November 2003
Environmental Technology
Verification Report
CHECKLIGHT, LTD.
TOXSCREEN-II
RAPID TOXICITY TESTING SYSTEM
Prepared by
Battelle
Batteiie
. . . Putting Technology To Work
Under a cooperative agreement with
tMr\ U.S. Environmental Protection Agency
ETV ETV ET
-------
November 2003
Environmental Technology Verification
Report
ETV Advanced Monitoring Systems Center
CheckLight, Ltd.
ToxScreen-ll
Rapid Toxicity Testing System
by
Ryan James
Amy Dindal
Zachary Willenberg
Karen Riggs
Battelle
Columbus, Ohio 43201
-------
Notice
The U.S. Environmental Protection Agency (EPA), through its Office of Research and
Development, has financially supported and collaborated in the extramural program described
here. This document has been peer reviewed by the Agency. Mention of trade names or
commercial products does not constitute endorsement or recommendation by the EPA for use.
11
-------
Foreword
The U.S. Environmental Protection Agency (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 seven 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. ep a. go v/et v/centers/center 1. html.
111
-------
Acknowledgments
The authors wish to acknowledge the support of all those who helped plan and conduct the
verification test, analyze the data, and prepare this report. Many thanks go to Battelle's Medical
Research and Evaluation Facility for providing the facilities for and personnel capable of
working with chemical warfare agents and biotoxins. We would also like to thank
Karen Bradham, U.S. EPA National Exposure Research Laboratory; Steve Allgeier, U.S. EPA
Office of Water; Ricardo DeLeon, Metropolitan Water District of Southern California; Yves
Mikol, New York City Department of Environmental Protection; and Stanley States, Pittsburgh
Water and Sewer Authority, for their careful review of the test/QA plan and this verification
report.
IV
-------
Contents
Page
Notice ii
Foreword iii
Acknowledgments iv
List of Abbreviations viii
1 Background 1
2 Technology Description 2
3 Test Design and Procedures 4
3.1 Introduction 4
3.2 Test Design 5
3.3 Test Samples 6
3.3.1 Quality Control Samples 6
3.3.2 Drinking Water Fortified with Contaminants 8
3.3.3 Drinking Water Fortified with Potential Interferences 8
3.4 Test Procedure 8
3.4.1 Test Sample Preparation and Storage 8
3.4.2 Test Sample Analysis Procedure 9
3.4.3 Stock Solution Confirmation Analysis 9
4 Quality Assurance/Quality Control 12
4.1 Quality Control of Stock Solution Confirmation Methods 12
4.2 Quality Control of Drinking Water Samples 12
4.3 Audits 13
4.3.1 Performance Evaluation Audit 13
4.3.2 Technical Systems Audit 13
4.3.3 Audit of Data Quality 14
4.4 QA/QC Reporting 14
4.5 Data Review 15
5 Statistical Methods and Reported Parameters 16
5.1 Endpoints and Precision 16
5.2 Toxicity Threshold 17
5.3 False Positive/Negative Responses 17
5.4 Field Portability 17
5.5 Other Performance Factors 17
-------
6 Test Results 19
6.1 Endpoints and Precision 19
6.1.1 Contaminants 19
6.1.2 Potential Interferences 26
6.1.3 Precision 28
6.2 Toxicity Threshold 28
6.3 False Positive/Negative Responses 28
6.4 Field Portability 30
6.5 Other Performance Factors 31
7 Performance Summary 32
8 References 35
Figures
Figure 2-1. ToxScreen-H Rapid Toxicity Testing System 2
Tables
Table 3-1. Contaminants and Potential Interferences 5
Table 3-2. Summary of Quality Control and Contaminant Test Samples 7
Table 3-3. Dose Confirmation Results 10
Table 3-4. Water Quality Parameters 11
Table 4-1. Summary of Performance Evaluation Audit 14
Table 4-2. Summary of Data Recording Process 15
Table 6-la. Aldicarb Percent Inhibition Results 20
Table 6-lb. Colchicine Percent Inhibition Results 21
Table 6-lc. Cyanide Percent Inhibition Results 22
Table 6-Id. Dicrotophos Percent Inhibition Results 23
VI
-------
Table 6-le. Thallium Sulfate Percent Inhibition Results 24
Table 6-If. Botulinum Toxin Percent Inhibition Results 24
Table 6-lg. Ricin Percent Inhibition Results 25
Table 6-lh. Soman Percent Inhibition Results 25
Table 6-li. VX Percent Inhibition Results 26
Table 6-2. Potential Interferences Results 27
Table 6-3. Toxicity Thresholds 29
Table 6-4. False Negative Responses 30
Table 7-1. Pro-Organic Buffer Results 32
Table 7-2. Pro-Metal Buffer Results 33
vn
-------
List of Abbreviations
AMS
ASTM
ATEL
DI
DDW
EPA
ETV
HOPE
ID
LD
\iL
mg
mL
NSDWR
%D
PE
QA
QC
QMP
SOP
TSA
Advanced Monitoring Systems
American Society for Testing and Materials
Aqua Tech Environmental Laboratories
deionized water
dechlorinated drinking water from Columbus, Ohio
U.S. Environmental Protection Agency
Environmental Technology Verification
high-density polyethylene
identification
lethal dose
microliter
milligram
milliliter
National Secondary Drinking Water Regulations
percent difference
performance evaluation
quality assurance
quality control
quality management plan
standard operating procedure
technical systems audit
Vlll
-------
Chapter 1
Background
The U.S. Environmental Protection Agency (EPA) supports the Environmental Technology
Verification (ETV) Program to facilitate the deployment of innovative environmental tech-
nologies 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 tech-
nologies 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 EPA's National Exposure Research Laboratory and its verification organization partner,
Battelle, operate the Advanced Monitoring Systems (AMS) Center under ETV. The AMS Center
recently evaluated the performance of the CheckLight, Ltd. ToxScreen-H rapid toxicity testing
system. Rapid toxicity testing systems were identified as a priority technology verification
category through the AMS Center stakeholder process.
-------
Chapter 2
Technology Description
The objective of the ETV AMS Center is to verify the performance characteristics of environ-
mental monitoring technologies for air, water, and soil. This verification report provides results
for the verification testing of ToxScreen-n. Following is a description of ToxScreen-n, based on
information provided by the vendor. The information provided below was not subjected to
verification in this test.
Luminous bacteria are self-maintained luminescent units that, under proper conditions, emit
high and steady levels of luminescence. Chemo-physical and biological toxicants that affect cell
respiration, electron transport systems, adenosine triphosphate generation, and the rate of protein
or lipid synthesis alter the level of luminescence. Similarly, agents that affect the cell's integrity,
and especially membrane function, have a strong effect on in vivo luminescence.
ToxScreen-n (Figure 2-1) uses a luminous bacteria, Photobacterium leiognathi and special
assay conditions to detect toxicants in water samples. When used in concurrent tests, two assay
buffers (one for detecting organic pollutants [pro-organic buffer] and the other for detecting
heavy metals [pro-metal buffer]) are used to discriminate between the presence of organic and
metal toxicants at sub milligram-per-liter concentrations. Inhibition of greater than 50% using
either buffer is considered a positive result. A positive result in the pro-organic buffer suggests
that the contaminant causing the toxicity is organic, while a positive result in the pro-metal
buffer suggests that the contaminant may be a metal.
First, a freeze-dried bioassay reagent is hydrated with the provided hydration buffer and, after
five minutes, transferred into the provided storage buffer. The suspended reagent is maintained
at 4°C until use (aliquots can be drawn for up to seven days). Next, pro-organic and pro-metal
concentrated assay buffers are added separately to
individual aliquots of source water samples, as well as
to two aliquots of the reference sample. Then, aliquots
of suspended reagent are rapidly dispensed into test
cuvettes. Finally, after 90 minutes of incubation at
ambient temperature, luminescence is measured using a
portable luminometer. Changes in luminescence
(compared with the reference sample) reflect water
toxicity.
Figure 2-1. ToxScreen-II Rapid
Toxicity Testing System
-------
The ToxScreen-H test kit contains stoppered vials holding freeze-dried luminous bacteria, empty
test tubes, hydration buffer, storage buffer, pro-organic concentrated assay buffer, pro-metal
concentrated assay buffer, and positive control solutions of copper chloride and sodium chloro-
acetate. A repeat dispenser (1 to 100 microliters [[J.L]) and 10- to 1,000-[J,L pipettes and tips are
required, but not provided in the kit. The price of the ToxScreen-H test kit (1,000 single tests) is
$300. The luminometer for reading the results of the bioassays costs $2,895.
The luminometer can be integrated with a personal computer for data acquisition, evaluation,
and storage. The luminometer is 5.9 x 11.0 x 6.7 inches and weighs approximately one pound.
An insulating styrofoam case is available for an additional cost to use while conducting field
tests.
-------
Chapter 3
Test Design and Procedures
3.1 Introduction
The objective of this verification test of rapid toxicity technologies was to evaluate their ability
to detect certain toxins and to determine their susceptibility to interfering chemicals in a
controlled experimental matrix. Rapid toxicity technologies do not identify or determine the
concentration of specific contaminants, but serve as a screening tool to quickly determine
whether water is potentially toxic. Rapid toxicity technologies use bacteria (e.g., Vibrio
fischeri), enzymes (e.g., luciferase), or small crustaceans (e.g., Daphnia magna) that either
directly, or in combination with reagents, produce a background level of light or use dissolved
oxygen at a steady rate in the absence of toxic contaminants. Toxic contaminants in water are
indicated by a change in the color or intensity of light produced or by a decrease in the dissolved
oxygen uptake rate in the presence of the contaminants.
As part of this verification test, ToxScreen-H was subjected to various concentrations of
contaminants such as industrial chemicals, pesticides, rodenticides, Pharmaceuticals, nerve
agents, and biological toxins. Each contaminant was added to separate drinking water samples
and analyzed. In addition to determining whether ToxScreen-H can detect the toxicity caused by
each contaminant, its response to interfering compounds in clean drinking water, such as water
treatment chemicals and by-products, was evaluated. Table 3-1 shows the contaminants and
potential interferences that were evaluated during this verification test.
This verification test was conducted according to procedures specified in the Test/QA Plan for
Verification of Rapid Toxicity Technologies.,(1) ToxScreen-n was verified by analyzing a
dechlorinated drinking water (DDW) sample from Columbus, Ohio, fortified with various
concentrations of the contaminants and interferences shown in Table 3-1. Hereafter in this
report, DDW will refer to dechlorinated drinking water from Columbus, Ohio. Where possible,
the concentration of each contaminant or potential interference was confirmed independently by
Aqua Tech Environmental Laboratories (ATEL), Marion, Ohio, or by Battelle, depending on the
analyte.
-------
Table 3-1. Contaminants and Potential Interferences
Category Contaminant
Carbamate pesticide aldicarb
Pharmaceutical colchicine
Industrial chemical cyanide
Organophosphate pesticide dicrotophos
Rodenticide thallium sulfate
Biological toxins botulinum toxin, ricin
Nerve agents soman, VX
Potential interferences aluminum, copper, iron, manganese, zinc,
chloramination by-products, and chlorination
by-products
ToxScreen-H was evaluated by
• Endpoint and precision—percent inhibition for all concentration levels of contaminants and
potential interfering compounds and precision of replicate analyses
• Toxicity threshold for each contaminant
• False negative responses—contaminants that were reported as producing inhibition results
similar to the negative control when the contaminant was present at lethal concentrations
• False positive responses—occurrence of inhibition of dechlorinated drinking water samples
significantly greater than the inhibition reported for unspiked American Society for Testing
and Materials (ASTM) Type n deionized (DI) water samples (zero inhibition)
• Field portability
• Ease of use
• Throughput.
3.2 Test Design
ToxScreen-H was used to analyze the DDW sample fortified with contaminants at
concentrations ranging from lethal levels to concentrations several orders of magnitude less than
the lethal dose. The lethal dose of each contaminant was determined by calculating the
concentration at which 250 mL of water would probably cause the death of a 154-pound person.
These calculations were based on toxicological data available for each contaminant. For soman,
the stock solution confirmation showed degradation in the water; therefore, the concentrations
-------
analyzed were less than anticipated. Whether the concentration is still a lethal dose, as is the
case for all contaminants, depends on the characteristics of the individual person and the amount
of contaminant ingested. Inhibition results (endpoints) from four replicates of each contaminant
at each concentration level were evaluated to assess the ability of ToxScreen-H to detect toxicity
at various concentrations of contaminants, as well as to measure the precision of ToxScreen-H
results.
The response of ToxScreen-H to compounds used during the water treatment process (identified
as potential interferences in Table 3-1) was evaluated by analyzing separate aliquots of DDW
fortified with each potential interference at approximately one-half of the concentration limit
recommended by the EPA's National Secondary Drinking Water Regulations (NSDWR)(2)
guidance. For analysis of by-products of the chlorination process, the unspiked DDW was
analyzed because Columbus, Ohio, uses chlorination as its disinfectant procedure. For the
analysis of by-products of the chloramination process, a separate drinking water sample from
St. Petersburg, Florida, which uses chloramination as its disinfection process, was obtained. The
samples were analyzed after residual chlorine was removed using sodium thiosulfate.
Sample throughput was measured based on the number of samples analyzed per hour. Ease of
use and reliability were determined based on documented observations of the operators and the
verification test coordinator. In addition to comprehensive testing in Battelle laboratories,
ToxScreen-H was operated in the basement of a Columbus, Ohio, home to test its ability to be
transported and operated in a non-laboratory setting.
3.3 Test Samples
Test samples used in the verification test included drinking water and quality control (QC)
samples. Table 3-2 shows the number and type of samples analyzed. QC samples included
method blanks and positive and negative control samples. The fortified drinking water samples
were prepared from a single drinking water sample collected from the Columbus, Ohio, system.
The water was dechlorinated using sodium thiosulfate and then fortified with various concen-
trations of contaminants and interferences. Using this DDW (Columbus, Ohio, dechlorinated
drinking water), individual solutions containing each contaminant and potential interference
were prepared and analyzed. The DDW containing the potential interferences was analyzed at a
single concentration level, while at least four dilutions (made using the DDW) were analyzed for
each contaminant using ToxScreen-n. Mixtures of contaminants and interfering compounds
were not analyzed. One concentration level of cyanide was analyzed in the field setting.
3.3.1 Quality Control Samples
QC samples included method blank samples, which consisted of ASTM Type n DI water;
positive control samples, which consisted of ASTM Type n DI water or DDW (depending on
vendor preference) fortified with a contaminant and concentration selected by the vendor; and
negative control samples, which consisted of the unspiked DDW. The method blank samples
were used to help ensure that no sources of contamination were introduced in the sample
handling and analysis procedures.
-------
Table 3-2. Summary of Quality Control and Contaminant Test Samples
Type of Sample
Sample
Characteristics
Concentration Levels
(mg/L)
No. of Sample Analyses
Quality control
Method blank
Positive control
Negative control
(unspiked DDW)
NS(a)
5 (sodium chloroacetate)
1 (copper chloride)
NS
9
10
24
24
44
Aldicarb
Colchicine
Cyanide
Dicrotophos
TATAV^ -c j Thallium sulfate
DDW fortified
with contaminants Botulinum toxin(b)
280; 28; 2.8; 0.28
240; 24; 2.4; 0.24; 0.024
250; 25; 2.5; 0.25; 0.025;
0.0025
1,400; 140; 14; 1.4; 0.14
2,400; 240; 24; 2.4
0.30; 0.030; 0.0030;
0.00030
4 per concentration level
4 per concentration level
4 per concentration level
4 per concentration level
4 per concentration level
4 per concentration level
Field location
DDW fortified
with potential
interferences
Disinfectant
by-products
Ricin(c)
Soman
VX
Cyanide
Aluminum
Copper
Iron
Manganese
Zinc
Chloramination by-
products
Chlorination by-
products
15; 1.5; 0.15; 0.015
0.068(d); 0.0068; 0.00068;
0.000068
0.22; 0.022; 0.0022;
0.00022
0.25
0.36
0.65
0.069
0.26
3.5
NS
NS
4 per concentration level
4 per concentration level
4 per concentration level
4
4
4
4
4
4
4
4
(a) NS = Samples not fortified with any contaminant or potential interference.
^ Lethal dose solution also contained 3 mg/L phosphate and 1 mg/L sodium chloride.
(c) Lethal dose solution also contained 3 mg/L phosphate, 26 mg/L sodium chloride, and 2 mg/L sodium azide.
-------
sample was used to set a background inhibition of the DDW, the matrix in which each test
sample was prepared.
3.3.2 Drinking Water Fortified with Contaminants
Approximately 150 liters of Columbus, Ohio, tap water were collected in a high-density
polyethylene (HDPE) container. The sample was dechlorinated with 0.5 milliliter (mL) of 0.4 M
sodium thiosulfate for every liter of water. All subsequent test samples were prepared from this
DDW and stored in glass containers to avoid chlorine leaching from HDPE containers.
A stock solution of each contaminant was prepared in ASTM Type n DI water at concentrations
above the lethal dose level. The stock solution was diluted in DDW to obtain one sample con-
taining the lethal dose concentration for each contaminant and three additional samples with
concentrations 10, 100, and 1,000 times less than the lethal dose. Exceptions to this were
colchicine, cyanide, and dicrotophos, contaminants that had to be diluted further because of
ToxScreen-IPs sensitivity to them at higher concentrations. Table 3-2 lists each concentration
level and the number of samples analyzed at each level.
3.3.3 Drinking Water Fortified with Potential Interferences
Individual aliquots of the DDW were fortified with one-half the concentration specified by the
EPA's NSDWR for each potential interference. Table 3-2 lists the interferences, along with the
concentrations at which they were tested. Four replicates of each of these samples were
analyzed. To test the sensitivity of ToxScreen-H to by-products of the chlorination process as
potential interferences, the unspiked DDW (same as the negative control) was used since the
water sample originated from a utility that uses chlorination as its disinfectant procedure. In a
similar test involving the by-products of the chloramination process, an additional water sample
was obtained from St. Petersburg, Florida, a city that uses chloramination as its disinfectant
procedure. The residual chlorine in both of these samples was removed using sodium thiosulfate,
and then the samples were analyzed in replicate with no additional fortification of contaminants.
3.4 Test Procedure
3.4.1 Test Sample Preparation and Storage
A drinking water sample was collected as described in Section 3.3.2 and, because free chlorine
kills the bacteria within the ToxScreen-H reagent and can degrade the contaminants during
storage, was immediately dechlorinated with sodium thiosulfate. Prior to preparing each stock
solution, dechlorination of the water sample was qualitatively confirmed by adding an n,n-
diethyl-p-phenylenediamine tablet to a 25-mL aliquot of the DDW. Once dechlorination was
confirmed, all the contaminant samples, potential interference samples, and negative control QC
samples were made from this DDW, while the method blank sample was prepared from ASTM
Type n DI water. The positive control samples were made from stock solutions provided by the
vendor using DDW as the dilution matrix. All QC samples were prepared prior to the start of the
testing and stored at room temperature for a maximum of 60 days. The aliquots of DDW con-
8
-------
taining the contaminants were prepared within seven days of testing and stored in the dark at
room temperature without chemical preservation. Aliquots to be analyzed by each technology
were placed in uniquely labeled sample containers. The sample containers were assigned an
identification (ID) number. A master log of the samples and sample ID numbers for each tech-
nology was kept by Battelle.
3.4.2 Test Sample Analysis Procedure
The ToxScreen-H protocol calls for the use of pro-organic and pro-metal concentrated assay
buffers in the analysis of each sample. The buffers were developed to enhance the sensitivity of
the test to a wide range of toxic agents. Inhibition of greater than 50% using either buffer should
be considered a positive result. A positive result in the pro-organic buffer would suggest that the
contaminant causing the toxicity is organic, while a positive result in the pro-metal buffer would
suggest that the contaminant may be a metal. Therefore, each water sample analyzed requires
two separate analyses, not including the reference samples. First, a freeze-dried bioassay reagent
was hydrated with the provided hydration buffer. Next, pro-organic and pro-metal concentrated
assay buffers were added separately to individual aliquots of the sample as well as to two
aliquots of the reference sample. Then, aliquots of suspended reagent were rapidly dispensed
into test cuvettes. Finally, after 90 minutes' incubation at ambient temperature, luminescence
was measured using the portable luminometer. The luminescence in the test samples was
compared with that of the reference sample to determine their relative toxicity.
For each contaminant, ToxScreen-H analyzed the lethal dose concentration and at least three
additional concentration levels four times. Only one concentration of potential interference was
analyzed. ToxScreen-H reported the absolute light units for each sample. To test the field
portability of ToxScreen-H, a single concentration level of cyanide, prepared in the same way as
the other DDW samples, was analyzed in replicate by ToxScreen-H in the basement of a
Columbus, Ohio, home. Sample analysis procedures were performed in the same way as during
testing in the laboratory. Two operators performed all the analyses using ToxScreen-H. Both
held bachelor's degrees in the sciences and spent approximately four hours with the vendor to
become accustomed to performing tests using ToxScreen-H.
3.4.3 Stock Solution Confirmation Analysis
The concentrations of the contaminant and interfering compound stock solutions were verified
with standard analytical methods, with the exception of colchicine, ricin, and botulinum toxin—
contaminants without standard analytical methods. Aliquots to be analyzed by standard methods
were preserved as prescribed by the method. In addition, the same standard methods were used
to measure the concentrations of each contaminant/potential interference in the unspiked DDW
so that background concentrations of contaminants or potential interferences were accounted for
within the displayed concentration of each contaminant/potential interference sample. Table 3-3
lists the standard methods used to measure each analyte; the results from the stock solution
confirmation analyses (obtained by reporting the lethal dose concentration for the contaminants
-------
Table 3-3. Dose Confirmation Results
Contaminant
Aldicarb
Colchicine
Cyanide
Dicrotophos
Thallium sulfate
Botulinum toxin
Ricin
Soman
VX
Method
EPA531.1(3)
(a)
EPA335.1(4)
EPASW846(8141A)(5)
EPA 200.8(6)
(a)
(a)
(c)
(c)
Average Concentration
± Standard Deviation N
= 4 (mg/L)
280 ± 28
NAW
250 ±15
1,400 ±140
2,400 ± 24
NA
NA
0.068(d) ± 0.001
0.22 ± 0.02
Background in
DDW Sample
(mg/L)
<0.0007
NA
0.008
<0.002
<0.001
NA
NA
<0.05
<0.05
Potential Interference
Aluminum
Copper
Iron
Manganese
Zinc
EPA 200.8
EPA 200.8
EPA 200.8
EPA 200.8
EPA 200.8
0.36 ±0.01
0.65 ±0.01
0.069 ± 0.008
0.26 ±0.01
3.5 ±0.35
<0.10
0.011
<0.04
<0.01
0.30
(a) No standard method available. QA audits and balance calibration assured accurately prepared solutions.
(b) NA = Not applicable.
(c) Purity analyses performed on chemical and biological agent materials using Battelle standard operating
procedures.
(d) The result of the dose confirmation analysis for soman was 23% of the expected concentration of 0.30 mg/L.
and the single concentration that was analyzed for the potential interferences); and the
background levels of the contaminants and potential interferences measured in the DDW
sample, which were all non-detect or negligible.
Standard methods were also used to characterize several water quality parameters such as the
concentration of trihalomethanes, haloacetic acids, and total organic halides; turbidity; dissolved
organic carbon content; pH; alkalinity; specific conductivity; and hardness. Table 3-4 lists these
measured water quality parameters for both the water sample collected in Columbus, Ohio,
representing a water system using chlorination as the disinfecting process, and the water sample
collected in St. Petersburg, Florida, representing a water system using chloramination as the
disinfecting process.
10
-------
Table 3-4. Water Quality Parameters
Parameter
Turbidity
Organic carbon
Specific conductivity
Alkalinity
pH
Hardness
Total organic halides
Total trihalomethanes
Total haloacetic acids
Dechlorinated Columbus,
Ohio, Tap Water (disinfected
Method by chlorination)
EPA180.1(7)
SM5310(8)
SM2510(8)
SM 2320(8)
EPA150.1(9)
EPA 130.2(9)
SM 5320B(8)
EPA 524.2(10)
EPA 552.2(11)
0.1 NTU(a)
2.5 mg/L
364 [imho
42 mg/L
7.65
112 mg/L
190 |ig/L
52.8 |ig/L
75.7 fig/L
Dechlorinated
St. Petersburg, Florida,
Tap Water (disinfected by
chloramination)
0.3 NTU
2.9 mg/L
460 [irnho
97 mg/L
7.95
160 mg/L
83fig/L
2.4 fig/L
13.5 fig/L
1 NTU = nephelometric turbidity unit.
11
-------
Chapter 4
Quality Assurance/Quality Control
QA/QC procedures were performed in accordance with the quality management plan (QMP) for
the AMS Center(12) and the test/QA plan for this verification test.(1)
4.1 Quality Control of Stock Solution Confirmation Methods
The stock solutions for aldicarb, cyanide, dicrotophos, and thallium sulfate were analyzed using
a standard reference method at ATEL. As part of ATEL's standard operating procedures (SOPs),
various QC samples were analyzed with each sample set. These included matrix spike,
laboratory control spike, and method blank samples. According to the standard methods used for
the analyses, recoveries of the QC spike samples analyzed with samples from this verification
test were within acceptable limits of 75% to 125%, and the method blank samples were below
the detectable levels for each analyte. For VX and soman, the confirmation analyses were
performed at Battelle using a Battelle SOP. Calibration standard recoveries of VX and soman
were always between 69% and 130%, and most of the time were between 90% and 100%.
Standard analytical methods for colchicine, ricin, and botulinum toxin were not available and,
therefore, were not performed. QA audits and balance calibrations assured that solutions for
these compounds were accurately prepared.
4.2 Quality Control of Drinking Water Samples
A method blank sample consisting of ASTM Type n DI water was analyzed once by
ToxScreen-H for approximately every 20 drinking water samples that were analyzed. These
samples were used to set a baseline light level for a clean water matrix. A negative control
sample (unspiked DDW) was analyzed for approximately every four samples. The light
produced from samples fortified with contaminants were compared with the light produced from
the negative control samples to calculate the percent inhibition caused by the contaminant. A
positive control sample also was analyzed once for approximately every 20 drinking water
samples. While performance limits were not placed on the results of the positive control sample,
the vendor informed Battelle that, if the positive control samples did not cause greater than
approximately 50% inhibition, it would indicate to the operator that ToxScreen-H was operating
incorrectly. Sodium chloroacetate was the positive control sample for the pro-organic buffer
samples and copper chloride for the pro-metal buffer samples. For 24 positive control samples of
each type, the average inhibition was 90% ± 10% for the pro-metal buffer and 88% ± 10% for
12
-------
the pro-organic buffer. This level of inhibition for the positive control samples indicated the
proper functioning of ToxScreen-H.
4.3 Audits
4.3.1 Performance Evaluation Audit
The concentration of the standards used to prepare the contaminant and potential interferences
was confirmed by analyzing solutions of each analyte prepared in ASTM Type n DI water from
two separate commercial vendors using the confirmation methods. The standards from one
source were used to prepare the stock solutions during the verification test, while the standards
from a second source were used exclusively to confirm the accuracy of the measured concentra-
tion of the first source. The percent difference (%D) between the measured concentration of the
performance evaluation (PE) sample and the prepared concentration of that sample was
calculated using the following equation:
M
%D = — xlOO%
A (1)
where M is the absolute value of the difference between the measured and the prepared concen-
tration and A is the prepared concentration. The %D between the measured concentration of the
PE standard and the prepared concentration had to be less than 25 for the measurements to be
considered acceptable. Table 4-1 shows the results of the PE audit for each compound. All %D
values were less than 25.
Given the lack of confirmation methodology for some of the contaminants in this verification
test, PE audits were not performed for all of the contaminants. PE audits were performed when
more than one source of the contaminant or potential interference was commercially available
and when were methods were available to perform the confirmation. To assure the purity of the
other standards, documentation, such as certificates of analysis, was obtained for colchicine,
botulinum toxin, and ricin. In the case of VX and soman, which were obtained from the U.S.
Army, the reputation of the source, combined with the confirmation analysis data, provided
assurance of the concentration analyzed.
4.3.2 Technical Systems Audit
The Battelle Quality Manager conducted a technical systems audit (TSA) to ensure that the
verification test was performed in accordance with the test/QA plan(1) and the AMS Center
QMP.(12) As part of the audit, the Battelle Quality Manager reviewed the contaminant standard
and stock solution confirmation methods, compared actual test procedures with those specified
in the test/QA plan, and reviewed data acquisition and handling procedures. Observations and
findings from this audit were documented and submitted to the Battelle verification test
coordinator for response. No findings were documented that required any significant action. The
records concerning the TSA are permanently stored with the Battelle Quality Manager.
13
-------
Table 4-1. Summary of Performance Evaluation Audit
Contaminant
Potential
interference
Aldicarb
Cyanide
Dicrotophos
Thallium sulfate
Aluminum
Copper
Iron
Manganese
Zinc
Average Measured
Concentration ±
Standard Deviation
(mg/L)
0.00448 ± 0.000320
0.207 ± 0.026
0.00728 ± 0.000699
0.090 ± 0.004
0.512 ±0.013
0.106 ±0.002
0.399 ± 0.004
0.079 ± 0.003
0.106 ±0.016
Actual
Concentration
(mg/L)
0.00500
0.200
0.00748
0.100
0.500
0.100
0.400
0.100
0.100
Percent
Difference
11
4
3
10
2
6
0.30
21
6
The EPA Quality Manager also conducted a TSA to ensure that the verification test was
performed in accordance with the test/QA plan(1) and the AMS Center QMP.(12) As part of the
audit, the EPA Quality Manager compared actual test procedures with those specified in the
test/QA plan and reviewed data acquisition and sample preparation records and procedures. No
significant findings were observed during the EPA TSA. The records concerning the TSA are
permanently stored with the EPA Quality Manager.
4.3.3 Audit of Data Quality
At least 10% of the data acquired during the verification test were audited. Battelle's Quality
Manager 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.
4.4 QA/QC Reporting
Each internal assessment and audit was documented in accordance with Sections 3.3.4 and 3.3.5
of the QMP for the ETV AMS Center.(12) Once the assessment report was prepared, the Battelle
verification test coordinator ensured that a response was provided for each adverse finding or
potential problem and implemented any necessary follow-up corrective action. The Battelle
Quality Manager ensured that follow-up corrective action was taken. The results of the TSA
were sent to the EPA.
14
-------
4.5 Data Review
Records generated in the verification test were reviewed before these records were used to
calculate, evaluate, or report verification results. Table 4-2 summarizes the types of data
recorded. The review was performed by a technical staff member involved in the verification
test, but not the staff member who originally generated the record. The person performing the
review added his/her initials and the date to a hard copy of the record being reviewed.
Table 4-2. Summary of Data Recording Process
Data to be
Recorded
Responsible
Party
Where
Recorded
How Often
Recorded
Disposition of Data(a)
Dates, times of test Battelle
events
Sample preparation Battelle
(dates, procedures,
concentrations)
Test parameters Battelle
(contaminant
concentrations,
location, etc.)
Laboratory
record books
Laboratory
record books
Start/end of test, and
at each change of a
test parameter
When each sample
was prepared
Laboratory
record books
When set or
changed
Used to organize/check
test results; manually
incorporated in data
spreadsheets as
necessary
Used to confirm the
concentration and
integrity of the samples
analyzed, procedures
entered into laboratory
record books
Used to organize/check
test results, manually
incorporated in data
spreadsheets as
necessary
Stock solution
confirmation
analysis, sample
analysis, chain of
custody, and
results
Battelle or
contracted
laboratory
Laboratory
record books,
data sheets, or
data acquisition
system, as
appropriate
Throughout sample
handling and
analysis process
Transferred to
spreadsheets/agreed
upon report
(a) All activities subsequent to data recording were carried out by Battelle.
15
-------
Chapter 5
Statistical Methods and Reported Parameters
The statistical methods presented in this chapter were used to verify the performance parameters
listed in Section 3.1.
5.1 Endpoints and Precision
The luminometer provided with the ToxScreen-H reported the absolute light units for each
sample analyzed. Each DDW sample containing contaminants was compared with a negative
control sample that, for this verification test, was unspiked DDW. This comparison was made by
accounting for the inhibition of the negative control in the calculation of the percent inhibition.
Therefore, the percent inhibition of the negative control sample within each sample set was
defined as approximately zero percent. The percent inhibition for each sample was calculated
using the following equation:
% inhibition = 1 - -- x 100%
n\
Where L is the absolute light units produced for the control and test samples. For this test, the
control sample was always DDW, except when the inhibition of the disinfectant by-products was
being determined, in that case, ASTM Type n DI water served as the control sample.
The standard deviation (S) of the results for the replicate samples was calculated, as follows, and
used as a measure of technology precision at each concentration.
s=p-i(/,-7)T (3)
_n — i k=i
where n is the number of replicate samples, Ik is the percent inhibition measured for the kth
sample, and / is the average percent inhibition of the replicate samples. Because the average
inhibitions were frequently near zero for this data set, relative standard deviations often would
have greatly exceeded 100%, making the results difficult to interpret. Therefore, the precision
results were left in the form of standard deviations so the reader could easily view the
uncertainty around the average for results that were both near zero and significantly larger than
zero.
16
-------
5.2 Toxicity Threshold
The toxicity threshold was defined as the lowest concentration of contaminant to exhibit a
percent inhibition significantly greater than the negative control. Also, each concentration level
higher than the toxicity threshold had to be significantly greater than the negative control, and the
inhibition produced by each lower concentration analyzed had to be significantly less than that
produced by the toxicity threshold concentration. Since the inhibition of the test samples was
calculated with respect to the inhibition of each negative control sample, the percent inhibition of
the negative control was always zero. An inhibition was significantly greater than the negative
control if the average inhibition plus or minus the standard deviation did not include zero.
5.3 False Positive/Negative Responses
A response would be considered false positive if an unspiked drinking water sample produced an
inhibition significantly greater than zero when determined with respect to ASTM Type IIDI
water. Depending on the degree of inhibition in the sample, toxicity due to subsequent con-
tamination of that sample may not be detectable or could be exaggerated as a result of the
baseline inhibition. To test for this possibility, the percent inhibition of the unspiked drinking
water was determined with respect to ASTM Type n DI water. Drinking water samples collected
from water systems using chlorination and chloramination as the disinfecting process were
analyzed in this manner. An inhibition was considered significantly different from zero if the
average inhibition, plus or minus the standard deviation, did not include zero.
A response was considered false negative when ToxScreen-H was subjected to a lethal concentra-
tion of some contaminant in the DDW and did not indicate inhibition greater than 50%. The
vendor's instructions stated that a result was not positive for toxicity unless it was at least 50%.
A difference was considered significant if the average inhibition plus or minus the standard
deviation did not encompass the value or range of values that were being compared.
5.4 Field Portability
The results obtained from the measurements made on DDW samples in the laboratory and in the
field setting were compiled independently and compared to assess the performance of the
ToxScreen-II under different analysis conditions. Means and standard deviations of the endpoints
generated in both locations were used to make the comparison. Also, qualitative observations of
ToxScreen-H in a non-laboratory setting were made by the verification test coordinator and
operators. Factors such as the ease of transport and set-up, demand for electrical power, and
space requirement were documented.
5.5 Other Performance Factors
Ease of use (including clarity of the instruction manual, user-friendliness of software, and overall
convenience) was qualitatively assessed throughout the verification test through observations of
17
-------
the operators and verification test coordinator. Sample throughput was evaluated quantitatively
based on the number of samples that could be analyzed per hour.
18
-------
Chapter 6
Test Results
6.1 Endpoints and Precision
Tables 6-la-i present the percent inhibition data for nine contaminants, and Table 6-2 presents
data for five potential interferences and the drinking water samples disinfected by both
chlorination and chloramination. Given in each table are the concentrations analyzed, the percent
inhibition results for each replicate at each concentration, and the average and standard deviation
of the inhibition of the four replicates at each concentration. Results are provided for each
contaminant analyzed in both the pro-organic and pro-metal buffers. Samples that produced
negative percent inhibition values indicated an increase in light production by the bacteria relative
to the negative control and were considered to be non-toxic.
6.1.1 Contaminants
The contaminants that were analyzed by ToxScreen-H during this verification test produced
results that differed depending on which buffer was used. Each contaminant was analyzed using
each buffer at all the concentration levels. Since the buffers were developed to enhance the
sensitivity of specific classes of compounds, the results were expected to show this difference.
Cyanide, dicrotophos, and thallium sulfate were the only three contaminants that exhibited
inhibitions that were significantly greater than the negative control in the pro-metal buffer.
Cyanide exhibited significant inhibitions down to 0.025 mg/L in that buffer while dicrotophos
was only significantly different than the negative control at the 1,400 mg/L level and thallium
sulfate for the 240 and 2,400 mg/L concentration levels.
In the pro-organic buffer, aldicarb exhibited inhibitions significantly greater than the negative
control for all of the concentration levels; but the average inhibition did not change in proportion
to the concentration of alicarb in each sample. The inhibition of 2.8 mg/L sample was greater than
the inhibition produced by the 28 mg/L sample and similar to the inhibition of the 280 mg/L
sample. A similar situation was true for colchicine, since the 0.24 mg/L sample exhibited higher
inhibitions than the 2.4 and 24 mg/L sample; however, the inhibition of the lethal dose of
colchicine (83% ± 3%) was significantly greater than all of the other concentration levels
analyzed. Also, in the pro-organic buffer, ToxScreen-H exhibited nearly compete inhibition (85%
to 99%) for cyanide down to a concentration of 2.5 mg/L and an inhibition of 68% ± 3% for 0.25
mg/L cyanide. The inhibition of dicrotophos increased from 41% at 0.14 mg/L and 1.4 mg/L to
99% at 140 mg/L and 1,400 mg/L in the pro-organic buffer. The inhibitions of the two highest
concentrations of botulinum toxin were significantly greater than the negative control, but not
19
-------
Table 6-la. Aldicarb Percent Inhibition Results
Concentration
(mg/L)
0.28
2.8
28
280
(Lethal Dose)
Inhibition
10
12
3
18
66
63
67
36
41
36
39
37
55
88
68
64
Pro-Organic Buffer
Average Standard
(%) Deviation (%)
11 6
58 15
38 2
69 14
Pro-Metal
Inhibition Average
-49
-51
-47 -46
-37
-20
-13
-15 -17
-22
-3
I '
13
-23
-21 -21
-21
-20
Buffer
Standard
Deviation
6
4
7
1
20
-------
Table 6-lb. Colchicine Percent Inhibition Results
Concentration
(mg/L)
0.024
0.24
2.4
24
240
(Lethal Dose)
Pro-Organic Buffer
Inhibition Average Standard
(%) (%) Deviation (%)
-73
-70
0
21
63
68
41
48
45
39
18
35
25
9
13
33
79
82
86
84
-31 48
55 13
34 11
20 11
83 3
Pro-Metal Buffer
Standard
Inhibition Average Deviation
0
-5 <
21
-70
-81 -79 7
-85
-78
-83
£ «
-88
-61
-77 9
-72
-70 '
-81
-50
-4° -42 7
-46 '
-34
21
-------
Table 6-lc. Cyanide Percent Inhibition Results
Concentration
(mg/L)
0.0025
0.025
0.25
2.5
25
250
(Lethal Dose)
0.25
(Field Location)
Pro-Organic Buffer
Inhibition Average Standard Deviation
29
-25
5
8
40
64
69
99
99
99 0
99
99
100
99 85 29
99
99
100 0
99
99
97
97 97 1
98 y/
96
Pro-Metal Buffer
Standard
Inhibition Average Deviation
-3
I '
3
37
30 31 8
20 31 8
37
26
34 35 7
37
42
65
(a)
87 8° B
88
36
58 48 9
50
99
99
99 1
98
98
65
61 6
54
1 Data point removed because of transcription error.
22
-------
Table 6-Id. Dicrotophos Percent Inhibition Results
Concentration
(mg/L)
0.14
1.4
14
140
1,400
(Lethal Dose)
Inhibition
30
39
29
67
30
57
43
34
77
92
65
89
99
99
99
99
98
99
99
99
Pro-Organic Buffer
Average Standard
(%) Deviation (%)
41 18
41 12
81 12
99 0
99 0
Pro-Metal Buffer
Standard
Inhibition Average Deviation
-76
-6° -52 23
-53 ^ *
-21
-16
-25
-43 -32 14
-46
-38
-43 -5° U
-63
-81
:™ -«
-86
25
23
31 9
43
33
23
-------
Table 6-le. Thallium Sulfate Percent Inhibition Results
Concentration
(mg/L)
2.4
24
240
2,400
(Lethal Dose)
Inhibition
(%)
-9
-49
-26
-43
-15
-24
-5
-3
31
100
26
51
-55
100
-82
-3
Pro-Organic Buffer
Average Standard
(%) Deviation (%)
-32 18
-12 10
52 34
-10 80
Pro-Metal
Inhibition Average
(%) (%)
-3
.°3 6
15
16
11 1fi
11 16
26
66
£ »
70
97
99
98
97
97
Buffer
Standard
Deviation
(%)
9
7
2
1
Table 6- If. Botulinum Toxin Percent Inhibition Results
Concentration
(mg/L)
0.00030
0.0030
0.030
0.30
(Lethal Dose)
Inhibition
(%)
7
15
-16
-36
-21
5
-26
-17
45
48
20
76
80
81
74
41
Pro-Organic Buffer
Average Standard
(%) Deviation (%)
-8 23
-15 14
47 23
69 19
Pro-Metal
Inhibition Average
(%) (%)
-100
-84 Q7
(a) -97
-108
-59
-63
-62
-43
18
1 '
-12
-47
-54
-12 -44
-62
Buffer
Standard
Deviation
(%)
12
9
14
22
(a) Data point removed because of transcription error.
24
-------
Table 6-lg. Ricin Percent Inhibition Results
Concentration
(mg/L)
0.015
0.15
1.5
15
(Lethal Dose)
Inhibition
-52
10
4
-4
26
30
22
41
27
28
-5
-11
56
53
46
43
Pro-Organic Buffer
Average Standard
(%) Deviation (%)
-10 28
30 8
10 21
50 6
Pro-Metal Buffer
Inhibition
-93
-96
-86
-100
-74
-85
-111
-102
-18
3
86
-93
-96
-86
-100
-74
Standard
Average Deviation
-89 12
-93 17
-5 74
-89 12
Table 6-lh. Soman Percent Inhibition Results
Concentration
(mg/L)
0.000068
0.00068
0.0068
0.068(a)
(Lethal Dose)
Pro-Organic Buffer
Inhibition Average Standard
(%) (%) Deviation (%)
-51
-35
9
10
-11
-20
1
-5
-10
-6
-5
16
-40
17
19
25
-17 31
-9 9
-1 12
5 30
Pro-Metal Buffer
Inhibition
-55
-53
-50
-64
-63
-57
-64
-64
-39
-28
-53
-68
-50
-57
-46
91
Standard
Average Deviation
-55 6
-62 3
-47 17
-15 71
Due to the degradation of soman in water, the stock solution confirmation analysis confirmed that the
concentration of the lethal dose was 23% of the expected concentration of 0.30 mg/L.
25
-------
Table 6-li. VX Percent Inhibition Results
Concentration
(mg/L)
0.00022
0.0022
0.022
0.22
(Lethal Dose)
Pro-Organic Buffer
Inhibition Average Standard
(%) (%) Deviation (%)
-53
-57
-60
-66
-15
29
-27
-41
6
-17
3
-12
-26
27
3
-34
-59 6
-13 30
-5 11
-8 28
Pro-Metal Buffer
Inhibition
-52
-19
-33
-21
-46
-13
-8
-18
-27
-14
-6
-30
-42
-43
4
-71
Standard
Average Deviation
-31 15
-21 17
-19 11
-38 31
significantly different from one another. Finally, the inhibition of ricin at 15 mg/L was
significantly greater than the negative control and the other concentration levels analyzed. The
inhibitions of thallium sulfate, soman, and VX did not meet the requirements of detectability, as
defined in Section 5.2, in the pro-organic buffer.
In general, of the contaminants that exhibited inhibition, those with more organic makeup
(aldicarb, dicrotophos, colchicine, ricin, botulinum toxin) were more likely to be detected in the
pro-organic buffer compared to the pro-metal buffer. Dicrotophos was detected in both buffers,
but was measured at a lower concentration in the pro-organic buffer. Thallium sulfate, a metallic
contaminant, was detected only in the pro-metal buffer. Inhibition caused by cyanide was
detected in both buffers because cyanide (made from potassium cyanide) has both metallic and
organic characteristics.
6.1.2 Potential Interferences
Table 6-2 presents the results from the samples that were analyzed to test the effect of potential
interferences on ToxScreen-n. In the pro-organic buffer, copper (59% ± 7%) and iron (29% ±
23%) were the only potential interferences that exhibited inhibitions significantly greater than
the negative control. In the pro-metal buffer, copper and zinc exhibited average inhibitions of
94% and 100%, respectively, and manganese exhibited a much smaller average inhibition of
22%. Each of these metals has the potential to interfere with ToxScreen-H measurements, but
especially copper and zinc in the pro-metal buffer.
26
-------
Table 6-2. Potential Interferences Results
Pro-Organic Buffer
Potential
Interferences
Aluminum
Copper
Iron
Manganese
Zinc
Chlorination
By-products
Chloramin-
ation By-
products
Concen-
tration
(mg/L)
0.36
0.65
0.069
0.26
3.5
NA(a)
NA
Standard
Inhibition Average Deviation
32
-24 16 41
65
-10
63
49
59 7
62
64
52
29 23
44
11
-22
45 -23 7
-22
-32
-f -68 79
-162
00 66 34
81
£
71
Pro-Metal Buffer
Inhibition Average
-35
2 -26
-37
-34
93
95 94
91
96
40
-6 3
13
-35
31
22
22
17
17
100
100
100 1UU
100
w 23
-57
£ -
-56
Standard
Deviation
19
2
32
7
0
15
20
(a) NA = Not applicable.
(b) Chlorination by-product data averaged over negative control data compared to ASTM Type IIDI water (N=44).
All of the contaminant and potential interference samples were prepared in the DDW and
compared with unspiked DDW. Therefore, any background inhibition in the DDW was corrected
by subtracting the inhibition caused by the negative control sample. To investigate whether
ToxScreen-H is sensitive to by-products of disinfecting processes, dechlorinated drinking water
samples from water systems that use Chlorination and chloramination were analyzed and
compared with ASTM Type n DI water as the control sample. This determination is crucial
because the ability of ToxScreen-H to detect toxicity is dependent on the bacteria's background
light production in a clean drinking water matrix. If clean drinking water produces 100%
inhibition of light, inhibition caused by contaminants could not be detected. In the pro-organic
buffer (over 42 samples), the sample from the water supply disinfected with Chlorination
exhibited inhibitions of 66% ± 34%, while the sample from the water supply disinfected by
chloramination exhibited inhibitions of 78% ± 11% on four replicates. This suggests that samples
that have been disinfected using either process are likely to interfere with ToxScreen-H results if a
27
-------
reference sample very similar to the sample matrix is not used because the inhibition caused by
the "clean" drinking water matrix left only one quarter to one third of the light to potentially be
inhibited by contamination. For the pro-metal buffer, the inhibition of the sample from the water
supply disinfected by chlorination was 23% ± 15%, and the inhibition of the sample from the
water supply disinfected by chloramination was -48% ± -20%. In the former case, this inter-
ference could cause slightly exaggerated inhibitions and, in the latter case, inhibitions could be
underestimated. For example, if a contaminant that exhibited approximately 50% inhibition was
placed in water from a chloraminated system, and ASTM Type n DI water was used as the
reference sample, the percent inhibition would be approximately zero percent.
Overall, because of the low level of light production when using the pro-organic buffer in the
DDW, water disinfected by chlorination may interfere with the ToxScreen-n results even if a
similar reference sample is used. As long as a similar reference sample is used when using the
pro-metal buffer, water disinfected using either process is not likely to interfere with the
ToxScreen-H results. However, if ASTM Type n DI water has to be used as the reference sample,
all the water types could potentially interfere with the results from ToxScreen-H.
6.1.3 Precision
Across all the contaminants and potential interferences, the standard deviation was measured and
reported for each set of four replicates to evaluate ToxScreen-H precision. Out of 96 opportunities,
the standard deviation of the four replicate measurements was less than 10% 42 times, between
10% and 20% 31 times, and greater than 20% 23 times. There was no clear reason for the high
variability of the results. The same two operators analyzed all the samples using the techniques
described in the instructions and provided personally during the vendor training.
6.2 Toxicity Threshold
Table 6-3 gives the toxicity thresholds, as defined in Section 5.2, for each contaminant. The
lowest toxicity threshold concentration was for cyanide at 0.025 mg/L when the pro-metal buffer
was used and botulinum toxin at 0.03 mg/L when the pro-organic buffer was used, indicating that
ToxScreen-H was most sensitive to cyanide and botulinum toxin, depending on the buffer used.
6.3 False Positive/Negative Responses
False positive responses were observed for unspiked drinking water samples from systems that use
chlorination and chloramination as their disinfectant processes when the pro-organic buffer was
used. As described in Section 6.1.2, for a clean water sample that had been disinfected using each
process, ToxScreen-H reported inhibitions of 66% ± 34% for chlorinated drinking water and 78 ±
11% for chloraminated drinking water. By-products of these processes apparently inhibited the
ToxScreen-H reagent in the pro-organic buffer. At times, the background inhibition caused by the
drinking water sample inhibited the light to the extent that it was questionable whether
28
-------
Table 6-3. Toxicity Thresholds
Concentration (mg/L)
Contaminant
Aldicarb
Colchicine
Cyanide
Dicrotophos
Thallium sulfate
Botulinum toxin
Ricin
Soman
VX
Pro-Organic Buffer
0.28
0.24
0.25
0.14
ND
0.030
15
ND
ND
Pro-Metal Buffer
ND(a)
ND
0.025
1,400
240
ND
ND
ND
ND
) ND = Significant inhibition was not detected.
enough light remained to detect inhibition due to contamination. Generally, the negative control
sample for the pro-organic buffer generated about 500 light units; in contrast, the negative
control in the pro-metal buffer typically would generate approximately 100,000 light units.
However, even with the low number of light units, the positive control sample always had an
inhibition of greater than 50% with respect to the negative control. At a minimum, to avoid the
risk of false positive responses, reference samples very similar to the test samples need to be
used. However, with inhibition to this extent, false positive responses are still possible. When
using the pro-metal buffer, the inhibition due to the sample from the water system disinfected by
chlorination was just 23% ± 15%. Therefore, there is potential for exaggerated inhibition, but
only if dissimilar reference samples are used. Since testing, the vendor has determined that
residual sodium thiosulfate from dechlorination may be the reason for the high inhibition in the
unspiked DDW. At any rate, the vendor has altered the storage buffer to sustain a higher light
level even in the presence of the disinfectant by-products.
The inhibition due to the sample from the water system disinfected by chloramination was -48%
± 20% when using the pro-metal buffer. This scenario introduces the possibility of a false
negative response if a reference sample similar to the water sample is not used. Because the
chloraminated water exhibited less inhibition than ASTM Type n DI water, a negative inhibition
was calculated. If ASTM Type n DI water was used as the reference sample, and the chloramin-
ated water sample was contaminated with cyanide or another contaminant that could cause
approximately a 50% inhibition, that inhibition would be calculated to an approximately zero
inhibition, therefore a false result. In this case, using a similar reference would solve the
problem, but the possibility of false negative responses must be considered if ASTM Type n
water has to be used as the reference.
A second type of false negative response was considered when a lethal dose of contaminant is
present in the water sample and the inhibition is not at least 50%, the lower limit for a positive
29
-------
response according to the vendor. Table 6-4 gives these results. Only soman and VX did not
exhibit at least 50% inhibitions at the lethal concentration in at least one of the two buffers used
by ToxScreen-n.
Table 6-4. False Negative Responses
False Negative Response
Lethal Dose
Concentration Inhibition > 50% with Inhibition > 50% with
Contaminant (mg/L) Pro-organic Buffer Pro-metal Buffer
Aldicarb
Colchicine
Cyanide
Dicrotophos
Thallium sulfate
Botulinum toxin
Ricin
Soman
VX
280
240
250
1,400
2,400
0.30
15
0.068(a)
0.22
no
no
no
no
yes
no
no
yes
yes
yes
yes
no
yes
no
yes
yes
yes
yes
-------
The ToxScreen-H reagent must be kept at approximately -20°C prior to reconstitution and, once
reconstituted and stored in storage buffer, can be maintained for up to one week at
approximately 4°C. These factors could be problematic in a long-term field deployment.
6.5 Other Performance Factors
The step-by-step pictorial instruction manual for ToxScreen-H was easy to understand, which
enabled operators to become quickly adept at analyzing multiple sample sets. ToxScreen-H was
straightforward to operate. Although the operators had scientific backgrounds, based on
observations of the verification test coordinator, operators with little technical training would
probably be able to successfully analyze sample sets using only the instruction manual. The
operators analyzed 25 samples per hour.
31
-------
Chapter 7
Performance Summary
Table 7-1 shows the performance verification results for the ToxScreen-n pro-organic buffer.
Parameter
Contaminants
inDDW
Potential
interferents in
DDW
Compound
Aldicarb
Colchicine
Cyanide
Dicrotophos
Thallium
sulfate
Botulinum
toxin(e)
Ricin(fl
Soman
VX
Interference
Aluminum
Copper
Iron
Manganese
Zinc
Lethal Dose
(LD) Cone.
(mg/L)
280
240
250
0.025(c)
1,400
2,400
0.30
15
0.068(E)
0.22
Cone.
(mg/L)
0.36
0.65
0.069
0.26
3.5
Average Inhibitions at Concentrations Relative to
the LD Concentration (%)
LD
69
83
99
14
99
-10
69
50
5
-8
LD/10
38
20
85
-6
99
52
47
10
-1
-5
LD/100
58
34
99
NA
81
-12
-15
30
-9
-13
LD/1,000
11
55
68
NA
41
-32
-8
-10
-17
-59
LD/10,000
NA(b)
-31
14
NA
41
NA
NA
NA
NA
NA
Average Inhibitions at a
Single Concentration (%)
16
59
29
-23
-68
Range of
Standard
Deviations
(%)
2-15
3^8
0-29
18-27
0-18
10-80
14-23
6-28
9-31
6-30
Standard
Deviation
(%)
41
7
23
7
79
Toxicity
Thresh.
(mg/L)(a)
0.28
0.24
0.25
NA
0.14
ND(d)
0.030
15
ND
^NT^^
False positive
responses
False positive responses were observed for both chlorinated and chloraminated water samples. To avoid
such results, reference samples similar to the test samples should be used. However, since the light
emitted for background water samples is greatly inhibited (66 ± 34% for chlorinated drinking water and
78% ±11% for chloraminated drinking water), a false positive response in these matrices may occur.
Residual sodium thiosulfate from dechlorination may have caused these responses. Storage buffer has
been altered to increase light output in presence of disinfectant by-products.
False negative
responses
The inhibition at the lethal concentration level was significantly larger than the negative control for all
contaminants except VX, soman, and thallium sulfate.
Field
portability
A single concentration of cyanide was analyzed in the field and the laboratory. Similar performance was
observed at both locations. Inhibition in the field was 97% ±1% and in the laboratory was 68% ± 3%.
ToxScreen-E was transported to the field in a vinyl bag provided by the vendor, but not included as part
of the test kit.
32
-------
Other
performance
factors
The pictorial instruction manual was useful, and instrument operation was straightforward. Although the
operators for this test had scientific backgrounds, operators with little technical training would probably
be able to successfully analyze sample sets. The operators analyzed approximately 25 samples per hour.
See Tables 6-la-I in the report for the precision around each individual inhibition result.
NA = Data not collected at this concentration level for this contaminant.
0.025 mg/L is not a lethal dose concentration, but it is used to describe additional concentrations of cyanide that were analyzed.
ND = Not detectable.
Lethal dose solution also contained 3 mg/L phosphate and 1 mg/L sodium chloride.
Lethal dose solution also contained 3 mg/L phosphate, 26 mg/L sodium chloride, and 2 mg/L sodium azide.
Due to the degradation of soman in water, the stock solution confirmation analysis confirmed that the concentration of the lethal
dose was 23% of the expected concentration of 0.30 mg/L.
Table 7-2 shows the performance verification results for the ToxScreen-H pro-metal buffer.
Parameter
Contaminants
inDDW
Potential
interferents in
DDW
Compound
Aldicarb
Colchicine
Cyanide
Dicrotophos
Thallium
sulfate
Botulinum
toxin(e)
Ricin(e
Soman
VX
Interference
Aluminum
Copper
Iron
Manganese
Zinc
Lethal
Dose (LD)
Cone.
(mg/L)
280
240
250
0.025(d)
1,400
2,400
0.30
15
0.068(E)
0.22
Cone.
(mg/L)
0.36
0.65
0.069
0.26
3.5
Average Inhibitions at Concentrations Relative to
the LD Concentration (%)
LD
-21
-42
99
31
31
98
-44
-89
-15
-38
LD/10
5
-72
48
3
-76
69
1
-5
-47
-19
LD/100
-17
-83
80
NA
-50
16
-57
-93
-62
-21
LD/1,000
-46
-79
35
NA
-32
6
-97
-89
-55
-31
LD/10,000
NA(b)
4
31
NA
-52
NA
NA
NA
NA
NA
Average Inhibitions at a
Single Concentration (%)
-26
94
3
22
100
Range of
Standard
Deviations
(%)
1-7
5-12
1-13
5-8
9-23
1-9
9-22
12-74
3-71
11-31
Standard
Deviation
(%)
19
2
32
7
0
Toxicity
Thresh.
(mg/L)«
ND(C)
ND
0.025
NA
1,400
240
ND
ND
ND
^NTJ
False positive
responses
The sample from the chlorinated water supply exhibited inhibition of 23% ± 15%. This could cause
slightly exaggerated results. Using a reference sample similar to the drinking water sample being tested
would protect from this possibility. Residual sodium thiosulfate from dechlorination may have caused
these responses. Storage buffer has been altered to increase light output in presence of disinfectant
by-products.
False negative
responses
Inhibition was -48% ± 20% for the sample from the chloraminated water supply, indicating the
possibility of false negative results. Adding a contaminant that caused 50% inhibition would result in an
inhibition of approximately 0%. Using a reference sample similar to the water sample treated by
dechloramination may avoid false negative results. Aldicarb, colchicine, botulinum toxin, ricin, soman,
and VX were indistinguishable from the negative control.
33
-------
Field
portability
A single concentration of cyanide was analyzed in the field and the laboratory. Similar performance was
observed at both locations. Inhibition in the field was 61% ± 6% and in the laboratory was 35% ± 7%.
ToxScreen-II was transported to the field in a vinyl bag provided by the vendor, but not included as part
of the test kit.
Other
performance
factors
The pictorial instruction manual was useful and instrument operation was straightforward. Although the
operators for this test had scientific backgrounds, operators with little technical training would probably
be able to successfully analyze sample sets. The operators analyzed approximately 25 samples per hour.
See Tables 6-la-I in the report for the precision around each individual inhibition result.
NA = Data not collected at this concentration level for this contaminant.
ND = Not detectable.
0.025 mg/L is not a lethal dose concentration, but it is used to describe additional concentrations of cyanide that were analyzed.
Lethal dose solution also contained 3 mg/L phosphate and 1 mg/L sodium chloride.
Lethal dose solution also contained 3 mg/L phosphate, 26 mg/L sodium chloride, and 2 mg/L sodium azide.
Due to the degradation of soman in water, the stock solution confirmation analysis confirmed that the concentration of the lethal
dose was 23% of the expected concentration of 0.30 mg/L.
34
-------
Chapter 8
References
1. Test/QA Plan for Verification of Rapid Toxicity Technologies, Battelle, Columbus, Ohio,
June 2003.
2. United States Environmental Protection Agency, National Secondary Drinking Water
Regulations: Guidance for Nuisance Chemicals, EPA/810/K-92/001, July 1992.
3. U.S. EPA Method 531.1, "Measurement of n-Methylcarbamoyloximes and
n-Methylcarbamates in Water by Direct Aqueous Injection HPLC with Post Column
Derivatization," in Methods for the Determination of Organic Compounds in Drinking
Water—Supplement III, EPA/600/R-95/131, 1995.
4. U.S. EPA Method 335.1, "Cyanides, Amenable to Chlorination," in Methods for the Chemical
Analysis of Water and Wastes, EPA/600/4-79/020, March 1983.
5. SW846 Method 8141 A, "Organophosphorous Compounds by Gas Chromatography: Capillary
Column Technique," Revision 1, September 1994.
6. U.S. EPA Method 200.8, "Determination of Trace Elements in Waters and Wastes by
Inductively-Coupled Plasma Mass Spectrometry," in Methods for the Determination of
Organic Compounds in Drinking Water, Supplement I, EPA/600/R-94/111, 1994.
7. U.S. EPA Method 180.1, "Turbidity (Nephelometric)," Methods for the Determination of
Inorganic Substances in Environmental Samples, EPA/600/R-93/100, 1993.
8. American Public Health Association, et al. Standard Methods for the Examination of Water
and Waste-water. 19th Edition, 1997. Washington, DC.
9. U.S. EPA, Methods for Chemical Analysis of Water and Wastes, EPA/600/4-79/020.
10. U.S. EPA Method 524.2, "Purgeable Organic Compounds by Capillary Column GC/Mass
Spectrometry," Methods for the Determination of Organic Compounds in Drinking
Water—Supplement III, EPA/600/R-95/131.
35
-------
11. U.S. EPA Method 552.2, "Haloacetic Acids and Dalapon by Liquid-Liquid Extraction,
Derivatization and GC with Electron Capture Detector," Methods for the Determination of
Organic Compounds in Drinking Water—Supplement IIIEPA/600/R-95/131.
12. Quality Management Plan (QMP)for the ETV Advanced Monitoring Systems Center,
Version 4.0, U.S. EPA Environmental Technology Verification Program, Battelle, Columbus,
Ohio, December 2002.
36
------- |