www.epa.gov/or
United States
Environmental Protection
Agency
LabLogic Radiation Detection
Online Water Quality
Monitoring System
Office of Research and Development
National Homeland Security Research Center
-------
EPA 600-R-13-228
May 2014
LabLogic Radiation Detection Online
Water Quality Monitoring System
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
26 Martin Luther King Drive
Cincinnati, OH 45268
-------
Acknowledgments
Contributions of the following individuals and organizations to perform the necessary equipment
installation, testing, evaluation, and the preparation of this document are acknowledged.
United States Environmental Protection Agency (EPA)
John Hall, NHSRC/ORD
Dan Mackney NAREL /OAR
Shaw Environmental & Infrastructure, Inc. (Shaw)
Srinivas Panguluri
Greg Meiners
Vendor (LabLogic Systems, Inc.) support during the installation phase is also acknowledged.
in
-------
Disclaimer
The U.S. Environmental Protection Agency (EPA), through its Office of Research and
Development's (ORD) National Homeland Security Research Center (NHSRC) collaborated
with EPA Office of Air and Radiation's (OAR) National Analytical Radiation Environmental
Laboratory (NAREL) to perform this evaluation. A portion of this evaluation was also
performed at the EPA's Test and Evaluation (T&E) Facility, funded by EPA NHSRC through
Contract No. EP-C-09-041 with Shaw Environmental & Infrastructure, Inc. (Shaw). This report
has been peer and administratively reviewed by the Agency and approved for publication as an
EPA document, but it does not necessarily reflect the Agency's views. Mention of trade names
or commercial products does not constitute endorsement or recommendation for use of a specific
product.
Questions concerning this document or its application should be addressed to:
John Hall
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
26 West Martin Luther King Dr.
Cincinnati, OH 45268
513-487-2814
hall.john@epa.gov
11
-------
Contents
Disclaimer ii
Acknowledgments iii
Contents iv
Abbreviations/Acronyms vi
Executive Summary vii
1.0 Introduction 1
1.1 Background 2
2.0 Technology and Testing Description 4
2.1 The Modified P-RAM Scintillation-Counting System 4
2.2 Installation and Testing at EPA NHSRC - Cincinnati 7
2.3 P-RAM Operating Cost Summary 9
2.4 Installation at EPA NAREL Laboratory -Montgomery, Alabama 10
2.5 P-RAM Calibration and Testing at EPA NAREL 11
3.0 Experimental Results 12
3.1 Background Determination 12
3.2 Sr-90 Test Run Results 13
3.3 Cs-137 Test Run Results 14
3.4 Am-241 Test Run Results 15
3.5 H-3 Test Run Results 16
3.6 Online Sample Cross Contamination Check 18
3.7 Event-based Automated Sampling 18
3.8 Overall Results Summary 18
4.0 Conclusions and Recommendations 20
4.1 Recommended System Enhancements and Future Testing 20
5.0 References 22
2
Appendix A - Dilute H Measurements Using a 5 mL Wilma Cell and ScintLogic LB - March
2012 (Reprinted with Permission -LabLogic Systems, Inc.) 23
iv
-------
Figures and Tables
Figure 1. Model 4 P-RAM scintillation-counting system and Walter fluid handling/ distribution
unit 4
Figure 2. Flow paths in Walter unit and the P-RAM system 5
Figure 3. Typical Wilma software operating cycle 6
Figure 4. P-RAM system initial (raw count) data 8
Figure 5. P-RAM system data (raw count) after replacement of the tubing 9
Table 1. P-RAM System Background Output (Counts) Water Analysis Summary 13
Table 2. P-RAM System Output (Counts) for Sr-90 (1,120 pCi/L) Water Analysis Summary .. 14
Table 3. P-RAM System Output (Counts) for Cs-137 (1,110 pCi/L) Water Analysis Summary 14
Table 4. P-RAM System Output (Counts) for Am-241 (2,480 pCi/L) Water Analysis Summary
15
Table 5. P-RAM System Output (Counts) for Am-241 (1,130 pCi/L) Water Analysis Summary
16
Table 6. P-RAM System Output (Counts) for H-3 (740 pCi/L) Water Analysis Summary 17
Table 7. P-RAM System Output (Counts) for H-3 (74 pCi/L) Water Analysis Summary 17
Figure 6. Overall Summary of P-RAM system testing. Counts (y-axis) versus test runs 19
-------
Abbreviations/Acronyms
P-RAM
Bq
°C
cm
cpm
DHS
dpm
EPA
FoM
HPLC
keV
mL
L
LAN
LLS
MCA
Modbus
PAG
PC
pCi
PMT
NAREL
NHSRC
O&M
OAR
ORD
QAPP
mrem
TA
T&E
USB
Beta RAM - Online Water Quality Radiation Monitor
Bequerel
degrees Celsius
centimeters
counts per minute
U.S. Department of Homeland Security
disintegrations per minute
U.S. Environmental Protection Agency
Figure of Merit
High Performance Liquid Chromatography
kilo electron volts
Milliliter
Liter
Local Area Network
LabLogic Systems, Inc.
Multi-Channel Analyzer
A serial communications protocol originally published by Modicon (now
Schneider Electric)
Protective Action Guide
Personal Computer
picocuries
Photomultiplier Tube
National Analytical Radiation Environmental Laboratory
National Homeland Security Research Center
Operations and Maintenance
Office of Air and Radiation
Office of Research and Development
Quality Assurance Project Plan
Milli-Roentgen Equivalent Man (units of radiation dose)
Technical Associates, Inc.
Test and Evaluation
Universal Serial Bus
Radionuclides
Am-241
Cs-137
H-3 or"
Sr-90
Y-90
'H
Americium-241
Cesium-137
Tritium
Strontium-90
Yittrium-90
VI
-------
Executive Summary
Detecting the presence of radiological
substance in drinking water is important
both from a consumer safety and a national
security perspective. The EPA's Office of
Research and Development /National
Homeland Security Research Center
(NHSRC) routinely acquires and evaluates
commercially available radiation detection
systems for water. This report summarizes
the information collected by EPA NHSRC
during the testing and evaluation of the P-
RAM scintillation-counting system
manufactured by LabLogic Systems, Inc.
(LLS).
The P-RAM Model 4 scintillation-counting
system (a.k.a. P-RAM) was procured by
EPA NHSRC and first operationally
evaluated at the EPA Test and Evaluation
(T&E) Facility in Cincinnati, Ohio.
Subsequently, the unit was tested with
radioactive materials at the EPA National
Analytical Radiation Environmental
Laboratory (NAREL) Facility in
Montgomery, Alabama.
The initial operational evaluation at the EPA
T&E Facility indicated that the system (after
some initial modifications) can be operated
and maintained by a typical water treatment
plant technician. The system was able to
function in a somewhat-rugged test
environment that is representative of a
typical drinking water monitoring site where
the unit can be potentially deployed. The
key focus of the testing at EPA NAREL was
to determine the system's ability to detect
radiation activity in water at levels near the
Department of Homeland Security's
Protective Action Guide (PAG) level, which
is based on a drinking water interdiction of
500 mrem/year. This 500 mrem/year dose
rate converts to the following PAG levels
for the isotopes tested: (1) Tritium H-3 -
4,540,000 pCi/L, (2) Strontium-90 (Sr-90) -
6,730 pCi/L, (3) Cesium-137 (Cs-137) -
13,800 pCi/L, (4) Americium-241 (Am-241)
- 908 pCi/L. The actual injected levels were
all well below PAG levels with the
exception of Am-241, which was slightly
above PAG level. The actual injected
activity levels are as follows: Sr-90 (1,120
pCi/L), Cs-137 (1,100 pCi/L), Am-241
(2,480 pCi/L), Am-241 (1,310 pCi/L), H-3
(740 pCi/L), and H-3 (74 pCi/L). Testing at
the NAREL Facility indicated that the P-
RAM system can be used as an online real-
time radiation monitor (response time < 30-
minutes).In its current state of development,
the P-RAM with the accompanying Wilma
software system demonstrated the ability to:
(1) detect contamination at the injected
levels (as stated above), (2) discharge a
sample for retention, and (3) provide an
alarm to the user.
The results of the testing also indicate that it
may be possible to detect lower-levels of
radioactivity in water than those used in
these tests. Section 2.0 of this document
presents an overview of the technology and
describes the operational evaluation and
testing. The detailed NAREL radioactivity
testing results are presented in Section 3.0.
Section 4.0 provides a summary of
conclusions and a listing of recommended
system enhancements for future testing.
vn
-------
1.0 Introduction
Detecting the presence of radiological
substance in drinking water is important
both from a consumer safety and national
security perspective. In some parts of the
U.S., groundwater naturally contains
elevated levels of radionuclides. For
example, radium-226 occurs at an elevated
level in the groundwater of the North-central
states (Zapecza and Szabo 1987, EPA 2000).
Similarly, uranium is found at elevated
levels in the groundwaters of the Colorado
plateau, the Western-central platform, the
Rocky mountain system, and the Pacific
mountain system (Zapecza and Szabo 1987,
EPA 2000). From a national security
perspective, pursuant to the Fukushima
nuclear plant disaster in Japan in March
2011, the U.S. Environmental Protection
Agency (EPA) temporarily stepped up
routine radiological monitoring of the
drinking water. The results (released on
April, 2011) from the monitoring of
drinking water indicated that only two of the
monitored locations detected radiation in
drinking water samples, but the values were
well below the levels of public-health
concern. Another potential security incident
occurred in December 2013, when a vehicle
containing radioactive cobalt was stolen, but
recovered by Mexican authorities (Romo et
al., 2013). Looking at these scenarios
collectively, one can say that the ability to
measure the presence of radiological
substance in drinking water in near real-time
is important for safeguarding public health
and national security. While a variety of
equipment is available to detect the presence
of radiation in air, the options for water are
somewhat limited. To fill the gaps in
technologies for keeping drinking water
safe, the EPA's Office of Research and
Development (ORD)/National Homeland
Security Research Center (NHSRC)
routinely acquires and evaluates such
commercially available radiological
detection systems.
In 2008, the EPA, Office of Air and
Radiation (OAR), National Analytical
Radiation Environmental Laboratory
(NAREL) committed to work with the EPA
NHSRC to evaluate the performance of
selected radiation detection systems,
purchased by NHSRC, relative to
manufacturer-supplied performance criteria.
The testing described in this report focuses
on the ability of the system to accurately
monitor and detect radioactive materials in
drinking water. After development and
acceptance of a proposal to execute the
desired tests, a Quality Assurance Project
Plan (QAPP) was developed and approved
in April 2009 to perform the tests with well
defined objectives and goals (EPA 2009,
Amended EPA 2012). The P-RAM
scintillation-counting system is
manufactured by LabLogic Systems, Inc.
(LLS) located in Brandon, Florida. The P-
RAM Model 4 scintillation-counting system
was procured by EPA and first operationally
evaluated at the EPA Test and Evaluation
(T&E) Facility in Cincinnati, Ohio.
Subsequently it was tested with radioactive
substances at the EPA NAREL Facility in
Montgomery, Alabama. The tests to
evaluate operational suitability at the T&E
Facility were conducted in an environment
that is representative of a typical water
distribution system. Following several
months of operational evaluation at the T&E
Facility, the P-RAM system was delivered to
NAREL. The radiological testing facilities
at NAREL allowed for testing the radiation
detection capabilities of the system by
-------
introduction of radioactive material at
designated activity levels. The reported
activity level by the P-RAM system and its
accuracy were monitored and evaluated.
This document describes the tests performed
at both testing facilities and provides some
additional information regarding the radio-
analytical theory and functional
requirements helpful for review and
understanding of the test results.
1.1 Background
Initially, LLS held discussions with EPA
NHSRC regarding the system specifications
and desired operational capabilities.
Subsequently, representatives from LLS and
EPA NHSRC arrived at NAREL with LLS's
existing commercially available P-RAM
scintillation-counting system. This original
P-RAM system is a bench-top offline flow
cell scintillation detection system which is
normally coupled with a High Performance
Liquid Chromatography (HPLC) unit, as
employed by radiopharmaceutical
companies and oncology facilities to detect
basic tracer isotopes such as tritium (H-3),
carbon-14, and phosphorus-32.
The LLS personnel demonstrated the
standard configuration and operation of their
original P-RAM system and requested
EPA's input for necessary modifications
required to convert this system into an
online radiation detection system suitable
for deployment in a water distribution
system. EPA suggested that the system meet
the following criteria:
1) The system should be able to be
operated by typical water
treatment plant operators and
technicians.
2) Maintenance of the system should
be simple and allow for routine
replenishment of supplies or
checking of instrument responses
without placing the instrument in
an "out-of-service" condition for
long periods (greater than 1 hour)
and should not require any
specialized training.
3) The system should be able to endure
the rugged environment of a
typical pump station facility.
This includes temperature
variation and humidity levels that
are often elevated.
4) A multi-channel analyzer (MCA)
should be incorporated into the
system with appropriate
software. The MCA should be
capable of providing information
to qualified individuals regarding
the nature of the ionizing
radiation: alpha decay or beta
decay. The MCA should be
available as an optional add-on
device or upgrade.
5) A means of calibrating and
checking the operational
readiness of the instrument
should be established, allowing
typical plant operators the ability
to check operational readiness of
the system.
6) The system should be capable of
displaying the detected level of
radiation and the time stamp; and
should provide an alarm at user-
defined activity levels to a
remote location.
7) The alarm set point should be
adjustable to an activity level
determined by the user based on
a site-specific background count
rate.
8) The system should have the ability
to automate the collection of the
water sample that created the
alarming condition for
verification analysis.
-------
9) The system should accommodate
user-selected and adjustable
count times with a maximum
polling interval of 30 minutes.
10) The system should have the ability
to be operated continuously
online as a flow-through sensor
without human intervention
except for alarm response,
reagent replacement, or normal
calibration and for Operations
and Maintenance (O&M)
requirements.
NAREL and NHSRC both agreed that
NHSRC would purchase and test the system
after LLS completed the system
modifications that would be required to meet
the aforementioned criteria. The testing
would follow the procedures previously
established for the Technical Associates,
Inc. (TA) Model SSS-33-5FT water
monitoring system (EPA 2009, EPA 2011).
A QAPP addendum for testing of the P-
RAM system was prepared and approved
prior to commencing the tests (EPA 2012).
In August of 2011, EPA received the
modified P-RAM system. Prior to delivery
of the system to EPA, LLS had tested the
system for tritium detection. These results
are included in Appendix A, Dilute H [H-
3]Measurements Using a 5 mL Wilma Cell
and ScintLogic LB.
-------
2.0 Technology and Testing Description
During decay, radioactive compounds emit
ionizing radiation in the form of gamma rays
and/or alpha or beta, particles. One type of
technology often used for detecting
radioactive compounds in liquid samples is
scintillation. A scintillator is a material that
exhibits scintillation (luminescence) when
excited by ionizing radiation. These
materials, when struck by ionizing radiation,
absorb its energy and scintillate (i.e., re-emit
the absorbed energy in the form of light).
The emitted light is often detected by using
an electronic sensor such as a
photomultiplier tube (PMT). The P-RAM
device described in this report uses a
proprietary scintillation liquid cocktail and a
PMT to detect radioactive compounds
present in water. The modified P-RAM
device (as tested by EPA) employs a flow-
through cell with a radio HPLC detector
combination. The features and performance
of the testing related to the P-RAM system
is described further in this report.
2.1 The Modified p-RAM Scintillation-
Counting System
Figure 1 shows a picture of the modified P-
RAM scintillation-counting system as tested
by EPA.
Figure 1. Model 4 P-RAM scintillation-counting system and Walter fluid handling/
distribution unit.
This system employs three separate
modules: (1) Walter - a fluid handling and
distribution unit (shown at the bottom of
Figure 1), (2) P-RAM - a scintillation-
counting system (shown at the top of Figure
®
1), and (3) a Lynx multi-channel analyzer
(MCA) from the Canberra Corporation,
Meriden, Connecticut (not shown in Figure
1). The entire system is controlled by
Wilma software, a computer program that
controls pump operation and valve
switching. The sample, scintillant, and wash
fluid flow paths set in the Walter module are
visualized through Wilma software and
illustrated in Figure 2.
-------
a Wilma - [Current Cyde]
-•'.. File Vie.v Measurement Tools Window Help
J i j* •* ' LJ :i& d ^ i J
-•
'
.
,
'| IkjkA ' l\ & '-lUSMJoJ ' ! 1 '
Valve |V6)
F*et(Fl] An Pump Check Valve
[ . .
Water Filter (F2) Sensor (S1)
(54)
{
Valve £v"l J
Standard Fillet (F 3)
Pump |P2|
Valve (V2)
|F41
Out
Pump (PI 1
Vatve (V9|
H^a^m*
BetafiAM
Scintilanl
Purrp(P3)
MixngTee
Waste LSCVial Vent Wast
Figure 2. Flow paths in Walter unit and the P-RAM system.
There are several preset operating cycles
included with the Wilma software. Each
operating cycle consists of a timed-sequence
of events that the instrument performs to
complete the measurement cycle. These
cycles are incorporated into methods. The
methods are essentially electronic files
containing the prescribed sequence of
instrument operation placed in file folders
on the computer hosting the Wilma software
for sets of common analyses. Both NHSRC
and NAREL methods were set by LLS for
the testing conducted at each facility.
Though different time cycles were employed
at each facility, they are all saved under the
respective facility-specific methods on the
PC hosting the Wilma software. A typical
cycle on the Wilma software is shown in
Figure 3. This cycle chart appears on the PC
screen during each cycle operation and
provides the user with a direct indication of
the current state of the cycle, the time
remaining in that portion of the cycle, and
the total elapsed time in the cycle.
-------
Event List
Configure the list of events
Event List
Name
Stint Pump
Trigger
Display Text
Wait
Display Text
Display Text
Count
Stop Count
Display Text
Empty
Close Vial Valve
Text
Wash
Empty again
Wash Again
Final Empty
.
Text
4
Time
00:00:00.000
00:02:45,000
00:00:00.000
00:00:00.000
00:00:00.000
00:00:00,000
00:07:00.000
00:00:00.000
00:00:00.000
00:00:00.000
00:10:00.000
00:00:00,000
00:00:00.000
00:00:00.000
00:00:00.000
00:03:30,000
00:00:00.000
00:00:00.000
00:00:00.000
00:00:00.000
00:00:00.000
00:00:00.000
00:01:05.000
00:00:00,000
00:00:00,000
00:00:00.000
00:00:00.000
00:01:30.000
00:00:00.000
00:00:00.000
00:00:00.000
00:00:00.000
00:01:05.000
00:00:00,000
00:00:00.000
00:00:00.000
00:00:00.000
00:02:30.000
00:00:00.000
00:00:00.000
00:00:00,000
00:00:00,000
Action
Set Device Status
Set Device Status
Set Device Status
On Threshold Run
Display on LCD [line 4)
Prepare Chromatograph
Wait
Display on LCD (line 4)
Display on LCD [line 1)
Start Chromatograph
Finish Chromatograph
Display on LCD (line 1)
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Display on LCD Oine 1)
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Set Device Status
Display on LCD Oine 1)
nil
Device
Pump 3
Pump 3
Valve 9
Threshold Exc
Valve 6
Valve 7
Pump 4
Pump 4
Valve 8
Valve 6
Valve 7
Valve 1
Pump 2
Pump 2
Valve 1
Valve 6
Valve 7
Pump 4
Pump 4
Valve 6
Valve 7
Valve 1
Pump 2
Pump 2
Valve 1
Valve 6
Valve?
Pump 4
Pump 4
Valve 6
Pump 1
Valve 7
Status
Run
Stop
Close
Open
Open
Run
Stop
Close
Close
Close
Open
Run
Stop
Close
Open
Open
Run
Stop
Close
Close
Open
Run
Stop
Close
Open
Open
Run
Stop
Close
Stop
Close
Value
7,000 ml/min
if > 10000.000 cp
7,000 ml/min
7,000 ml/min
7.000 ml/min
7,000 ml/min
7. 000 ml/min
1
Description
Wait for m
Counting Sample...
Emptying Cell...
Washing cell...
Waiting.,.
i
A.
T
Click Next to continue.
Figure 3. Typical Wilma software operating cycle.
-------
2.2 Installation and Testing at EPA
NHSRC - Cincinnati
In September 2011, NHSRC received the
Model 4 P-RAM system at the EPA T&E
Facility in Cincinnati. On September 14,
2011, the LLS representatives arrived at the
T&E Facility to set up the system and
provide operational instructions to the
NHSRC and Shaw Environmental &
Infrastructure, Inc. project staff.
Operational set-up was fairly
straightforward. Both Walter and the P-
RAM modules require a Universal Serial
Bus (USB) connection to the PC on which
the Wilma software resides. The Canberra
Lynx MCA module requires either a USB or
Local Area Network (LAN) connection to
the PC. In addition, there are other
interconnecting cables that connect these
devices. A number of small tubes are used
as fluid lines in the Walter module to draw
water from the sample source, the wash fluid
source (50% mix of water and methanol),
the scintillation fluid source, and associated
sample capture/drain lines. The pumps and
valves located in the Walter module perform
all the operations to deliver and discharge
fluid from the P-RAM module. The P-RAM
module has only one set of fluid lines
connecting to and from the flow cell
chamber used for measuring the radiation
activity. As mentioned previously, the
radiation activity is measured via a
scintillation counter, which is an instrument
for detecting light/photon counts. The
Scintillator generates photons of light in
response to incident radiation; these photon
counts are cascaded using a PMT. The
measured cascaded photon counts are
referred to as "raw counts."
At the T&E Facility, the three components
of the radiation detection system, i.e., the
Model 4 P-RAM scintillation-counting
system unit, the Walter fluid handling and
distribution unit, and the Wilma software,
were operational. Although the Canberra
Lynx MCA hardware was provided by LLS,
the software license for utilizing the MCA
was initially not provided to EPA. During
the initial discussions, the LLS
representatives informed EPA that they were
in the process of developing their own MCA
and wanted to avoid the cost related to
acquiring the Canberra software until
necessary. As no radiological testing was
proposed to be conducted at the EPA T&E
Facility, LLS agreed that either the Canberra
software or the alternative LLS product (if
developed and ready) would be provided
when the unit was ready for shipment to the
EPA NAREL laboratory for radiological
testing. Meanwhile, a dummy MCA graphic
spectrum was provided as a place-holder in
the software setting. At the completion of
the initial setup in Cincinnati, the only
available interpretable water quality output
from the system was the raw counts.
After the installation of the system was
deemed complete, EPA proceeded to
evaluate the first three criteria previously
identified as Operational Parameters 1
through 3 in Section 1.1 of this report. The
potential option for evaluating Operational
Parameter 6 (i.e., displaying the alarm and
other sample information to a remote
location) was also discussed during the
initial meeting. The specific options
discussed to facilitate Operational Parameter
6 included: (1) file transfer via the existing
LAN, and (2) using a standard online
instrument data transfer protocol such as
(R) (R)
Modbus . (The Modbus protocol is a
serial communications protocol originally
published by Modicon, which is now owned
by Schneider Electric, Palatine, Illinois.)
The LLS representative advised that, under
normal operations, the LAN card should be
connected to the Canberra Lynx MCA. Any
future data transfer option would either
-------
require a serial Modbus protocol or
additional LAN card to be installed on the
Wilma-PC for interfacing with the Lynx
MCA. As the current LAN port was not
utilized for Lynx connectivity, to potentially
test for simple file data transfer, the PC
(loaded with Wilma software) was enabled
to utilize the LAN for network access so that
data could be transferred to a remote
location.
While the initial evaluation was underway at
the T&E Facility, it was noted during
September that there were several days
when the instrument raw counts spiked
intermittently from the background levels
between 0 and 10 to over 100 (Figure 4).
LLS representatives were contacted via
email about this problem. After some
troubleshooting efforts, LLS determined that
these peaks were the result of light leaks and
suggested covering the tubes connecting
Walter unit and the P-RAM scintillation-
counting system with black electrical tape.
Light leaks increase the background photon
counts, which is unrelated to the tested
radiation activity present in the sample. The
black electrical tape was applied and the
system was continually monitored during
October 2011.
Water Activity «o,
180-
170-
160-
150
140-
130-
120
110-
» 100
I 9°
0 SO-
70-
60-
50-
40
30-
20-
10-
D3D
,1
P — Unknown
P — Unknown
(v* — Unknown
[^ — Unknown
|7 — Unknown
|7 — Unknown
M — Unknown
& — Unknown
]*? — Unknown
|7 — Start Up
9/12/2011 9/14/2011 9/16/2011 9/18/2011 9GO/2Q11 9/22/2011 9/24/2011 9/26/2011 9/28/2011
Date & Time
BBIter
X Axis scale C
late -r
From 9:00:00 AM ;f:'
To 5:30:00 PM \
On 8/23/2011 UK
On 9/29/2011 O"
Figure 4. P-RAM system initial (raw count) data.
In November 2011, EPA informed LLS that
the P-RAM module was not communicating
with the Wilma software and the module
display was malfunctioning. Subsequently,
LLS serviced the system by installing and
upgrading the software. The signal processor
was re-seated which corrected the display
malfunction. In addition, LLS provided
documentation for serial Modbus interface
to collect data remotely. Implementing the
Modbus serial interface on the existing data
logger at the T&E Facility required some
upgrades; hence, the Modbus
implementation was temporarily delayed.
An additional system malfunction (bad
display) was observed in December 2011.
-------
In response, LLS shipped a replacement
module. However, the periodic spikes
continued to be observed in the data. LLS
provided EPA with new black Teflon-coated
tubing (see Figure 1) to eliminate the light
leaks, which significantly improved the data
quality as shown in Figure 5. As a
final/permanent resolution, LLS redesigned
the unit with stainless steel interconnecting
tubes to eliminate the light leaks. The
stainless steel interconnecting tubes were
received and installed in January 2012.
Water Activity J£
250
240
230
220
210
200
90
80
70
60
50
40
30
20
10
00
90
« BO
1 70
U 60
50
40
30
20
10
0
-10-
-20
-30
-40
-SO
-60
-70
-80
-90
-100
-110
-120
1
lj*v^w-s-— s^
, (Lr&^Ar*
K/T_fL-w-jl_r*^-
A^^VJ_ ^
-•-" "- — "
2/15SQ11 12/16/2011 12/16/2011 12/17/2011 12/17/2011 12/18/2011 12/18/2011 12/19E011 12/1912011
Date 8 Time
(7 - Unknown
K7 - Unknown
R - Unknown
17 - Unknown
!7 - Unknown
17 - Unknown
!7 - Unknown
17 - Unknown
17 - SlartUp
Figure 5. P-RAM system data (raw count) after replacement of the tubing.
The design modification of using stainless
steel interconnecting tubes resulted in the
elimination of periodic light leaks. As
shown in Figure 5, the normal background
counts were reduced close to single digits
and remained relatively flat without any
background spikes in the raw data when
compared to the previously shown Figure 4.
The reagent/fluid use was monitored and
replaced as needed. After over a month of
stable background raw count observation
and trouble-free operation, it was
determined now the system was in a form
that could be operated and maintained by a
typical water treatment plant technician.
Subsequently, the unit was prepared for
shipment to the NAREL laboratory for
radiological testing.
2.3 P-RAM Operating Cost Summary
Besides the nominal electric use of the
system and the associated PC (which were
not monitored), the main operational costs
were the scintillant reagent and the methanol
-------
for preparing the wash water. The unit, as
tested at the EPA T&E Facility, was
configured to use these fluids at ~ 5 ml/min
and the cycle time/use varied with the
overall sampling rate. The unit was
programmed to sample once every half hour
and it used approximately one gallon per
week of lab grade methanol (i.e., one case of
methanol per month at a cost of ~
$361/month) and one gallon of scintillant
reagent per month (~ $195/month). The
scintillant (Flowlogic U from LLS, Brandon,
Florida) used for this testing is considered
non hazardous per the MSDS sheets. The
total monthly cost ($556) is based on the 5
ml/min and !/2 hour cycle times as tested in
Cincinnati at the EPA's T&E facility. Once
the unit was fully operational, it did not
require much oversight other than the
weekly (methanol)/monthly (scintillant)
replacement of the fluids. The fluid
replacement frequency could be decreased
by using fluid storage containers greater
than 1 gallon. The exact consumption rate of
the fluids would depend upon the
operational input requirements. For
example, to obtain cleaner signals, the wash
cycle between measurements could be
doubled leading to increase in methanol use
and the monthly operational costs. The cost
data presented above were based on the
reagent prices quoted at the time of testing
in the year 2011.
2.4 Installation at EPA NAREL
Laboratory - Montgomery,
Alabama
In May of 2012, NHSRC transferred the
system to NAREL for radiological testing.
LLS representatives assisted EPA NAREL
in setting up and initializing the system for
operation. Several challenges resulted from
the storage and transfer of the system from
EPA Cincinnati operations to EPA NAREL,
which are listed below:
1. The instrument had to be cleaned
out. Several weeks of non-operation
had resulted in bio-fouling of the
instrument lines.
2. The instrument sample pump had
failed as a result of bio-fouling,
sealing the micro pump passages.
The pump was replaced with a larger
pump.
During the enabling process of the MCA, it
was observed that the MCA spectrum signal
was low compared to the P-RAM output.
Subsequent troubleshooting activities
discovered that the third-party MCA chosen
by LLS (Canberra - Lynx model) required a
pre-amplifier when used in conjunction with
the instrument's PMT output. The
configuration the manufacturer uses with
their own brand of scintillation detectors
incorporates a pre-amplifier into the output
of the PMT. However, at the time this
testing was conducted, LLS did not have a
pre-amplifier in-line with the output signal
to the MCA. Therefore, to improve the Lynx
MCA output to match the P-RAM output,
the gain setting of the MCA was changed
from a "times 4 setting" to a "times 256
setting." By increasing the gain setting, a
sufficient increase in pulse count was
achieved to allow for a distinction between
background and sample activity. LLS is
incorporating a modification with the next
generation system to place the MCA
internally into the P-RAM and to have a pre-
amplifier included. The radiological testing
of the system at NAREL began in August
2012.
10
-------
2.5 P-RAM Calibration and Testing at
EPA NAREL
EPA NAREL proceeded to test the system's
capabilities previously identified as
Operational Parameters 4 through 10 in
Section 1.1 of this report. The key focus of
the testing was to determine the system's
ability to detect radiation activity in water at
levels near the Department of Homeland
Security's (DHS) Protective Action Guide
(PAG) level, which is based on a drinking
water interdiction of 500 mrem/year (DHS
2008). This 500 mrem/year dose rate
converts to the following PAG levels for the
isotopes tested: (1) Tritium H-3 - 4,540,000
pCi/L, (2) Strontium-90 (Sr-90) - 6,730
pCi/L, (3) Cesium-137 (Cs-137) - 13,800
pCi/L, (4) Americium-241 (Am-241) - 908
pCi/L.
Testing for radiological response consisted
of the following steps: 1) ensuring that
background values are stable, 2) determining
the response to a calibration source material
(Sr-90), and 3) introducing prepared
radioactive solutions of known activities and
determining the response relative to the
calibration and measured background level.
The solutions of the following four
radioisotopes were employed: H-3, Sr-90,
Cs-137, and Am-241. Counts were repeated
a minimum of 20 times to determine
statistical variation in the counts. For the
purposes of this testing, a response is
considered to be above background level if
it is at least three standard deviations above
the background level (EPA and DHS, 2008).
NAREL began using the P-RAM to
determine the total background counts in
clean water. The background sample counts
of 25 runs resulted in a mean count rate of
18.5 ± 5.7 cpm (counts per minute). The
reported error range represents three
standard deviations (see Section 3.0 for raw
data and further discussion). During the
next 25 runs, a prepared solution (with
known activity level) of Sr-90 was
introduced into the system. With decay
correction, the solution had a stated value of
413.76 Bq/L or 11,180 pCi/L of Sr-90. Sr-90
decays to Y-90 with a half-life of 64 hours.
The Sr-90 solution was produced in 2007
and therefore is at full equilibrium with Y-
90. The Sr-90 has an average beta decay
energy of 200 keV, while Y-90 has an
average beta decay energy of 931 keV.
These energy levels bracket many of the
typical beta energies measured in
environmental samples. The maximum beta
energy of Y-90 is 2,245 keV, which is at the
lower end of the energy emitted by potential
alpha decays. This peak enables a qualified
individual to identify potential cross-over
measurements of high energy betas being
detected as low energy alphas.
It should be noted that the MCA provides a
measure of pulse height and does not record
the pulse timing. The alpha decay time is
often 150 to 200 nanoseconds delayed from
a beta/gamma peak, which is not
differentiated by this device. Therefore, it is
possible to erroneously count a delayed
alpha light peak in the MCA spectrum.
Pulse height will not show this difference. In
order to properly calibrate for alpha beta
analysis, an alpha source (Am-241 solution)
was also prepared and used. Calibrations of
this manner should be performed by an
individual appropriately trained in radio
analytical measurements beyond that
required of a typical operator of the system.
Calibration of the instrument was performed
by pumping the prepared solutions through
the sample lines and waiting for a repetitive
stable response. It is feasible that this
calibration could be performed in the field
by simply swapping the existing sample cell
with a static cell containing the calibration
source and cocktail solution.
11
-------
3.0 Experimental Results
As with the calibration/background level
determination procedures described in
Section 2.5, solutions of known activity
levels were prepared and introduced into the
system and analyzed between 20 and 25
times each to ensure stable readings. The
following solutions were prepared by EPA
NAREL onsite personnel using NIST
traceable source materials and verified at the
listed activity levels:
• Sr-90atl,129pCi/L
• Cs-137 at 1,100 pCi/L
• Am-241 at 2,480 pCi/L and 1,310
pCi/L
• H-3 at 740 pCi/L and 74 pCi/L
The results of these tests are described in
this section. The P-RAM counts are
measured/detected using two sensors: Radio
1 and Radio 2. The instrument outputs
provide both values. The Radio 1 output is a
processed signal, which uses proprietary
chemi-luminescence and background
subtraction algorithms. The Radio 2 output
is simply the raw counts without any data
processing. The results presented include
instrument output from Radio 1 and Radio 2.
However, only the processed signal values
from Radio 1 have been interpreted in this
section as recommended by LLS. The P-
RAM system was setup to count for 10
minutes, to get results in near real-time as
will be required for a typical drinking water
contamination warning system. The
tabulated 10-minute counts of Radio 1 are
converted to counts per minute (cpm) to
evaluate the system performance.
The remaining solutions of Sr-90, Cs-137,
Am-241, and H-3 not used for this testing
remain under the control and ownership of
NAREL at the Gunter Air Force Base in
Montgomery, Alabama per the Interagency
Agreement used to perform this work. The
solutions, methanol, and scintillant used
during this testing were disposed in
accordance with the applicable permits for
the NAREL Facility.
3.1 Background Determination
Table 1 summarizes the raw results from the
25 background runs (112 through 136).
12
-------
Table 1. P-RAM System Background Output (Counts) Water Analysis Summary
Run#
112
113
114
115
116
117
118
119
120
121
122
123
124
Radio-1
194
218
159
201
178
157
204
157
186
162
184
196
216
Radio-2
373
364
346
387
374
342
355
365
351
377
371
340
326
Run#
125
126
127
128
129
130
131
132
133
134
135
136
Average
Std. Dev
Radio-1
160
178
156
186
185
207
196
190
180
179
195
209
185.3
18.9
Radio-2
414
390
369
394
363
384
373
346
340
364
394
406
As shown in Table 1, the average Radio 1
reported values over the programmed 10-
minute period was 185.3 raw counts with a
standard deviation of 18.9 raw counts for the
background runs. This value, divided by the
10-minute counting period, results in an
average background level of 18.5 cpm with
a standard deviation of 1.9 cpm. For the
purposes of this report, an error range of
three standard deviations from the mean was
used an indication of radiation detection.
Therefore, any detected level over 24.2 cpm
(=18.5 + 3 x 1.9) in this operation mode
(using 10-minute counts) would be
considered as detectable.
3.2 Sr-90 Test Run Results
Table 2 summarizes the raw results from the
23 Sr-90 test runs (89 through 111) at an
activity level 1,120 pCi/L.
13
-------
Table 2. p-RAM System Output (Counts) for Sr-90 (1,120 pCi/L) Water Analysis
Summary
Run#
89
90
91
92
93
94
95
96
97
98
99
100
101
Radio-1
715
738
714
705
735
688
686
717
766
666
679
683
677
Radio-2
2250
2208
2280
2210
2334
2234
2292
2163
2284
2261
2303
2282
2310
Run#
102
103
104
105
106
107
108
109
110
111
Average
Std. Dev
Radio-1
741
714
964
701
820
651
685
687
675
633
714.8
67.2
Radio-2
2435
2303
2412
2289
2371
2387
2367
2350
2379
2345
As shown in Table 2, the average Radio 1
reported values over the programmed 10-
minute period was 714.8 raw counts with a
standard deviation of 67.2 raw counts for the
Sr-90 runs. This value, divided by the 10-
minute counting period, results in an
average level of 71.5 cpm. This mean value
is well above the established detection level
of background plus three-sigma (24.2 cpm)
at the tested activity level.
3.3 Cs-137 Test Run Results
Table 3 summarizes the raw results from the
23 Cs-137 test runs (137 through 159) at an
activity level 1,100 pCi/L.
Table 3. p-RAM System Output (Counts) for Cs-137 (1,110 pCi/L) Water Analysis
Summary
Run#
137
138
139
140
141
142
143
144
145
146
147
148
149
Radio-1
487
503
537
543
550
484
513
486
529
525
532
546
498
Radio-2
782
855
796
871
831
837
794
723
819
821
747
757
786
Run#
150
151
152
153
154
155
156
157
158
159
Average
Std. Dev
Radio-1
510
511
540
479
538
522
532
537
532
501
518.9
21.9
Radio-2
833
715
747
774
765
802
861
776
799
849
14
-------
As shown in Table 3, the average Radio 1
reported values over the programmed 10-
minute period was 518.9 raw counts with a
standard deviation of 21.9 raw counts for the
Cs-137 runs. This value, divided by the 10-
minute counting period, results in an
average level of 51.9 cpm. This mean value
is well above the established detection level
of background plus three-sigma (24.2 cpm)
at the tested activity level.
3.4 Am-241 Test Run Results
Table 4 summarizes the raw results from the
23 Am-241 test runs (160 through 182) at an
activity level 2,480 pCi/L.
Table 4. p-RAM System Output (Counts) for Am-241 (2,480 pCi/L) Water Analysis
Summary
As shown in Table 4, the average Radio 1
reported values over the programmed 10-
minute period was 297.1 raw counts with a
standard deviation of 68.8 raw counts for the
Am-241 runs at 2,480 pCi/L level. The
standard deviation is high because Test Run
171 reported a high count of 574. This
value, divided by the 10-minute counting
period, results in an average level of 29.7
cpm. This mean value is above the
Run#
160
161
162
163
164
165
166
167
168
169
170
171
172
Radio-1
290
276
291
263
248
291
314
290
255
265
269
574
302
Radio-2
1302
1284
1266
1231
1388
1342
1257
1332
1252
1251
1323
1249
1252
Run#
173
174
175
176
177
178
179
180
181
182
Average
Std. Dev
Radio-1
360
281
307
237
259
296
352
219
317
277
297.1
68.8
Radio-2
1303
1232
1265
1292
1334
1249
1256
1282
1259
1321
established detection level of background
plus three-sigma (24.2 cpm) at the tested
activity level.
Am-241 was also tested at an activity level
of 1,310 pCi/L, which is closer to the PAG
level. Table 5 summarizes the raw results
from these 20 Am-241 test runs (183
through 202).
15
-------
Table 5. p-RAM System Output (Counts) for Am-241 (1,130 pCi/L) Water Analysis
Summary
Run#
183
184
185
186
187
188
189
190
191
192
193
194
195
Radio-1
356
319
376
422
337
376
316
308
376
383
301
399
358
Radio-2
1061
1012
981
981
1063
999
1069
973
942
980
960
1018
970
Run#
196
197
198
199
200
201
202
Average
Std. Dev
Radio-1
410
424
378
291
384
356
382
362.6
39.5
Radio-2
1021
1015
1000
1004
990
999
1017
As shown in Table 5, the average Radio 1
reported values over the programmed 10-
minute period was 362.6 raw counts with a
standard deviation of 39.5 raw counts for the
Am-241 runs at a 1,130 pCi/L level. The
overall response is counter intuitive when
compared to the other Am-241 test runs.
Those tests were conducted using higher
Am-241 activity level stock, but they
reported lower overall counts. However,
counting efficiencies of scintillation
detection methods can vary for many
reasons, such as differences in sample and
scintillation cocktail compositions. Poor
counting efficiency may result from lower
energy to light conversion rate, referenced
as scintillation efficiency. Regardless, both
sets of values are above the established
detection level of background plus three-
sigma (24.2 cpm), and therefore Am-241
would be considered as detectable at both of
these activity levels.
3.5 H-3 Test Run Results
Table 6 summarizes the raw results from the
23 test runs (203 through 225) of H-3 at an
activity level 740 pCi/L.
16
-------
Table 6. p-RAM System Output (Counts) for H-3 (740 pCi/L) Water Analysis Summary
As shown in Table 6, the average Radio 1
reported values over the programmed 10-
minute period was 499.6 raw counts with a
standard deviation of 103.8 raw counts for
the H-3 runs at 740 pCi/L level. This value,
divided by the 10-minute counting period,
results in an average level of 50 cpm. This
mean value is above the established
Run#
203
204
205
206
207
208
209
210
211
212
213
214
215
Radio-1
477
459
487
466
497
452
375
406
437
405
400
483
447
Radio-2
667
717
701
734
692
695
682
689
732
656
736
707
740
Run#
216
217
218
219
220
221
222
223
224
225
Average
Std. Dev
Radio-1
385
442
478
449
590
662
711
708
614
661
499.6
103.8
Radio-2
732
690
701
686
798
837
872
933
846
869
detection level of background plus three-
sigma (24.2 cpm) at the tested activity level.
H-3 was also tested at an activity level of 74
pCi/L. Table 7 summarizes the raw results
from the 23 test runs (227 through 250) for
H-3 at this level.
Table 7. P-RAM System Output (Counts) for H-3 (74 pCi/L) Water Analysis Summary
Run#
227
228
229
230
231
232
233
234
235
236
237
238
239
Radio-1
246
260
303
263
323
337
278
291
286
313
273
290
284
Radio-2
677
649
652
632
668
671
670
631
612
665
638
660
600
Run#
240
241
242
243
244
246
247
248
248
250
Average
Std. Dev
Radio-1
342
317
292
339
328
337
359
369
354
315
308.7
33.9
Radio-2
627
651
667
606
622
620
670
689
673
650
17
-------
As shown in Table 7, the average Radio 1
reported values over the programmed 10-
minute period was 308.7 raw counts with a
standard deviation of 33.9 raw counts for the
H-3 runs at 74 pCi/L level. This value,
divided by the 10-minute counting period,
results in an average level of 30.9 cpm. This
mean value is above the established
detection level of background plus three-
sigma (24.2 cpm) at the tested activity level.
3.6 Online Sample Cross Contamination
Check
Tests were conducted to determine how long
the contamination remained in the system
after analysis, during flushing, and
subsequent measurements. Under normal
operation, the system has a continuous
sample flow running through the sample
system pump up to the sample divergence
valve (previously illustrated in the list of
events in Figure 3). However, during the
tests conducted at NAREL, this normally
continuously running sample flow was
terminated to preserve the sample.
Subsequently, it provided an opportunity to
observe how the system responded to
changing contamination levels. The dead
volume was noted to pass through the
detector with a return to baseline after the
third sampling of normal water. There was
no indication of retained contamination in
the plumbing or the sample cell. It would be
expected that, if the sample contamination is
cleared out during analysis as the sample is
flushed through the sample line, the only
retained contamination would be that
contained downstream of the divergence
valve, a volume less than that of the cell.
Monitored activity levels would then drop in
the first count after sample analysis.
3.7 Event-based Automated Sampling
A test of the automated event-based water
sampling system was performed. The P-
RAM system allows the user to set a level of
activity in cpm at which the sample,
complete with cocktail solution, will be
discharged automatically to an awaiting
container for further analysis. The system is
still under design with respect to the type of
container to be used for collection of the
sample. As the sample is essentially a
scintillation sample, a scintillation cell or
container is considered the likely candidate.
Multiple cells could be placed into a
carousel that could rotate the cells after
filling to allow for a flush prior to filling the
next cell. One concern noted with the
sample collection procedure is the lack of
flushing of the line after saving the sample.
The line contains microliters of sample
media that could be inadvertently added to
the next sample, causing some cross-
contamination. This effect should not greatly
change the contaminant alarm capability of
the device, but would likely affect the
accuracy of the sample count.
3.8 Overall Results Summary
The results show that all isotopes measured
would be detectable near the PAG levels.
The overall response of activity in raw 10-
minute counts (Radio 1 values), as reported
by the instrument (as shown in Tables 1
through 7), are illustrated in Figure 6. It
should be noted that these are simple tests of
the system; a more extensive testing of the
next generation P-RAM system is
recommended to determine the lowest
detection limits for each compound.
18
-------
10 00
900
800
700
i? 00
500
400
300
200
100
10
20
t Background
-*-Arn-241 (1310 pCi/L)
r-90 (1120 pCi/L!
H-3 (740 pCi/L)
•Cs-137(1100pCi/L)
-H-3(74pCi/L)
25
-Am-241 (2480 pCi/L)
Background-i- 3 Sigma
30
Figure 6. Overall Summary of P-RAM system testing. Counts (y-axis) versus test runs
19
-------
4.0 Conclusions and Recommendations
The test results indicate that the P-RAM
system can be used as an online real-time
(response less than 30-minutes) radiation
monitor. As shown previously in Figure 6,
the P-RAM system consistently reported raw
count values above the background level
plus three-sigma for selected isotopes near
or below the PAG activity levels. Based on
the system improvement performed during
the initial operational evaluation in
Cincinnati indicates that the system can be
operated and maintained by a typical water
treatment plant technician. The system was
able to function in a somewhat-rugged test
environment.
The LLS P-RAM with the Wilma operating
system operated satisfactorily, as
designed/modified by the vendor. The
system provided an indication of the
presence of radioactive material at the
measured background level plus three-
sigma. The level of detection will be
dependent upon the background radiation
level of the drinking water supply being
tested. In its current state of development,
the system clearly demonstrated the ability
to (1) detect contamination at the injected
levels, (2) discharge the sample for
retention, and (3) provide an alarm to the
user. The results of the testing also indicate
that it may be possible to detect lower levels
of activity than those used in this test.
Based on the testing and evaluation data
collected during the study, it appears that the
instrument is well suited as an online water
quality monitor for detecting the presence of
radiological contamination at the tested
levels. From a field deployment perspective,
it is a substantial improvement and can be
considered as the "state of the art" for an
online radiological water quality monitor at
the time when the testing was performed.
However, several system enhancements are
recommended to make the device more
readily acceptable to the market.
4.1 Recommended System
Enhancements and Future Testing
The following is a listing of recommended
system improvements:
• The MCA module needs further
development and integration prior to
additional testing. LLS is continuing
to refine the integration of the MCA
module.
• The industrial robustness of the
fluidics needs to be further improved
to fit the capabilities of a typical user
(or a water technician).
• Vendor must continue to find ways
to lower the count times and level of
detections from the current tested
levels.
• A radiation source and radiation
expertise is necessary to calibrate the
P-RAM system. These types of
facilities (e.g., water utilities) may
not be permitted to store or use
radioactive calibration sources.
Therefore, a "factory calibration"
procedure or factory service option
may need to be provided along with
the sale of the device.
• Although the system provides the
capability for programming a set
alarm level, further improvements
are necessary to fully automate the
sample collection system.
• A redesign of the device with
"industrial use" focus for the nuclear
industry might further expand the
market for the device.
• Additional long-term testing needs to
be performed for use as a
20
-------
contamination warning device; use of this technology. Based upon the
especially, in areas with high encouraging test results of this technology
background levels of radioactivity in presented in this report, the P-RAM
water. technology should continue to be tested and
commercialized for wide-spread field
This list of recommendations is not meant to applications.
be exhaustive, but more to illustrate the
general areas of improvements to further the
21
-------
5.0
References
Romo. R., Parker N. and Castillo, M., 2013.
Mexico: Stolen radioactive material found
http://www.cnn.com/2013/12/04/world/amer
i cas/mexi co-radi oacti ve-theft/
U.S. Department of Homeland Security
(DHS). 2008. Planning Guidance for
Protection and Recovery Following
Radiological Dispersal Device and
Improvised Nuclear Device Incidents.
Federal Register Vol. 73, No. 149, pp 45029
-45048, August 1,2008.
U.S. Environmental Protection Agency
(EPA). 2000. Radionuclides Notice of Data
Availability Technical Support Document,
Targeting and Analysis Branch, Standards
and Risk Management Division, Office of
Ground Water and Drinking Water, EPA
March, 2000.
U.S. Environmental Protection Agency
(EPA). 2009. Testing for real time in line
radioactivity, water monitoring systems.
EPA QAID 879 approved April 08, 2009.
U.S. Environmental Protection Agency
(EPA). 2011. Test of an In Line Radiation
Monitoring System for Drinking water. A
summary of the Tests and Responses of
Technical Associates SSS-33-5FT Radiation
Monitoring System. EPA 600/R-l 1/005.
Available through Water Information
Sharing and Analysis Center (WaterlSAC) .
Membership is required for access
U.S. Environmental Protection Agency
(EPA). 2012. Letter to Amend QA ID 879 to
include LLS P-RAM Testing. Dated
November 16, 2012. Approved by Ramona
Sherman, EPA, Office of Research and
Development, Cincinnati, Ohio.
U.S. Environmental Protection Agency
(EPA) and U.S. Department of Homeland
Security (DHS). 2008. All Hazards Receipt
Facility Screening Protocol. EPA 600 R-08-
105.
Zapecza, O. S., and Z. Szabo, 1986. Natural
radioactivity in ground water- a review. U.S.
Geological Survey National Water Summary
1986, Ground-Water Quality: Hydrologic
Conditions and Events, U. S. Geological
Survey Water Supply Paper 2325. pp. 50-57.
22
-------
Appendix A - Dilute H Measurements Using a 5 mL Wilma Cell and
ScintLogic LB - March 2012 (Reprinted with Permission - LabLogic
Systems, Inc.)
Samples
Five H-glycerol samples were prepared by
Unilever Research and the activity rates were
determined at the point of production using static
Table 1. 3H Sample Activities
counting (10:1 cocktail:sample ratio). Three 1 mL
aliquots of each sample were measured over a
period of 20 minutes to determine the dpm
(counting efficiency ca. 40%). The sample properties
are summarized in Table 1:
Sample Ref
3H Oil 1
3H Oil 2
3H Oil 3
3H Oil 4
3H Oil 5
Target Activity (dpm/mL)
<50
100
350
600
1000
Actual Activity (dpm/mL)
42.5±0.2
121.9±24.2
372.0±29.2
622.8±13.9
1003.5±7.6
[dpm, disintegrations per minute]
Sample Preparation
Prior to measurement, each of the five samples was
prepared as follows: 3 mL of ScintLogic LB was
dispensed into a 7 mL vial using a 2.5 mL
disposable needle syringe. 3 mL of sample was
then added to the cocktail and the two mixed
together vigorously for 60 s. The sample/cocktail
mixture was then left to stand for 15 min to allow
any air bubbles to diffuse out. When ready for
measurement, 5 mL of the sample was syringed into
the 5 mL Wilma cell, the cell then being placed into
the (3-RAM used for all measurements. Each sample
was then left to dark adapt in the instrument for a
minimum of three hours to ensure that any unwanted
luminescent effects were kept to a minimum.
Between each sample, the measurement cell was
cleaned thoroughly using a water-diluted methanol
solution. The syringes were also changed for each
step involving a new sample in order to prevent
contamination.
Measurements of Counting Efficiency
Prior to the measurement of the H samples,
the background characteristics of the ScintLogic
LB cocktail was measured using a 1:1 cocktail :H2O
sample over a period of 15 minutes. Five replicates
were measured for two different background
samples. The background rate was found to be
8.9±0.4 cpm for Channel 1 and 25.2±0.5 cpm for
Channel 2.
Each sample was then measured statically in the (3-
RAM within the recommended counting window of
5-70 in order to determine the average count rate.
Both counting channels were used in order to
determine the effect of the cross-talk subtraction
firmware algorithm. For each sample, three 15 min
runs were made, followed by one 30 min and 60 min
in order to confirm that the measured count rate did
not vary significantly with the counting time.
The counting efficiency for each sample was then
determined using the average count rate value
obtained from the three 15 min measurements and
using the relationship:
Counting Efficiency (Eff%) = (cpm/dpm)/100
This calculation was made for both channels and
with/without background subtraction (BS). Using the
calculated value of the efficiency, the Figure of
Merit (FoM), E /B was also determined. The results
are summarized in Table 2, the values quoted with
their associated errors.
23
-------
Table 2. Counting Efficiency and FoM Values for the Five Samples
Sample
3H Oil !Ch.l;NoBS
3H_Dil_l Ch.2; No BS
3H Oil 1 Ch.l;BS
3H Oil 1 Ch.2; BS
3H Oil 2Ch.l;NoBS
3H Oil 2 Ch.2; No BS
3H_Dil_2Ch.l;BS
3H Oil 2 Ch.2; BS
3H Oil 3 Ch.l;NoBS
3H Oil 3 Ch.2; No BS
3H Oil 3Ch.l;BS
3H_Dil_3 Ch.2; BS
3H Oil 4Ch.l;NoBS
3H Oil 4 Ch.2; No BS
3H Oil 4Ch.l;BS
3H Oil 4 Ch.2; BS
3H_Dil_5Ch.l;NoBS
3H Oil 5 Ch.2; No BS
3H Oil 5Ch.l;BS
3H Oil 5 Ch.2; BS
Average Count
Rate (cpm)
12.4±0.4
28.4±0.7
3.2±0.02
3.2±0.08
42.5±2.1
54.2±3.1
33.3±1.7
29.0±1.7
66.4±0.8
72.7±1.4
57.2±0.7
47.5±0.9
99.0±2.4
100.9±2.2
89.8±2.2
75.7±1.7
154.8±4.2
155.7±4.1
145.6±4.0
130.5±3.4
Expected Activity
Rate in 2.5mL (dpm)
106.3±0.5
106.3±0.5
106.3±0.5
106.3±0.5
304.7±60.5
304.7±60.5
304.7±60.5
304.7±60.5
929.9±73.0
929.9±73.0
929.9±73.0
929.9±73.0
1557.0±34.8
1557.0±34.8
1557.0±34.8
1557.0±34.8
2508.8±19.0
2508.8±19.0
2508.8±19.0
2508.8±19.0
Eff. (%)
11.7±0.4
26.7±0.4
3.0±0.02
3.0±0.08
13.9±2.9
17.8±3.7
10.9±2.2
9.5±2.0
7.1±0.6
7.8±0.6
6.2±0.5
5.1±0.4
6.4±0.2
6.5±0.2
5.8±0.2
4.9±0.2
6.2±0.2
6.2±0.2
5.8±0.2
5.3±0.2
E2/B
14.9±0.8
21±0.5
0.97±0.04
0.35±0.01
21±4.4
12.6±2.6
12.9±2.7
3.6±0.8
5.5±0.5
2.4±0.2
4.2±0.4
l.OiO.l
4.5±0.3
1.7±0.1
3.7±0.2
1.0±0.03
4.2±0.2
1.5±0.05
3.7±0.2
1.U0.04
:pm, counts per minute; dpm, disintegrations per minute]
Discussion
The measurements have shown that the
o
Wilma cell can be used to measure H
samples with levels of activity below 50
dpm/mL and the results are reproducible.
Channel 1 shows a greater impact in
terms of count subtraction for low-level
samples; the effect is less evident above
around 350 dpm/mL.
The counting efficiencies and hence figures of merit
are rather low, being less than
10% in general for the former and less than 10 for the
latter.
This may be due to a number of factors that
need to be optimized such as the
sample :cocktail ratio and also the width of
the counting window. The latter needs to be
investigated further in order to maximize the
SNR [signal-to-noise ratio].
Dr. TomDeakin, March 2012
24
-------
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
------- |