA solution and continue AR
rotation at 100 rpm.
1
CCA 3
Repeat the step above until no Cs-137 is detected on the surface of the
coupon, or until no coupons remain. Collect a CCA solution sample
from the AR with each coupon.
6
2.3.1 Ammonium Chloride Chemical Cleaning Agent Evaluation
At the start of the experiment, the drinking water in the AR was replaced with a chemical
cleaning agent consisting of 1 molar (M) ammonium chloride (NH4CI), and the inner cylinder of
the AR was set to rotate at 100 rpm. The chemical solution remained in contact with the surface
of the coupons for 24 hrs, after which point the AR was drained and one coupon was removed
and its activity counted using a LSC. Additionally, the same two coupons were removed from
the AR and counted using the NaISS, then returned to the AR. The AR rotated at 100 rpm for
the duration of the experiment and was only stopped to remove coupons and exchange the
chemical. The process of draining the AR, removing three coupons (one measured via LSC and
two measured via NaISS), and then refilling every 24 hrs was repeated for four days for the
NH4CI solution. Each time a coupon was removed from the AR, a sample of the NH4CI solution
in the AR was also collected. The NaISS coupons were returned the AR after measurement.
The %P was calculated using these results in the same manner as for the PE.
2.3.2 Potassium Chloride Chemical Cleaning Agent Evaluation
As during the NH4CI experiment, the coupons in the ARs had biofilms cultivated and the
CFU/coupon was calculated for each pipe material. Background measurements were taken and
then the coupons were contaminated with a DI water solution of Cs-137 at a concentration of 100
microcuries per liter (|iCi/L) for the ARs containing PVC and copper, and 10 |iCi/L in the AR
containing the concrete-lined coupons. The rotation was set for 100 rpms and started for 24
hours. The contamination solution was sampled and drained. The same two coupons from each
6
-------
AR previously measured by NaISS were measured again to establish the contamination activity
on the coupons. Being a nondestructive method, the same two coupons were measured by
NaISS each day and returned to the ARs. Two contamination coupons were also removed from
the ARs and their activity measured by LSC. The ARs were filled with deionized (DI) water and
drained to ensure the contamination solution was removed from the ARs. Then they were filled
with fresh drinking water and rotation of the AR at 100 rpm was restarted (Day 1 Tap Water).
After 24 hrs of exposure to drinking water, the two NalSS-designated coupons were removed
from the AR and their activity measured by NaISS. Again, two coupons were removed for LSC
analysis.
The drinking water in the AR was then replaced with a chemical solution consisting of 1
M KC1 and the AR was set to rotate at 100 rpm. This chemical solution remained in contact with
the surface of the coupons for 24 hrs. After 24 hours, the AR was drained, two coupons were
removed, and their activity counted using a LSC (Day 1 KC1). Additionally, the original two
coupons were removed from the AR and counted using the NaISS, then returned to the AR. The
AR rotated at 100 rpm for the duration of the test and was only stopped to remove coupons and
exchange the chemicals. The process of draining the AR, removing four coupons (two measured
via LSC and two measured via NaISS), and then refilling every 24 hrs was repeated for five days
for the copper and PVC pipe materials and six days for the concrete-lined coupons. Each time a
coupon was removed from the AR, a sample of the KC1 solution in the AR was collected. The
%P was calculated using these results in the same manner as for the PE.
3.0 QUALITY ASSURANCE/QUALITY CONTROL
3.1 Experimental Controls
Table 3 summarizes the controls included in the experiments. The controls are important
because the results of the persistence and decontamination experiments are dependent on the
original concentration of contaminant on the surface of the control coupons. In addition, the AR
has more available surface area that could become contaminated, thus becoming a source of
secondary contamination.
7
-------
Table 3. Experimental Controls
Type of Control
Description
Non-contaminated
control coupons blanks
Taken from set of coupons following biofilm growth; allows comparison
of coupons that have been contaminated from the bulk solution and those
that have not
Contaminated control
coupons
Removed from AR after contamination of coupons, but before PE testing
in clean water; allows comparison of coupons containing a "diminished"
amount of contaminant (due to the shear from rotating within drinking
water) with coupons containing "total" amount of contaminant (not
influenced by the shear from rotating within drinking water)
3.2 Measurement Methods
3.2.1 Liquid Scintillation Counting
The analytical method used for the Tri-Carb 3110TR Low Activity (Perkin-Elmer,
Waltham, MA) liquid scintillation analyzer is based on EPA Method 900.0 "Gross Alpha and
Gross Beta Radioactivity in Drinking Water"7 Cs-137 was used as calibration standards and
calibration followed the manufacturers guidelines. The method for liquid analysis was very
straight forward as 1 mL of each sample was added to a LSC vial containing 9 mL of Bio-Safe
LSC II cocktail (Research Products International, Mount Prospect, IL), placed in the auto
sampler and programmed for analysis. The method for the coupon analysis was similar;
however, the coupons were cut into thirds, placed into the LSC vial with 18 mL of the Bio-Safe
II LSC cocktail, and programmed for analysis. The analysis took 1 minute per sample and the
results were presented in disintegrations per minute (DPM).
3.2.2 Sodium Iodide Spectroscopy
The analytical method used for the NaISS (Maestro, EG&G Ortec, Oak Ridge, TN) was
based on EPA Method 901.1 "Gamma emitting Radionuclides in Drinking Water"8. Cs-137 was
used as calibration standards and calibration followed the manufacturers guidelines. Non-
destructive analysis was performed by placing the surface coupons within the detection chamber
in a common geometry. Counting took place for 15 minutes.
3.2.3 Enumeration ofBiofilm Growth
This method was used to determine the extent of biofilm growth on the coupons. A sterile
0.05% solution (by volume) of Triton X-100 (Fisher # BP151-500, Fisher BioReagents,
8
-------
Waltham, MA) in phosphate buffered saline was prepared. The coupons were placed in sterile
centrifuge tubes (VWR # 89004-364, VWR, West Chester, PA) containing 30 mL of the 0.05%
Triton X solution, mixed using a vortex mixer, and then decanted. Thereafter, two tenfold
dilutions of that decanted solution were prepared using the 0.05% Triton X solution as the
diluent. Each of those solutions were plated in triplicate by dispensing 100 microliter ([J.L) onto
agar plates (Teknova #T0134, Teknova, Hollister, CA). Using a spreader, the aliquot was evenly
distributed on the plate and placed in an incubator for 48 hours at 35-37 degrees Celsius (°C).
After incubation, the distinguishable colonies on each plate were counted. In order to be
considered a viable plate count, the number of colonies on a given plate should be between 30
and 300. If the number of colonies was below 30, a more concentrated dilution was re-plated in
triplicate. If the number of colonies was more than 300, a less concentrated dilution was re-
plated in triplicate. The CFU/coupon was obtained by dividing the average number of colonies
by the plated volume and then adjusting for the dilution factor.
3.3 Quality Control
Quality control samples for the contaminant reference method included continuing
calibration checks and laboratory blanks. The data quality objectives for each of these quality
control (QC) samples are provided in Table 4. The acceptable ranges limit the error introduced
into the experimental work. Both instruments operated within the QC requirements for
continuing calibration checks and laboratory reagent blanks during this evaluation.
9
-------
Table 4. Data Quality Objectives for Sodium Iodide
and Liquid Scintillation Analyses
Method
Sample Type
QC Requirement
Corrective Action
Liquid
Scintillation
Counter (LSC)
Continuing
calibration/efficiency
check
90-110% of known; daily
Repeat sample analysis; if
still outside of range
repeat calibration
Laboratory reagent
blank
-------
At least 10% of the data acquired during the evaluation were audited. These data include
the biofilm measurements, water quality measurements, LSC and NaISS measurements. The
Battelle QA manager traced the data from the initial acquisition, through analysis, to final
reporting, to ensure the integrity of the reported results. All calculations performed on the data
undergoing the audit were checked. No significant adverse findings were noted in this audit.
4.0 RESULTS
Note that data for both the NaISS and LSC measurement techniques are presented in this section.
However, for simplicity, percent persistence and discussion of decontamination effectiveness of
the NaISS data is presented herein. Note that any background activity was subtracted from all
samples presented in this section.
4.1 Evaluation of Contaminant Persistence
The persistence evaluation results are presented in Figure 2 for each coupon surface type.
This graph presents the average result of the two NaISS measurements for the same two coupons
on each testing day. The vertical axis (%P) show the percent of the Cs-137 remaining on the
coupon after each experimental day (shown on the horizontal axis). The columns at the far-left
side of the graphs (pre-decontamination) represent the initial contamination level (as measured
on the contaminated control coupons) and each successive column represents the number of days
of exposure to clean water.
The persistence evaluation results show a similar initial decrease in Cs-137 of
approximately 60-70% for all coupon types. The residual Cs-137 on the copper coupons
continued to decrease through Day 3, but ceased decreasing thereafter, remaining at
approximately 25 %P. The PVC and concrete coupons continued to decrease each day,
culminating at a %P of just less than 10% on the last two days of testing. These calculations
were based on the initial activity levels on each coupon. The starting activity levels (in counts for
the NaISS and DPM for the LSC) for each type of coupon are shown in Table 5.
11
-------
100%
fa L L
¦ Cu Nal
¦ PVC Nal
100%
100%
40%
38%
32%
21%
24%
15%
27%
10%
25%
9%
¦ Con Nal
100%
33%
18%
11%
9%
7%
Day 3
24%
15%
11%
Day 4
27%
10%
9%
Day 5
25%
9%
7%
Figure 2. Persistence evaluation for clean water exposure - average percent persistence
(%P) of Cs-137 by coupon type (Cu, PVC, concrete-lined [Con]); Sodium iodide
spectroscopy (Nal) method.
Table 5. Initial Cs-137 Activity Levels for the Persistence Evaluation
Analysis Method
Coupons
Copper
PVC
Concrete
NaISS (counts)
1,870 Cul
15,364 PVC1
58,233 Conl
1,793 Cu2
8,838 PVC2
57,057 Con2
LSC (DPM)
5,441 Cu3
36,901 PVC3
221,694 Con3
4,638 Cu4
46,148 PVC4
180,811 Con4
Con. concrete-lined; DPM, disintegrations per minute; LSC, liquid scintillation counting;
NaISS, sodium iodide spectroscopy
These trends are shown in more detail in the graphs in Figure 3. The individual NaISS
and LSC results are presented for the different coupon types separately. Note that the NaISS
results are from daily measurements of the same two coupons and the LSC results are
measurements of different coupons exposed to the same contamination and decontamination
conditions. Because of the limited space for replicate coupons in the AR and to ensure enough
coupons were available for the decontamination evaluation, only one coupon was removed for
LSC analysis on some days (no purple bar is present on days 3 to 5 in Figure 3). Overall, the
NaISS and LSC show similar trends. Based on the NaISS data in Figure 2, flushing with clean
water during the persistence evaluation reduced the contamination level by approximately 75%
12
-------
for the copper pipe material, and 91% and 93% for the PVC and concrete pipe materials,
respectively.
PE Results for PVC Coupons
100%
80%
60%
? I l i J ,u -i
Pre-decon
Day 1
Day 2
Day 3
Day 4
Day 5
¦ PVC1 Nal
100%
39%
23%
16%
11%
9%
¦ PVC2 Nal
100%
36%
20%
13%
8%
8%
¦ PVC3 LSC
100%
121%
48%
32%
18%
24%
¦ PVC4 LSC
100%
58%
27%
PE Results for Copper Coupons
100%
80%
lui.
Pre-decon
Day 1
Day 2
Day 3
Day 4
Day 5
¦ Cul Nal
100%
39%
32%
17%
27%
26%
¦ Cu2 Nal
100%
41%
32%
31%
27%
24%
¦ Cu3 LSC
100%
64%
43%
42%
41%
28%
¦ Cu4 LSC
100%
56%
57%
PE Results for Concrete Coupons
Pre-decon
Day 1
Day 2
Day 3
Day 4
Day 5
¦ Conl Nal
100%
26%
11%
6%
4%
3%
¦ Con2 Nal
100%
41%
24%
17%
13%
10%
¦ Con3 LSC
100%
20%
12%
4%
4%
3%
¦ Con 4 LSC
100%
23%
10%
Figure 3. Persistence evaluation for clean water exposure - percent
persistence (%P) of Cs-137 of liquid scintillation counting (LSC) and sodium
iodide spectroscopy (Nal) results on copper, PVC, and concrete-lined (Con)
coupons individually.
13
-------
4.2 Chemical Cleaning Agent Evaluations
4.2.1 Ammonium Chloride
Directly after the persistence evaluation, the NH4CI chemical cleaning experiment was
conducted. Results are presented in Figure 4 for each coupon surface type. This graph presents
the average result of the two NaISS measurements for the same two coupons on each testing day.
The bars to the far-left side of the graphs present the initial contamination level for this portion
of the testing, which began one day after the final day (Day 5) of the PE experiment. The ARs
were not re-spiked with Cs-137 before this portion of the testing, and the coupons sat overnight
between the PE and cleaning phases of the experiment. Therefore, the amount of activity on the
coupons at the beginning of this experiment (Table 6) was less than what remained on Day 5 of
the persistence evaluation. The remaining bars represent the %P of Cs-137 after exposure to the
NH4CI chemical solution noted along the horizontal axis.
The NH4CI cleaning results for copper and concrete coupons had a similar initial Cs-137
activity decrease of approximately 75%. The Cs-137 was more persistent on the PVC coupons
in the NH4CI solution, decreasing by 50%; however, after four days, the PVC coupons had the
least amount of contamination. Percent P was calculated from the initial activity levels presented
in Table 6 (the activity present at Day 5 of the PE experiment).
Table 6. Initial Cs-137 Activity Levels for the NH4C1 Cleaning
Analysis Method
Coupons
Copper
PVC
Concrete
NaISS (counts)
502 Cul
1,031 PVC1
1,515 Conl
444 Cu2
575 PVC2
4,771 Con2
LSC (DPM)
1,725 Cu3
6,446 PVC3
4,103 Con3
DPM, disintegrations per minute; LSC, liquid scintillation counting; NaISS, sodium iodide
spectroscopy
14
-------
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Cu Nal Ave
I PVC Nal Ave
Con Nal Ave
Day 1
31%
50%
22%
Day 2
24%
38%
22%
Day 3
11%
18%
Day 4
11%
9%
15%
Figure 4. NH4CI chemical cleaning agent evaluation - average (Ave) percent persistence
(%P) of Cs-137 by coupon type (Cu, PVC, concrete-lined [Con]); sodium iodide
spectroscopy (Nal) method.
After the first day of exposure to the NH4CI
chemical cleaning solution, the bulk solution in the AR
with the copper coupons was a brilliant blue. After
draining the AR, it became clear that the NH4CI was
reacting with the copper to form a blue/green precipitate
throughout the AR, and especially on the copper
coupons. Because of this precipitate, the copper coupons
were extremely difficult to remove from the AR slots
and, therefore, the NaISS coupons were unable to be
removed and measured on Day 3. Figure 5 shows two
pictures of the copper coupons with the precipitate at
Day 1 and Day 4 respectively. The wet precipitate looked
blue, as did the bulk solution. Once the coupons were
removed from the AR and placed into the drying rack,
the precipitate appeared green.
The results of the NH4CI chemical cleaning are
shown in more detail in the graphs in Figure 6. The individual NaISS and LSC results are
presented for the different coupon types separately.
Figure 5. Copper coupons with
green and blue precipitate from the
reaction of NII4CI with the copper
coupons. Left photo: One day of
exposure to NH4CI.
Right photo: Four days of exposure
to NH4CI.
15
-------
NH4CI Cleaning Results for Copper Coupons
Pre-decon
Day 1 NH4CI
Day 2 NH4CI
Day 3 NH4CI
Day 4 NH4CI
NH4CI Cleaning Results for PVC Coupons
1UU/0
90%
80%
70%
6°%
S 50%
40%
30%
20%
10%
0%
¦
Day 1
NH4CI
Pre-decon
Day 2 NH4CI
Day 3 NH4CI
Day 4 NH4CI
¦ PVC1 Nal
100%
54%
40%
4%
1%
¦ PVC2 Nal
100%
46%
36%
17%
17%
¦ PVC3 LSC
100%
22%
22%
7%
7%
NH4CI Cleaning Results for Concrete Coupons
Pre-decon
Day 1 NH4CI
Day 2 NH4CI
Day 3 NH4CI
Day 4 NH4CI
¦ Conl Nal
100%
17%
18%
16%
15%
¦ Con2 Nal
100%
27%
25%
19%
16%
¦ Con3 LSC
100%
22%
14%
15%
10%
Figure 6. Ammonium chloride chemical cleaning agent evaluation - percent
persistence (%P) on of Cs-137 on copper, PVC, and concrete-lined (Con)
coupons; sodium iodide spectroscopy (Nal) and liquid scintillation counting
(LSC) methods.
16
-------
In general, the NaISS and LSC results agree and show similar trends; however there appears to
be more variability in the results compared to the PE results, possible due to the precipitate. The
precipitate or corrosion could cause physical damage to a home plumbing system with copper
piping. The %P following 4 days of NH4CI treatment on the copper, PVC and concrete piping
materials was 11%, 9% and 15% (89%, 91% and 85% removal), respectively, based on the Nal
measurements in Figure 4. However, when adding together the Cs-137 activity removal
(measured by NaISS) achieved during exposure to clean water for 5 days (persistence
evaluation) and the NH4CI cleaning experiment, total removals of 97%, 99% and 99% for
copper, PVC and concrete were observed. This demonstrates that flushing with clean water will
remove 75-93% of the Cs-137 activity (depending on pipe material), but chemical cleaning is
needed to obtain more removal.
4.2.2 Potassium Chloride
As described in section 2.3.2, the coupons were exposed to clean tap water for 24 hours
after spiking with Cs-137 before the cleaning with KC1 began. Table 7 presents the initial NaISS
and LSC activity measurements for each coupon type immediately after spiking. After exposure
to clean water for 24 hrs, the average %P on the copper coupons was 20% for the NaISS (12%
for LSC) results. For the PVC coupons, the average %P with the tap water exposure was 11 %
for the NaISS (9% for LSC) results. Finally, the average %P for the concrete coupons with tap
water exposure was 19% for the NaISS (11% for LSC) results.
Table 7. Initial Cs-137 Activity Levels before Tap Water Introduction
Analysis Method
Coupons
Copper
PVC
Concrete
NaISS (counts)
2,806 Cul
2,573 PVC1
39,242 Conl
2,930 Cu2
2,390 PVC2
71,847 Con2
LSC (DPM)
8,458 Cu3
14,984 PVC3
307,417 Con3
8,947 Cu4
9,149 PVC4
238,010 Con4
Con. concrete; DPM, disintegrations per minute; LSC, liquid scintillation counting; NaISS,
sodium iodide spectroscopy
After exposure to clean water for 24 hrs, the KC1 solution was added for the following 5
days (6 days for the concrete coupons). Cs-137 activities on the coupons at the beginning of the
17
-------
KC1 cleaning experiment are presented in Table 8. These activity levels are what remained on
the coupons after exposure to clean water for 24 hrs. Figures 7 and 8 shows the %P (vertical
axis) results for Cs-137 decontamination using the 1 M KC1 solution on the three coupon surface
types. The bars to the far-left side of the graph present the residual activity after 24 hours of
exposure to clean tap water. The remaining bars along the horizontal axis represent the %P of
Cs-137 after exposure to the KC1.
Table 8. Initial Cs-137 Levels for the KC1 Cleaning
Analysis Method
Coupons
Copper
PVC
Concrete
NaISS (counts)
598 Cul
271 PVC1
9,085 Conl
562 Cu2
267 PVC2
10,775 Con2
LSC (DPM)
1,171 Cu3
794 PVC3
13,731 Con3
865 Cu4
1,189 PVC4
39,975 Con4
DPM, disintegrations per minute; LSC, liquid scintillation counting; NaISS, sodium iodide spectroscopy
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
LlLL
Predecon Day 1KCI Day 2 KCI Day3KCI Day4KCI Day 5 KCI
¦ Cu Nal Ave
100%
81%
50%
34%
59%
54%
¦ PVC Nal Ave
100%
33%
8%
10%
14%
12%
¦ Con Nal Ave
100%
12%
10%
9%
8%
9%
Results after Day 2 are at or below the detection limits for the PVC coupons.
Figure 7. KCI chemical cleaning agent evaluation - average (Ave) percent
persistence (%P) of Cs-137 by coupon type (Cu, PVC, and concrete-lined
[Con]); sodium iodide spectroscopy (Nal) method.
The KCI cleaning results show initial decreases in Cs-137 activity of approximately 19%
for copper, 67% for PVC, and 88% for concrete coupons with the first 24 hours of KCI exposure
18
-------
(%P of 81%, 33% and 12% respectively). The Cs-137 persisted on the copper coupons at
approximately 50% P for the remainder of the experiment. The PVC and concrete coupons
decreased to roughly 10 %P at 2 days after introduction of the KC1 solution, and remained at that
level for the duration of the experiment.
These trends are shown in more detail in the graphs in Figure 8. The individual NaISS and LSC
results are presented for the different coupon types separately. Overall, the NaISS and LSC
results show similar trends, although there is some variation in the copper coupon results.
KCI Cleaning Results for Copper Coupons
100%
80%
60%
40%
20%
¦ llJj
U/O
Predecon
Day 1 KCI
Day 2 KCI
Day 3 KCI
Day 4 KCI
Day 5 KCI
¦ Cul Nal
100%
85%
47%
30%
64%
52%
¦ Cu2 Nal
100%
76%
54%
38%
53%
57%
¦ Cu3 LSC
100%
52%
96%
38%
45%
70%
¦ Cu4 LSC
100%
159%
115%
78%
80%
101%
KCI Cleaning Results for PVC Coupons
100%
U/o
Predecon
Day 1 KCI
Day 2 KCI
Day 3 KCI
Day 4 KCI
Day 5 KCI
¦ PVCl Nal
100%
32%
4%
23%
1%
18%
¦ PVC2 Nal
100%
34%
11%
0%
27%
5%
¦ PVC3 LSC
100%
38%
14%
7%
3%
4%
¦ PVC4 LSC
100%
17%
10%
5%
2%
14%
19
-------
KCI Cleaning Results for Concrete Coupons
Predecon Day 1 KCI Day 2 KCI Day 3 KCI Day 4 KCI Day 5 KCI Day 6 KCI
¦ Conl Nal
100%
14%
16%
15%
14%
15%
13%
¦ Con2 Nal
100%
10%
4%
4%
3%
2%
2%
¦ Con3 LSC
100%
8%
9%
9%
3%
3%
9%
¦ Con 4 LSC
100%
5%
4%
1%
1%
1%
1%
Figure 8. Potassium chloride chemical cleaning agent evaluation - percent persistence
(%P) on of Cs-137 on copper, PVC, and concrete-lined (Con) coupons; sodium iodide
spectroscopy (Nal) and liquid scintillation counting (LSC) methods.
In conclusion, the KCI solution removed approximately 90% of the Cs-137 activity (10 %P)
from the PVC and concrete pipe materials (based on the NaISS results in Figure 7). For PVC
and concrete, the 90% decrease in Cs-137 activity occurred one day after introduction of the
KCI, and the activity levels were stable thereafter. This contrasts with the NH4CI cleaning
solution results where Cs-137 activity decreased continuously over 4 days. The copper coupons
retained approximately 50 % of the contamination after 5 days of being exposed to the KCI
solution, although % P fluctuated between 34 and 60% (NaISS results, Figure 7) over the
decontamination period. It should be noted that when adding together the Cs-137 activity
removal achieved during exposure to clean water for 24 hrs (persistence evaluation) and the KCI
cleaning, total removals of 89%, 99% and 99% for copper, PVC and concrete, respectively, were
observed. This demonstrates that flushing with clean water will remove 80-89% of the Cs-137
activity (with one day of clean water exposure, depending on pipe material), but chemical
cleaning is needed to obtain more removal.
4.3 Biofilm Measurements
20
-------
Biofilm was grown and counted on the three types ol
coupons before the experiments began. Figure 9 is a photo
of the coupons removed from the biofilm recirculating
system after 7 days. Table 9 presents the HPC results for
each coupon type. Biofilm grew on all three coupons types,
with concrete and PVC returning similar results and copper
holding the least biofilm on the coupon surface. The CFU
per coupon ranged from 6.3E+04 to 6.6E+07. There was a
span of time between the biofilm measurement and the Cs-
137 contamination that allowed for continued biofilm
formation, which was not measured. The coupon CFU
count for the persistence evaluation was measured on
November 11, 2016 and testing began seven days later on
November 18, 2016. Similarly, the CFU on the coupons
were measured on February 17, 2017 for the KC1 cleaning and testing began 10 days later on
February 27, 2017. The coupons remained in the dark in the biofilm recirculating system until
the first day of testing.
Table 9. Colony Forming Units Grown on Coupon Surfaces
Coupon Type
CFU/Coupon
11/11/2016
2/17/2017
Copper
5.1E+05
6.3E+04
PVC
3.3E+07
6.6E+07
Concrete
2.5E+07
1.7E+07
4.4 Water Quality Measurements
Throughout this evaluation the pH, conductivity, and free and total chlorine
concentrations of the tap water were all measured on a batch basis. The pPI of the tap water was
on average 7.5 ± 0.26 with an average conductivity of 566 ± 56 Siemen (S)/cm. The average
free chlorine concentration of the tap water was measured to be 0.17 ± 0.39 milligram per liter
(mg/L).
Figure 9. Testing coupons with
biofilm growth: concrete-lined
(left), PVC (middle), and copper
(right).
21
-------
4.5 Summary and Conclusions
The objective of this study was to evaluate the persistence of radioactive Cs-137 on
concrete-lined, copper, and PVC pipe surfaces, and possible chemical decontamination
approaches. During this study, conditions within operational drinking water pipes were
simulated using ARs. Concrete-lined, copper, and PVC coupons (samples) with biofilm growth
were used as representative pipe materials. Radioactive Cs-137 was used to contaminate the
ARs, and its persistence was calculated after exposure to clean water and decontamination with
two chemical solutions (in separate experiments).
Persistence results showed that continued exposure to clean tap water reduced the
adhered Cs-137 activity level by 75% for the copper pipe material, and 91% and 93% for the
PVC and concrete pipe materials, respectively. When 1 M NH4CI was used to decontaminate the
coupons, there appeared to be more variability in the results on the copper coupons compared to
the other pipe material tested, possibly due to a precipitate that formed. The precipitate, or
corrosion on the copper piping material, may cause physical damage to a home plumbing water
system. The %P following 4 days of NH4CI treatment on the copper, PVC and concrete piping
materials was 11%, 9% and 15% (89%, 91% and 85% removal). However, when adding
together the Cs-137 activity removal (measured by NaISS) achieved during exposure to clean
water for 5 days (persistence evaluation) and the NH4CI solution, total removals of 97%, 99%
and 99% for copper, PVC and concrete, respectively, were observed.
The 1 M KC1 chemical cleaning results show that Cs-137 activity remained persistent on
copper coupons at 50 %P after five days of KC1 exposure, although % P fluctuated between 34
and 60%. There was no damage to the copper coupons from a precipitate or corrosion when
using KC1. The KC1 solution removed approximately 90% of the remaining Cs-137 activity (10
%P) from the PVC and concrete pipe materials. For PVC and concrete, the 90% decrease in Cs-
137 activity occurred one day after introduction of the KC1, and the activity levels were stable
thereafter. This contrasts with the NH4CI cleaning results where Cs-137 activity decreased
continuously over 4 days. It should be noted that when adding together the Cs-137 activity
removal achieved during exposure to clean water for 24 hrs (persistence evaluation) and the KC1
solution, total removals of 89%, 99% and 99% for copper, PVC and concrete, respectively, were
observed.
22
-------
These results suggest that KC1 would be a better decontamination agent compared to
NH4CI since the decontamination performance is comparable across materials, KC1 acts faster on
PVC and concrete, and KC1 does not corrode the copper pipe. However, it should also be
considered that exposure to clean water alone removed 75% of the Cs-137 activity on copper
pipe material, and 91% and 93% of the activity adhered to PVC and concrete pipe materials,
respectively. When using a decontamination chemical, these removal totals increased to 97%
(NH4CI) and 89%) (KC1) on copper, and 99% for all other pipe material and cleaning agent
combination. Following a Cs-137 contamination event in a water system, responders and
decision makers may want to consider whether flushing with clean water alone is an adequate
decontamination measure.
23
-------
REFERENCES
1. Szabo, J. G., E. W. Rice, and P. L. Bishop. 2006. Persistence of Klebsiella pneumoniae on
simulated biofilm in a model drinking water system. Environ. Sci. Technol. 40:4996-5002.
2. Szabo, J.G., Persistence and decontamination of surrogate radioisotopes in a model drinking
water distribution system, Water Research (2009), doi:10.1016/j.watres.2009.08.012.
3. ASTM C 150-07 Standard, 2007, "Standard Specification for Portland Cement," ASTM
International, West Conshohocken, PA, www.astm.org.
4. American Water Works Association C104-03 Standard, 2004, "Standard for Cement-Mortar
Lining for Ductile-Iron Pipe and Fittings for Water" Denver, CO, www.awwa.org.
5. Welter, G., M. Lechevallier, S. Spangler, J. Cotruvo, R. Moser, Guidance for
Decontamination of Water System Infrastructure. Denver, CO: AWWA Research
Foundation, 2009.
6. ASTM Standard D1784, 2011, " Standard Specification for Rigid Poly(Vinyl Chloride)
(PVC) Compounds and Chlorinated Poly (Vinyl Chloride) (CPVC) Compounds," ASTM
International, West Conshohocken, PA, www.astm.org.
7. U.S. EPA, EMSL. "Method 900.0: Gross Alpha and Gross Beta Radioactivity in Drinking
Water." Prescribed Procedures for Measurement of Radioactivity in Drinking Water,
EPA/600/4/80/032, 1980.
8. U.S. EPA, EMSL. "Method 901.1: Gamma Emitting Radionuclides in Drinking
Water." Prescribed Procedures for Measurement of Radioactivity in Drinking Water,
EPA/600/4/80/032, 1980.
-------
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
------- |