May 2011
Environmental Technology
Verification Report
BEACON
MICROCYSTIN TEST KITS
TUBE KIT
PLATE KIT
Prepared by
Battelle
Battelle
1 he Business of Innovation
Under a cooperative agreement with
U.S. Environmental Protection Agency
ET1/ET1/ET1/
-------
May 2011
Environmental Technology Verification
Report
ETV Advanced Monitoring Systems Center
BEACON
MICROCYSTIN TEST KITS
TUBE KIT
PLATE KIT
by
Ryan James, Anne Gregg, and Amy Dindal, Battelle
John McKernan, U.S. EPA
-------
Notice
The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed, or partially funded and collaborated in, the research described herein. It
has been subjected to the Agency's peer and administrative review. Any opinions expressed in
this report are those of the author(s) and do not necessarily reflect the views of the Agency,
therefore, no official endorsement should be inferred. Any mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
11
-------
Foreword
The EPA is charged by Congress with protecting the nation's air, water, and land resources.
Under a mandate of national environmental laws, the Agency strives to formulate and implement
actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, the EPA's Office of Research and
Development provides data and science support that can be used to solve environmental
problems and to build the scientific knowledge base needed to manage our ecological resources
wisely, to understand how pollutants affect our health, and to prevent or reduce environmental
risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA to
verify the performance characteristics of innovative environmental technology across all media
and to report this objective information to permitters, buyers, and users of the technology, thus
substantially accelerating the entrance of new environmental technologies into the marketplace.
Verification organizations oversee and report verification activities based on testing and quality
assurance protocols developed with input from major stakeholders and customer groups
associated with the technology area. ETV consists of six environmental technology centers.
Information about each of these centers can be found on the Internet at http://www.epa.gov/etv/.
Effective verifications of monitoring technologies are needed to assess environmental quality
and to supply cost and performance data to select the most appropriate technology for that
assessment. Under a cooperative agreement, Battelle has received EPA funding to plan,
coordinate, and conduct such verification tests for "Advanced Monitoring Systems for Air,
Water, and Soil" and report the results to the community at large. Information concerning this
specific environmental technology area can be found on the Internet at
http ://www. epa.gov/etv/centers/centerl .html.
in
-------
Acknowledgments
The authors wish to acknowledge the contribution of the many individuals, without whom, this
verification testing would not have been possible. Quality assurance (QA) oversight was
provided by Michelle Henderson, U.S. EPA, and Zachary Willenberg, Battelle. We thank David
Schumacher and John Lund of the Nebraska Department of Environmental Quality and Robert
Waters from the New York, Suffolk County Department of Health Services for their support of
this verification by providing recreational water samples, and also for peer review of the test/QA
plan. We acknowledge the support from Daniel Snow and David Cassada from the University of
Nebraska Water Sciences Laboratory for developing and validating the reference method in
addition to analyzing the samples for the verification. Finally, we want to thank Andrew Lincoff,
U.S. EPA and Daniel Snow of the University of Nebraska for their review of the test/QA plan
and verification report.
IV
-------
Contents
Page
Foreword iii
Acknowledgments iv
List of Abbreviations ix
Chapter 1 Background 1
Chapter 2 Technology Description 2
2.1 Microcystins Tube and Plate Kit 2
Chapter 3 Test Design and Procedures 4
3.1 Test Overview 4
3.2 Experimental Design 4
3.3 Test Procedures 5
3.3.1 QC Samples 6
3.3.2 PT Samples 6
3.3.3 RW Samples 7
Chapter 4 Quality Assurance/Quality Control 9
4.1 Reference Method Quality Control 9
4.2 Audits 11
4.2.1 Performance Evaluation Audit 11
4.2.2 Technical Systems Audit 12
4.2.3 Data Quality Audit 12
Chapters Statistical Methods 14
5.1 Accuracy 14
5.2 Linearity 14
5.3 Precision 15
5.4 Method Detection Limit 15
5.5 Inter-Kit Lot Reproducibility 15
5.6 Matrix Effects 15
Chapter 6 Test Results for the Beacon Microcystin Tube Kit 16
6.1 Beacon Test Kit Summary 16
6.2 Test Kit QC Sample 16
6.3 PT Samples 17
6.3.1 Accuracy 17
6.3.2 Precision 21
6.3.3 Linearity 23
6.3.4 Method Detection Limit 24
6.3.5 Inter-Kit Lot Reproducibility 24
6.3.6 Matrix Effect 25
6.4 RW Sample Results 28
6.5 Operational Factors 29
6.5.1 Ease of Use 29
6.5.2 Cost and Consumables 30
Chapter 7 Test Results for the Beacon Plate Kit 31
-------
7.1 Beacon Plate Kit Summary 31
7.2 Test Kit QC Sample 31
7.3 PT Samples 32
7.3.1 Accuracy 32
7.3.2 Precision 36
7.3.3 Linearity 38
7.3.4 Method Detection Limit 39
7.3.5 Inter-Kit Lot Reproducibility 40
7.3.6 Matrix Effects 40
7.4 RW Sample Results 44
7.5 Operational Factors 45
7.5.1 Ease of Use 45
7.5.2 Cost and Consumables 46
Chapter 8 Performance Summary for the Beacon Tube and Plate test Kits 47
8.1 Performance Summary for the Beacon Tube Test Kit 47
8.2 Performance Summary for the Beacon Plate Test Kit 48
Chapter 9 References 50
APPENDIX A Reference Laboratory Method Detection Limit Memo 51
APPENDIX B Beacon Test Kit Raw Data 53
VI
-------
Tables
Table 1. Summary of Test Samples 7
Table 2. DQIs and Summary of Reference Method QC Results 10
Table 3. Summary of Reference Method CCV Percent Recoveries 11
Table 4. PEA Results: Analytical Comparison of Microcystin Standards 11
Table 5. PEA Results: Evaluation of Extracted Low Level Water Sample 12
Table 6. RB Sample Results for the Beacon Tube Kit 17
Table 7. Positive Control Sample Results for the Beacon Tube Kit 17
Table 8. Beacon Tube Kit Sample Results and Reference Method Results for LR 18
Table 9. Beacon Tube Kit Sample Results and Reference Method Results for LA 19
Table 10. Beacon Tube Kit Sample Results and Reference Method Results for RR 20
Table 11. Beacon Tube Kit Precision Results 22
Table 12. Detection Limit Results for the Beacon Tube Kit 24
Table 13. Inter-kit lot Comparison of Kit Calibration Standards for the Beacon Tube Kit 25
Table 14. RW Matrix Interferent Sample Results for the Beacon Tube Kit 26
Table 15. Chlorophyll-a Interferent Sample Results for the Beacon Tube Kit 27
Table 16. Statistical Comparisons between Interference Samples 28
Table 17. Recreational Water Sample Results for the Beacon Tube Kit 29
Table 18. RB Sample Results for the Beacon Plate Kit 32
Table 19. Positive Control Sample Results for the Beacon Plate Kit 32
Table 20. Beacon Plate Kit Sample Results and Reference Method Results for LR 33
Table 21. Beacon Plate Kit Sample Results and Reference Method Results for LA 34
Table 22. Beacon Plate Kit Sample Results and Reference Method Results for RR 35
Table 23. Beacon Plate Kit Precision Results 37
Table 24. Detection Limit Results for the Beacon Plate Kit 39
Table 25. Inter-kit lot Comparison of Kit Calibration Standards for the Beacon Plate Kit 40
Table 26. RW Matrix Interferent Sample Results for the Beacon Plate Kit 42
Table 27. Chlorophyll-a Interferent Sample Results for the Beacon Plate Kit 43
Table 28. Statistical Comparisons between Interference Samples 44
Table 29. Recreational Water Sample Results for the Beacon Plate Kit 45
Table 30. Beacon Tube Test Kit Performance Summary 47
Table 31. Beacon Plate Test Kit Performance Summary 48
vn
-------
FIGURES
Figure 1. Beacon Microcystin Tube Kit 3
Figure 2. Beacon Microcystin Plate Kit 3
Figure 3. Linearity for the Beacon Tube Kit for LR 23
Figure 4. Linearity for the Beacon Tube Kit for LA 23
Figure 5. Linearity for the Beacon Tube Kit for RR 24
Figure 6. Linearity for the Beacon Plate Kit for LR 38
Figure 7. Linearity for the Beacon Plate Kit for LA 38
Figure 8. Linearity for the Beacon Plate Kit forRR 39
Vlll
-------
List of Abbreviations
AMS Advanced Monitoring Systems
ASTM American Society for Testing and Materials
CCV continuing calibration verification
°C degrees Celsius
CR cross reactivity
CV coefficient of variation
DQI data quality indicator
DQO data quality objective
DI Deionized
EPA Environmental Protection Agency
ETV Environmental Technology Verification
ELISA Enzyme-Linked Immunosorbent Assay
LFM laboratory fortified matrix
LC-MS-MS liquid chromatography tandem mass spectrometry
LOQ Limit of quantification
MB method blank
MDL method detection limit
mg/L milligram per liter
mL Milliliter
Nm Nanometer
NDEQ Nebraska Department of Environmental Quality
NRC National Research Council
OD optical density
Ppb parts per billion
%D percent different
PEA performance evaluation audit
PT performance test
QA quality assurance
QAO quality assurance officer
QC quality control
QMP quality management plan
r2 coefficient of determination
RB reagent blank
RW recreational water
RSD relative standard deviation
S standard deviation
SOP standard operating procedure
SPE solid phase extraction
TQAP Test/Quality Assurance Plan
TSA technical systems audit
(ig/L Microgram per linter
WSL Water Sciences Laboratory
IX
-------
Chapter 1
Background
The U.S. Environmental Protection Agency (EPA) supports the Environmental Technology
Verification (ETV) Program to facilitate the deployment of innovative environmental
technologies through performance verification and dissemination of information. The goal of the
ETV Program is to further environmental protection by accelerating the acceptance and use of
improved and cost-effective technologies. ETV seeks to achieve this goal by providing high-
quality, peer-reviewed data on technology performance to those involved in the design,
distribution, financing, permitting, purchase, and use of environmental technologies.
ETV works in partnership with recognized testing organizations; with stakeholder groups
consisting of buyers, vendor organizations, and permitters; and with the full participation of
individual technology developers. The program evaluates the performance of innovative
technologies by developing test plans that are responsive to the needs of stakeholders,
conducting field or laboratory tests (as appropriate), collecting and analyzing data, and preparing
peer-reviewed reports. All evaluations are conducted in accordance with rigorous quality
assurance (QA) protocols to ensure that data of known and adequate quality are generated and
that the results are defensible. The definition of ETV verification is to establish or prove the
truth of the performance of a technology under specific, pre-determined criteria or protocols and
a strong quality management system. The highest-quality data are assured through
implementation of the ETV Quality Management Plan. ETV does not endorse, certify, or
approve technologies.
The EPA's National Risk Management Research Laboratory (NRMRL) and its verification
organization partner, Battelle, operate the Advanced Monitoring Systems (AMS) Center under
ETV. The AMS Center recently evaluated the performance of two technologies offered by
Beacon Analytical Systems, Inc.: Microcystin Plate Kit and Microcystin Tube Kit.
-------
Chapter 2
Technology Description
This verification report provides results for the verification testing of two Beacon Analytical
Systems Microcystin Test Kits. Following are descriptions of the Microcystin Plate and Tube
Test Kits based on information provided by the vendor. The information provided below was
not verified in this test.
2.1 Microcystins Tube and Plate Kit
Beacon Microcystin Tube Kit, Figure 1, is an immunological laboratory test for the
quantification of microcystins in water. The tube kit uses a polyclonal antibody that binds both
microcystins and a microcystin-enzyme conjugate. Microcystins in the sample compete with the
microcystin-enzyme conjugate for a limited number of antibody binding sites. The assay
procedure includes the following steps:
• Add microcystin-enzyme conjugate and a sample for analysis of microcystins to a test tube,
followed by antibody solution. The conjugate competes with any microcystins in the sample
for the same antibody binding sites. The test tube is coated with anti-rabbit immunoglobulin
G (IgG) to capture the rabbit anti-microcystin added.
• Wash away any unbound molecules, after incubating this mixture for 20 minutes.
• Add clear substrate solution to each tube. In the presence of bound microcystin-enzyme
conjugate, the substrate is converted to a blue compound. One enzyme molecule can convert
many substrate molecules.
Since the same number of antibody binding sites is available in every tube, and each tube
receives the same number of microcystin-enzyme conjugate molecules, a sample containing a
low concentration of microcystins allows the antibody to bind many microcystin-enzyme
conjugate molecules. The result is a dark blue solution. Conversely, a high concentration of
microcystins allows fewer microcystin-enzyme conjugate molecules to be bound by the
antibodies, resulting in a lighter blue solution. The color is analyzed using a colorimeter or
spectrophotometer to obtain optical density (OD) values at 450 nanometers (nm). Reader
software or a spreadsheet is used to generate a standard curve and interpolate the sample values
from that curve.
There are approximately 80 different variants of microcystins present in the environment and the
kits tested are not able to detect the difference between microcystin variants. Results from the
kits tested are calibrated with respect to the microcystin-LR variant. However, other microcystin
variants are known (based on information provided by Beacon) to react to different extents with
the antibodies used for detection. Cross reactivity values provided by Beacon are used to
quantify results for different variants based on the LR calibration.
The Beacon Microcystin Plate Kit (Figure 2) is also an immunological laboratory test for the
quantification of microcystins in water. The plate kit uses the same principles as the tube kit but
-------
employs the use of a 96 well assay plate to allow for processing a larger number of samples. The
plate is a break-apart strip design allowing for assays of any number of wells to be employed
from 1 to 96. The manufacturer recommends significant familiarity with the technology,
incorporation of additional assay controls and the use of multichannel pipettes for "whole plate"
(96 well) assays.
The price of the tube kit at the time of this verification test is $200 for a 40-tube kit (24 samples).
The plate kit costs $275 per plate to analyze a maximum of 84 samples. Both kits measure 6.25
x 5.125 x 3.75 inches (15.9 x 13.0 x 9.5 centimeters), with the plate kit weighing llounces (312
grams) and the tube kit is 17 ounces (483 grams).
Figure 1. Beacon Microcystin Tube Kit
Figure 2. Beacon Microcystin Plate Kit
-------
Chapter 3
Test Design and Procedures
3.1 Test Overview
This verification test was conducted according to procedures specified in the Test/Quality
Assurance Plan for Verification of Microcystin Test Kits1 (TQAP) and adhered to the quality
system defined in the ETV AMS Center Quality Management Plan (QMP)2. It assessed the
performance of microcystin test kits relative to key verification parameters including accuracy,
precision, and method detection limit. This verification test took place from July 26 through
August 12, 2010. The reference analysis was performed the week of August 16th, 2010.
This verification report has been reviewed by experts in the field related to microcystin
detection. The following experts have provided input to the TQAP that guided this testing as
well as the verification report and verification statement.
• Andy Lincoff, U. S. EPA Region 9
• Keith Loftin, U.S. Geological Survey
• Daniel Snow, University of Nebraska
The responsibilities of verification test stakeholders include:
• Participate in technical discussions as a part of the test designing process,
• Review and provide input to the TQAP, and
• Review and provide input to the verification report and verification statements.
The AMS Center Water Stakeholder Committee has considered the technology category of
microcystin immunoassay kits a priority area since 2005. The Battelle Verification Test
Coordinator presented the fundamentals of the test design in a stakeholder committee
teleconference in November 2009 to gather input from the stakeholders on the approach.
3.2 Experimental Design
The objective of this verification test was to evaluate the performance of the microcystins test
kits against known concentrations of microcystin in ASTM International Type II deionized (DI)
water, as well as in natural recreational water samples. Battelle conducted this verification test
with recreational samples provided from the Nebraska Department of Environmental Quality
(NDEQ), document review by the Suffolk County Department of Health Services (SCDHS),
with the University of Nebraska Water Sciences Laboratory (WSL) providing reference analyses.
The technologies were used to analyze a variety of water samples spiked with the variants
microcystin-LR, microcystin-LA, and microcystin-RR. Because none of the three technologies
can specify between the different variants, the samples were spiked with individual variants. The
quantitative results from the microcystin test kits were compared to the results from the reference
method by calculating percent differences between the results. The reference method for
microcystin was a liquid chromatography tandem mass spectrometry (LC-MS-MS) method for
the determination of microcystins3. To attain lower levels of detection, a sample preparation
4
-------
method4 was used to extract the microcystins from the water samples and concentrate the
samples using solid phase extraction (SPE). The enzyme-linked immunosorbent assay (ELISA) tube
kit and plate kits were evaluated in terms of:
• Accuracy - comparison of test kit results (samples prepared in DI) to results from a reference
method;
• Precision - repeatability of test kit results from three sample replicates analyzed in DI water,
matrix interference, and recreational water samples;
• Linearity - determination of whether or not the test kit response increases in direct proportion
to the known concentration of microcystin;
• Method detection limit - the lowest quantity of toxin that can be distinguished from the
absence of that toxin (a blank value) at a 95% confidence level;
• Inter-kit lot reproducibility - determination of whether or not the test kit response is
significantly different between two different lots of calibration standards within the kits;
• Matrix interference - evaluation of the effect of natural recreational matrices and chlorophyll-
a on the results of the test kits; and
• Operational and Sustainability factors such as general operation, data acquisition, set-up, and
consumables.
Test kits were operated according to the vendor's instructions by a vendor-trained Battelle
technician. Water samples were tested according to the kit instructions and in compliance with
the TQAP.
3.3 Test Procedures
The ability of each microcystin test kit to determine the concentration of microcystin was
challenged using quality control (QC) samples, performance test (PT) samples and recreational
water (RW) samples. These sample results were also compared to reference method results.
Table 1 presents the test samples analyzed during this verification test.
QC, PT, and RW samples were prepared by Battelle technical staff the day before testing began.
The test samples were prepared in glass volumetric flasks and stored in amber glass vials at 4
degrees Celsius (°C) ± 3 °C. The reference samples that were aliquotted from the test samples
were stored in amber glass bottles at < -10°C until analysis approximately two weeks later.
Replicate samples for the test kits were taken from the same sample bottle. The QC, PT, and
RW samples were prepared blindly for the operator by coding the sample labels to ensure the
results were not influenced by the operator's knowledge of the sample concentration and variant.
Because the reference method is mass specific for different variants, the PT samples for the three
different variants at each spiking concentration were combined into a volumetric flask and
brought up to a known volume with DI water before being sent to the reference laboratory. Then
the calculated dilution factor was applied to the reference method result to determine the PT
sample concentration of each variant. The RW samples were sent for reference analysis without
dilution.
-------
3.3.1 QC Samples
Reagent Blank (KB) samples were prepared from DI water and were exposed to identical
handling and analysis procedures as other prepared samples, including the addition of all
reagents. These samples were used to ensure that no sources of contamination were introduced
in the sample handling and analysis procedures. At least 10% of all the prepared samples were
RBs. As specified in the test kit procedure, at least one positive was analyzed with each ELISA
plate.
3.3.2 PT Samples
PT samples were used to verify the accuracy, precision, linearity, method detection limit, and
inter-kit lot reproducibility of the test kits. All PT samples were prepared at Battelle using DI
water as the water source. PT samples were individually spiked with microcystin-LR,
microcystin-LA, and microcystin-RR and analyzed in triplicate. The concentration levels were
0.10, 0.50, 1.0, 2.0, and 4.0 parts per billion (ppb). These concentration levels were used for
microcystin-LR, and because of the cross reactivity (CR) of the LA and RR microcystin variants,
a 7.0 ppb concentration level was also included to evaluate the dynamic range of the test kits for
these two variants. EPA Guidelines5 were followed to determine the method detection limit
(MDL) of the quantitative test kits. In doing so, a solution with a concentration five times the
vendor's reported detection limit was used. Seven replicate analyses of this solution were made
individually for each variant to obtain precision data with which to determine the method
detection limit.
Additional performance testing was performed to verify the impact of possible interferences on
the performance of the test kits. Two types of possible interferences were tested, the possible
interference of RW water and chlorophyll. Testing was performed using a RW sample with a
low level of native microcystin concentration (based on information from NDEQ). This RW
sample was serial diluted by a factor of 10 with DI water to provide a less concentrated level of
the RW matrix. Then both the original RW sample and diluted RW samples were fortified with
4.0 ppb (tube test kit) or 2.0 ppb (plate test kit) of microcystin LR, LA, or RR. The spike level
chosen was dependent on the detection range of each kit. The test kit results in each of the
matrices were compared to determine the impact of the matrix concentration on the test kit
results. In addition, the results from the matrix samples were compared with the PT sample in
DI water of the same microcystin concentration.
To evaluate the effect of chlorophyll-a as a possible interferent, a DI water sample that was
fortified with 10 milligram/Liter (mg/L) of chlorophyll-a (Sigma Aldrich, Cat # C5753-5MG
Chlorophyll-a from spinach) was prepared by adding known amount of chlorophyll-a into a
volumetric flask and diluting to volume. The chlorophyll was insoluble. Therefore the resulting
solutions were clear solutions containing small black pieces of solid chlorophyll-a. These
solutions were then treated in an identical fashion as the above RW sample. The solution of
chlorophyll-a was serial diluted by a factor of 10 to provide solutions of 10 and 1 mg/L
chlorophyll-a. Then, each of these concentration levels were fortified with 4.0 or 2.0 ppb of
microcystin-LR, -LA, or -RR. The test kit results in each of the matrices were compared to
determine the impact of the chlorophyll-a on the test kit results.
-------
Table 1. Summary of Test Samples
Type of Sample
QC Samples - Kit
Positive Controls
QC Samples- Laboratory
Reagent Blank (RB)
PT Samples - DI Water
PT Samples - RW Matrix
Interference Samples:
ND RW sample and
tenfold dilution
PT Samples -
Chlorophyll-a Matrix
Interference Samples:
Chlorophyll-a sample and
tenfold dilution
PT Samples - Inter-kit lot
reproducibility
RW Samples- Through
freeze-thaw lysing
procedure
RW Samples- Through
the vendor recommended
procedure
Microcystin
Variant
LR
none
LR
LA
RR
LR
LA
RR
LR
LA
RR
LR
LA
RR
Microcystin
Concentration
(ppb)
1.0
0
0.10,0.50,1.0,2.0,
4.0
0.50, 1.0,2.0,4.0,
7.0
0.50, 1.0,2.0,4.0,
7.0
5 times the vendor
stated MDL
5 times the vendor
stated MDL
5 times the vendor
stated MDL
4.0 or 2.0*
4.0 or 2.0*
4.0 or 2.0*
4.0 or 2.0*
4.0 or 2.0*
4.0 or 2.0*
Replicates
1
3
3
3
3
7
7
7
3
3
3
3
3
3
Total Number
of Samples per
Test Kit
1
10% of total test
samples, 2
15
15
15
7
7
7
6
6
6
6
6
6
A second set of vendor provided calibration standards from a different lot
analyzed following the vendor's procedure
Unknown
Unknown
3 samples >20 ppb, 3
samples >10 ppb, 3
samples ND
3 samples at
unknown
concentrations
3
3
27
9
*concentration that is within the calibration range of the test kit
Lastly, the calibration standards provided with the microcystin test kits from different lots could
cause variability in the results across test kits. Therefore, two separate lots of calibration
standards were analyzed using the kits and compared to determine the inter-kit lot
reproducibility.
-------
3.3.3 RWSamples
RW samples were obtained from lakes in and around Lincoln, Nebraska to assess kit
performance in recreational waters. The procedure for collecting and preparing the samples for
verification testing and reference analysis is described in the NDEQ standard operating
procedure for microcystin analysis (SOP# SWS-2320.1A)6. In summary, staff from NDEQ
collected the water samples from lakes where there is a potential for human exposure to
microcystins. The RW samples were collected in brown plastic bottles with head space
remaining and returned to the laboratory where they were frozen and thawed three times to lyse
the cyanobacteria and free the microcystin into solution, making it available for analysis. Then
the samples were split for verification testing and reference analysis. Using analytical data
generated by NDEQ, samples used for ETV testing were selected from lakes that had both
detectable and not-detectable microcystin concentrations. Because not all possible variants are
monitored by the reference method, there could be a discrepancy between the test kit results and
the total microcystin determined by the reference method.
-------
Chapter 4
Quality Assurance/Quality Control
QA/quality control (QC) procedures were performed in accordance with the AMS Center QMP
and the TQAP for this verification test. QA level III, Applied Research was specified for this
test by the EPA Project Officer. These procedures and results are described in the following
sub chapters.
4.1 Reference Method Quality Control
To ensure that this verification test provided suitable data for a robust evaluation of performance,
a variety of data quality objectives (DQOs) were established for this test. The DQOs indicated
the minimum quality of data required to meet the objectives of the verification test. The DQOs
were quantitatively defined in terms of specific data quality indicators (DQIs) and their
acceptance criteria. The quality of the reference method measurements were assured by
adherence to these DQI criteria and the requirements of the reference methods, including the
calibration and QA/QC requirements of the method. Blank samples were required to generate
results below the detection limit and the Laboratory Fortified Matrix (LFM), duplicate, and
Performance evaluation audit (PEA) sample results were required to be within 30% of the
expected results. Continuing calibration verification (CCV) samples were required to be within
20% of the expected result. Battelle visited the reference laboratory prior to initiation of the
reference analysis and audited the data package provided by the reference laboratory following
analysis. More details about the audits are provided in Section 4.2. Table 2 presents these DQIs
and the reference method QC sample results. A total of 22 samples were analyzed by the
reference method, so in cases where the frequency required was one per 20 samples, only one
sample was analyzed to assess the DQI.
The calibration of the LC-MS/MS method was verified by the analysis of a CCV every 10
samples. All of the calibration standards were used as CCVs and were interspersed throughout
the run every five samples. The percent recoveries (%R) of CCVs were calculated from the
following equation:
C
o/0# = ^xlOO (1)
s
where Cs is the measured concentration of the CCV, s is the spiked concentration. If the CCV
analysis differed by more than 20% from the true value of the standard (i.e., % R values outside
of the acceptance window of 80-120%), the instrument was recalibrated before continuing the
analysis. As shown in Table 3, all reference CCV analyses were within the required range.
Spiked samples were analyzed to assess the efficiency of the extraction method. There was a
laboratory fortified matrix (LFM) spike performed every 20 samples; it was assessed by
calculating the spike percent recovery (%Rs) as shown below.
-------
%Rs=-
C -C
•xlOO
(2)
Cis the measured concentration of the spiked sample, C is the measured concentration of the
unspiked samples, and s is the spiked concentration. The spike percent recovery was required to
be within 30% of the spiked amount. The LFM sample results were within this range.
Duplicate samples were analyzed to assess the precision of the reference analysis. There was an
analytical duplicate performed at least every 20 samples and these were expected to be within
30% of each other. The relative percent difference (RPD) of the duplicate sample analysis was
calculated from the following equation:
RPD = -
(C + CD)/2
•xlOO
(3)
Where C is the concentration of the sample analysis, and CD is the concentration of the duplicate
sample analysis. If the RPD was greater than 30%, then the extraction method and the analytical
methods were investigated. As shown in Table 3, the RPD for the duplicate sample analyses was
calculated from the duplicate 30 and 60 ppb CCVs. All RPD were within the acceptable range
for duplicate analysis.
Table 2. DQIs and Summary of Reference Method QC Results
DQI
Performance
Evaluation Audit
(PEA)
Method
contamination
check
Method
Calibration
Check
Method precision
Method accuracy
Method of Assessment
(Frequency)
PEA Samples (Once before testing
begins)
Method Blank (MB) (Once every
20 samples)
Continuing Calibration Verification
(CCV) (Once every 5 samples)
Laboratory Duplicates (Once
every 20 samples)
Laboratory Fortified Matrix (LFM)
Spikes (Once every 20 samples)
Acceptance
Criteria for
Microcystins
70% - 130%
Recovery
< Lowest
Calibration
Standard
80% - 120%
Recovery
< 30% RPD
70% - 130%
Recovery
Results
See Tables 4 and 5 in Section
4.2.1
ND for all three variants
See Table 3
See Table 3
93% LR
79% LA
97% RR
10
-------
Table 3. Summary of Reference Method CCV Percent Recoveries and Method Precision
CCV Cone.
(ppb)
10
30
30
60
60
75
<
LR
99.5
109
96.5
97.6
103
98.7
Yo Recover;
LA
98.2
104
97.1
94.2
109
91.8
f
RR
96.1
112
98.7
93.5
108
101
Duplii
R
LR
NA
12%
5%
:ate CCV S
esults (RPl
LA
NA
7%
14%
ample
>;
RR
NA
13%
14%
NA | NA | NA
NA-not applicable, no duplicate measurements made
4.2 Audits
Three types of audits were performed during the verification test, a performance evaluation audit
(PEA), a technical systems audit (TSA) of the verification test procedures, and an audit of data
quality audit (ADQ). Audit procedures and results are described further below.
4.2.1 Performance Evaluation Audit
A PEA was conducted to assess the quality of the reference measurements made in this
verification test. National Institute of Standards and Technology (NIST) traceable standards of
microcystin are not available; however, the Canadian National Research Council (NRC) offers
standards that have gone through the most validation of any commercially available standards
and were recognized by the vendors and stakeholders reviewing the TQAP as the most reliable
standards. The microcystin-LA variant was not available through the Canadian NRC and
therefore was obtained from Abraxis. The standards from Abraxis also undergo a high level of
scrutiny and are considered a reliable source. The approach of using the microcystin-LA variant
standard from Abraxis was approved by all participating vendors prior to use. The standards
obtained from both sources were prepared at 50 ppb in DI and sent blindly to the reference
laboratory for analysis. These PEA samples were analyzed directly (i.e., without additional
preparation) and were in the mid-level of the calibration range of the reference method. The
standards used to prepare the calibration standards by the reference laboratory were obtained
from EMD Biosciences (microcystin-LR), Sigma Aldrich (microcystin-LA), and ENZO Life
Sciences (microcystin-RR). The results from the analysis are presented in Table 4.
Table 4. PEA Results: Analytical Comparison of Microcystin Standards
Standard
Source
NRC
Canada
Abraxis
#of
Replicates
2
8
2
8
Analysis
Date
27-May
9-Jun
27-May
9-Jun
MC-LR
150% ±3%
135% ±7%
129% ±2%
121% ±6%
MC-LA
Not available
Not available
86% ±2%
86% ± 5%
MC-RR
192% ± 1%
194% ±12%
144% ±0%
153% ±10%
Shading indicates results outside acceptable 30% tolerance based on TQAP
11
-------
The over-recoveries of these standards revealed that the reference laboratory method (using the
standards from alternate sources) did not determine either the NRC or Abraxis standards within
±30%. Therefore it was discussed with the stakeholders, and accepted by the vendors and the
EPA Project Officer, that the reference laboratory would use the two available NRC standards
(LR and RR) as well as LA from Abraxis for their calibration solutions. Therefore, the same
standards were used for test solutions and reference calibration. It is not a common practice for
calibration standards and test solutions to be generated from the same source, but since the
objective was to generate comparable vendor and reference data, it was deemed appropriate for
this verification test due to the difficulties in obtaining certified microcystin standards.
To achieve the low detection limits required to analyze the test samples, an SPE extraction
method was used by the reference laboratory for samples expected to be below 5 ppb. The MDL
of this method was determined using eight solutions of LR, LA, and RR at 0.38 ppb which were
extracted and analyzed. The reference method MDLs for LR, LA, and RR were determined to
be 0.10 ppb, 0.14 ppb, and 0.13 ppb, respectively. Appendix A is the memo from the reference
laboratory presenting the MDL data.
A second PEA was performed to evaluate the extraction method efficiency and the analytical
method at a lower concentration relevant for this verification test. Battelle provided WSL with a
blind spiked DI sample at 0.25 ppb that was extracted in triplicate. The results from the second
PEA are presented in Table 5.
TableS. PEA Results: Evaluation of Extracted Low Level Water Sample
0.25 ppb Spiked
Sample
Replicate 1
Replicate 2
Replicate 3
Average
Standard Deviation
LR
Cone. %
(ppb) Recovery
0.23 92%
0.25 100%
0.23 92%
0.24 95%
0.01 5%
LA
Cone. %
(ppb) Recovery
0.21 84%
0.23 92%
0.22 88%
0.22 88%
0.01 4%
RR
Cone. %
(ppb) Recovery
0.24 96%
0.22 88%
0.26 104%
0.24 96%
0.02 8%
4.2.2 Technical Systems Audit
Battelle's Quality Assurance Officer (QAO) conducted a TSA to ensure that the verification test
was being conducted in accordance with the TQAP and the AMS Center QMP. As part of the
TSA, test procedures were compared to those specified in the TQAP, and data acquisition and
handling procedures as well as the reference method procedures were reviewed. Two
observations on storage of test records and sample handling and custody were documented and
submitted to the Battelle Verification Test Coordinator for response. None of the observations
from the TSA required corrective action. TSA records are permanently stored with the QAO.
4.2.3 Data Quality Audit
Two Audits of Data Quality (ADQ) were performed for this verification test. The first was for the
data collected on the first day of testing and the second was on the complete data package
12
-------
generated during verification test preparation and execution. During the audits, test kit data were
reviewed and verified for completeness, accuracy and traceability.
Because the EPA Project Officer designated this as an EPA Category III verification test, at least
10% of the data acquired were audited. The QAO traced the data from the initial acquisition,
through reduction and statistical analysis, to final reporting to ensure the integrity of the reported
results. All calculations performed on the data undergoing the audit were checked.
Observations and findings (mostly related to test record documentation) were reported and
submitted to the Battelle Verification Test Coordinator for response.
13
-------
Chapter 5
Statistical Methods
The statistical methods used to evaluate the quantitative performance factors listed in Section 3.2
are presented in this chapter. Qualitative observations were also used to evaluate verification test
data.
The microcystin test kits being verified report total microcystin and are also calibrated against
microcystin-LR. Because of this, the kit data was converted from microcystin-LR equivalents to
compare the test kit results to the reference method results for all PT samples. Using cross
reactivity data provided by Beacon, the microcystin-LR equivalents were converted to
microcystin concentration by variant as follows:
r
^
where CiReqmv is the test kit result in equivalents of microcystin-LR and CR is the mass-based
cross reactivity of the variant.7
For the RW samples, each variant identified through analysis by the reference method was
converted to LR-equivalents, and added together to calculate the total microcystins. The total
microcystin-LR equivalents from the RW reference analyses were compared to the total
microcystin results from the test kits. Because not all possible variants are monitored by the
reference method, there could be a discrepancy between the test kit results and the total
microcystin determined by the reference method.
5.1 Accuracy
Accuracy of the test kits verified was assessed relative to the results obtained from the
reference analyses. The results for each set of analyses were expressed in terms of a percent
difference (%D) as calculated from the following equation:
C -C
%D = — - - x 100 (5)
CR
where CT is the microcystin-LR equivalent results from the test kits being verified and CR is the
concentration as determined by the reference method.
5.2 Linearity
Linearity was determined by linear regression with the toxin concentration measured by the
reference method as the independent variable, and the test kit result being verified as the
dependent variable. Linearity was expressed in terms of the slope, intercept, and the coefficient
of determination (r2). In addition, plots of the observed and predicated concentration values were
constructed to depict the linearity for each variant of microcystin being tested.
14
-------
5.3 Precision
The standard deviation (S) of the results for the replicate samples were calculated and used as a
measure of test kit precision at each concentration. S was calculated from the following
equation:
s =
(6)
where n is the number of replicate samples, Ck is the concentration measure for the kth sample,
and C is the average concentration of the replicate samples. The kit precision at each
concentration was reported in terms of the relative standard deviation (RSD) presented below as
equation 7.
RSD = = x 100 (7)
\C
5.4 Method Detection Limit
Method detection limit (MDL) was determined by seven replicate analyses of a fortified sample
with the toxin concentration of five times the vendor's estimated detection limit. The MDL was
calculated from the following equation:
MDL = txS (8)
where t is the Student's value for a 95% confidence level, and S is the standard deviation of the
replicate samples.
5.5 Inter-Kit Lot Reproducibility
Inter-kit lot reproducibility was assessed by calculating the RPD (Equation 3) between OD
results from two lots of calibration standards.
5.6 Matrix Effects
Matrix interference effects also were assessed by using a t-test to compare the microcystin test
kit results generated from samples made by spiking undiluted and diluted interference matrices
with the PT sample results at the same spiked concentration (either 2 or 4 ppb spike
concentration). Each paired t-test was performed using the replicate data from each type of
sample. The null hypothesis is that there is no difference between the two sets of data.
Therefore, the resulting probability (p)-value gives the likelihood of observing a difference as
large as is seen in the data, or a larger difference, if the null hypothesis were true. Therefore, at
the 95% confidence level, p-values less than 0.05 will indicate there is evidence against the null
hypothesis being true and therefore a significant difference between the two sets of data.
15
-------
Chapter 6
Test Results for the Beacon Microcystin Tube Kit
The following sections provide the results of the quantitative and qualitative evaluations of this
verification test for the Beacon microcystin tube kit.
6.1 Beacon Test Kit Summary
As discussed in Chapter 2, the tube kit quantifies total microcystins in water based on an LR
calibration. Other variants of microcystins bind differently to the immunosorbent (i.e., cross
reactivity). Therefore, the relative ability for other microcystins to bind has been experimentally
determined by the vendor. For the tube kit, the CR of microcystin LA is 2% and the CR of
microcystin RR is 73%. In this report, the test kit data have been reported in both test kit results
as LR equivalents and in CR corrected results by variant, based on Equation 4.
Each tube kit contains five calibration solutions including a blank (0 ppb) standard. Following
the analysis method, the tube reader measured the absorbance containing the calibration
solutions at 450 nm wavelengths and the calibration curve was generated based on the OD of
each standard. These results were plotted against concentrations using a vendor-provided
spreadsheet that generated a four parameter curve to quantify the samples. The data from a batch
of samples was considered acceptable when the positive control was recovered within 80% and
130% of 1.0 ppb. According to Beacon, if the data result of a sample was out of range it was
determined to be either above or below the calibration range and either diluted into the linear
range or reported as less than limit of quantification (< LOQ) or non-detectable (ND). The
results below the calibration curve were reported as < LOQ when the OD value was greater than
the lowest standard OD value but less than the negative control sample OD value. A sample was
reported as a ND when the OD value was greater than the negative control sample OD value.
6.2 Test Kit QC Sample
As described in Section 3.3.1, the QC samples analyzed with the tube kit included RB samples
and the positive controls included in the test kit. Ten percent of all samples analyzed were RB
samples, and the results were used to verify that no contamination was introduced during sample
handling. All RB sample results were reported as ND or < LOQ for the tube kit, as presented in
Table 6. Two RB samples were analyzed by the reference method and were determined to be <
LOQ for all three variants.
16
-------
Table 6. RB Sample Results for the Beacon Tube Kit
QC Sample ID
RB 1
RB2
RB3
RB4
RB5
RB6
Plate
1
1
1
o
J
3
3
Mean Concentration (ppb)
ND
ND
< LOQ
ND
ND
ND
The positive controls for the tube kit are presented in Table 7. The vendor stated acceptable
range for recovery of the positive control is between 80% and 130%. At least one positive
control was analyzed at the end of each batch of 20. The positive control for Batch 2 was a 5.0
ppb calibration solution rather than the 1.0 ppb sample used for the rest of the plates. As shown
in Table 7, all tube kit batches used for testing produced a positive control result within the
acceptable range.
Table 7. Positive Control Sample Results for the Beacon Tube Kit
QC Sample ID
1
2a
3
4
5
6
7
8
9
10
11
12
Plate
1
2
3
4
5
6
7
8
9
10
11
12
Mean Concentration (ppb)
0.91
5.8
0.94
0.98
0.98
1.1
0.91
0.82
1.1
1.1
1.1
0.98
Percent Recovery (%)
91%
120%
94%
98%
98%
110%
91%
82%
110%
110%
110%
98%
a 5.0 ppb positive control standard from a different lot
6.3 PT Samples
Tables 8, 9, and 10 present the results for the PT samples for the three variants of microcystin
used during this verification test. In addition, the tables present the sample concentration
corrected for the microcystin cross reactivity, the reference method results and the accuracy
results by variant for the PT samples prepared in DI water. All samples were analyzed in
triplicate.
6.3.1 Accuracy
Tables 8, 9, and 10 present the accuracy results for the tube kit, expressed as percent difference
(%D) between the tube kit concentrations and reference concentrations. As shown in Equation 5
(Section 5.1), the reference method value was used for calculation of accuracy. For LR, the
reference method was within 10% of the spike concentration. For LA and RR, the reference
value was 5-45% lower than the spike concentration depending upon the sample. All data are
17
-------
provided so that the calculation of % D can be calculated relative to the spike value as well as the
reference method if desired by the reader.
Table 8. Beacon Tube Kit Sample Results and Reference Method Results for LR
Sample
Description
0.1 LR
Avg ± SD
0.5 LR
Avg ± SD
1.0 LR
Avg ± SD
2.0 LR
Avg ± SD
4.0 LR
Avg ± SD
Kit Results: LR
Equivalents
(ppb)
-------
Table 9. Beacon Tube Kit Sample Results and Reference Method Results for LA
Sample
Description
0.5 LA
Avg ± SD
LOLA
Avg ± SD
2.0 LA
Avg ± SD
4.0 LA
Avg ± SD
7.0 LA
Avg ± SD
Kit Results: LR
Equivalents (ppb)
0.23
0.28
0.24
0.25 ±0.03
0.28
0.36
0.35
0.33 ±0.04
0.40
0.40
0.37
0.39 ±0.02
0.46
0.46
0.49
0.47 ± 0.02
0.53
0.57
0.52
0.54± 0.02
CR Corrected
Cone, by
Variant (ppb)
12
14
12
13±1.3
14
18
17
16 ±2.0
20
20
19
20± 1.0
23
23
24
23 ±0.77
27
28
26
27 ±1.2
Accuracy by
Variant (%
Difference)
28
34
29
3000% ±3 10%
19
25
24
2200% ±3 10%
11
11
9.9
1100% ±51%
6.7
6.6
7.1
680% ± 26%
4.6
5.0
4.5
470% ±25%
Reference
Concentration
(ppb)
0.40
0.70
1.7
3.0
4.7
19
-------
Table 10. Beacon Tube Kit Sample Results and Reference Method Results for RR
Sample
Description
0.5 RR
Avg ± SD
1.0 RR
Avg ± SD
2.0 RR
Avg ± SD
4.0 RR
Avg ± SD
7.0 RR
Avg ± SD
Kit Results: LR
Equivalents
(ppb)
0.38
0.39
0.45
0.41 ±0.04
1.2
0.88
1.1
1.1 ±0.15
1.7
1.7
2.4
1.9 ±0.40
2.7
2.5
2.6
2.6 ±0.10
3.9
4.0
4.2
4.0±0.10
CR Corrected
Cone, by
Variant (ppb)
0.52
0.53
0.62
0.56 ±0.06
1.6
1.2
1.6
1.5 ±0.21
2.3
2.3
3.3
2.7 ±0.60
3.7
3.5
3.6
3.6 ±0.10
5.4
5.4
5.7
5.5 ±0.10
Accuracy by
Variant (%
Difference)
36%
40%
63%
46% ± 15%
190%
120%
190%
170% ±38%
47%
45%
110%
67% ± 36%
17%
8%
13%
13% ±4%
20%
22%
28%
23% ± 4%
Reference
Concentration
(ppb)
0.38
0.54
1.6
3.2
4.5
For the LR spiked samples, the reference method results were approximately 17% lower than the
spike value. For LR, the percent difference between the tube kit results and the reference method
results ranged from -76% to 21%, with overall average percent difference values ranging from
-76% to 17%. The 0.10 ppb samples were determined as being < LOQ so no %D was calculated.
One sample was greater than the LOQ for the 0.5 ppb samples with a %D of -76%. For the 1.0
ppb samples, the %D ranged from 5% to 21%, corresponding to a maximum absolute difference
from the reference concentration of 0.17 ppb. Similarly, for the 2.0 ppb samples, the %D ranged
from 15% to 21% and the maximum absolute difference from the reference concentration was
0.40 ppb. For the 4.0 ppb samples, the %D ranged from -1% to 21% and the maximum absolute
difference from the reference concentration was 0.80 ppb.
For the LA spiked samples, the reference method results were approximately 15% to 33% lower
than the spike value. For LA, the percent difference between the tube kit results and the reference
method results ranged from 450% to 3400%. These %Ds were calculated based on the
concentration being corrected for the CR of LA of 2%. The concentrations in LR equivalents
more closely tracked the spiked concentrations, suggesting that the CR for LA may have a
different value than was provided by Beacon.
For the RR spiked samples, the reference method results were approximately 20% to 45% lower
than the spike value. For RR, the percent difference between the tube kit results and the reference
method results ranged from 8% to 190%, with overall average percent difference values ranging
20
-------
from 13% to 168%. For the 0.50 ppb samples, the %D ranged from 36% to 63%, corresponding
to an absolute maximum difference from the reference concentration of 0.24 ppb. For the 1.0
ppb samples, the %D ranged from 120% to 190% and the maximum absolute difference from the
reference concentration was 1.1 ppb. The reference result for the 1.0 ppb PT sample was 53% of
the target concentration; the lowest recovered reference measurement. For the 2.0 ppb samples,
the %D ranged from 45% to 110% corresponding to a maximum absolute difference from the
reference concentration of 1.7 ppb. For the 4 ppb samples, the %D ranged from 8% to 17%,
corresponding to a maximum absolute difference from the reference concentration of 0.50 ppb.
For the 7 ppb samples, the %D ranged from 20% to 28%, corresponding to a maximum absolute
difference from the reference concentration of 1.2 ppb.
6.3.2 Precision
Precision results for the tube kit are presented in Table 11. The RSD was determined as a
percentage according to Equation 7 (Section 5.3) for all DI water, matrix interferent and
recreational water samples. The RSDs ranged from 3% to 10% for the LR variant. For the LA
variant, the RSDs ranged from 0% to 18% and from 3% to 22% for the RR variant. The
precision for the RW sample sets ranged from 2% to 99%, however, all except two samples sets
had RSDs less than 14% and within these two sample sets two replicates were very similar to
one another and the third replicate was somewhat different, thus causing the high standard
deviations.
21
-------
Table 11. Beacon Tube Kit Precision Results
Variant
LR
LA
RR
Unknown
Sample Concentration in DI
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
4.0 ppb LR in 1 mg/L Chlorophyll-a DI
4.0 ppb LR in 10 mg/L Chlorophyll-a DI
4.0 ppb LR in lOx dilution of RW Matrix
4.0 ppb LR in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
7.0 ppb
4.0 ppb LA in 1 mg/L Chlorophyll-a DI
4.0 ppb LA in 10 mg/L Chlorophyll-a DI
4.0 ppb LA in lOx dilution of RW Matrix
4.0 ppb LA in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
7.0 ppb
4.0 ppb RR in 1 mg/L Chlorophyll-a DI
4.0 ppb RR in 10 mg/L Chlorophyll-a DI
4.0 ppb RR in lOx dilution of RW Matrix
4.0 ppb RR in RW Matrix
RW1
RW2
RW3
RW4
RW5
RW6
RW7
RW8
RW9
Precision (%RSD)
NA
NA
7%
3%
10%
7%
5%
9%
9%
10%
13%
4%
3%
4%
7%
0%
14%
18%
10%
14%
22%
4%
3%
7%
9%
17%
17%
14%
2%
45%
4%
5%
7%
NA
11%
99%
NA - Result was < LOQ so no calculation of RSD
22
-------
6.3.3 Linearity
The linearity of the tube kit measurements was assessed by performing a linear regression of the
tube kit results against the reference method results for the five PT samples ranging from
0.10 ppb to 4.0 ppb of microcystin LR in DI water and four PT samples ranging from 0.50 ppb to
4.0 ppb for microcystin LA and RR in DI water. Figures 3, 4, and 5 present the results of the
linear regressions for LR, LA, and RR respectively. The slope, intercept, and coefficient of
determination (r2) for each regression equation are shown on the charts. The linear regressions
compared to the reference method results had coefficients of determination of 0.98 for LR, 0.90
for LA and RR.
5.0
4.0
si
a.
a.
3.0
01
.Q
2.0
o
u
as
01
00
y=1.1932x-0.1879
R2 = 0.9792
1.0
0.0
0.0 1.0 2.0 3.0 4.0
Reference Cone, (ppb)
Figure 3. Linearity for the Beacon Tube Kit for LR
5.0
30
y = 3.0906X + 13.254
R2 = 0.8987
10 15 20
Reference Cone, (ppb)
25
30
Figure 4. Linearity for the Beacon Tube Kit for LA
23
-------
o
o
ro
01
00
0.1
y = 0.0625X + 0.2678
R2 = 0.8988
0.0 1.0 2.0 3.0 4.0
Reference Cone, (ppb)
5.0
6.0
Figure 5. Linearity for the Beacon Tube Kit for RR
6.3.4 Method Detection Limit
The MDL was assessed by analyzing at least seven replicates of a sample spiked at
approximately five times the vendor-stated detection limit for the microcystin test kit (which was
0.30 ppb). Table 12 lists the replicate results, the standard deviations for the replicate results, and
shows the calculated MDLs for the three variants. The calculated MDL values were 0.18, 0.34,
and 0.52 ppb for LR, LA, and RR respectively.
Table 12. Detection Limit Results for the Beacon Tube Kit
Variant
Sample Concentration
(ppb)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
Standard Deviation
t (n=7)5
MDL
LR
Mean Cone.
(ppb)
.5
.6
.7
.5
.6
.7
.5
0.09
1.9
0.18
LA
Mean Cone, (ppb LR
equivalents)
0.27
0.56
0.26
0.21
0.66
0.24
0.28
0.18
1.9
0.34
RR
Mean Cone, (ppb LR
equivalents)
.3
.2
.4
.5
.1
.9
.6
0.27
1.9
0.52
6.3.5 Inter-Kit Lot Reproducibility
Two sets of kit calibration standards were analyzed on the sample plate to compare whether or
not the calibration standards from different lots were similar. The data are presented in Table 13.
The OD values were compared by calculation of the RPD between each pair of OD
measurements. In addition, the RPD for each pair of OD results are shown. All RPDs were less
than 14%.
24
-------
Table 13. Inter-kit lot Comparison of Kit Calibration Standards for the Beacon Tube Kit
Standard (ppb)
0
0.30
0.80
2.0
5.0
OD Values
Set A Set B
1.21 1.11
0.820 0.735
0.591 0.658
0.407 0.370
0.275 0.261
RPD
14%
11%
-11%
10%
5%
6.3.6 Matrix Effect
Matrix interference effects were assessed by using a t-test to compare the tube kit results
generated from samples made by spiking undiluted and diluted interference matrices with the PT
sample results at the same concentration. The two possible interfering matrices included a RW
sample both undiluted and after undergoing a tenfold dilution and addition of chlorophyll-a at 10
mg/L and 1.0 mg/L. Tables 14 and 15 provide the tube kit sample results for the RW matrix
interference samples and chlorophyll-a interference samples, respectively, including the average
and SD for each sample. Because this comparison is made to evaluate only the impact of the
matrix on the sample result, LR equivalents are used.
Each paired t-test was performed using the replicate data from each type of sample. The null
hypothesis is that there is no difference between the two sets of data. The resulting probability
(p)-value gives the likelihood of observing a difference as large as is seen in the data, or a larger
difference, if the null hypothesis were true. Therefore, at the 95% confidence level, p-values less
than 0.05 indicate there is evidence against the null hypothesis being true and therefore a
significant difference between the two sets of data.
Table 16 summarizes the results of a paired t-test for both sets of interference data by showing
the p-values associated with each of the applicable comparisons across both types of possible
interfering matrices. Across both the chlorophyll-a and RW results, two out of 18 comparisons
resulted in statistically significant differences. The first statistically significant difference was
between the diluted RW 4.0 ppb LR spikes and 4.0 ppb DI water with a p-value of 0.016.
Table 14 shows that the 4.0 ppb spike into DI water generated an average result of 4.2 ppb
compared with an average result of 3.5 ppb. In addition, the 4.0 ppb undiluted RW LR spikes were
very close to also being significantly different (p=0.054). The other statistically significant difference
was between the LA DI spikes and LA spikes into diluted RW (p=0.041). None of the
comparisons of the chlorophyll-a samples had significant differences.
There are p-values from 18 tests reported in Table 16 and only two of them is smaller than 0.05.
At a significance level of 5%, we would expect one test out of every 20 to have a p-value below
0.05 just by chance, even if the null hypothesis were true in each case. A formal multiple
comparisons adjustment is not needed here because a performance standard is not being
evaluated as this is more of an exploratory test to determine if there is any difference caused by
the matrix. However, a conservative Bonferroni correction, for example, would set the p-value
associated with a significant result at 0.05 divided by 18, corresponding to a p-value of 0.0028
for the individual tests.
25
-------
Table 14. RW Matrix Interferent Sample Results for the Beacon Tube Kit
Variant
Unknown
LR
Sample
Description
UnspikedRW
Matrix (RW 9)
4.0ppbLRinDI
4.0 ppb LR in lOx
dilution of RW
Matrix
4.0ppbLRinRW
Matrix
4.0 ppb LA in DI
LA
RR
4.0 ppb LA in lOx
dilution of RW
Matrix
4.0 ppb LA in RW
Matrix
4.0ppbRRinDI
4.0 ppb RR in lOx
dilution of RW
Matrix
4.0ppbRRinRW
Matrix
Mean Kit Average
Results: LR Result
Equivalents (ppb) (ppb) SD
0.21 0.48 0.47
1.0
0.19
4.3 4.2 0.42
4.5
3.7
3.6 3.5 0.33
3.9
3.2
3.7 3.3 0.31
3.2
3.1
0.46 0.47 0.02
0.46
0.49
0.66 0.75 0.24
1.0
0.56
0.77 0.97 0.18
1.1
1.0
2.7 2.6 0.10
2.5
2.6
4.3 3.8 0.64
3.8
3.1
2.7 3.0 0.51
3.6
2.7
CR Corrected
Cone. By
Variant (ppb)
4.2
3.6
3.9
3.2
3.7
3.2
3.1
"
33
51
28
39
56
51
..
5.9
5.3
4.2
3.7
4.9
3.8
26
-------
Table 15. Chlorophyll-a Interferent Sample Results for the Beacon Tube Kit
Variant
LR
LA
RR
Sample
Description
4.0ppbLRinDI
4.0ppbLRinl.O
mg/L Chlorophyll-a
DI
4.0 ppb LR in 10
mg/L Chlorophyll-a
DI
4.0 ppb LA in DI
4.0 ppb LA in 1.0
mg/L Chlorophyll-a
DI
4.0 ppb LA in 10
mg/L Chlorophyll-a
DI
4.0ppbRRinDI
4.0 ppb RR in 1.0
mg/L Chlorophyll-a
DI
4.0 ppb RR in 10
mg/L Chlorophyll-a
DI
Mean Kit Average
Results: LR Result
Equivalents (ppb) (ppb) SD
4.3 4.2 0.42
4.5
3.7
3.9 3.9 0.29
4.2
3.6
3.4 3.6 0.17
3.7
3.6
0.46 0.47 0.02
0.46
0.49
0.42 0.45 0.03
0.44
0.47
0.50 0.50 0
0.50
0.50
2.7 2.6 0.10
2.5
2.6
3.3 3.1 0.23
3.2
2.9
2.8 3.1 0.28
3.3
3.1
CR Corrected
Cone. By
Variant (ppb)
:
3.9
4.2
3.6
3.4
3.7
3.6
23
21
22
24
25
25
25
3.6
4.6
4.4
4.0
3.8
4.5
4.3
27
-------
Table 16. Statistical Comparisons between Interference Samples
Description of Comparison
4.0 ppb in DI compared with 4.0 ppb in lOx
dilution of RW
4.0 ppb in DI compared with 4.0 ppb in
undiluted RW
4.0 ppb in undiluted RW compared with lOx
dilution of RW
4.0 ppb in DI compared with 4.0 ppb in 1.0
mg/L Chlorophyll-a DI
4.0 ppb in DI compared with 4.0 ppb in 10
mg/L Chlorophyll-a DI
4.0 ppb in 1.0 mg/L Chlorophyll-a DI
compared with 4.0 ppb in 10 mg/L
Chlorophyll-a DI
p-value (D-different, ND-not different)
LR
0.016 (D)
0.054 (ND)
0.404 (ND)
0.150(ND)
0.168(ND)
0.184(ND)
LA
0.202 (ND)
0.041 (D)
0.201 (ND)
0.199(ND)
0.078 (ND)
0.086 (ND)
RR
0.088 (ND)
0.399 (ND)
0.244 (ND)
0.060 (ND)
0.194(ND)
0.772 (ND)
Shading indicates a statistically significant difference
6.4 RW Sample Results
Table 17 presents the RW results for the tube kit and the reference analysis. The concentrations
were determined by the reference method for only three of the approximately 80 variants that are
naturally occurring in recreational waters. The total microcystins measured by the tube kit may
have other variants present that would not have been detected by the reference method.
Therefore, no quantitative comparison was made between the tube kit and the reference method
results. The reference data were converted into LR-equivalents according to the tube kit cross
reactivity for the variants. In general, the samples were in agreement when comparing the tube
kit to the reference method. In particular, results from RW 1, RW 3, and RW 6 were within 1
ppb of the reference method result. This indicates that the LR, LA, and RR variants make up a
considerable proportion of the microcystins that are measurable by the tube kit.
28
-------
Table 17. Recreational Water Sample Results for the Beacon Tube Kit
Sample
Description
RW 1 (lOx
dilution)
RW 2 (lOx
dilution)
RW 3 (lOx
dilution)
RW4
RW 5 (4x
dilution)
RW6
RW 6 (2x
dilution)
RW7
RW 8
RW 9(RW
Matrix)
Kit Re suits: LR
Equivalents
(Ppb)
2.4
2.0
2.6
0.49
0.50
0.52
1.5
0.70
0.72
0.62
0.67
0.66
0.10
0.11
0.10
1.6
1.6
1.9
0.86
0.90
0.82
-------
6.5.2 Cost and Consumables
According to the vendor, once the analysis is complete, the remaining solutions and tube
contents may be flushed down the drain with no hazardous waste being generated for disposal.
Since waste disposal requirements vary from state-to-state, the reader is encouraged to consult
with the appropriate state government agency for proper waste disposal requirements.
The listed price for the tube kit at the time of the verification test was $200 for a 40 tube kit that
will analyze 24 samples. The kit has a 6-month shelf life as received and should be stored at 4 -
8 °C. Other consumables not included in the kit are pipettes, pipette tips, and distilled or DI
water.
30
-------
Chapter 7
Test Results for the Beacon Microcystin Plate Kit
The following sections provide the results of the quantitative and qualitative evaluations of this
verification test for the Beacon microcystin plate kit.
7.1 Beacon Microcystin Plate Kit Summary
As discussed in Chapter 2, the plate kit quantifies total microcystins in water based on an LR
calibration. Other variants of microcystins bind differently to the immunosorbent. Therefore,
the relative ability for other microcystins to bind has been experimentally determined by the
vendor and is published in the vendor literature as the cross reactivity (CR) of the microcystin.
For the plate kit, the CR of microcystin LA is 2% and the CR of microcystin RR is 73%. The
published CR value was determined using a different source of LA than was used for this study
and CR values can vary with microcystin concentration which can impact the quantitative
results. In this report, the test kit data have been reported in both test kit results as LR
equivalents and in CR corrected results by variant, based on Equation 4.
The plate kit requires that each standard and sample be analyzed in duplicate, and the raw data
output from the plate reader software reports a mean concentration of the duplicate analyses.
Therefore, a sample indicated in Table 1 would have three replicates that corresponded to six
wells being filled as part of the plate kit. Each plate kit plate contains five calibration solutions,
including a blank (0 ppb) standard. Following the analysis method, the plate reader measured
the absorbance of the wells containing the calibration solutions at 450 nm wavelengths and the
calibration curve was generated based on the OD of each well. These results were plotted
against concentrations using a 4-parameter curve to quantify the rest of the samples. The results
below the calibration curve were reported as < LOQ when the OD value was greater than the
lowest standard OD value but less than the negative control sample OD value. A sample was
reported as a ND when the OD value was greater than the negative control sample OD value.
The coefficient of variation (CV) of the duplicate analyses was reported as a gauge for accurate
quantification of microcystins. According to Beacon, the plate was acceptable when the positive
control was recovered within 80% and 130% of a 1.0 ppb positive control and the calibration
standard %CVs were less than 10%.
7.2 Test Kit QC Sample
As described in Section 3.3.1, the QC samples analyzed with the plate kit included RB samples
and the positive control included in the test kit. Ten percent of all samples analyzed were RB
samples, and the results were used to verify that no contamination was introduced during sample
handling. All RB sample results were < LOQ for the plate kit and are presented in Table 18.
Two RB samples were analyzed by the reference method and determined to be < LOQ for all
three variants.
-------
Table 18. RB Sample Results for the Beacon Plate Kit
QC Sample ID
RB 1
RB2
RB3
RB4
RB5
RB6
Plate
1
1
1
3
o
J
3
Mean Concentration (ppb)
-------
can be calculated relative to the spike value as well as the reference method if desired by the
reader.
Table 20. Beacon Plate Kit Sample Results and Reference Method Results for LR
Sample
Description
Kit Results:
LR
Equivalents
(ppb)
CR Corrected
Cone, by Variant
(ppb)
Accuracy by
Variant (%
Difference)
Reference
Concentration
(ppb)
0.1 LR
Avg ± SD
0.5 LR
Avg ± SD
1.0 LR
Avg ± SD
2.0 LR
Avg ± SD
0.13
0.15
0.18
0.16 ±0.02
0.60
0.50
0.49
0.60
0.62
0.72
0.67
0.53
0.49
0.50
0.57 ±0.08
1.1
1.2
1.0
1.1 ±0.10
2.6
2.3
2.3
2.4 ±0.20
0.13
0.15
0.18
0.16 ±0.02
0.60
0.50
0.49
0.60
0.62
0.72
0.67
0.53
0.49
0.50
0.57 ±0.08
1.1
1.2
1.0
1.1 ±0.10
2.6
2.3
2.3
2.4 ±0.20
34%
51%
81%
55% ±24%
42%
20%
16%
43%
48%
72%
59%
26%
18%
19%
36% ± 19%
27%
47%
26%
33% ±12%
38%
21%
22%
27% ±9%
0 10
0.42
0.83
1.9
33
-------
Table 21. Beacon Plate Kit Sample Results and Reference Method Results for LA
Sample
Description
Kit Results:LR
Equivalents
(ppb)
CR Corrected
Cone. By Variant
(ppb)
Accuracy by
Variant (%
Difference)
Reference
Concentration
(ppb)
0.5 LA
Avg ± SD
LOLA
Avg ± SD
2.0 LA
Avg ± SD
4.0 LA
Avg ± SD
7. OLA
Avg ± SD
0.18
0.19
0.18
0.19
0.17
0.18
0.19
0.19
0.21
0.24
0.24
0.20 ±0.02
0.21
0.21
0.29
0.30
0.31
0.27
0.26 ±0.04
0.28
0.22
0.33
0.32
0.34
0.36
0.31 ±0.05
0.38
0.30
0.34
0.46
0.44
0.44
0.39 ±0.06
0.39
0.42
0.35
0.55
0.52
0.45
0.45 ±0.08
8.8
9.3
9.2
10
8.6
9.0
9.4
9.3
10
12
12
9.8 ±1.2
11
11
15
15
15
13
13 ±2.0
14
11
16
16
17
18
16 ±2.5
19
15
17
23
22
22
20 ±3.0
20
21
18
28
26
22
22 ± 4.0
2100%
2200%
2200%
2300%
2000%
2100%
2200%
2200%
2500%
2900%
2900%
2300% ± 290%
1400%
1400%
2000%
2000%
2100%
1800%
1800% ±300%
720%
560%
870%
840%
900%
970%
8 10% ±150%
530%
410%
470%
660%
640%
630%
560% ±100%
320%
350%
270%
490%
450%
370%
370% ±81%
0.40
0.70
1.7
3.0
4.7
34
-------
Table 22. Beacon Plate Kit Sample Results and Reference Method Results for RR
Sample
Description
Kit Results:
LR
Equivalents
(ppb)
CR Corrected
Cone. By Variant
(ppb)
Accuracy by
Variant (%
Difference)
Reference
Concentration
(ppb)
0.5 RR
Avg ± SD
1.0 RR
Avg ± SD
2.0 RR
Avg ± SD
0.42
0.41
0.57
0.71
0.66
0.75
0.64
0.62
0.64
0.57
0.60 ±0.11
1.1
1 l
1.2
1.1 ±0.10
2.0
2.3
1.9
2.1 ±0.20
0.58
0.56
0.78
0.97
0.90
1.0
0.88
0.85
0.87
0.77
0.82 ±0.15
1.5
1 4
1.6
1.5 ±0.10
2.8
3.2
2.5
2.8 ±0.30
51%
49%
100%
160%
140%
170%
130%
120%
130%
100%
120% ± 40%
180%
170%
190%
180% ±12%
74%
100%
59%
77% ±21%
0.38
0 54
1 6
For the LR spiked samples, the reference method results ranged from 0% - 17% less than the
target concentration. For LR, the percent difference ranged from 16% to 81%, with overall
average percent difference values ranging from 27% to 55% between the plate kit and the
reference method. For the 0.1 ppb samples, the %D ranged from 34% to 81%, corresponding to
an absolute maximum difference from the reference concentration of 0.08 ppb. For the 0.50 ppb
samples, the %D ranged from 16% to 72%, but the absolute difference from the reference
concentration was no more than 0.30 ppb. For the 1.0 ppb samples, the %D ranged from 26% to
47%, corresponding to a maximum absolute difference from the reference concentration was
0.37 ppb. Similarly, for the 2.0 ppb samples, the %D ranged from 21% to 38% and the maximum
absolute difference from the reference concentration was 0.70 ppb. No replicates are given for
the 4.0 ppb samples as the samples were above the calibration range of the plate kit. The sample
was not diluted because lower concentrations had already been analyzed.
For the LA spiked samples, the reference method results were 15% - 33% less than the target
concentration. For LA, the percent difference ranged from 270% to 2900%. These %Ds were
calculated based on the concentration being corrected for the CR of LA. The LR equivalents
were closer to the spiked concentration, suggesting that the actual CR for LA may have a
different value than was published in Beacon's instruction booklet. The published CR value was
determined using a different source of LA than was used for this study. Also, CR values can
vary with microcystin concentration which may have contributed to the large %Ds observed
here. Calculation of a range of concentrations based on a CR determined at a single point of the
dose response curve (50% preferential binding of microcystin) used to generate the published CR
is not recommended by the vendor.
35
-------
For the RR spiked samples, the reference method results were 20% - 46% less than the target
concentration. For RR, the percent difference ranged from 49% to 181%, with overall average
percent difference values ranging from 77% to 180%. For the 0.5 ppb samples, the %D ranged
from 49% to 170%, corresponding to an absolute maximum difference from the reference
concentration of 0.62 ppb. For the 1 ppb samples, the %D ranged from 168% to 192% and the
maximum absolute difference from the reference concentration was 1.1 ppb. The reference
result for the 1.0 ppb PT sample was only 53% of the target concentration, the lowest recovered
reference measurement. For the 2.0 ppb samples, the %D ranged from 59% to 100%
corresponding to a maximum absolute difference from the reference concentration of 1.6 ppb.
No replicates are given for the 4.0 and 7.0 ppb samples as the samples were above the calibration
range of the plate kit. The sample was not diluted because lower concentrations had already
been analyzed.
7.3.2 Precision
Precision results for the plate kit are presented in Table 23. The RSD was determined as a
percentage according to Equation 7 (Section 5.3) for all DI water, matrix interferent and
recreational water samples. The RSDs ranged from 1% to 15% for the LR variant. For LA, the
RSDs ranged from 3% to 16%, and from 4% tol 8% for the RR variant. The precision for the
RW samples sets ranged from 3% to 59%. The highest RSD at 59% is from RW 4; however, all
other RW RSDs were below 9%.
36
-------
Table 23. Beacon Plate Kit Precision Results
Variant
LR
LA
RR
Unknown
Sample Concentration in DI
O.lOppb
0.50 ppb
1.0 ppb
2.0 ppb
2.0 ppb LR in 1.0 mg/L Chlorophyll-a DI
2.0 ppb LR in 10 mg/L Chlorophyll-a DI
2.0 ppb LR in lOx dilution of RW Matrix
2.0 ppb LR in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
4.0 ppb
7.0 ppb
2.0 ppb LA in 1.0 mg/L Chlorophyll-a DI
2.0 ppb LA in 10 mg/L Chlorophyll-a DI
2.0 ppb LA in lOx dilution of RW Matrix
2.0 ppb LA in RW Matrix
0.50 ppb
1.0 ppb
2.0 ppb
2.0 ppb RR in 1.0 mg/L Chlorophyll-a DI
2.0 ppb RR in 10 mg/L Chlorophyll-a DI
2.0 ppb RR in lOx dilution of RW Matrix
2.0 ppb RR in RW Matrix
RW1
RW2
RW3
RW4
RW 4 (4x dilution)
RW5
RW6
RW7
RW8
RW9
Precision (%RSD)
15%
14%
9%
7%
13%
1%
9%
6%
12%
16%
16%
15%
16%
5%
10%
3%
5%
18%
4%
12%
7%
9%
17%
17%
4%
2%
2%
7%
18%
3%
7%
NA
3%
8%
NA - Result was < LOQ so no calculation of RSD
37
-------
7.3.3 Linearity
The linearity of the plate kit measurements was assessed by performing a linear regression of the
plate kit results against the reference method results for the five PT samples ranging from
0.10 ppb to 4.0 ppb of microcystin LR in DI water and four PT samples ranging from 0.50 ppb to
4.0 ppb for microcystin LA and RR in DI water. Figures 6, 7, and 8 present the results of the
linear regressions for LR, LA, and RR respectively. The slope, intercept, and coefficient of
determination (r2) for each regression equation are shown on the charts. The linear regressions
compared to the reference method results had coefficients of determination of 0.99, 0.76, and
0.91 for LR, LA, and RR respectively.
3.0
^ 2.5
a.
a.
•y 2.0
8
1-5
o
-------
3.5
3.0
3"
£2.5
2.0
u
8
V
I 1-5
c
o
01
00
1.0
y = 1.6324X + 0.2873
R2 = 0.9116
0.5
0.0
0.0
1.0 2.0
Reference Cone, (ppb)
3.0
Figure 8. Linearity for the Beacon Plate Kit for RR
7.3.4 Method Detection Limit
The MDL was assessed by analyzing at least seven replicates of a sample spiked at
approximately five times the vendor-stated detection limit for the microcystin test kit (which was
0.10 ppb). Table 24 lists the replicate results, the %CV of the duplicate plate kit analysis for each
individual replicate, the standard deviations for the replicate results, and shows the calculated
MDLs for the three variants. The calculated MDL values were 0.15, 0.04, and 0.20 ppb for LR,
LA, and RR respectively.
Table 24. Detection Limit Results for the Beacon Plate Kit
Variant
Sample
Concentration (ppb)
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
Standard Deviation
t value
n
MDL
LR
Mean
Cone.
(ppb)
0.60
0.50
0.49
0.60
0.62
0.72
0.67
0.53
0.49
0.50
NA
0.082
1.8
10
0.15
%cv
7.7
5.7
3.1
8.4
1.0
1.1
15
7.0
13
2.4
NA
LA
Mean Cone.
(ppb LR
Equivalents)
0.18
0.19
0.18
0.19
0.17
0.18
0.19
0.19
0.21
0.24
0.24
0.024
1.8
11
0.043
%cv
13
10
4.0
3.7
1.0
7.8
11
1.4
5.7
14
1.5
RR
Mean Cone.
(ppb LR
Equivalents)
0.42
0.41
0.57
0.71
0.66
0.75
0.64
0.62
0.64
0.57
NA
0.11
1.8
10
0.20
%cv
4.9
2.5
2.7
9.3
14
6.1
2.8
16
11
5.4
NA
39
-------
7.3.5 Inter-Kit Lot Reproducibility
Two sets of kit calibration standards were analyzed on the sample plate to compare whether or
not the calibration standards from different lots were similar. The OD values were compared by
calculation of the RPD between each pair of OD measurements. The RPD for each pair of OD
results are shown along with the OD data in Table 25. The RPDs were less than 14%.
Table 25. Inter-kit lot Comparison of Kit Calibration Standards for the Beacon Plate Kit
Standard
Std 0 ppb
StdO.l
Std 0.3 ppb
Std 0.8 ppb
Std 2.0 ppb
OD
Set A
1.35
1.43
1.15
1.13
0.830
0.830
0.546
0.555
0.398
0.390
Values
SetB
1.31
1.37
1.01
1.02
0.738
0.747
0.527
0.495
0.362
0.340
RPD
3%
4%
13%
10%
12%
11%
4%
11%
9%
14%
7.3.6 Matrix Effects
Matrix interference effects were assessed by using a t-test to compare the plate kit results
generated from samples made by spiking undiluted and diluted interference matrices with the PT
sample results at the same concentration. The two possible interfering matrices included a RW
sample both undiluted and after undergoing a tenfold dilution and addition of chlorophyll-a at 10
mg/L and 1.0 mg/L. Tables 26 and 27 give the plate kit sample results for the RW matrix
interference samples and chlorophyll-a interference samples, respectively, including the average
and SD for each sample. Because this comparison is made to evaluate only the impact of the
matrix on the sample result, LR equivalents are used.
Each paired t-test was performed using the replicate data from each type of sample. The null
hypothesis is that there is no difference between the two sets of data. The resulting probability
(p)-value gives the likelihood of observing a difference as large as is seen in the data, or a larger
difference, if the null hypothesis were true. Therefore, at the 95% confidence level, p-values less
than 0.05 will indicate there is evidence against the null hypothesis being true and therefore a
significant difference between the two sets of data.
Table 28 summarizes the results of a paired t-test for both sets of interference data by showing
the p-values associated with each of the applicable comparisons across both types of possible
interfering matrices. Across both the RW and chlorophyll-a results, five out of 18 comparisons
resulted in statistically significant differences. The 2.0 ppb LA spike into DI water was
significantly different from the 2.0 ppb LA spike into the tenfold diluted RW samples (p=0.006).
Table 26 shows that the 2.0 ppb spike into DI water generated an average result of 0.31 ppb
compared with an average result of 0.77 ppb the spike into 1.0 mg/L chlorophyll-a samples. The
other statistically significant difference with the RW matrix was between the RR spikes into
undiluted and diluted RW (p=0.006). These two samples were not significantly different from
40
-------
the PT sample spike in DI water, but they were different from each other with average
concentrations of 0.77 ppb for the diluted RW and 0.35 ppb for the undiluted RW.
The 2.0 ppb LR spike into DI water was significantly different from the 2.0 ppb LR spike into
both 10 mg/L (p=0.002) and 1 mg/L (p=0.003) chlorophyll-a. Table 27 shows that the 2.0 ppb
spike into DI water generated an average result of 2.4 ppb compared with an average result of
0.45 and 0.53 ppb for the spike into 1.0 mg/L and 10 mg/L chlorophyll-a, respectively. For LA,
the 1.0 mg/L and 10 mg/L chlorophyll-a average results were different by 0.01 ppb and when
compared to the DI water results, the p-value for the 1.0 mg/L chlorophyll-a solution was 0.05
and therefore considered to be significantly different from the DI water spike. The 10 mg/L
chlorophyll-a results were also very close to being significantly different at the 95% confidence
interval (p= 0.066). The 2.0 ppb spike into DI water generated an average result of 0.53 ppb
compared with an average result of 0.14 ppb for the spikes into 1.0 mg/L and 10 mg/L
chlorophyll.
There are p-values from 18 tests reported in Table 28 and five of them are smaller than 0.05. At
a significance level of 5%, one would expect one test out of every 20 to have a p-value below
0.05 by random chance, even if the null hypothesis were true in each case. A formal multiple
comparisons adjustment is not needed here because a performance standard is not being
evaluated as this is more of an exploratory test to determine if there is any difference caused by
the matrix. However, a conservative Bonferroni correction, for example, would set the p-value
associated with a significant result at 0.05 divided by 18, corresponding to a p-value of 0.0028
for the individual tests.
Given that the molecular basis on which the test kits operate is well-characterized and
o
understood from the literature , Table 27 provided unexpected results. Two variants (LR and
LA) demonstrated an interference effect but the third variant (RR) did not. This could have been
caused by a number of factors, such as chlorophyll-a source and stability and as mentioned in
Section 3.3.2, the fact the chlorophyll-a was not in solution when analyzed. However, due to the
limited number of replicates that were analyzed, additional testing would be required to provide
a better understanding as to whether there is matrix interference due to chlorophyll-a, or another
variable not investigated in this verification testing.
41
-------
Table 26. RW Matrix Interferent Sample Results for the Beacon Plate Kit
Variant
Unknown
LR
RR
Sample
Description
UnspikedRW
Matrix (RW 9)
2.0ppbLRinDI
2.0 ppb LR in lOx
dilution of RW
Matrix
2.0ppbLRinRW
Matrix
2.0 ppb LA in DI
2.0 ppb LA in lOx
dilution of RW
Matrix
2.0 ppb LA in RW
Matrix
2.0ppbRRinDI
2.0 ppb RR in lOx
dilution of RW
Matrix
2.0ppbRRinRW
Matrix
Mean Kit
Results: LR Average
Equivalents Result
(ppb) (ppb) SD
0.32 0.33 0.02
0.36
0.31
2.6 2.4 0.20
2.3
2.3
2.1 2.2 0.20
2.2
2.5
2.2 2.2 0.10
2.4
2.1
0.28 0.31 0.05
0.22
0.33
0.32
0.34
0.36
0.83 0.77 0.05
0.73
0.74
0.36 0.35 0.02
0.37
0.34
2.7 2.6 0.10
2.5
2.6
4.3 3.7 0.60
3.8
3.1
2.7 3.0 0.50
3.6
2.7
CR Corrected
Cone. By
Variant (ppb)
2.4
2.1
2.2
2.5
2.2
2.4
2.1
15
41
36
37
18
18
17
3.6
5.9
5.3
4.2
3.7
4.9
3.8
42
-------
Table 27. Chlorophyll-a Interferent Sample Results for the Beacon Plate Kit
Variant
LR
LA
RR
Sample
Description
2.0ppbLRinDI
2.0ppbLRinl.O
mg/L Chlorophyll-
aDI
2.0 ppb LR in 10
mg/L Chlorophyll-
aDI
2.0 ppb LA in DI
2.0 ppb LA in 1.0
mg/L Chlorophyll-
aDI
2.0 ppb LA in 10
mg/L Chlorophyll-
aDI
2.0ppbRRinDI
2.0 ppb RR in 1.0
mg/L Chlorophyll-
aDI
2.0 ppb RR in 10
mg/L Chlorophyll-
aDI
Mean Kit
Results: LR Average
Equivalents Result
(ppb) (ppb) SD
2.6 2.4 0.20
2.3
2.3
0.52 0.45 0.06
0.53
0.37
0.42
0.43
0.45
0.53 0.53 0.01
0.53
0.52
0.28 0.31 0.05
0.22
0.33
0.32
0.34
0.36
0.14 0.13 0.01
0.14
0.13
0.14
0.12
0.15 0.14 0.01
0.15
0.13
2.7 2.6 0.10
2.5
2.6
3.3 3.1 0.20
3.2
2.9
2.8 3.1 0.30
3.3
3.1
CR Corrected
Cone. By
Variant (ppb)
2.4
0.52
0.53
0.37
0.42
0.43
0.45
0.53
0.53
0.52
15
7.2
6.8
6.7
6.9
6.2
7.7
7.4
6.4
3.6
4.6
4.4
4.0
3.8
4.5
4.3
43
-------
Table 28. Statistical Comparisons between Interference Samples
Description of Comparison
2.0 ppb in DI compared with 2.0 ppb in lOx
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in
undiluted RW
2.0 ppb in undiluted RW compared with lOx
dilution of RW
2.0 ppb in DI compared with 2.0 ppb in 1.0
mg/L Chlorophyll-a DI
2.0 ppb in DI compared with 2.0 ppb in 10
mg/L Chlorophyll-a DI
2.0 ppb in 1.0 mg/L Chlorophyll-a DI
compared with 2.0 ppb in 10 mg/L
Chlorophyll-a DI
p-value (D-different, ND-not different)
LR
0.470 (ND)
0.289 (ND)
0.912 (ND)
0.002 (D)
0.003 (D)
0.384 (ND)
LA
0.006 (D)
0.194(ND)
0.006 (D)
0.045 (D)
0.066 (ND)
0.494 (ND)
RR
0.088 (ND)
0.399 (ND)
0.244 (ND)
0.060 (ND)
0.194(ND)
0.772 (ND)
Shading indicates a statistically significant difference
7.4 RW Sample Results
Table 29 presents the RW results for the plate kit and the reference analysis. The concentrations
were determined by the reference method for only three of the approximately 80 variants that are
naturally occurring in recreational waters. The total microcystins measured by the plate kit may
have other variants present that would not have been detected by the reference method.
Therefore, no quantitative comparison was made between the plate kit and the reference method
results. The reference data have been converted into LR-equivalents according to the plate kit
cross reactivity for the variants. In general, the samples that were determined to have higher
total concentrations by the plate kit had higher total concentrations as determined by the
reference method. All of the plate kit total microcystin results were greater than the reference
method results, which were consistent with the likelihood that all of the microcystins were not
being measured by the reference method, which only measured three variants. However, the
results of the plate kit were usually within 25% of the reference method, indicating that the LR,
LA, and RR variants make up a considerable proportion of the microcystins that are measurable
by the plate kit.
44
-------
Table 29. Recreational Water Sample Results for the Beacon Plate Kit
Sample
Description
RW 1 (20x
dilution)
RW 2 (20x
dilution)
RW 3 (20x
dilution)
RW4
RW 4 (4x
dilution)
RW5
RW 6 (2x
dilution)
RW7
RW8
RW 9 (RW
Matrix)
Kit Results:
LR
Equivalents
(ppb)
1.6
1.5
1.6
1.6
0.67
0.67
0.65
0.67
0.68
0.65
0.38
0.35
0.33
0.23
0.27
0.33
.0
.1
.0
.1
.2
.3
-------
above the quantification range. The procedure includes two incubation periods that are 30
minutes each. Previous knowledge or training on the use of micro-pipettes and or multi-channel
pipettes with 96-well plates is recommended for consistent readings. The Battelle operator that
was trained by the vendor had experience with ELISA kits and pipetting. A spectrophotometer
plate reader is necessary for obtaining the spectrophotometric readings that are then analyzed
using any commercial ELISA evaluation program (for example, 4-parameters, Logit/Log or
alternatively point to point).
7.5.2 Cost and Consumables
According to the vendor, once the analysis is complete, the remaining solutions and tube/plate
contents may be flushed down the drain with no hazardous waste being generated for disposal.
Since waste disposal requirements vary from state-to-state, the reader is encouraged to consult
with the appropriate state government agency for proper waste disposal requirements.
At the time of the verification test, the list price for the plate kit that will analyze 84 samples was
$275. According to the vendor, the kits have a 6-month shelf life as received and should be
stored at 4 - 8 °C. Other consumables not included in the kit are pipettes, pipette tips, and
distilled or DI water.
46
-------
Chapter 8
Performance Summary for the Beacon Tube and Plate Test Kits
8.1 Performance Summary for the Beacon Tube Test Kit
The verification of the Beacon tube test kit is summarized by the parameters described in Table
30.
Table 30. Beacon Tube Test Kit Performance Summary
Verification Parameters | LR | LA | RR
Accuracy (ppb, range of %D)
0.10
-------
Operational Factors. The test kit operator reported that the tube kit was easy to use. Solution or
sample preparation is minimal, involving diluting the wash solution or the samples that are above
the quantification range. The procedure includes two incubation periods that are 20 minutes
each. A spectrophotometer tube reader is necessary for obtaining the spectrophotometric
readings that are then analyzed using any commercial ELISA evaluation program (for example,
four parameters, Logit/Log or alternatively point to point).
The listed price for the tube kit at the time of the verification test was $200 for a 40 tube kit that
will analyze 24 samples. The kit has a 6-month shelf life as received and should be stored at 4 -
8 °C. Other consumables not included in the kit are pipettes, pipette tips, and distilled or DI
water.
8.2 Performance Summary for the Beacon Plate Test Kit
The verification of the Beacon Plate Test Kit is summarized by the parameters described in
Table 31.
Table 31. Beacon Plate Test Kit Performance Summary
Verification Parameters
LR
LA
RR
Accuracy (ppb, range of %D)
0.10
34% to 81%
0.50
1.0
2.0
16% to 72%
26% to 47%
21% to 3 8%
4.0
7.0
270% to 2900% The
LR equivalent values
were closer to the
spiked values
suggesting that the 2%
CR for LA may differ
from those provided by
Beacon.
49% to 170%
170% to 190%
59% to 100%
Precision (range of %RSD)
Precision (RW samples)
Linearity (y=)
Method Detection Limit (ppb)
l%to 15%
3% to 16%
4% to 18%
All RSD results < 9%, except one at 59%
1.2x + 0.052
r2 = 0.99
0.15
2.9x + 9.8
r2 = 0.76
0.043
1.6x + 0.29
r2 = 0.91
0.20
Inter-kit lot reproducibility. Calibration standards from two different lots were measured and
the RPD of the resulting optical densities were all less than 14%.
Matrix Interference. Matrix interference effects were assessed by using a t-test to compare the
plate kit results generated from samples made by spiking undiluted and diluted interference
matrices with the PT sample results at the same concentration. For chlorophyll-a and RW
matrices, five comparisons resulted in statistically significant differences: 1) 4 ppb LA spike into
DI water was significantly different from the 4 ppb LA spike into the tenfold diluted RW
samples (p=0.006); 2) the RW matrix was between the RR spikes into undiluted and diluted RW
(p=0.006); 3) 4 ppb LR spike into DI water was significantly different from the 4 ppb LR spike
into the 10 mg/L (p=0.002); 4) the 1 mg/L (p=0.003) chlorophyll-a; and 5) for LA, the 1 mg/L
and 10 mg/L chlorophyll-a solutions average results were different by 0.007 ppb and when
compared to the DI water results. The 1.0 mg/L chlorophyll-a solution results were statistically
different (p = 0.045) and the 10 mg/L chlorophyll-a results were very close to being significant
at the 95% confidence interval (p= 0.066).
48
-------
Recreational Water (RW). Because the reference method did not measure all possible
microcystin variants, no quantitative comparison was made between the plate kit and the
reference method results. The reference data were converted into LR-equivalents according to
the plate kit cross reactivity for the variants. In general, the samples that were determined to
have higher total concentrations by the plate kit had higher total concentrations as determined by
the reference method. All of the plate kit total microcystin results were greater than the
reference method results, this was consistent with the likelihood that all of the microcystins were
not being measured by the reference method, which only measured three variants. However, the
results of the plate kit were usually within 25% of the reference method, indicating that the LR,
LA, and RR variants make up a significant proportion of the microcystins that are measurable by
the plate kit.
Operational Factors. The test kit operator reported that the plate kit was easy to use. Solution
or sample preparation is minimal, mostly involving diluting the wash solution or the samples that
are above the quantification range. The procedure includes two incubation periods that are 30
minutes each. Previous knowledge or training on the use of micro-pipettes and or multi-channel
pipettes with 96-well plates is recommended for consistent readings. A spectrophotometer plate
reader is necessary for obtaining the spectrophotometric readings that are then analyzed using
any commercial ELISA evaluation program (for example, four parameters, Logit/Log or
alternatively point to point).
At the time of the verification test, the list price for the plate kit that will analyze 84 samples was
$275. According to the vendor, the kits have a 6-month shelf life as received and should be
stored at 4 - 8 °C. Other consumables not included in the kit are pipettes, pipette tips, and
distilled or DI water.
49
-------
Chapter 9
References
1. Test/Quality Assurance Plan for Verification ofMicrocystin Test KitsTest/Quality
Assurance Plan for Verification ofMicrocystin Test Kits. U.S. Environmental
Technology Verification Program, Battelle, July 2010.
2. Quality Management Plan for the ETVAdvancedMonitoring Systems Center, Version 7.
U.S. Environmental Technology Verification Program, Battelle, November 2008.
3. Hollrah, M., Standard Operating Procedure (SOP) Determination ofalgaltoxin residues
in water extracts by liquid chromatography (LC)- atmospheric pressure electrospray
ionization tandem mass spectrometry (MS/MS). December, 2005, Water Sciences
Laboratory, University of Nebraska.
4. Cong, L.H., B.; Chen, Q.; Lu, B.; Zhang, J.; Ren, Y., Determination of trace amount of
microcystins in water samples using liquid chromatography coupled with triple
quadrupole mass spectrometry. Anal. Chim. Acta, 2006. 569 (1-2): p. 157-168.
5. "Guidelines Establishing Test Procedures for the Analysis of Pollutants. ", USEP A,
Editor. 2000, U.S. Code of Federal Regulations.
6. SOP# SWS-2320.1A: Microcystin Analysis Using the Abraxis ELISA (Enzyme-Linked
Immuno-Sorbent Assay) Method. Nebraska Department of Environmental Quality.
7. Loftin, K.A., et al., Comparison of Two Cell Lysis Procedures for Recovery of
Microcystins in Water Samples from Silver Lake in Dover, Delaware, with Microcystin
Producing Cyanobacterial Accumulations., in USGS Open-File Report 2008 -1341. 2008.
p. 9.
50
-------
APPENDIX A
Reference Laboratory Method Detection Limit Memo
July 14, 2010
To: Anne Gregg and Ryan James, Battelle Laboratories
From: Daniel Snow and David Cassada, UNL Water Sciences Laboratory
Re: Summary of Microcystin SPE method validation - July 13-14, 2010
Microcystins LA, LR and RR were spiked into water and extracted using solid phase extraction (SPE) to
evaluate method accuracy and precision, and method detection limits. The method described in Cong et
al. 2006 was modified to allow for extraction of a larger sample by using higher capacity polymeric
(Waters Oasis, HLB) SPE cartridges. Briefly, 400-milliliter (mL) of purified reagent water was fortified
with 1500 uL of a diluted mixed stock (0.1 ng/uL) obtained from Battelle to produce 0.375 ug/L of each
analyte. Nodularin (1600 uL of a 0.1 ng/uL solution) was also added to produce a concentration of 0.40
ug/L . Eight 50 mL portions of this fortified water were weighed into 125 mL amber glass bottles and
each portion separately spiked with 100 uL of the enkephalin-Leu internal standard (IS) solution (0.1
ng/uL) to give a concentration of 2.0 ug/L. A single method blank was prepared by spiking with IS and
surrogate only.
After capping and shaking each solution to equilibrate, samples were drawn under vacuum through pre-
conditioned 200 mg Oasis HLB SPE cartridges at a rate of approximately 10 mL/min. When the sample
had completely passed through the cartridge, it was allowed to air-dry under vacuum, removed from the
extraction apparatus and prepared for elution. Ten (10) milliliters of high purity methanol (Fisher Optima
Grade) were used to elute analyte, IS and surrogate compounds from the cartridges. The methanol was
evaporated under nitrogen to approximately 0.4 mL and the extracts transferred to low volume inserts for
analysis on the LCQ ion trap tandem LC/MS system. Calibration solutions (5, 10, 30, 60 and 75 ng/mL)
were prepared in water from the same mixed stock as the spiking solutions. A table summarizing the
results of the validation is copied below (Table A-l.).
A second 10-mL aliquot of methanol was passed through 4 of SPE cartridges and collected separately to
check for completeness of analyte elution. These second aliquots were blown down to the same 0.4 mL
volume as the MDLs eluants and analyzed. The resulting absolute areas of the analyte, surrogate, and
internal standard peaks obtained were approximately 1% of the areas obtained in the first portion. This
suggests that lower elution volumes can result in decreased analyte recovery.
References
Cong, L.; Huang, B.; Chen, Q.; Lu, B.; Zhang, J.; Ren, Y. (2006) Determination of trace amount of
microcystins in water samples using liquid chromatography coupled with triple quadrupole mass
spectrometry. Anal. Chim. Acta, 569 (1-2), 157-168.
51
-------
Table A-l. Average, standard deviation, method detection limits (MDL = S x tN-1) and recoveries
of microcystins obtained from extraction and analysis of 8 fortified reagent water (0.375 jig/L)
samples.
50 mL
sample
Aliquot
MDL 1
MDL 2
MDL 3
MDL 4
MDL 5
MDL 6
MDL 7
MDL 8
AVG
STD DEV
MDL
%REC
Expected
value
Amount obtained (ng)
Nodularin
23.994
24.647
22.716
23.157
26.361
19.618
20.254
19.889
22.580
2.460
7.371
112.9
20.0
MC-RR
19.679
19.661
17.660
19.715
19.731
18.214
14.533
15.247
18.055
2.113
6.333
96.3
18.75
MC-LR
18.084
21.985
20.524
21.022
20.462
18.322
20.046
17.518
19.745
1.586
4.753
105.3
18.75
MC-LA
19.913
21.752
18.404
20.304
21.182
18.393
21.490
14.614
19.507
2.360
7.072
104.0
18.75
Concentration (ug/L)
Nodularin
0.480
0.493
0.454
0.463
0.527
0.392
0.405
0.398
0.452
0.049
0.147
112.9
0.4
MC-RR
0.394
0.393
0.353
0.394
0.395
0.364
0.291
0.305
0.361
0.042
0.127
96.3
0.375
MC-LR
0.362
0.440
0.410
0.420
0.409
0.366
0.401
0.350
0.395
0.032
0.095
105.3
0.375
MC-LA
0.398
0.435
0.368
0.406
0.424
0.368
0.430
0.292
0.390
0.047
0.141
104.0
0.375
52
-------
APPENDIX B
Beacon Test Kit Raw Data
Table B-l. Beacon Tube Kit Raw Data
Sample Description
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Positive Control 1
Positive Control 10
Positive control 1 1
Positive control 12
Positive Control 3
Positive Control 4
Positive Control 5
Positive Control 6
Positive Control 7
Positive Control 8
Positive Control 9
Negative DiffLot
StdO.Sppb DiffLot
Std 0.8 ppb DiffLot
Std 2.0 ppb DiffLot
Std 5.0 ppb DiffLot
0.1 LR
0.1 LR
0.1 LR
0.5 LA
0.5 LA
0.5 LA
0.5 LR
0.5 LR
0.5 LR
0.5 RR
0.5 RR
0.5 RR
I.OLA
I.OLA
I.OLA
Variant
RB
RB
RB
RB
RB
RB
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
OD Value
1.245
1.253
1.131
1.172
1.162
1.163
0.503
0.589
0.526
0.54
0.568
0.546
0.524
0.474
0.483
0.509
0.319
1.053
0.735
0.658
0.37
0.261
1.122
1.125
1.066
0.964
0.916
0.952
1.069
1.018
1.037
0.76
0.753
0.72
0.915
0.845
0.858
Cone, (ppb)
Range?
Range?
0.061
Range?
Range?
Range?
0.908
1.059
1.102
0.984
0.941
0.982
0.983
1.046
0.909
0.821
1.118
0.076
0.438
0.608
2.451
5.82
0.051
0.05
0.081
0.23
0.278
0.241
0.066
0.1
0.087
0.376
0.389
0.452
0.279
0.361
0.345
53
-------
Table B-l. Beacon Tube Kit Raw Data Continued
Sample Description
1.0 LR
1.0 LR
1.0 LR
1.0 RR
1.0 RR
1.0 RR
1.5 LA
1.5 LA
1.5 LA
1.5 LA
1.5 LA
1.5 LA
1.5 LA
1.5LR
1.5LR
1.5LR
1.5LR
1.5LR
1.5LR
1.5LR
1.5RR
1.5RR
1.5RR
1.5RR
1.5RR
1.5RR
1.5RR
2.0 LA
2.0 LA
2.0 LA
2.0 LR
2.0 LR
2.0 LR
2.0 RR
2.0 RR
2.0 RR
4.0 LA
4.0 LA
4.0 LA
4.0 LR
4.0 LR
4.0 LR
Variant
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
OD Value
0.48
0.512
0.49
0.511
0.57
0.515
0.832
0.674
0.834
0.879
0.637
0.852
0.824
0.455
0.446
0.429
0.458
0.451
0.435
0.456
0.435
0.443
0.419
0.407
0.462
0.386
0.414
0.814
0.816
0.837
0.341
0.34
0.334
0.403
0.405
0.348
0.772
0.778
0.759
0.267
0.264
0.323
Cone, (ppb)
1.006
0.873
0.962
1.154
0.882
1.133
0.266
0.556
0.263
0.207
0.655
0.239
0.276
1.522
1.594
1.743
1.499
1.553
1.688
1.514
1.256
1.208
1.365
1.457
1.104
1.891
1.615
0.402
0.4
0.371
2.178
2.194
2.292
1.715
1.696
2.432
0.464
0.455
0.485
4.308
4.474
3.672
54
-------
Table B-l. Beacon Tube Kit Raw Data Continued
Sample Description
4.0 RR
4.0 RR
4.0 RR
7.0 LA
7.0 LA
7.0 LA
7.0 RR
7.0 RR
7.0 RR
4ppb Chloro lOx LA
4ppb Chloro lOx LA
4ppb Chloro lOx LA
4ppb Chloro lOx RR
4ppb Chloro lOx RR
4ppb Chloro lOx RR
4ppb Chloro LA
4ppb Chloro LA
4ppb Chloro LA
4ppb Chloro LR
4ppb Chloro LR
4ppb Chloro LR
4ppb Chloro LR lOx
4ppb Chloro LR lOx
4ppb Chloro LR lOx
4ppb Chloro RR
4ppb Chloro RR
4ppb Chloro RR
4ppb Matrix lOx LA
4ppb Matrix lOx LA
4ppb Matrix lOx LA
4ppb Matrix lOx LR
4ppb Matrix lOx LR
4ppb Matrix lOx LR
4ppb Matrix lOx RR
4ppb Matrix lOx RR
4ppb Matrix lOx RR
4ppb Matrix LA
4ppb Matrix LA
4ppb Matrix LA
4ppb Matrix LR
4ppb Matrix LR
4ppb Matrix LR
Variant
RR
RR
RR
RR
LA
LA
RR
RR
RR
LA
LA
LA
RR
RR
RR
LA
LA
LA
LR
LR
LR
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LR
LR
LR
OD Value
0.333
0.343
0.337
0.688
0.733
0.713
0.296
0.316
0.312
0.696
0.677
0.658
0.279
0.283
0.292
0.673
0.673
0.673
0.301
0.294
0.324
0.289
0.311
0.324
0.297
0.28
0.285
0.57
0.454
0.622
0.326
0.268
0.283
0.251
0.319
0.341
0.526
0.432
0.454
0.272
0.282
0.287
Cone, (ppb)
2.733
2.525
2.646
0.522
0.53
0.566
3.913
3.972
4.167
0.416
0.444
0.474
3.346
3.194
2.897
0.498
0.498
0.498
3.361
3.673
3.632
3.934
4.219
3.632
2.754
3.307
3.122
0.661
1.019
0.555
3.556
3.854
3.194
4.34
3.838
3.071
0.771
1.119
1.019
3.652
3.23
3.054
55
-------
Table B-l. Beacon Tube Kit Raw Data Continued
Sample Description
4ppb Matrix RR
4ppb Matrix RR
4ppb Matrix RR
RWl(lOxdil)
RWl(lOxdil)
RWl(lOxdil)
RW2(10xdil)
RW2(10xdil)
RW2(10xdil)
RW3(10xdil)
RW3(10xdil)
RW3(10xdil)
RW4
RW4
RW4
RW 5 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW6
RW6
RW6
RW6 (2xdil)
RW6 (2xdil)
RW6 (2xdil)
RW7
RW7
RW7
RW8
RW8
RW8
RW9 Matrix
RW9 Matrix
RW9 Matrix
Variant
RR
RR
RR
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
OD Value
0.29
0.264
0.354
0.44
0.471
0.424
0.73
0.723
0.717
0.528
0.637
0.631
0.457
0.436
0.439
1.049
1.041
1.054
0.45
0.45
0.418
0.584
0.573
0.598
1.176
1.198
1.185
0.691
0.696
0.745
0.910
0.596
0.928
Cone, (ppb)
2.687
3.589
2.742
2.372
1.981
2.613
0.49
0.503
0.515
1.45
0.7
0.717
0.62
0.668
0.66
0.104
0.111
0.1
1.576
1.576
1.873
0.864
0.903
0.816
0.015
0.001
0.009
0.569
0.558
0.462
0.213
1.022
0.193
56
-------
Table B-2. Beacon Plate Kit Raw Data
Sample Description
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Reagent Blank
Positive Control 1
Positive Control 2a
Positive Control 2b
Positive Control 4
Positive control 5a
Positive control 5b
Positive Control 6
Positive Control 6
Positive Control 6
StdODiffLot
StdO.lDiffLot
StdO.SDiffLot
StdO.SDiffLot
Std2.0DiffLot
0.1 LR
0.1 LR
0.1 LR
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
Variant
RB
RB
RB
RB
RB
RB
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
Mean Cone.
(ppb)
0.033
0.032
0.054
0.025
0.035
0.054
0.98
1.123
1.267
1.069
1.268
1.336
0.965
1.064
1.281
0.02
0.164
0.397
0.966
3.16
0.134
0.151
0.181
0.175
0.186
0.184
0.192
0.172
0.18
0.188
0.185
0.205
0.24
0.239
0.597
0.503
0.488
0.602
0.62
0.721
0.668
0.528
0.494
0.501
0.42
0.412
0.566
0.709
0.657
Standard
Deviation (ppb)
0.008
0.007
0.007
0.007
0.018
0.007
0.004
0.095
0.006
0.003
0.128
0.297
0.134
0.113
0.028
0.017
0.005
0.008
0.105
0.612
0.001
0.018
0.027
0.023
0.019
0.007
0.007
0.002
0.014
0.021
0.003
0.012
0.034
0.004
0.046
0.029
0.015
0.05
0.006
0.008
0.102
0.037
0.063
0.012
0.02
0.01
0.015
0.066
0.094
cv%
23.3
22.4
13.2
29.4
52
13.2
0.5
8.5
0.5
0.3
10.1
22.2
13.9
10.6
2.2
85.1
2.9
2.1
10.9
19.4
0.8
11.9
14.9
13.3
10.1
4
3.7
1
7.8
11
1.4
5.7
14.1
1.5
7.7
5.7
3.1
8.4
1
1.1
15.3
7
12.8
2.4
4.9
2.5
2.7
9.3
14.3
57
-------
Table B-2. Beacon Plate Kit Raw Data Continued
Sample Description
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
.OLA
.OLA
.OLA
.OLA
.OLA
.OLA
.OLR
.OLR
.OLR
.ORR
.ORR
.ORR
2.0 LA
2.0 LA
2.0 LA
2.0 LA
2.0 LA
2.0 LA
2.0 LR
2.0 LR
2.0 LR
2.0 RR
2.0 RR
2.0 RR
4.0 LA
4.0 LA
4.0 LA
4.0 LA
4.0 LA
4.0 LA
4.0 LR
4.0 LR
4.0 LR
4.0 RR
4.0 RR
4.0 RR
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
0.1 LR
Variant
RR
RR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LR
Mean Cone.
(ppb)
0.745
0.639
0.622
0.636
0.565
0.213
0.211
0.291
0.295
0.306
0.269
.054
.217
.046
.108
.055
.151
0.279
0.223
0.328
0.321
0.339
0.364
.312
.153
.163
.014
.169
0.926
0.38
0.304
0.339
0.457
0.442
0.438
125.744
Range?
Range?
9.029
170.126
Range?
0.393
0.424
0.352
0.553
0.52
0.445
0.134
Standard
Deviation (ppb)
0.046
0.018
0.1
0.072
0.031
0.011
0.034
0.047
0.035
0.066
0.022
0.04
0.237
0.092
0.016
0.011
0.246
0.012
0.047
0.001
0.037
0.001
0.082
0.01
0.206
0.08
0.009
0.068
0.067
0.06
0.002
0.018
0.005
0.005
0.014
0
Range?
Range?
0
0
Range?
0.074
0.019
0.083
0.007
0.032
0.007
0.001
cv%
6.1
2.8
16
11.3
5.4
5.3
16.3
16.3
11.7
21.5
8.1
3.8
19.5
8.8
1.4
1.1
21.4
4.5
21.3
0.3
11.7
0.4
22.6
0.7
17.9
6.9
0.9
5.8
7.2
15.8
0.7
5.4
1
1.1
3.2
0
Range?
Range?
0
0
Range?
18.9
4.5
23.5
1.3
6.2
1.5
0.8
58
-------
Table B-2. Beacon Plate Kit Raw Data
Sample Description
0.1 LR
0.1 LR
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LA
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 LR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
0.5 RR
.OLA
.OLA
.OLA
.OLA
.OLA
.OLA
.OLR
.OLR
.OLR
.ORR
.ORR
.ORR
2.0 LA
Variant
LR
LR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
Mean Cone.
(ppb)
0.151
0.181
0.175
0.186
0.184
0.192
0.172
0.18
0.188
0.185
0.205
0.24
0.239
0.597
0.503
0.488
0.602
0.62
0.721
0.668
0.528
0.494
0.501
0.42
0.412
0.566
0.709
0.657
0.745
0.639
0.622
0.636
0.565
0.213
0.211
0.291
0.295
0.306
0.269
.054
.217
.046
.108
.055
.151
0.279
Standard
Deviation (ppb)
0.018
0.027
0.023
0.019
0.007
0.007
0.002
0.014
0.021
0.003
0.012
0.034
0.004
0.046
0.029
0.015
0.05
0.006
0.008
0.102
0.037
0.063
0.012
0.02
0.01
0.015
0.066
0.094
0.046
0.018
0.1
0.072
0.031
0.011
0.034
0.047
0.035
0.066
0.022
0.04
0.237
0.092
0.016
0.011
0.246
0.012
cv%
11.9
14.9
13.3
10.1
4
3.7
1
7.8
11
1.4
5.7
14.1
1.5
7.7
5.7
3.1
8.4
1
1.1
15.3
7
12.8
2.4
4.9
2.5
2.7
9.3
14.3
6.1
2.8
16
11.3
5.4
5.3
16.3
16.3
11.7
21.5
8.1
3.8
19.5
8.8
1.4
1.1
21.4
4.5
59
-------
Table B-2. Beacon Plate Kit Raw Data Continued
Sample Description
2.0 LA
2.0 LA
2.0 LA
2.0 LA
2.0 LA
2.0 LR
2.0 LR
2.0 LR
2.0 RR
2.0 RR
2.0 RR
4.0 LA
4.0 LA
4.0 LA
4.0 LA
4.0 LA
4.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
7.0 LA
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Chloro 10X
2.0 LA Matrix
2.0 LA Matrix
2.0 LA Matrix
2.0 LA Matrix lOx
2.0 LA Matrix lOx
2.0 LA Matrix lOx
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro
2.0 LR Chloro lOx
2.0 LR Chloro lOx
Variant
LA
LA
LA
LA
LA
LR
LR
LR
RR
RR
RR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LR
LR
LR
LR
LR
Mean Cone.
(ppb)
0.223
0.328
0.321
0.339
0.364
1.312
1.153
1.163
1.014
1.169
0.926
0.38
0.304
0.339
0.457
0.442
0.438
0.393
0.424
0.352
0.553
0.52
0.445
0.393
0.424
0.352
0.553
0.52
0.445
0.153
0.147
0.127
0.143
0.135
0.134
0.138
0.124
0.357
0.368
0.335
0.827
0.729
0.742
0.53
0.531
0.52
0.523
0.528
Standard
Deviation (ppb)
0.047
0.001
0.037
0.001
0.082
0.01
0.206
0.08
0.009
0.068
0.067
0.06
0.002
0.018
0.005
0.005
0.014
0.074
0.019
0.083
0.007
0.032
0.007
0.074
0.019
0.083
0.007
0.032
0.007
0.001
0.029
0
0.012
0.007
0.001
0.012
0.004
0.013
0.008
0.022
0.069
0.079
0.165
0.051
0.015
0.02
0.007
0.063
cv%
21.3
0.3
11.7
0.4
22.6
0.7
17.9
6.9
0.9
5.8
7.2
15.8
0.7
5.4
1
1.1
3.2
18.9
4.5
23.5
1.3
6.2
1.5
18.9
4.5
23.5
1.3
6.2
1.5
0.5
19.7
0.3
8.4
5.1
0.6
8.8
3.6
3.7
2.2
6.6
8.3
10.9
22.3
9.5
2.8
3.9
1.3
12
60
-------
Table B-2. Beacon Plate Kit Raw Data Continued
Sample Description
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Chloro lOx
2.0 LR Matrix
2.0 LR Matrix
2.0 LR Matrix
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 LR Matrix lOx
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Chloro lOx
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix
2.0 RR Matrix lOx
2.0 RR Matrix lOx
2.0 RR Matrix lOx
RWl(20xdil)
RWl(20xdil)
RWl(20xdil)
RWl(20xdil)
RW2 (20x dil)
RW2 (20x dil)
RW2 (20x dil)
RW3 (20x dil)
RW3 (20x dil)
RW3 (20x dil)
RW4
RW4
RW4
RW 4 (4x dil)
RW 4 (4x dil)
RW 4 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW 5 (4x dil)
RW6 (2xdil)
RW6 (2xdil)
RW6 (2xdil)
RW7
RW7
RW7
RW8
Variant
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
RR
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone.
(ppb)
0.373
0.424
0.428
0.449
2.196
2.355
2.113
2.093
2.156
2.481
1.957
2.072
1.943
1.913
1.932
1.943
2.267
2.428
2.122
2.018
2.22
2.176
1.59
1.489
1.594
1.639
0.669
0.67
0.647
0.674
0.681
0.653
0.376
0.35
0.327
0.231
0.269
0.329
.042
.106
.043
.144
.179
.304
0.053
0.059
0.06
0.686
Standard
Deviation (ppb)
0.022
0.001
0.01
0.01
0.146
0.177
0.037
0.034
0.066
0.198
0.062
0.246
0.014
0.26
0.08
0.244
0.179
0.14
0.126
0.043
0.055
0.413
0.243
0.139
0.338
0.192
0.066
0.004
0.021
0.012
0.028
0.043
0.019
0.011
0.017
0.026
0.002
0.058
0.031
0.046
0.095
0.078
0.011
0.145
0.001
0.005
0.008
0
cv%
5.8
0.2
2.3
2.2
6.7
7.5
1.7
1.6
3
8
3.2
11.9
0.7
13.6
4.1
12.6
7.9
5.8
5.9
2.1
2.5
19
15.3
9.3
21.2
11.7
9.8
0.6
3.3
1.8
4.2
6.6
5
3.1
5.2
11.4
0.6
17.5
3
4.1
9.1
6.8
1
11.1
2.4
8.1
14
0
61
-------
Table B-2. Beacon Plate Kit Raw Data Continued
Sample Description
RW8
RW8
RW 9 Matrix
RW 9 Matrix
RW 9 Matrix
Variant
Unknown
Unknown
Unknown
Unknown
Unknown
Mean Cone.
(ppb)
0.646
0.666
0.323
0.358
0.31
Standard
Deviation (ppb)
0.069
0.133
0
0.002
0.025
cv%
10.7
20
0.1
0.4
8.2
62
------- |