v>EPA
United States
Environmental Protection
Agency
Office of Solid Waste and
Emergency Response
Washington DC 20460
EPA540 R-93.518
June 1993
Superfund Innovative
Technology Evaluation (SITE)
Program Evaluation
Report for ANTOX BTX
Water Screen (BTX
IMMUNOASSAY)
SUPERFUND INNOVATIVE
TECHNOLOGY EVALUATION
-------
-------
NOTICE
The information in this document has been funded by the
U.S. Environmental Protection Agency under Contract Number
68-CO-0049 to Lockheed Engineering & Sciences Company. It has
been subject to the Agency's peer and administrative review, and
it has been approved for publication as an EPA document. Mention
of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
The results of a demonstration of a portable immunoassay for the detection of benzene, toluene,
and xylene(s) (BTX) are described in this report. The BTX immunoassay was developed by Antox,
Inc., (South Portland, ME) and is intended as a screening technology, with a claimed detection level
of 25 ppb BTX. This demonstration was conducted under the Superfund Innovative Technology
Evaluation (SITE) Program, under which measurement and monitoring technologies of interest to
the U.S. Environmental Protection Agency are evaluated.
The demonstration was designed to investigate the ability of the immunoassay to perform as a
portable, on-site screening method for BTX contaminated ground water samples. Samples from
monitoring well fields at four sites in the Las Vegas valley provided a range of concentration levels
for gasoline contaminated ground water. Sample splits were analyzed on-site by the BTX
immunoassay and in the laboratory by gas chromatography (GC). The confirmatory GC analysis
followed the procedures in EPA Method 8020.
The BTX immunoassay was rapid and simple to use, and it performed well in identifying high
level contamination and gasoline contaminated samples having BTX concentrations greater than
100 ppb. The results from duplicate immunoassay analysis of 79 field samples were consistent
with GC results for most concentrations. Only two samples were found to generate consistent
false negative immunoassay results (each sample had BTX concentrations between 25 and 100
ppb). Two samples also gave consistent false positive immunoassay results. However, all false
positive immunoassay results were on samples having distinct low-level gasoline signatures in their
chromatographs. Both field sample and various performance evaluation sample results showed a
significant tendency to generate false negative results in the 25 to 100 ppb BTX region. A more
sensitive formulation of the immunoassay may need to be developed if the immunoassay is to be
used for critical decisions involving samples with these levels of BTX.
This evaluation is submitted in partial fulfillment of Contract Number 68-CO-0049 by Lockheed
Engineering & Sciences Company under the sponsorship of the EPA.
in
-------
TABLE OF CONTENTS
Section page
Notice "
Abstract i"
List of Figures vi
List of Tables vi
Abbreviations and Acronyms vii
Acknowledgments viii
Executive Summary ix
1. Introduction 1
Overview of the Immunoassay Program 1
Overview of the SITE Program 3
Overview of the BTX SITE Demonstration 3
Project Organization 4
2. Descriptions of Technologies 6
BTX Immunoassay 6
Gas Chromatography Analysis 6
Gas Chromatography/Mass Spectrometry Analysis B
3. BTX Immunoassay Demonstration Design 10
Predemonstration Testing and Planning 10
Cross Reactivity 10
Colorimeter Calibration 10
Variability of Optical Density Readings 11
Study Design for the SITE Demonstration and Evaluation 13
Site Selection 13
Sample Collection Procedures 13
Quality Assurance Design 14
Data Quality Objectives 14
QA Problems and Resolutions 15
Changes to the QA Plan 15
4. BTX Demonstration Results and Discussion 17
Gas Chromatography Results 17
Field BTX Immunoassay Versus GC 17
iv
-------
False Negatives 20
False Positives 20
Variability of BTX Immunoassay Field Analysis Results 20
Quality Assurance and Quality Control Sample Results 22
An Assessment in Terms of Five Data Quality Elements 23
Results of the On-Site Audit 27
Additional QA/QC Observations and Conclusions 27
5. Conclusions and Recommendations 29
Conclusions 29
Advantages of the BTX Immunoassay 29
Limitations of the BTX Immunoassay 29
Minimizing False Negative and False Positive Results 30
Limitations of this Demonstration 30
Recommendations 31
Recommendations Specific to the BTX Demonstration 31
Recommendations for Future Studies 31
References 32
Appendices
Appendix A - A Description of the BTX Immunoassay Procedure 34
Appendix B — Procedural Details and Quality Assurance Summary for
Gas Chromatography and Purge-and-Trap Methods 41
Appendix C ~ Modified Procedure for Preparation of Trip
Audit/Performance Audit Samples 63
Appendix D -- Field Audit Report 66
Appendix E - Participating Personnel 70
Appendix F - Data Tables for Field Immunoassay, Laboratory
Immunoassay, and Gas Chromatography Analyses 73
-------
LIST OF FIGURES
Figure
1 . Organizational structure for the SITE Program demonstration of the BTX immunoassay. ... 5
2. BTX immunoassay test kit and field test equipment ............................. 7
3. BTX immunoassay test procedure ......................................... 8
4. Colorimeter versus spectrophotometer comparison ............................ 12
5. Gas chromatographs for typical field samples ................................ 18
6. First replicate immunoassay S/R vs GC BTX concentration ....................... 19
7. Second replicate immunoassay S/R vs GC BTX concentration ..................... 19
8. Replicate vs original S/R ratio ........................................... 21
9. BTX immunoassay results for field performance evaluation samples ................. 26
10. Reference cuvette optical density vs time .................................. 28
LIST OF TABLES
Table page
1. BTX immunoassay cross-reactivity data 11
2. Cuvette variability results 13
3. GC results for field prepared QA samples 23
4. Laboratory BTX immunoassay results of performance evaluation samples 23
VI
-------
ABBREVIATIONS AND ACRONYMS
BTX benzene, toluene, xylene
BTEX benzene, toluene, ethylbenzene, xylene
CECS Converse Environmental Consultants Southwest
ELISA enzyme-linked immunosorbent assay
EMSL-LV Environmental Monitoring Systems Laboratory - Las Vegas
EPA Environmental Protection Agency
GC gas chromatography
GC/MS gas chromatography/mass spectrometry
HP Hewlett-Packard
HRP horseradish peroxidase
MCL maximum contaminant level
MCLG maximum contaminant level goal
MDL method detection limit
MMTP Monitoring and Measurement Technologies Program
MTBE methyl t-butyl ether
OD optical density
PEM personal exposure monitor
PIP photoionization detector
ppb parts per billion
ppm parts per million
QA quality assurance
QC quality control
SARA Superfund Amendments and Reauthorization Act
SITE Superfund Innovative Technology Evaluation Program
S/R sample to reference
TMB tetramethylbenzidine
VOA volatile organic analysis
UV/vis ultraviolet/visible
VII
-------
ACKNOWLEDGMENTS
We gratefully express our appreciation to John Worlund, Eric Noack, Dean Alford, and their
co-workers at Converse Environmental Consultants Southwest (Las Vegas, Nevada) for their
generosity in providing site access and sampling support during this study. Also, special thanks to
Pat Amick and Neil Amick (Lockheed) for analytical analysis, Gracie Martucci (Lockheed) for data
base management, John Curtis and John Zimmerman (Lockheed) for field sampling, and Vicki Ecker
and Neil Amick (Lockheed) for Quality Assurance input.
VIII
-------
EXECUTIVE SUMMARY
This evaluation report presents the results of a demonstration designed to assess the
performance of a field-portable immunoassay. The immunoassay was developed by Antox, Inc.
(South Portland, ME) for the analysis of benzene, toluene, and xylene(s) (BTX) in environmental
water samples. The demonstration was conducted under the Monitoring and Measurement
Technologies Program of the U.S. Environmental Protection Agency (EPA) Superfund Innovative
Technology Evaluation (SITE) Program. The demonstration was performed under the guidance of
the EPA Environmental Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-LV).
The demonstration took place between January and April, 1992 at four sites in the Las Vegas
valley. Each site contained a well field established to monitor groundwater contaminated with
gasoline. Groundwater contamination from leaking underground storage tanks, pipelines, or other
sources is a significant problem of EPA's interest. The EPA and various state and local agencies
are interested in preventing, detecting, and correcting problems of gasoline contamination in ground
water.
The BTX immunoassay demonstration involved the assay of 79 field samples. These samples
were analyzed on-site in duplicate, and sample splits were analyzed in the laboratory using gas
chromatography (GO as the confirmatory method. GC analysis following EPA Method 8020
produced quantitative estimates for benzene, toluene, and xylene. Quality control of the GC
analysis was maintained by following the requirements and meeting the specifications in the
method description. In addition, a variety of quality control and performance evaluation samples
were included in the study design. Selected samples from each site were also analyzed by gas
chromatography/mass spectrometry (GC/MS) to evaluate sample components unidentified by the
GC analysis.
Immunoassays are analytical techniques based on protein molecules (antibodies). The binding
of a specific antibody to its target analyte can be used to quantitatively or qualitatively determine
the extent of contamination in environmental samples. Specific antibodies can be developed to
detect single analytes or groups of compounds. The Antox BTX Immunoassay is a competitive
enzyme-linked immunosorbent assay (ELISA), and is meant to provide a qualitative estimate of the
concentration of several small aromatic compounds in environmental water samples.
The test is rapid and easy to use. It takes thirty minutes to analyze a group of up to four
samples. The test is performed by placing reference and test solutions in separate antibody-coated
polystyrene cuvettes in which competition for binding to the antibody occurs. After a 10 minute
incubation step, the cuvettes are washed to remove unbound enzyme conjugate. Substrate and
chromogen are then added to the cuvettes, and a colored reaction product is formed during a
second incubation period. The enzymatic reaction is stopped by the addition of a few drops of 1 N
sulfuric acid. The color intensity of the enzymatic product is inversely proportional to the BTX
hydrocarbon level in the test sample. The optical density (OD) of the reference and sample
cuvettes is read on a portable digital colorimeter, and the ratio of sample to reference OD values is
used to estimate whether the aromatic hydrocarbon levels are above or below 25 ppb BTX. Ratios
IX
-------
below 0.85 are associated with high BTX concentrations and ratios above 0.85 are associated with
low BTX concentrations.
Results from the GC analysis gave 36 samples with BTX concentrations above 25 ppb BTX
and 43 samples with concentrations below 25 ppb BTX. Using the immunoassay, two samples in
the range from 25 to 100 ppb BTX were consistently misidentified as false negatives and two
samples with concentrations below 25 ppb were consistently founc to be false positives.
Interestingly, all false positive immunoassay results were associated with samples having low-level
gasoline contamination.
An evaluation of sample to reference (S/R) OD ratios for repeated analysis of performance
evaluation samples having concentrations of 2.5, 25, 40, and 100 ppb BTX showed that the BTX
immunoassay performed relatively poorly in the region from 25 to 100 ppb BTX. Most of the 25
and 40 ppb performance samples were classified as below 25 ppb by the immunoassay. The
variability of the S/R OD ratio in this region leads one to expect a relatively high number of false
negatives in this concentration range. GC analysis of sample splits from these performance
samples confirmed that they were correctly prepared and did not represent loss due to improper
dilution or volatilization.
The BTX immunoassay was simple to perform, and, for the most part, provided results
consistent with those from later GC analysis. If critical decisions are dependent on analysis of
samples in the range from 25 to 100 ppb BTX, then a more sensitive formulation of the test should
be pursued. However, for certain applications, such as mapping the distribution of contaminants at
a site or monitoring changes in concentration over time, the current level of sensitivity would be
adequate.
-------
SECTION 1
INTRODUCTION
The performance of an immunoassay field kit technique was evaluated during a U.S.
Environmental Protection Agency (EPA) Superfund Innovative Technology Evaluation (SITE)
demonstration. This semi-quantitative method was developed by Antox Inc. (South Portland, ME),
to detect benzene, toluene, and xylene contaminant levels in environmental water samples. The
evaluation compared results from BTX immunoassays performed in the field to quantitative results
obtained by gas chromatographic analysis using EPA Method 8020 (U.S. EPA, 1986a).
There are several million underground storage tanks (USTs) in the United States that contain
petroleum or hazardous chemicals. The EPA has estimated that approximately 300,000 suspected
releases will be identified at UST facilities over the next five years. Leaking USTs can cause fires
or explosions and can contaminate ground water, which is a primary source of drinking water. In
the interest of public health and safety, the EPA and various state and local agencies have issued
regulations aimed at preventing, detecting, and correcting problems associated with leaking USTs
(U.S. EPA, 1990a). The EPA has set regulatory standards for toluene, xylene, and benzene under
the Safe Drinking Water Act (U.S. EPA, 1990b). For toluene and xylene, the proposed National
Primary Drinking Water maximum contaminant levels (MCL) are 1 and 10 parts per million (ppm)
respectively, and the proposed National Secondary Drinking Water maximum contaminant level
goals (MCLG) are 0.04 and 0.02 ppm, respectively. The National Primary Drinking Water MCL for
benzene is 0.005 ppm and the MCLG is 0 ppm.
Many government agencies require that laboratory analyses be performed on soil and water
samples collected during tank closures, from suspected leaking UST sites, or to measure progress
of corrective action. The cost and delays in decision making associated with laboratory analyses
has prompted a search for field monitoring methods that are less expensive and are capable of
providing rapid, on-site results. In many situations, real-time results facilitate better protection of
public health through quicker corrective action decisions (Midwest Research Institute, 1990). An
immunoassay test for BTX has the potential for use as a quick and inexpensive screening method
for the detection of gasoline components in ground and surface water.
This document includes descriptions of the BTX immunoassay method and the theory behind
it (Section 2), the design of the SITE demonstration and QA plans (Section 3), and an evaluation of
the results compared to a standard EPA method (Section 4). Conclusions and recommendations for
the BTX immunoassay are discussed in Section 5.
OVERVIEW OF THE IMMUNOASSAY PROGRAM
The immunochemistry program at the Environmental Monitoring Systems Laboratory, Las
Vegas (EMSL-LV) consists of the following major components: identification of need for an
immunochemical method, identification of existing technologies, development of new technologies,
-------
adaptations and evaluations of existing technologies, field demonstration of portable technologies,
and finally, technology transfer.
Immunoassays are increasingly being recognized as cost-effective and expedient alternatives
to chromatographic and spectroscopic procedures for use in large-scale environmental monitoring
studies. Immunoassay techniques have been applied to the analysis of many hazardous substances
and possess several attributes that make them suitable for field screening methods. In general,
immunoassays have proven to be sensitive, selective, precise, rapid, cost-effective, and applicable
to a wide range of contaminants. A variety of immunoassay methods and formats can be applied
to environmental analysis problems.
Once an immunoassay is presented to the EPA it may undergo an evaluation. Evaluation
studies are guided by xns parameters of the individual immunoassay system. If an assay is not well
developed or characterized by the developer, the EPA may undertake a preliminary evaluation.
However, the assay must be for a compound of high priority to the EPA and there" must be a well-
defined need for the method. Results from the preliminary studies are made known to the
developer. The developer can use this information to base further assay development to eventually
provide an assay suitable for EPA's use. Assays that are mature are evaluated in more detail than
those assays still in the developmental stage. If a field-portable immunoassay system has
successfully undergone a laboratory evaluation, the technology is demonstrated in the field. Such
demonstrations are conducted under the SITE provisions of the Superfund Amendment and
Reauthorization Act of 1986 (SARA).
The majority of the immunochemistry program at the EMSL-LV is currently directed towards
the development and evaluation of specific immunoassay technologies. However, as needs have
been identified, the program is expanding into other areas, including: immunoaffinity personal
exposure monitors (PEMs), immunoaffinity chromatography sample preparations, fiber optic
immunosensors (biosensors), immunoassays for biological indicators (biomarkers) of exposure, and
the more effective development of specific antibodies.
The overall goal of the immunochemistry program is to provide quality assured methods to
augment the agency's monitoring capabilities. The objectives of the program are listed below:
• Provide standardized immunoassays with proper quality assurance
and standard operating procedures.
• Identify existing technologies that can be applied off-the-shelf or
with minimal modification.
• Identify monitoring needs that cannot be adequately addressed with current
technologies; and initiate a methods development program.
• Convert existing technologies into usable monitoring and site charac-
terization methods.
• Demonstrate those technologies that successfully emerge from the
development phases of the program.
• Prepare standard operating protocols and guidelines for the imple-
mentation of methods.
-------
• Transfer the new technology to the user community and encourage
the commercialization of resulting products.
• Foster the development of a respected Center of Excellence for
Immunochemistry - both within and outside of the EPA.
The EPA Environmental Monitoring Systems Laboratory at Las Vegas, Nevada (EMSL-LV), is
responsible for developing and evaluating immunoassays for specific environmental applications.
According to EPA guidelines for methods evaluations, this process requires the determination of
performance parameters such as precision, between method bias, method detection limits,
interferences, and ruggedness of the method.
To be effective as rapid screening tools, immunoassays must provide timely, cost-effective
results that complement conventional analytical methods. They must be capable of measuring the
target analyte with sufficient accuracy and precision to identify that samples are clearly above or
below a critical concentration range.
OVERVIEW OF THE SITE PROGRAM
Under SARA, site monitoring and characterization activities are expected to increase. As a
direct result, the costs associated with collecting, transporting, and analyzing environmental
samples will escalate. Field analysis can significantly reduce the time and cost associated with
sample collection and analysis by providing timely information about the presence or absence of
environmental .pollutants. The EPA established the SITE program to accelerate the development,
demonstration, and use of alternative and innovative technologies at Superfund sites. The
Monitoring and Measurement Technologies Program (MMTP) was created as a complementary
component of the SITE program to provide rapid, low-cost field methods to support hazardous
waste site monitoring and characterization activities. An overview of the MMTP is given by Koglin
(1990).
The SITE program addresses two broad categories of innovative technologies: (1) monitoring
and measurement technologies, rugged enough for use under field conditions, for analysis of
environmental pollutants of interest in site characterization; and (2) alternative treatment technolo-
gies used in remediation at Superfund sites. The three main objectives of the MMTP element of
the SITE program are to select appropriate technologies for inclusion in the program, to provide the
mechanisms for demonstrating each selected technology, and to ensure that the demonstrated
technology is evaluated properly.
OVERVIEW OF THE BTX SITE DEMONSTRATION
The main purpose of this demonstration was to evaluate the BTX immunoassay technology
for on-site detection of benzene, toluene, and xylene (as a group) under field conditions. Attention
was focused on the BTX immunoassay's performance under ambient environmental conditions, but
laboratory tests were also a significant component of the study. This demonstration was designed
to assess whether the technology is capable of yielding rapid, accurate, and cost-effective analysis
of the above pollutants in aqueous samples.
The demonstration took place from January to April 1992, at four sites in the Las Vegas
valley, NV. Each of these sites was known to have ground water contaminated with one or more
hydrocarbon fuels, usually gasoline. Each site also had an existing field of sampling wells, which
-------
provided an opportunity to generate numerous independent samples having different concentrations
of contaminants. Some of the wells were expected to monitor ground water outside the contami-
nant plume, which provided samples with contaminant concentrations well below the detection
level of the BTX immunoassay. Thus, the sample sites represented a variety of sample sources.
Since the laboratory performing the comparative analysis was also within the Las Vegas valley,
effects due to sample handling and shipping were minimal. And finally, logistical costs were
minimized with this design.
PROJECT ORGANIZATION
The success of the immunoassay evaluation depended on the combined efforts of three
primary organizations: EMSL-LV, Antox Inc., and Converse Environmental Consultants Southwest
(CECS). The organizational structure for this study is shown in Figure 1.
The responsibilities of EMSL-LV, with assistance from its prime contractor, Lockheed, include:
• Designing, overseeing, and ensuring the implementation of the
elements of the demonstration and quality assurance (QA) plan.
• Performing on-site BTX immunoassay analysis.
• Performing off-site analysis by GC.
• For selected samples, performing off-site analysis with the BTX
immunoassay and GC/MS.
• Preparing and distributing quality assurance (QA) and quality control (QC) samples.
• Evaluating and reporting on the performance of the technologies.
Antox Inc., the developer of the immunoassay, was responsible for:
• Performing preliminary testing to assess immunoassay performance.
• Supplying a sufficient number of field kits to perform the analysis required to
conduct the demonstration.
• Providing technical assistance.
CECS was responsible for:
• Site access and for providing sample splits from each well.
-------
SUPERFUND INNOVATIVE TECHNOLOGY EVALUATION
DEMONSTRATION PROGRAM
MONITORING AND MEASUREMENTS TECHNOLOGIES
PROGRAM
• Environmental Monitoring Systems
Laboratory - Las Vegas
• Matrix Manager - E. N. Koglin
• Lockheed
- SITE Projects Coordinator - N. F. D. 0' Leary
BTX IMMUNOASSAY DEMONSTRATION
• Environmental Monitoring Systems
Laboratory - Las Vegas
- Project Manager - J. M. Van Emon
• Lockheed
- Project Technical Lead - R. W. Gerlach
- Project Analytical Lead - R. J. White
• Antox Inc.
- President - R. Piasio
- Research Scientist - K. Prouty
• Converse Environmental Consultants Southwest
- Principal Environmental
Engineer - J. Worlund
Figure 1. Organizational structure for the SITE Program demonstration of the BTX immunoassay.
5
-------
SECTION 2
DESCRIPTIONS OF TECHNOLOGIES
A brief description of the immunoassay technology and of the confirmatory laboratory
analysis methods is presented in this section. Figure 2 shows the complete BTX immunoassay kit
supplied by Antox and various ancillary equipment needed to utilize the kit. The kit is a portable,
self-contained package containing all reagents and supplies required for field use. Confirmatory GC
and GC/MS laboratory analysis were traditional laboratory techniques, using approved EPA
laboratory methods (EPA Methods 8020 and 8240, U.S. EPA, 1986a,b).
BTX IMMUNOASSAY
The BTX immunoassay is a competitive enzyme-linked immunosorbent assay (ELISA), which
uses an antibody-coated polystyrene cuvette as the solid phase. The immunoassay uses horserad-
ish peroxidase (HRP) as the enzyme marker, hydrogen peroxide as the substrate, and tetramethyl-
benzidine (TMB) as a chromogen. The rabbit polyclonal antibody binds specifically to the BTX or
other closely associated aromatic hydrocarbons. A hapten-enzyme conjugate mimics the free
analyte (BTX compounds) and competes with aromatic compounds in the sample for binding sites
on the immobilized antibody.
The test is performed by placing reference and test solutions in separate coated cuvettes in
which competition for binding to the antibody occurs. Figure 3 is a graphical presentation of the
individual steps involved in the test. After a 10 minute incubation step, the cuvettes are washed to
remove unbound enzyme conjugate. Substrate and chromogen are then added to the cuvettes and
a colored enzymatic reaction product is formed during a second incubation period. The enzymatic
reaction is stopped by the addition of a few drops of 1 N sulfuric acid. A complete description of
the method is given in Appendix A.
The color intensity of the enzymatic product is inversely proportional to the BTX hydrocarbon
level in the test sample. If the level of BTX hydrocarbon in the sample is high, most of the
antibody binding sites will be occupied with the sample analyte, thus preventing the hapten-enzyme
conjugate from binding. For less concentrated samples, more of the antibody binding sites are
occupied with the hapten-enzyme conjugate, resulting in the development of more intense color in
the cuvette. The optical density (OD) of the colored enzymatic product in the reference and sample
cuvettes is read on a portable digital colorimeter equipped with a filter which will pass light at 450
nm (near the wavelength of peak absorption for the colored reaction product). The ratio of sample
to reference OD values is used to estimate whether the aromatic hydrocarbon level in the sample is
above or below 25 parts per billion (ppb).
GAS CHROMATOGRAPHY ANALYSIS
Gas chromatography analysis following the general guidelines of EPA Method 8020 (U.S.
EPA, 1986a) was used as the primary confirmatory method. One modification was the use of a 30
-------
-------
ANTOX BTX ASSAY PRINCIPLE
Reference
Hibe
Sample
Tiibe
Step 1
Oeionized
Water
(no analyte)
Antibody./ -<
Coating
4-*A>
BTX-Analog
Enzyme
Conjugate
Step 2
Incubate 10 minutes
Wash 4 Times
(delonlzed water)
Step 3
Substrate/
Chromogen
o o o o o
o o
Step 4
Incubate 5 Minutes-
Step 5
Add 4 Drops of
Stop Solution
Step 6
Colorimeter
Read Absorbance
at 450 nm
Step 7
Calculate Sample/Reference
Absorbance Ratio
Figure 3. BTX immunoassay test procedure.
8
-------
m J&W DB-624 chromatography column. Purge-and-trap sample preparation followed EPA Method
5030 (U.S. EPA, 1986c). Details of the gas chromatography and purge-and-trap procedures are
given in Appendix B.
GAS CHROMATOGRAPHY/MASS SPECTROMETRY ANALYSIS
Selected samples from each site were subjected to GC/MS analysis following the general
procedure given in EPA Method 8240 (U.S. EPA, 1986b). These analyses were carried out to help
identify unassigned peaks from the GC analysis. Cross-reacting compounds may cause high
immunoassay results compared to what is expected based on GC detection for only benzene,
toluene, and xylene(s).
-------
SECTION 3
BTX IMMUNOASSAY DEMONSTRATION DESIGN
PREDEMONSTRATION TESTING AND PLANNING
Preliminary testing of the BTX immunoassay was conducted prior to starting the field
sampling. This included testing a variety of compounds and mixtures for cross-reactivity, checking
the performance of the colorimeter, and determining the variability associated with reading the
optical density using different cuvettes.
Cross Reactivity
Table 1 summarizes the cross reactivity data of the BTX immunoassay. Compounds are listed
in descending order by strength of cross-reactivity. Percentage cross-reactivity is calculated
relative to regular gasoline (leaded, from pump). These percentages are intended to give a rough
estimate of the relative strength of cross-reactivity. Since these values are calculated from a single
point and based on the simplifying assumptions of approximate linearity and parallelism of the
standard curves, they should be considered only as rough estimates. When S/R is greater'than
1.0, the above assumptions are not valid, and the cross-reactivity is assumed to be negligible.
Note that the antibody interacts most strongly with BTX hydrocarbons and closely related
compounds which are substituted with nitro-, chloro-, and hydroxyl groups. It is also notable that
the antibody tends to cross-react moderately with some chlorinated alkanes and alkenes (e.g., 1,4-
dichlorobutane, c/s-1,3-dichloropropene). The antibody does not cross-react appreciably with most
of the alkanes. The difference in cross-reactivity between regular (leaded) gasoline, white gasoline,
and kerosene may reflect the relative percentages of BTX compounds in each.
Colorimeter Calibration
Another predemonstration activity involved investigation of bias associated with the portable
colorimeter. A set of serially diluted samples of N-2,4-dinitrophenylglycine in water were prepared
and read in both the colorimeter and a Hewlett-Packard Mode; S452A diode array UV/vis spectro-
photometer, at 450 nm. Both instruments were zero-adjusted with cuvettes containing deionized
water. The data, plotted in Figure 4, indicate low bias with respect to the spectrophotometer at
high OD values. The non-uniform bias observed in Rgure 4 might introduce systematic error in
estimating the Sample OD/Reference 00 ratio.
There will only be a small region where this bias might effect the classification of samples.
Samples will be biased toward misclassification if the reference OD is high (near 2.0) and the
sample OD is slightly smaller. In all other cases the bias will not affect the classification or it will
enhance the probability of correct classification. If the bias at high OD readings (a bias of 0.73) is
compared to the OD bias at 0.85 of a high OD reading (a bias of Q.79), we can estimate a modified
S/R OD ratio. The modified ratio represents an alternate decision level for the BTX immunoassay.
10
-------
Table 1. BTX immunoassay cross-reactivity data*.
Compound(s) S/Rb ratio Percent cross-reactivity6
BTEXd(1: 1:1:1 by volume)
BTX"(1:1:1 by volume)
o-Cresol (2-Methylphenol)
Regular gasoline*
Chlorobenzene
Nitrobenzene
1 ,2-Dichlorobenzene
Phenol
White gasoline
1 ,4-Dichlorobutane
c/s-1 ,3-Dichloropropene
2,4-Dinitrotoluene
1 ,2-Dichloroethane
Iso-octane
Ethylene glycol
Pentane
trans-1 ,2-Dichloroethylene
1 , 1 -Dichloroethane
1 , 1 ,2-Trichloromethane
Kerosene
c/s-1 ,2-Dichloroethylene
Methylene chloride
0.18
0.25
0.27
0.31
0.34
0.43
0.43
0.52
0.53
0.72
0.77
0.79
1.03
1.04
1.06
1.06
1.11
1.12
1.12
1.12
1.18
1.26
172
124
115
100
91
72
72
60
58
43
40
41
0
0
0
0
0
0
0
0
0
0
' All cross-reactivity results were determined at 100 ppm.
b S/R = Sample absorbance/Reference absorbance.
0 Percentage cross-reactivity = 100 (0.31/compound S/R ratio). The cross-reactivity is assumed to
be 0 when S/R is 1.0 or greater.
d Benzene/toluene/ethylbenzene/xylene.
' Benzene/toluene/xylene.
' Regular (leaded gas = 100%).
When the S/R ratio is 0.85, the bias of the S/R ratio is 0.73 / 0.79 = 0.92, from which an
alternative decision level of 0.85 X 0.92 = 0.78 can be calculated. The results section will show
that this change in the decision level will only slightly affect the field sample analyses results.
Variability of Optical Density Readings
The background variability associated with different cuvettes was determined by measuring
the OD for three sets of 10 cuvettes. One set was left empty, one set contained distilled water,
and one set contained a solution of N-2,4-dinitrophenylglycine. The volume of solution was equal
to that expected when using the standard immunoassay procedure. A summary of results is given
in Table 2. The standard deviations are small, ranging from 0.004 to 0.014 OD units. All
colorimeter readings were repeated after re-orienting the cuvettes in the reader slots. No practical
difference in mean or standard deviation was observed compared to the initial readings. However,
it was observed that repeated use of a plastic cuvette with the colorimeter might lead to the
11
-------
Colorimeter Calibration vs HP Spectrometer
0>
o
(0
.0
o
0)
.0
0)
o
o
o
Line of Perfect
Agreement
1.0 2.0
HP Model 8452 UV/vis Spectrometer (Absorbance)
Figure 4. Colorimeter versus spectrophotometer comparison.
12
-------
Table 2. Cuvette variability results.
Solution
Empty
Deionized water
N-2>4-dinitrophenylglycine
n
10
10
10
Mean
(OD units)
0.044
-0.001
0.68
Standard Deviation
(OD units)
0.004
0.007
0.014
development of scratches, which could affect the OD reading. This suggests that one should
periodically discard cuvettes being repeatedly used to zero the colorimeter. To obtain the best
performance, one should always inspect cuvettes for scratches or other blemishes before use.
STUDY DESIGN FOR THE SITE DEMONSTRATION AND EVALUATION
The primary focus of the study design was the evaluation of the accuracy, precision, and
other key performance characteristics of the BTX immunoassay, under field conditions. Samples
from a variety of locations around hydrocarbon plumes contaminating groundwater were included
to provide representative data over a range of concentrations.
Site Selection
For this study an arrangement was negotiated with Converse Environmental Consultants
Southwest (CECS) for sample acquisition from four different well fields located within the Las
Vegas valley. Three of these sites represented gasoline sources and the fourth site had wells that
might also be contaminated with other hydrocarbon mixtures, such as jet or diesel fuel. Each of
these sites had well monitoring fields already in place, with several wells having no or very low
levels of hydrocarbons present. The exact location of the sample wells is confidential due to an
agreement with CECS, which controls access to the sites and represents the interests of site
owners. The site owners are reluctant, for various reasons, to have their names divulged.
The four sites have been coded A, B, C, and D. Eleven wells were sampled at site A at three
different times, during January 22-23, February 24-25, and April 15-16. Site B had 18 wells
sampled on February 25-28. Site C, with 16 wells sampled, was visited on March 9-13. Site D
had 12 wells sampled on April 17-22. A total of 79 well samples were obtained.
Sample Collection Procedures
All field samples were collected from pre-existing wells that had been previously developed to
remove non-native materials from the well bore and penetrated aquifer in order to ensure that
representative ground-water samples were obtained. Purging at sites A, B, and D involved the use
of a purging pump. However, due to the 30-40 m depth to the water table at site C, dedicated
sampling pumps were used for both purging and acquiring samples. Each well was purged of three
well volumes of water prior to sampling to remove stagnant water from the well and surrounding
aquifer. After purging, samples were taken by placing a clean, disposable 5.1-cm diameter teflon
bailer slowly down the well. After slowly bringing the bailer to the surface, a clean Voss vial
13
-------
pipette was placed on the bottom of the bailer. Clear 40-mL VOA (Volatile Organic Analysis) vials
with teflon septums were filled from the bottom after triple rinsing each vial with formation water.
VOA vial splits from a given well were analyzed on-site, or held at 4°C until analyzed at the
laboratory- To maintain a record of sample collection, transfer, shipment, and receipt; a chain-of-
custody form was filled out for each day's collection of samples. Sample tracking was accom-
plished by recording the sample site, well number, and date of sampling.
QUALITY ASSURANCE DESIGN
To ensure that the data resulting from the SITE demonstration are of known quality and that
a sufficient number of critical measurements are taken, a quality assurance project plan (QAPjP)
was prepared (Gerlach et al., 1992). The QAPjP for this demonstration describes procedures
appropriate for a Category il project, as specified by EPA (U.S. EPA, 1987). Category II projects
apply to measurement activities when the results are not directly used in regulatory, enforcement,
legal, or policy matters, where data quality objectives may be semi-quantitative, and for which
precision, accuracy, completeness, reproducibility, and representativeness may not be easily
defined.
Data Quality Objectives
In order to make sound decisions about the BTX immunoassay field kit, the conclusions of the
demonstration must be based on sound interpretations of the resulting data. To achieve this end,
objectives were established for data quality based on the proposed end uses of the data (Stanley
and Verner, 1985). The QA component of the demonstration was designed to maximize the
probability that the resulting data would be suitable for its intended use. EPA has established data
quality objectives (DQOs) for five indicators of data quality (Stanley and Verner, 1983): represent-
ativeness, completeness, comparability, bias, and precision. These data quality objectives are
provided as guidelines. If the objectives were not met, it does not necessarily mean that the data
would not meet the needs of the project.
Three analytical methods were used in this study: BTX immunoassay, GC, and GC/MS. The
BTX immunoassay is the focus of the evaluation, and in this sense the entire project is a study of
the performance characteristics of this method. Consideration of whether the BTX immunoassay
performed well is based on a composite evaluation of all aspects of the study, including ease of
use, evaluation of the protocol, performance with respect to audit (see Gerlach et al., 1992 and
Section 4) and field samples, etc. On a more basic level, the study design required at least 60
environmental well samples to be analyzed on-site by the immunoassay. The study design also
specified that at least 90 percent of the analysis should be free of identifiable errors associated
with the immunoassay itself (for example, cuvettes which fail the manufacturer's QA/QC criteria).
Several types of quality assurance and quality control samples were used in this demonstra-
tion. Trip blanks and 40 and 100 ppb BTX trip audit samples were prepared in the field and
transported with the field samples to check for contamination or loss from transportation or storage
effects. Individual VOA vial splits of these samples were analyzed in the laboratory by GC and
immunoassay. Performance audit samples were prepared at 2.5, 25, and 100 ppb BTX to aid in
monitoring both immunoassay and GC results. These samples were prepared immediately prior to
analysis. In the field, one set was run at the beginning and another set was prepared and run at
the end of the day. Separate sets of performance audit samples were also prepared and run at the
time of GC analysis and for laboratory immunoassay analysis. The above audit samples served to
check and confirm the analysis results reported for field samples by both immunoassay and GC.
14
-------
The most important DQOs are associated with the GC analysis, as this is the confirmatory
method against which the BTX immunoassay results are compared. GC analysis was run in
accordance with the requirements spelled out in EPA Method 8020 (U.S. EPA, 1986a). A standard
curve was initially developed for each of four analvtes; benzene, ethylbenzene, toluene, and
o-xylene. Each standard curve was verified daily by injection of one or more working standards.
The responses were checked to verify that they did not vary by more than ±15 percent. If so, a
new calibration curve was established. Method 8020 states that it will generate estimates that
represent greater than 90 percent recovery with relative inter laboratory standard deviations from
18 to 26 percent. These bias and variability estimates were considered adequate to evaluate the
performance of the BTX immunoassay.
The GC/MS analysis was used primarily to identify GC peaks which might correspond to
compounds that cross-react in the immunoassay. At least one positive sample (by immunoassay)
from each of the sites (A, B, C, and D) was assayed by GC/MS. The samples were run following
the protocol and requirements in EPA Method 8240 (U.S. EPA, 1986b).
Data quality objectives were established for the five usual data quality indicators; representa-
tiveness, completeness, comparability, bias, and precision, as well as for detection limit. These
objectives are discussed in the relevant sub-sections of Section 4. Because their assessment
required an analysis of both field and QA/QC sample data, it was not practical to generate an
independent section of this report containing all DQO results. As has been stated previously, any
study whose main function is a method evaluation is nothing more than an exercise devoted to the
evaluation of quality assurance and quality control samples. In this sense field samples are merely
another type of QA/QC sample.
QA Problems and Resolutions
The procedure for generating trip and performance audit samples outlined in the demonstra-
tion plan had some deficiencies. Due to the volatility of the target analvtes, there was no way of
determining how close the actual final concentrations were to the target values. In addition, some
of the immunoassay results on splits of audit samples assayed directly after preparation and after
several hours, indicated possible changes with time. For this reason, the procedure for preparing
and analyzing the audit samples was modified to allow one of the splits of each audit sample to be
analyzed by GC. In addition, a holding time study was conducted to check for changes to analyte
concentration over time for audit samples stored in VOA vials in coolers. The section following
entitled "Changes to the QA Plan" gives further details on these modifications.
Changes to the QA Plan
The procedure for generating trip audit samples and performance audit samples was modified
between the February and March sampling events. Site A had been sampled once in January, and
sites A and B had both been sampled in late February. There were several reasons for modifying
the audit sample procedure. The BTX immunoassay results from the original protocol for the
performance audit suggested there might be a time lag associated with generating the correct
performance audit sample concentration. That is, the low level performance audit samples might
not reflect an accurate concentration until after some period of time necessary to achieve
equilibrium. In addition, there was concern about the comparability of field and laboratory results
for performance evaluation samples because the initial plan involved preparing separate sets of
standards for each location. With the original design, site-specific sample preparation effects might
15
-------
be the source of bias between field and laboratory results. And finally, the field crew was having
to spend too much of their time with the preparation of audit samples.
The plan was modified to have the trip audits made from an aliquot of the same solutions
prepared for the performance audits. This meant that there were three positive trip audits; at 2.5,
25, and 100 ppb prepared from the toluene standard. Since these samples served as both trip
audit and performance audit samples, they were assigned a new sample code, W1 for the 2.5 ppb,
W2 for the 25 ppb, and W3 for the 100 ppb. In addition, splits of these same solutions were used
as the performance audit samples analyzed at the beginning and end of each day. A detailed
description of the modified audit sample preparation protocol is given in Appendix C.
A second modification was made to carry out an additional laboratory holding time study in
order to determine if holding time had any effect on the BTX immunoassay results for performance
evaluation samples with concentrations in the 25 to 100 ppb BTX range. This study was initiated
because numerous incorrect assignments for the 25 ppb BTX samples were being made with the
BTX immunoassay at the beginning of the evaluation. There was concern that the sample
concentration was not stable at these levels. To address these concerns, several sample splits of
25 and 100 ppb toluene standards were prepared. These sample splits were run immediately after
preparation and at several other times on the day of preparation and on the following day. There
was no evidence of any change in response due to a change in sample composition. Instead, it
appears from examination of the complete data set for the demonstration that the problems with
the 25 ppb BTX standards resulted from the lower than expected sensitivity of the immunoassay.
16
-------
SECTION 4
BTX DEMONSTRATION RESULTS AND DISCUSSION
Several sets of data were developed as part of the evaluation. The primary data set consists
of duplicate field BTX immunoassay results for 79 groundwater samples and corresponding
laboratory GC results for splits of these samples. Several secondary data sets were generated from
different types of QA sample analyses. Sample types such as trip blanks, trip positives, and
performance evaluation standards were generated to evaluate the analytical methods and to
monitor procedures. A separate set of QA/QC data was generated for the GC method in order to
ensure the confirmatory method was in control. The results from each of these types of data sets
are presented and discussed in this section.
Gas Chromatooraphv Results
Representative chromatographs for three different samples are displayed in Figure 5. The
chromatograph on the bottom is for a typical high level sample, and was run at a five-fold dilution.
Peaks for benzene, toluene, and xylene are easily distinguished in the first half of the chromato-
graph. Two internal standards, c/s-1,2-dichloroethene and fluorobenzene, are also present on this
chromatograph. A characteristic fingerprint associated with gasoline contaminated samples is
present at the end of the run. The middle chromatograph is from an almost clean sample; barely
detectable peaks can be discerned. The chromatograph on the top is thought to be an example
from a weathered location. The compounds at the beginning of the chromatograph are barely
detectable, perhaps they have evaporated since the gasoline came in contact with the groundwa-
ter. However, the higher boiling point constituents in the fingerprint region are still present at
easily detectable levels. The BTX concentration estimates used in the following discussion are the
sum of the analytical results for benzene, toluene, and xylene(s) from the GC analysis. Ethyl-
benzene was also quantitated by GC, and was expected to cross-react to some extent in the assay.
However, it was found that adding the ethylbenzene component to the BTX value would not have
changed the classification of any sample as positive or negative relative to the 25 ppb decision
level. GC/MS results from selected field samples did not reveal any other compounds at levels
which would affect the classification of samples by the BTX immunoassay.
Field BTX Immunoassav Versus GC
A summary of the relationship between the field BTX immunoassay results and the GC results
is shown in figures 6 and 7, where the letters A, B, C, and D represent the sampling sites (see
Section 3). Figure 6 uses the initial field BTX immunoassay result, and Figure 7 uses the duplicate
field BTX immunoassay value. For samples where the OD reading exceeded the performance
criteria of the colorimeter (recorded as >2) the S/R ratio was arbitrarily set to 1.5 so that these
samples remained correctly classified but were distinguishable from the other sample results. In
both figures the horizontal line is at the Sample OD / Reference OD (S/R) decision level of 0.85 and
the vertical line is at 25 ppb BTX, the detection level claimed by the developer. One can see that
17
-------
UJ
8 III
dj 5
ill
*S
J g g
5
2 3
ti 2
a
o
CM
a
S
uj z
m -~
SAMPLE ID LGB0701
SAMPLED ON
FEBRUARY 26, 1992
SAMPLE ID LGA0401
SAMPLED ON
JANUARY 22. 1992
SAMPLE ID LGB0301
SAMPLED ON
FEBRUARY 26, 1992
5fe
if
\i
1 II
:
^1 . 1 fl n J 1 . .....
L uu<\A. JLAJ^aMu^-
Figure 5. Gas chromatographs for typical field samples.
18
-------
O.I
1.5-
1.0-
.85
0.5-
n. n-
A A AA
A
D R B
A
B
B C
B
' ">\
A V
R c
^ A A
^ C?cc
0.10 1 10 25 100 1000 10000 100000
Target BTX (ppb), log scale
Figure 6. First replicate immunoassay S/R vs GC BTX concentration.
1.5-
1.0-
CO
0.85
0.5-
0.0-J—r
3 ^
c c
e
db '
a
a c
0.10 1 10 25 100 1000 10000 100000
Target BTX (PPB), log scale
Figure 7. Second replicate immunoassay S/R vs GC BTX concentration.
19
-------
both plots show similar relationships, the sigmoidal shape mirrors a typical immunoassay calibration
curve. The data can be analyzed by considering the four quadrants separated by the 25 ppb BTX
and 0.85 S/R lines. Samples in the upper left and lower right quadrants were correctly determined
by the immunoassay with respect to the GC results. Samples in the upper right quadrant are false
negatives, and samples in the lower left quadrant are false positives.
False Negatives
Of the 79 field samples, 36 samples were determined to be above 25 ppb BTX components
by GC analysis. Two of these samples (6 percent) produced negative results on both replicate
analysis with the BTX immunoassay. One sample had a BTX concentration of 35.5 ppb and the
other a concentration of 74.8 ppb. If we include the ethylbenzene component, these estimates rise
to 36 and 80.6 ppb respectively.
False Positives
Based on the BTX estimates from the GC analysis, 43 samples had concentrations below the
25 ppb criteria of the immunoassay. Overall, there was a seven percent false positive rate for
these field samples (assuming replicates are independent), though only five percent of the positive
samples were consistently classified incorrectly. Two of the 43 samples (five percent) were
estimated as higher than 25 ppb by the initial field immunoassay determination, and four samples
(including the two in the first determination), a nine percent rate, were estimated as above 25 ppb
by the second immunoassay determination.
It is interesting to note that in each of the above six instances of a false positive by the BTX
immunoassay, an examination of the associated gas chromatograph showed the characteristic
fingerprint of gasoline contamination. An example of a chromatograph for these samples is given
in Figure 5 for sample LGB0701, sampled on February 26, 1992. In this sense, the BTX immuno-
assay appears to be robust with respect to identifying gasoline contaminated samples. One
example of a false positive occurred from a well that was sampled on three different months. The
false positive was for the second sampling date, and the GC BTX results for these three dates were
0.1, 3.6, and 36.5 ppb, respectively. In this case, the use of the BTX immunoassay as a screening
tool would have forwarded the second sample for further analysis and the identification of low level
gasoline contamination might have provided useful information for monitoring the changing water
quality of this well. However, because the contamination due to the gasoline fingerprint region is
at so low a level, cross-reactivity cannot explain the false positive result in this instance. Loss
during sample preparation would explain the discrepancy, but data from audit samples (see below)
do not support this theory. Nevertheless, for other samples it is possible that higher molecular
weight aromatic compounds (e.g. polyaromatic hydrocarbons or diethylbenzene) are causing a
combined cross-reaction signaling the presence of gasoline.
Variability of BTX Immunoassav Field Analysis Results
A qualitative evaluation of the measurement variability associated with the BTX immunoassay
can be obtained from the field replicate measurements. Figure 8 shows the replicate S/R measure-
ment plotted against the original S/R value. The vertical and horizontal lines represent the 0.85 S/R
decision level for the BTX immunoassay. Points in the lower left region represent samples with
high BTX concentrations, and points in the upper right represent samples with low BTX concentra-
tions. Samples which produced inconsistent results are located in the upper left or lower right
20
-------
1.6-
1.2-
o
CO
55
-------
quadrants. One can see that the variability about the line of perfect agreement is lowest for high
level samples, and becomes larger as the BTX sample concentration becomes smaller. In agree-
ment with the information in figures 6 and 7, one can also see that two samples lie in the off-
diagonal quadrants. As a practical matter, we note that wherever the decision level is, a large
distribution of samples near the decision level will increase the likelihood of getting an indetermi-
nate result (for replicate analysis) or false positive and false negative results (for single analysis
protocols).
QUALITY ASSURANCE AND QUALITY CONTROL SAMPLE RESULTS
Several types of QA samples were prepared in the field and transported back to the laborato-
ry for analysis by GC. These included trip blanks, trip lows (40 ppb) and trip highs (100 ppb)
during the initial portion of the evaluation, and three trip positives at 2.5, 25, and 100 ppb. All
concentrations are in terms of toluene. Table 3 summarizes the results from these analyses, with
the results grouped by their expected target concentration.
The mean and standard deviation obtained for the field prepared QA samples can be
compared to those expected from a single analyst using EPA Method 8020 (U.S. EPA, 1986a).
Except for a relatively high standard deviation for the 40 ppb samples, this study's results are
relatively close to the single analyst expectations. The consistently higher variability is probably
due to the fact that the single analyst expectations were determined from a round robin analysis of
split samples made up at a single time while each individual sample in this study was produced
independently. This suggests that a protocol for generating performance evaluation standards at
the time of analysis is quite feasible. The results also show that sample integrity has not been
compromised during transportation or storage. From these GC results, samples in this study had
no significant or systematic loss or gain of volatile organics.
Laboratory BTX immunoassay results of sample splits from the samples with positive target
values discussed in the preceding paragraph show an interesting trend. The data are summarized
in Table 4 on both a per individual immunoassay basis and on a per sample basis. For the per
sample results, "correct" or "incorrect" means that all individual immunoassays agreed when
compared to the GC results, and "indeterminate" means that some immunoassay results agreed
with and some disagreed with the GC results. Most immunoassay and GC analysis were in
agreement for the 2.5 and 100 ppb standard samples. Only one of 20 (5 percent) of the 100 ppb
samples was incorrectly determined. Ten of 12 of the 2.5 ppb samples were correctly assigned;
with two samples having indeterminate results. However, for the 25 and 40 ppb samples, less
than 50 percent of the individual immunoassay analyses, and fewer than 50 percent on a per
sample basis, were in agreement with the GC results. It appears that the BTX immunoassay
performs relatively poorly in the region near the 25 ppb classification level.
Laboratory immunoassay of the performance evaluation standards P1, P2, and P3 (2.5, 25,
and 100 ppb toluene, respectively) carried out in the first half of the evaluation generated results in
line with those above. Recall (from Section 3) that these samples were generated in the laborato-
ry, while the samples discussed above were generated in the field. Only four samples of each type
were assayed. One incorrect classification was found for each of the 2.5 and 100 ppb sample
groups, while 3 of 4 samples having 25 ppb toluene were incorrectly classified. Once again, the
immunoassay does not appear sensitive enough near the 25 ppb level.
22
-------
Table 3. GC results for field prepared QA samples.
Field Prepared QA Sample Results
EPA Method 8020
Single Analyst
Expectations
Target
Value
(ppb)
0
2.5
25
40
100
n
17
10
10
7
17
Mean
(ppb)
0.7
4.0
29
30
101
Standard
Deviation
(ppb)
1.1
2.4
3.2
13
14
Range
(ppb)
0-3.5
2.4 - 9.4
24-34
4-39
72-129
Mean
(ppb)
0.6
3.0
24
38
95
Standard
Deviation
(ppb)
0.5
0.7
2.7
4
9.5
Table 4. Laboratory BTX immunoassay results
of performance evaluation samples.
Target
Value
(ppb)
2.5
25
40
100
Per Assay
n
Assays
26
28
13
40
Results
n
Correct
24
9
4
38
Correct
10
1
1
19
Per Sample Results
Indeterminate
2
6
3
0
Incorrect
0
6
3
1
An Assessment in Terms of Five Data Quality Elements
Accuracy
Analysis of trip blanks provided one way of evaluating accuracy and checking for contamina-
tion due to sample shipping and storage. The GC analysis was often carried out several days after
the field immunoassay analysis, but well within the allowed 14-day holding time. No preservative
was added to any sample. However, the samples were cooled and maintained at 4°C until analysis;
which is the recommended procedure for volatile organic analysis of aqueous samples (U.S. EPA,
1986a). All thirty-three trip blanks were correctly identified by immunoassay in the laboratory.
Only one determination resulted in an S/R ratio below 0.85. However, this sample was reanalyzed
twice, with ratios of 1.02 and 1.01 resulting. Thus, no trip blanks were incorrectly classified by
23
-------
the laboratory immunoassay. These results suggest that sample contamination was not a
significant problem.
A select number of sample splits were analyzed by immunoassay in the laboratory to provide
additional laboratory-to-field comparison data. These samples were analyzed at approximately the
same time as the GC runs. Of 49 determinations from which comparisons could be made, only
four were inconsistent with the GC results. The S/R ratios for these were 0.59, 0.60, 0.70, and
0.83. The last two results were associated with indeterminate results; that is, only one of two
replicate analysis were not in agreement with the GC results. In summary, only one sample gave a
false positive immunoassay result and two samples produced indeterminate immunoassay results.
Comparison of laboratory to field immunoassay results showed one pair of inconsistent
laboratory results for a sample giving inconsistent field analysis results, one pair of positive
laboratory results paired with a set of inconsistent field results, and three individual laboratory
determinations inconsistent with the field results. However, two of these laboratory immunoassay
values were associated with a sample which gave results that were consistent with the GC values.
In summary, the laboratory immunoassay results from selected field samples do not give one
reason to question the integrity of sample transportation and holding time. These factors do not
appear to bias the results in any important way.
The false negative and false positive results were reassessed with respect to the bias
measured between the colorimeter and the laboratory spectrophotometer. Classification of samples
with high OD measurements for both sample and reference cuvettes was not affected by the bias.
Classification of samples where the reference OD is high and the sample OD is low would also not
be affected. Neither would the classification of samples with a low reference OD and a high
sample OD. In all the above situations, the effect of bias is low, or it enhances the probability of
correct classification. Only situations where the reference OD is high and the sample OD is slightly
lower will be subject to misclassification due to bias in the OD reading. Using the alternative
sample OD/reference OD ratio of 0.78 from Section 3 as the classification level, very little
difference was found in the false negative and false positive rates for the field BTX immunoassay
results. The alternative ratio gave six false positive determinations, which represent both replicate
analyses for each of three samples. This compares with six false positives (both replicates for two
samples, and one replicate for each of two other samples) for the original ratio. The alternative
ratio yielded six false negatives, two samples with both replicates incorrect and two samples with
only one replicate incorrect. This compares with four false negatives (both replicates for two
samples) for the original ratio. Based on replicate analysis, only one more sample was consistently
misidentified using the alternative S/R ratio.
Further results bearing on accuracy are presented in the following section on precision.
Precision
Precision of the BTX immunoassay analysis was assessed from tlie analysis results for field
prepared performance evaluation standards. Three standard levels, at 2.5, 25, and 100 ppb
toluene were prepared each day by the field analyst from amputated 1000 ppm toluene standards.
Usually, one set of analysis was performed before and another set was performed after the well
samples were analyzed. Fresh sets of samples were prepared for each day's analysis, therefore
variances reflect both dilution error and error in the assay.
There were 36 determinations (involving 19 separate days) for each concentration. The mean
S/R ratio for the 2.5 ppb standards was 0.999, with a standard deviation of 0.056. All analyses
had S/R ratios greater than 0.85. For the 25 ppb standard, only four analysis were less than 0.85.
24
-------
The mean S/R was 0.943, with a standard deviation of 0.108. The BTX immunoassay was not
sensitive enough to correctly classify most of these samples. The 100 ppb standards showed nine
determinations with S/R ratios greater than 0.85. The mean ratio was 0.766, with a standard
deviation of 0.104. A graphical summary is given in Figure 9. Figure 9 shows curves correspond-
ing to the mean S/R ratio and the mean ratio plus or minus one and two standard deviations. This
clearly shows why this formulation of the BTX immunoassay tends to produce a significant number
of false negative and false positives in the concentration range from just below 25 ppb to just
above 100 ppb.
Representativeness
The field samples were representative of typical groundwater samples. All field samples were
obtained from pre-established monitoring wells. The well fields were at four different sites in the
Las Vegas valley and represented sites where the aquifer was contaminated with gasoline.
Naturally, these sites do not represent all the different groundwater characteristics present across
the nation. For instance, variability due to differences in hardness or pH were not studied.
However, it is felt that the results from this study are generally applicable to water contaminated
with gasoline. Antox results with spiked sea water samples indicated that the test is not signifi-
cantly affected by mineral content.
Completeness
The completeness criteria of 54 samples (90 percent of the originally specified 60 well
samples) was met. Seventy-nine well samples were analyzed both by the BTX immunoassay in the
field and by the laboratory GC method.
The performance of the BTX immunoassay was also assessed in terms of the percentage of
cuvettes which generated acceptable results. Data quality objectives of 95 percent acceptable
performance for reference and sample cuvettes had been specified in the demonstration plan.
However, the developer had not provided limits on acceptable performance for the cuvettes.
For reference cuvettes the only qualification provided was that the optical density should be
between 1.0 and 1.9 for optimal performance. However, this does not mean that ODs outside this
range mean unacceptable performance. During the field and laboratory evaluation of the water
samples 144 reference cuvettes were used. Seven had ODs less than 1.0, 99 had ODs between
1.0 and 1.9, and 38 had ODs greater than 1.9. Reference cuvettes with ODs below 1.0 were
believed to be indicative of poor performance, and constituted 5 percent of the total. Reference
cuvettes with results above 1.9 were accepted as valid, as it appeared the high values were a
function of the levels of enzyme conjugate and antibody supplied in the kit. Antox provided
instructions for modifying reagent levels and timing of assay steps to lower reference sample OD
levels (see Appendix A).
No identifiable rules for determining acceptable sample cuvette behavior were provided by
Antox. In this study we could not identify one problem associated with the 552 sample cuvettes
used for the analysis of field or QA/QC samples. Since the results can span the OD range, it does
not appear that any a priori criteria can be specified. It appears that the only types of sample
cuvette problems one might find are associated with visible defects such as scratches or other
marks on the cuvettes.
25
-------
cc
W
0.8-
0.4-
— Mean
— ±1 Standard Deviation
•— ±2 Standard Deviations
.0.85 Decision
Level
10 100
Concentration of BTX (ppb) (Logarithmic Scale)
Figure 9. BTX immunoassay results for field performance evaluation samples. Mean S/R ± one
and two standard deviations are shown.
26
-------
Comparability
The comparability of results between GC and immunoassay was addressed by the use of a
variety of QA/QC samples and through close adherence to written protocols. The QA/QC
procedures described in EPA Method 8020 (U.S. EPA, 1986a) were followed for gas chromatogra-
phy analysis. The results are presented in Appendix B, showing that this study was within the
guidelines for the method. The results of quality control samples (presented elsewhere in this
section) also demonstrated that the GC was providing high quality data. Trip audit samples,
including both blank and positive BTX samples, showed that sample integrity was maintained
between field and laboratory.
GC/MS analysis of representative samples indicated that a large unassigned GC peak found in
several samples was methyl t-butyl ether (MTBE), an octane booster for gasoline. MTBE is one of
several oxygenated compounds used to boost oxygen content of gasoline in the Las Vegas valley in
the winter months. For high level samples, GC/MS confirmed the assignments of benzene, toluene,
ethylbenzene, and xylene as the major aromatic compounds present in the samples. For very high
level samples, much smaller amounts of other (unidentified) aromatic compounds were also
reported. However, the relative amounts of these compounds were small, and would not have
changed the classification of the samples with respect to the 25 ppb detection level used in this
study.
Results of the On-Site Audit
A field audit was conducted on April 16, 1992. Only minor deviations from the demonstra-
tion plan were reported. The dilution protocol for preparing audit samples was different, and
sample identification codes were changed for the field-analyzed performance evaluation standards.
The sample identification codes were changed due to modifications to the demonstration plan after
the evaluation was in progress. The dilution protocols were being followed correctly, but the
auditor had misinterpreted the discussion in the demonstration plan. The auditor noted that "These
small changes in procedure should not influence the results of the study". The auditor's report is
included in this document as Appendix D.
ADDITIONAL QA/QC OBSERVATIONS AND CONCLUSIONS
Optical densities for the reference cuvettes are plotted as a function of time in Figure 10.
Most of the reference OD values were between 1.7 and 2.0. Figure 10 shows that the range of
OD values narrowed over the course of the demonstration. It is not known whether this effect is
due to increased experience and reproducibility on the part of the analyst, or to changes in the
reagents. Because of the non-linearity of the colorimeter in this range, it is recommended that the
antibody and enzyme conjugate titers be adjusted to give reference OD values in the 1.0 to 1.5
range. It may be necessary to develop formulations of coated cuvettes and enzyme conjugates
that have better stability with lower initial reference OD values.
According to the field analyst, there were indications toward the end of the study that the
colorimeter drift was more pronounced. This has been associated with decline in the performance
of the light source. Since the colorimeter was re-blanked before and after each run, the signifi-
cance of this drift was minimal.
27
-------
10
CO
20
c
0)
Q
"to
.o
Q.
O
o> 1.5-
0)
o>
<4—
&
1.0-
*;.* • • *. •••*•...,.>.
. ••• .. .*. ...-. ....**-. • • * . *
***** iA *
* * * * * *
* *
•*
20 40 60
Time-indexed Run Number
80
Figure 10. Reference cuvette optical density vs time.
-------
SECTION 5
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
The BTX Immunoassay performed well for most samples evaluated in this study. Only five
percent of the samples were misclassified based on replicate determinations. The false negative
results suggest that the current formulation of the BTX immunoassay is not robust enough to
accurately deal with samples in the range of 25 to 100 ppb BTX. If one wants to be very accurate
in this range, a more sensitive format should be developed. The immunoassay was easy to use and
provided a quick estimate of BTX concentration that would prove useful in certain applications.
Advantages of the BTX Immunoassav
• Rapid - Up to four samples can be analyzed on-site within 30 minutes, allowing rapid field
screening of environmental water samples.
• Portable - The field kit can be set up in the space of a small table. The portable
colorimeter can be run from either a 120 VAC wall outlet or from batteries.
However, one needs to have a cooler available to keep most of the reagents at 4°C.
• Easy to use • No extensive training is needed to use the field kit. The instructions
are clear and easy to follow.
• Inexpensive - As a screening tool, use of the BTX immunoassay is much less
expensive than requiring that all samples be analyzed by conventional laboratory
analytical techniques.
• Effective - For certain screening functions, the BTX immunoassay appears to be
very effective. Samples above approximately 100 ppb BTX were accurately
identified. Low S/R ratios were related to contaminated samples, even for samples
with gasoline contamination below the 25 ppb detection claim of the test. This
method could prove useful in mapping the distribution of contaminants at a site or
for monitoring changes in contaminant levels over time.
Limitations of the BTX Immunoassav
• Sensitivity - Results from this evaluation indicate that the BTX immunoassay does
not meet the developer's claim for sensitivity. The test appears to allow a signifi-
cant percentage of false negative results from samples in the range from 25 to 100
ppb BTX. If critical decisions are relying on analysis of samples with these concen-
tration ranges, this test would not be appropriate. A sample would have to be
29
-------
above 100 ppb to be in a range where there was a five percent or lower probability of
getting a false negative result.
• Performance standards - Despite efforts by the developer, the kit comes with no
performance standards. It is difficult to prepare stable low-level (in this case, 10 to
50 ppb) standards of volatile organic compounds. In this evaluation, low level
standards were prepared on-site during each day's analysis by dilution from high-
level ampulated standards. Preparation of these standards requires volumetric
pipettes and glassware and careful attention to detail to prevent errors due to
dilution or contamination. The use of these procedures with the kit would require
the addition of several items of analytical glassware and more training, as well as a
clear, written protocol. The user would also need access to an appropriate source
of water free from BTX for making dilutions.
• Battery powered use - Despite the fact that the colorimeter is only used approxi-
mately every half-hour, leaving it on all the time drains the .batteries. This study left
the colorimeter on continuously during the day to avoid any instability that might
have occurred from repeatedly turning it on and off. It was found that the batteries
lasted about one and a half days (12 hours). Batteries were replaced at the
beginning of each day to avoid having to monitor for unusual measurement proper-
ties during the second day, as there was no clue other than "unexpected" or
"strange" readouts to show that the batteries were beginning to fail.
Minimizing False Negative and False Positive Results
In the absence of bias one can minimize the number of false positives and false negatives
through two actions. One way is to reduce the variability associated with the immunoassay. As
seen from Figure 8, low variability will reduce the probability that a sample will fall on the wrong
side of the 0.85 S/R decision value simply from analysis error (variability). The other factor in
determining false positive and false negative rates is the distribution of BTX concentrations one
expects from a study. If the sample concentrations are expected to be mostly well removed from
the decision level, then the rate of false negatives and false positives will be low. Alternatively, for
a given study one may wish to minimize either false positive or false negative results by shifting the
decision level. However, lower false negative rates will always be compensated by higher false
positive rates, and vice versa.
Limitations of this Demonstration
The principal limitation of this SITE demonstration has to do with the limited matrix types
evaluated. All the sites represented ground water in the Las Vegas valley, and contamination at
each site was primarily due to gasoline contamination.. Thus, the BTX immunoassay results
reported here may not be fully applicable at sites with different or more complex organic contami-
nation matrices, or at sites with large concentrations of inorganic compounds.
A second limitation of this evaluation is that it did not necessarily reflect the variability one
might expect from use &v a large number of field analysts. This study had only one field analyst.
The evaluation did no; test the robustness of the directions and procedures with respect to
individuals not receiving direct training by personnel familiar with immunoassay and analytical
chemistry techniques.
30
-------
RECOMMENDATIONS
Recommendations Specific to the BTX Demonstration
• Quality assurance guidelines should be made very specific for an analysis which
might be used by those with little experience. For the BTX immunoassay, specific
guidance with respect to the minimal OD for the reference sample is needed.
Though the developer told us verbally that a reference OD of less than 0.7 was
unacceptable, the written guidance with respect to trouble shooting mentions only
that reference ODs "should always be in the range 1.0—1.9 for optimal perfor-
mance." The developer should provide more detailed and specific QC acceptance
criteria based on statistically valid performance evaluations.
• It appears that the kit is not quite as sensitive as claimed. The developer needs to
improve the kit's ability to perform as claimed in the 25 to 100 ppb region.
• A simple protocol for generation of performance standards needs to be developed.
• The colorimeter should be modified so as to display a warning light when battery
power is too low for adequate performance. Alternately, the test procedure could
include samples with known concentration. One would then be able to check the
performance of the colorimeter as part of a check of the entire kit immunoassay
procedure.
• Improve kit reagent stability so that reagent levels can be adjusted to give lower OD
values at the beginning of the kit's shelf-life.
Recommendations for Future Studies
• Development of a more sensitive format would be advantageous. This is especially
true for the detection of benzene, which has a maximum contaminant level of 5 ppb
under the Safe Drinking Water Act (U.S. EPA, 1990b).
• Develop stable low-level BTX performance evaluation standards. Currently a field
analyst needs extra standards and analytical glassware in order to generate perfor-
mance evaluation standards.
• Evaluate the immunoassay on a variety of ground water matrices.
• Conduct a multi-laboratory or multi-site study to evaluate interlaboratory variability
and assess operator dependent responses.
• Conduct additional studies with selected matrix interferences and cross-reacting
substances.
31
-------
REFERENCES
1. Gertach, R. W., R. J. White, N. F. D. 0' Leary, G. I. Martucci, and J. M. Van Emon.
1992. Superfund Innovative Technology Evaluation (SITE) Program Demonstration Plan
for Antox BTX Water Screen (BTX Immunoassay). U.S. Environmental Protection
Agency, Environmental Monitoring Systems Laboratory, Las Vegas, NV. Internal
Report.
2. Koglin, E. N. 1990. Advanced Field Monitoring Methods Program. EPA/600/X-90/063. U.S.
Environmental Protection Agency, Environmental Monitoring Systems Laboratory. Las Vegas,
NV, 1 p.
3. Midwest Research Institute. 1990. How do you get "Relief" from Lab Analysis?.
Prepared for the United States Environmental Protection Agency, Office of Underground
Storage Tanks, Washington, D.C.
4. Stanley, T. W. and S. S. Verner. 1983. Interim Guidelines and Specifications for
Preparing Quality Assurance Project Plans. EPA/600/4-83/004. U.S. Environmental
Protection Agency, Washington, D.C.
5. Stanley, T. W. and S. S. Verner. 1985. The U.S. Environmental Protection Agency's
Quality Assurance Program. Jni J. K. Taylor and T. W. Stanley (Eds), (1985). Quality
Assurance for Environmental Measurements. ASTM STP 867, pp 12-19. American
Society for Testing and Materials. Philadelphia, PA.
6. U.S. Environmental Protection Agency. 1986a. Method 8020 - Aromatic Volatile Organics
by Gas Chromatography. Manual SW 846, U.S. Environmental Protection Agency, Office of
Solid Waste and Emergency Response, Washington, D.C.
7. U.S. Environmental Protection Agency. 1986b. Method 8240 - Volatile Organics by Gas
Chromatography/Mass Spectrometry (GC/MS): Packed Column Technique. Manual SW 846,
U.S. Environmental Protection Agency, Office of Solid Waste and Emergency Response,
Washington, D.C.
8. U.S. Environmental Protection Agency. 1986c. Method 5030 - (Purge and Trap) - Test
Methods for Evaluating Solid Waste. Manual SW 846, U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, D.C.
9. U.S. Environmental Protection Agency. 1987. Quality Assurance Plan. EPA/600/X-
87/241. U.S. Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, Las Vegas, NV.
10. U.S. Environmental Protection Agency. 1990a. Must for USTS - A Summary of the
Regulations for Underground Storage Tank Systems. EPA/530/UST-88/008. U.S.
Environmental Protection Agency, Office of Underground Storage Tanks, Washington, D.C.
32
-------
11. U.S. Environmental Protection Agency. 1990b. Drinking Water Regulations Under the
Safe Drinking Water Act. Fact Sheet. U.S. Environmental Protection Agency, Criteria
and Standards Division, Office of Drinking Water, Washington, D.C.
33
-------
APPENDIX A
A DESCRIPTION OF THE BTX IMMUNOASSAY PROCEDURE
34
-------
ANTOX BTX WATER SCREEN
25 ppb SENSITIVITY
Intended Use
The Antox BTX Water Screen is an enzyme immunoassay for the qualitative detection of
BTX (Benzene, Toluene and Xylene) and similar compounds. This test is designed to
reduce the number and frequency samples requiring expensive and time-consuming analytical
testing by identifying those samples that have significant levels of BTX contamination.
Summary and Explanation
BTX are aromatic hydrocarbons widely used in industrial processes and products. The
compounds can become environmental pollutants if misused or improperly disposed of. The
detection of BTX compounds in water samples is important in determining contamination
and effectiveness of remediation processes.
Principle of the Procedure
The Antox BTX Water Screen is a sensitive, fast and easy-to-use enzyme immunoassay
which employs polyclonal antibodies to identify BTX and similar compounds in water
samples. Using this test, a qualitative determination of BTX is possible.
The Antox BTX Water Screen is based upon immunological principles governing antigen-
antibody reactions. The kit features a cuvette coated with antibody specific for BTX and
an enzyme-conjugated BTX (antigen) reagent. Sample is added to the antibody coated
cuvette after which the enzyme-conjugated BTX reagent is added. BTX or similar
compounds present in the sample will compete with the enzyme-conjugated BTX for binding
sights on the antibody-coated cuvette. The cuvette is then rinsed to remove excess reactants.
Substrate and chromogen are then added which, in the presence of enzyme, turn a color that
can be measured on a photometer. The amount of BTX in the sample is indirectly
proportional to the color development. A comparison to a reference sample determines the
relative concentration of BTX or similar compounds in the sample.
Materials Provided in the kit:
Antibody Coated Cuvettes
Buffer Solution (Gray Cap)
Enzyme Solution (Red Cap)
Color Developer 1 (Black Cap)
Color Developer 2 (Blue Cap)
Terminating Solution (Purple Cap)
Disposable Syringes/Transfer Pipettes
35
-------
Materials not Provided:
VOA Transport Vials
Reference Solution (Distilled Water)
Felt Tip Pen \
Liquid Dispenser
Quality Control Samples (Available from Antox)
Timer
Caution
Do not mix reagents from different lots.
Do not use components beyond expiration date.
Do not mix reagent bottle caps.
FOR ENVIRONMENTAL TESTING ONLY.
Storage
All reagents should be stored refrigerated (2-S°C) when not in use except color developer
2 (Blue Cap) which should be stored at room temperature (20°-28°C). Prior to use, place
the Buffer Solution (gray cap), Enzyme Solution (red cap), Reference Solution (distilled
water) into the ice chest with the samples. Allow Color Developer 1 (black cap), Color
Developer 2 (blue cap) and bagged cuvettes to come to ambient temperature.
DO NOT FREEZE
SAMPLE COLLECTION, STORAGE AND PREPARATION
Ground water samples are best, and testing should be done as the samples are collected.
Testing on site is recommended due to the volatile nature of BTX and similar compounds.
The sample and reference solution (distilled water) should be maintained in a cold
environment, such as an ice water bath, prior to testing. This is due to the volatility of these
compounds.
Procedural Notes
Quality Control Samples Refer to Enclosed QC Procedure
Temperature Limits
REFERENCE SOLUTION, SAMPLE SOLUTION, ENZYME SOLUTION, AND
BUFFER SOLUTION SHOULD BE BROUGHTTO EQUIVALENT TEMPERATURES
IN AN ICE WATER BATH. IF AMBIENT TEMPERATURE IS LESS THAN 75°F BUT
36
-------
MORE THAN 32°F THEN THE ASSAY CAN BE RUN AT ROOM TEMPERATURE.
IF THE TEMPERATURE IS GREATER THAN 75°F STEPS 1-8 SHOULD BE
PERFORMED IN AN ICE WATER BATH.
Each reagent should be added to the reference cuvette immediately followed by the sample
cuvette. This should be done as quickly as possible so there is a minimum amount of time
between the addition of reagents to the reference and sample cuvette.
All components should be added to the cuvettes by holding the dropper vial in an upright
inverted position so the dropper nozzle is in the center of the test cuvette. When adding
the Color Developing Solutions, take care to dispense the drops into the bottom of the
cuvette.
When washing the cuvettes, discard the contents of both cuvettes with a flicking motion.
Using the rinse bottle provided, rinse the cuvettes four times with distilled water. Forcefully
fill the cuvettes to overflowing to overflowing and forcefully empty after each wash. Flick
the cuvettes well after the last addition to eliminated residual water.
The timing aspects of this test are important. DO NOT WALK AWAY!
ASSAY PROCEDURE:
( 1.) LABEL ONE ANTIBODY COATED CUVETTE "R" for Reference.
( 2.) Appropriately label an ANTIBODY COATED CUVETTE for each sample (up to
four).
NOTE: *****COMPLETE STEPS 3 THROUGH 6 AS QUICKLY AS POSSIBLE
( 3.) Use a disposable syringe or transfer pipette to dispense 4 ml of Reference Solution
(or distilled water) into the "R" cuvette.
( 4.) Use the same technique to measure 4 ml from sample for the VOA vial into the
appropriately labeled cuvette. BE SURE TO USE A CLEAN TRANSFER PIPETTE FOR
EACH SAMPLE TO AVOID CROSS CONTAMINATION.
( 5.) Add 4 drops of the Buffer Solution (Gray cap) to all cuvettes.
( 6.) Add 4 drops of the Enzyme Solution (Red Cap) to all cuvettes. Cap and invert
cuvettes.
( 7.) Incubate 10 minutes.
( 8.) Using distilled water (NOT COLD) and wash bottle, wash and decant the cuvettes
4 times. Be sure to remove as much water from the cuvettes as possible (blot if necessary).
37
-------
( 9.) Add 4 drops of Color Developer 1 (Black cap) to each cuvette. Add 4 drops Color
Developer 2 (Blue cap) to each cuvette. Rap cuvettes gently to assure all solution is in the
bottom of the cuvettes.
(10.) Incubate 5 minutes.
(11.) Add 4 drops terminating solution (Purple cap) to each cuvette. Mix by swirling
gently.
(12.) Read the cuvettes at 450 nm. Wipe the- lower portion of each cuvette with a
Kimwipe before reading.
Performance Characteristics
Sensitivity
Assay sensitivity is 25 ppb for water samples. This sensitivity will vary based on the specific
compounds that are being tested, possible interfering substances or matrix effects.
Sensitivity characteristics for individual compounds may need to be determined.
Specificity
The antibody is directed at BTX (Benzene, Toluene and Xylehe). Substantial cross-reactivity
should be anticipated due to the fact many compounds possess similar antigenic properties.
Calculation and Interpretation of Results
Absorbance of Sample (S)
R
Absorbance of Reference (R)
At the 95% confidence:
If S/R is equal to of greater then .850, then the sample contains 25 ppb (parts per billion)
or less of BTX.
If S/R is less than .850, then the sample contains 25 ppb or more of BTX.
38
-------
Example:
Sample abs. = .600
Reference abs. = 1.000
.600
1.00 = .600
.600 is less than .850. The sample contains 25ppb or more BTX.
Manufactured By:
Antox, Inc.
95 Darling Avenue
South Portland, Maine 04106
(207) 772-3544 - (800)323-3199
39
-------
Antox BTX Water Screen
Trouble Shooting Addendum #1 : Off
Scale Absorbance Values
At the time of Quality control testing the Antox BTX Water Screen
has a Reference Absorbance that averages 1.3 units at 450 nm. The
Reference Absorbance values should always be in the range 1.0 - 1.9
for optimal performance.
The following suggestions are meant to assist the user when Absorbance
values are consistently greater than 1.9. It is assumed that the
photometer used has been checked for effieciency.
BEFORE MANIPULATION OF THE TEST PROCEDURE. RULE OUT THE FOLLOWING:
1) PROPER WASHING It is EESENTIAL THAT ALL NONIMMUNOLOGICALLY BOUND
ENZYME SOLUTION be renoved before adding the color developers.
Using a laboratory wash bottle and a FORCEFUL jet of water the
cuvettes should be filled to overflowing and emptied at least 4 times.
The cuvettes should be empties with a sharp "flicking " motion to
assure removal of all rinse water.
2) CLEAN TIPS ON COLOR DEVELOPERS If the vial tips on the color
developers become contaminated with enzyme solution, the
enzyme conjugate could react with the color developers. This should
not happen if the user practices care in replacing the caps on the proper
vials. An easy way to check this is: place 2 drops of each color
developer in a disposable test tube. If any enzyme conjugate is
present color will develope immediately.
3) TIMING Both incubation times should be adhered to. The reactions
are continuous and longer incubations will give higher Absorbance values
IF THE ABOVE ARE RULED OUT, THE FOLLOWING MANIPULATIONS MAY BE DONE
1) Reduce the Enzyme Solution from 4 to 3 drops.
2) Reduce the 2nd incubation from 5 to 4 minutes.
3) Dilute the terminated test solutions with distilled water. A
1:2 dilution should be sufficient (E.g. the terminated test
solution in the cuvette has a volume of approximately 500 uL. Add
500 uL distilled water for a 1:2 dilution.
EQUATE™ INNOVATIVE TEST SYSTEMS
40
-------
APPENDIX B
PROCEDURAL DETAILS AND QUALITY ASSURANCE SUMMARY FOR
GAS CHROMATOGRAPHY AND PURGE-AND-TRAP METHODS
Table of Contents
pace
GC Method Used in the BTX SITE Demonstration 43
Instrumentation 43
Calibration Curve Generation 43
Daily Calibration Standards 44
Retention Time Windows 44
Method Detection Limits 44
Method Precision 45
List of Figures
Figure page
B-1. Benzene standard curve (3-12-92) 49
B-2. Toluene standard curve (3-12-92) 49
B-3. Ethylbenzene standard curve (3-12-92) 50
B-4. o-Xylene standard curve (3-12-92) 50
B-5. Benzene calibration control chart (32 ppb) 52
B-6. Toluene calibration control chart (32 ppb) 52
B-7. Ethylbenzene calibration control chart (32 ppb) 53
B-8. o-Xylene calibration control chart (32 ppb) 53
B-9. Benzene calibration control chart (60 ppb) 55
B-10. Toluene calibration control chart (60 ppb) 55
B-11. Ethylbenzene calibration control chart (60 ppb) 56
B-12. o-Xylene calibration control chart (60 ppb) 56
41
-------
List of Tables
Table page
B-1. HP 5890/PID detector calibration on 1-14-92 (Mean of three injections) 46
B-2. HP 5890/PID detector calibration on 2-24-92 (Mean of three injections) 47
B-3. HP 5890/PID detector calibration on 3-12-92 (Mean of three injections) 48
B-4. HP 5890/PID detector continuing calibrations - (32 ppb). 51
B-5. HP 5890/PID detector continuing calibrations - (60 ppb) 54
B-6. HP 5890/PID detector - retention time windows
within day precision (analyzed 1-14-92) 57
B-7. HP 5890/PID detector - retention time windows
precision over a two week period (analyzed 1-14-92 to 1-29-92) 58
B-8. HP 5890/PID detector - method detection limits in water
spiked at 1 //g/L (ppb) (analyzed 1-15-92) 59
B-9. HP 5890/PID detector - method detection limits in water
spiked at 1 //g/L (ppb) (analyzed 2-25-92) 60
B-10. HP 5890/PID detector - method detection limits in water
spiked at 1 //g/L (ppb) (analyzed 3-13-92) 61
B-11. HP 5890/PID detector - precision study replicates
analyzed at 20 //g/L (analyzed 1-09-92) 62
42
-------
GC METHOD USED IN THE BTX SITE DEMONSTRATION
Water samples were analyzed by gas chromatography to provide comparison results for
replicate samples analyzed by the Antox BTX immunoassay. The analytical procedures used were
EPA Method 8020 (GC with PID detector) and EPA method 5030 (Purge and Trap).
Instrumentation
The following analytical instruments were used:
Gas Chromatograph: Hewlett-Packard Model 5890 - series 2
GC Column: DB 624 30 m X 0.54 mm ID (J&W Scientific)
Purge and Trap: O.I. Analytical Model 3340 with O.I. Model MPM-16 AutoSampler
Detector: O.I. Model 3320 P.I.D. Detector
Data Integrator: Hewlett-Packard Model 3396A
The instrument conditions were as follows:
Gas Chromatograph:
Carrier Gas - Helium
Flow Rate - 10 cc/min
Oven Temperature - 40°C for 2 min.
Ramped at 4°C / min
100"C final
Injector Temperature - 200°C
Detector Temperature - 2508C
PID Lamp 10 eV at intensity 4
Purge and Trap:
Purge Gas - Helium
Purge Flow - 40 cc/min
Purge Time - 11 min
Purge Temperature - 28°C (ambient)
Desorb Temperature - 180°C
Desorb Time - 2 min
Calibration Curve Generation
The GC was calibrated using the external standard calibration procedure as outlined in SW-
846. Calibration standards were prepared at a minimum of five concentrations by adding volumes
of stock standards to a volumetric flask and diluting with methanol. The analytical reference
standards were obtained from the EPA Quality Assurance Materials Bank at Research Triangle Park,
NC. The following standards were used:
Benzene 5,000//g/mL Lot #04-01-06
Toluene 1,000//g/mL Lot # 84-01-08
Ethylbenzene 10,000 jt/g/mL Lot #36-01-04
o-Xylene 5,000 ^g/mL Lot # 201-01-05
43
-------
The above solutions were diluted into a single solution containing 5 ug/tnL of each compound. This
working standard was used to spike deionized water to concentrations of between 1 to 200 //g/L
(ppb). These water spikes were analyzed by purge-and-trap in the same manner as the actual
samples.
The instrument was calibrated three times over the course of the study; on 1-14-92, 2-14-
92, and 3-12-92. The results of these calibrations are presented in Tables B^1, B-2, and B-3. The
response factors were calculated by dividing the concentration of the water standard by the area of
the peak. The areas were obtained from the HP 3396A integrator. The method states linearity can
be assumed and the average calibration factor can be used if the relative standard deviation
(percent RSD) of the calibration factor is less than 20 percent over the working range of the
method. As shown in the data tables, this was achieved for all analytes for all three dates. Tables
B-1, B-2, and B-3 also show the squared correlation coefficient for each calibration and the slope of
the prediction equation. Examples of the calibration curves for the four analytes are shown in
figures B-1 to B-4, for the calibration performed on 3-12-92.
Daily Calibration Standards
EPA method 8020 requires the calibration to be verified on each working day by injection of
at least one calibration standard. For this study at least two standards were injected on each day
samples were analyzed, one before any samples were analyzed and one following the last sample
analyzed. In addition, a standard was analyzed for each set of 12 samples. This QC standard was
near the mid point of the calibration curve. The method states the analysis is in control if the
response for any analyte varies from the predicted response by less than ±15 percent. If the
response varies by more than ±15 percent, a new calibration curve is to be prepared.
From 1-15-92 to 3-02-92, the QC standard was a water sample spiked to contain 32 ppb of
each target analyte. The analytical results are presented in Table B-4 for these daily QC standards.
These results are shown as quality control charts in figures B-5 to B-8. The lines indicate the upper
and lower control limits at ± 15 percent of the expected concentration. There was only one
analysis on March 2 which exceeded the control limits (up to 23 percent high). Prior to running
any further samples, a calibration curve was prepared on 3-12-92 with the concentration range
expanded to 200 ppb. Therefore, the QC standard analyzed after this date was increased to 60
ppb. The results of the daily 60 ppb QC standards are presented in Table B-5, and shown as QC
charts in figures B-9 to B-12. None of the 60 ppb QC standards exceeded the control limits.
Retention Time Windows
The method requires the establishment of retention time windows for the target analytes.
The procedure defines the window as three times the standard deviation of the absolute retention
time for each analyte analyzed. Each standard deviation was obtained by replicate injection of a
standard. Retention time precision was calculated for a single day, and over a two week period
(1-14-92 to 1-29-92). These data are presented in tables B-6 and B-7. The data show very good
retention time precision. Single day retention time windows were less than 0.01 minutes and
retention time windows over two weeks were less than 0.05 minutes for each analyte.
Method Detection Limits
The method detection limit (MDL) was determined each time a new calibration was prepared.
The MDL was determined by analyzing replicate low level standards. The low level standard was
1.0 ppb. The MDL was defined as the concentration obtained by a peak area of three times the
standard deviation of the replicate low level standard. Data is presented in tables B-8, B-9, and B-
44
-------
10 for the MDL calculations. All analytes show an MDL of less than 1 ppb; well below the data
quality objective of 10 ppb specified for this study.
Method Precision
The overall precision of the method was determined by analyzing standards at 20 ppb. A
standard was analyzed on each port of the autosampler. The results are presented in Table B-11
The percent RSD for each of the four standards was four.
45
-------
Concentration
U/g/U
4
16
32
60
80
100
120
140
4
16
32
60
80
100
120
140
Mean
RSD (%)
Benzene
35859
125097
233663
440296
550769
69461 2
969963
1139242
1.115E-04
1 .279E-04
1 .369E-04
1 .363E-04
1 .453E-04
1 .440E-04
1 .237E-04
1 .229E-04
1.31E-04
8.3
Area U/V X
Toluene
30112
111768
201471
384755
488408
618535
891411
1030770
Response
1 .328E-04
1 .432E-04
1 .588E-04
1 .559E-04
1 .638E-04
1 .61 7E-04
1 .346E-04
1 .358E-04
1 .48E-04
8.2
sec / 8)
Ethylbenzene
27058
103456
188228
363435
461050
588340
842367
972185
Factors
1 .478E-04
1 .547E-04
1 .700E.04
1.651 E-04
1 .735E-04
1 .700E-04
1 .425E-04
1 .440E-04
1 .58E-04
7.5
o-Xylene
26786
102712
185906
356390
450415
578171
839200
972051
1 .493E-04
1 .558E-04
1.721 E-04
1 .684E-04
1 .776E-04
1 .730E-04
1 .430E-04
1 .440E-04
1 .60E-04
8.2
Regression Results
RA2
Slope
0/g/L)/Area
0.9862
1 .261 E-04
0.9821
1 .383E-04
0.9831
1.461 E-04
0.9799
1 .465E-04
Table B-1. HP 5890/PID detector calibration on 1-14-92
(Mean of three injections).
46
-------
Concentration
0/g/D
4
16
32
60
100
4
16
32
60
100
Mean
RSD (%)
Benzene
72109
242937
48501 1
888909
1393942
5.55E-05
6.59E-05
6.60E-05
6.75E-05
7.17E-05
6.53E-05
8.2
Area U/V X
Toluene
60900
214254
423830
777131
1216043
Response
6.57E-05
7.47E-05
7.55E-05
7.72E-05
8.22E-05
7.51E-05
7.2
sec / 8)
Ethylbenzene
52798
197431
388253
707516
1099833
Factors
7.576E-05
8.104E-05
8.242E-05
8.480E-05
9.092E-05
8.30E-05
6.0
o-Xylene
53579
188352
375529
688894
1092868
7.466E-05
8.495E-05
8.521E-05
8.710E-05
9.150E-05
8.47E-05
6.5
Regression Results
RA2
Slope
Oug/L)/Area
0.9986
7.22E-05
0.9984
8.27E-05
0.9979
9.14E-05
0.9991
9.20E-05
Table B-2. HP 5890/PID detector calibration on 2-24-92
(Mean of three injections).
47
-------
Area
-------
2000000-
1500000H
M
4J
0 1000000-
«J
0)
500000H
50 100 150
Concentration (ppb)
200
Figure B-1. Benzene standard curve (3-12-92).
(fl
4J
C
OS
0)
2000000
1500000-
1000000-
500000H
0 50 100 150
Concentration (ppb)
Rgure B-2. Toluene standard curve (3-12-92).
200
49
-------
10
-p
c
3
o
o
«J
0)
2000000
1500000-
1000000-
500000-
0 50 100 150
Concentration (ppb)
Figure B-3. Ethylbenzene standard curve (3-12-92).
200
w
-p
§
o
(C
0)
2000000
1500000-
1000000-
500000-
50 100 150
Concentration (ppb)
Fioure B-4. o-Xylene standard curve (3-12-92).
200
50
-------
Concentration (JJQ/L)
Date
1-15-92
1-17-92
1-21-92
1-27-92
1-27-92
1-27-92
1-28-92
1-28-92
2-25-92
2-25-92
2-26-92
2-26-92
2-28-92
2-28-92
2-28-92
2-29-92
2-29-92
3-01-92
3-01-92
3-01-92
3-02-92
3-02-92
3-02-92
3-02-92
3-02-92
3-02-92
Run
Number
136
191
209
256
274
278
281
297
633
635
675
653
724
739
740
750
764
767
781
782
795
796
798
812
813
814
Benzene
33.1
33.7
34.1
31.8
33.5
31.4
31.8
33.0
34.9
33.3
36.3
32.7
35.5
31.9
34.5
32.3
32.2
34.0
32.3
34.5
35.1
33.4
34.2
35.9
38.2
30.9
Toluene
33.9
34.1
35.1
33.2
32.9
35.5
33.0
33.7
34.5
34.2
36.9
33.3
36.1
32.8
34.6
33.6
33.4
35.2
32.7
35.2
36.1
34.2
35.3
36.8
39.3
32.3
Ethylbenzene
33.7
33.5
34.9
33.0
32.8
35.3
32.9
33.8
34.4
33.1
36.4
32.4
34.4
29.8
32.3
31.8
31.0
33.5
30.8
32.6
33.8
32.2
34.0
35.2
37.6
30.8
o-Xylene
33.0
32.4
34.5
32.4
32.4
34.7
32.2
33.1
35.5
34.3
36.5
33.2
35.5
27.9
33.1
32.7
31.8
34.3
31.4
33.7
34.6
32.7
34.6
36.0
38.4
31.8
Table B-4. HP 5890/PID detector continuing calibrations - (32 ppb).
51
-------
4O
38-
^ 36-
O
3 32
i
O 28-
U
30-
26-
24-
0
O
OQ
1 3 5 7 9 11 13 15 17 19 21 23 25
Run Index
Figure B-5. Benzene calibration control chart (32 ppb).
40-
38-
~ 36-
I
3 34H
§ _,
32H
o o o
o
o o
o " o
O
§ 28-
O
26-
24-
1 3 5 7 9 11 13 15 17 19 21 23 25
Run Index
Figure B-6. Toluene calibration control chart (32 ppb).
52
-------
40-
til
38-
"* 34-
o
3
-------
Date
3-13-92
3-13-92
3-14-92
3-14-92
3-15-92
3-16-92
3-16-92
3-16-92
3-17-92
3-17-92
3-18-92
3-23-92
3-23-92
4-20-92
4-20-92
4-20-92
4-20-92
4-21-92
4-23-92
4-23-92
4-23-92
4-24-92
4-24-92
Run
Number
1053
1068
1071
1086
1094
1109
1113
1120
1124
1131
1134
1201
1210
1413
1426
1427
1441
1445
1461
1475
1490
1495
1499
Concentration (//g/L)
Benzene
57.0
61.7
54.3
60.1
64.2
60.7
61.4
62.3
67.6
61.9
58.4
58.7
60.4
53.5
57.2
58.2
57.8
55.1
53.0
55.6
54.3
57.1
59.0
Toluene
55.1
58.5
52.4
59.0
62.1
60.3
62.1
62.4
64.7
59.9
56.7
56.4
57.6
53.2
53.5
56.2
54.5
52.6
51.3
53.0
51.5
55.5
56.3
Ethylbenzene
56.6
60.9
54.9
57.6
64.3
59.0
60.7
61.5
68.5
63.6
59.5
59.1
60.2
56.3
56.5
59.5
56.5
55.7
54.3
55.8
54.3
57.3
59.5
o-Xylene
56.9
59.8
54.4
57.5
63.9
58.6
60.0
61.4
67.2
62.6
58.6
57.8
59.1
54.9
55.7
58.1
55.1
54.6
53.4
54.9
53.1
57.7
58.3
Table B-5. HP 5890/PID detector continuing calibrations - (60 ppb).
54
-------
75-
70-
I
£ 65H
§
•H
8
50-
1 35 7 9 11 13 15 17 19 21 23
Run Index
Rgure B-9. Benzene calibration control chart (60 ppb).
70
* .
-' 65-
§
•H
« 60H
s
8
8
50-
45
O O
1 3 5 7 9 11 13 15 17 19 21 23
Run Index
Rgure B-10. Toluene calibration control chart (60 ppb).
55
-------
75-
i i
70-
65H
§
•H
4)
o
I-
50
o
o
o0°
o o
0 o o o 0
0 00
1 3 5 7 9 11 13 15 17 19 21 23
Run Index
Rgure B-11. Ethylbenzene calibration control chart (60 ppb).
« 70"
2"
c
65-
°- o
4*
2
50-
4
O
O
o
o o
0 ° «
o ° °
o
5JH o
1 '3 5 ' 7 *9 ' 11 ' 13 ' 15 ' 17 ' 19 ' 21 ' 23
Run Index
Rgure B-12. o-Xylene calibration control chart (60 ppb).
56
-------
Retention Time (min)
Run
Number
109
111
112
113
115
118
119
120
121
122
124
125
126
127
128
129
130
133
Mean
SD
RSD (%)
Benzene
3.022
3.015
3.016
3.015
3.016
3.020
3.016
3.018
3.019
3.019
3.019
3.012
3.016
3.015
3.019
3.015
3.017
3.019
3.017
0.002
0.08
Toluene
5.345
5.340
5.340
5.340
5.342
5.345
5.343
5.346
5.345
5.345
5.346
5.340
5.344
5.344
5.345
5.345
5.345
5.346
5.344
0.002
0.04
Ethylbenzene
7.850
7.843
7.845
7.843
7.843
7.849
7.848
7.850
7.849
7.848
7.848
7.846
7.849
7.850
7.850
7.850
7.851
7.853
7.848
0.003
0.04
o-Xylene
8.786
8.777
8.780
8.779
8.779
8.785
8.783
8.785
8.786
8.785
8.786
8.785
8.786
8.785
8.786
8.785
8.785
8.789
8.784
0.003
0.03
Retention Time Windows
3*SD
Percent
0.007
0.2
0.007
0.1
0.009
0.1
0.009
0.1
Table B-6. HP 5890/PID detector - retention time windows
within day precision (analyzed 1-14-92).
57
-------
Retention Time (min)
hun
Number
109
126
130
256
274
278
281
297
315
Mean
SD
RSD (%)
Benzene
3.022
3.016
3.017
3.030
3.045
3.045
3.035
3.041
3.040
3.032
0.012
0.38
Toluene
5.345
5.344
5.345
5.367
5.376
5.376
5.370
5.370
5.376
5.363
0.014
0.27
Ethylbenzene
7.850
7.849
7.851
7.877
7.882
7.884
7.879
7.878
7.887
7.871
0.016
0.20
o-Xylene
8.786
8.786
8.785
8.816
8.817
8.818
8.815
8.813
8.825
8.807
0.016
0.18
Retention Time Windows
3*SD
Percent
0.035
1.1
0.043
0.8
0.048
0.6
0.049
0.6
Table B-7. HP 5890/PID detector - retention time windows
precision over a two week period
(analyzed 1-14-92 to 1-29-92).
58
-------
Hun
Number
142
144
145
146
147
148
149
150
151
Mean
SD
RSD {%)
Benzene
12006
10335
10107
9808
12704
9189
10468
9332
9853
10422
1117
10.7
Area U/V X
Toluene
9317
10020
10297
10025
13085
9515
9939
10045
9785
10225
1049
10.3
sec / 8)
Ethylbenzene
7102
7933
7972
7892
11132
7603
7629
7149
7470
7987
1151
14.4
o-Xylene
7185
8103
8449
8266
12037
7588
7879
7702
7716
8325
1360
16.3
Method Detection Limits
3*SD
MDL (ppb)
3352
0.322
3147
0.308
3454
0.432
4080
0.490
Table B-8. HP 5890/PID detector - method detection limits in water
spiked at 1 //g/L (ppb) (analyzed 1-15-92).
59
-------
Hun
Number
642
643
644
645
646
648
649
650
651
Mean
SD
RSD {%)
Benzene
22579
30941
21331
22365
21414
21964
27709
24352
26192
24316
3138
12.9
Area (//V X
Toluene
16394
16240
17621
16696
17217
17234
17359
17815
18367
17216
647
3.8
sec / 8)
Ethylbenzene
22516
—
19826
31269
20022
20745
11256
20863
28222
21840
5608
25.7
o-Xylene
16769
12248
18652
27580
18467
18392
13867
20351
29876
19579
5460
27.9
Method Detection Limits
3*SD
MDL (ppb)
9413
0.39
1941
0.11
16825
0.77
16380
0.84
Table B-9. HP 5890/PID detector - method detection limits in water
spiked at 1 //g/L (ppb) (analyzed 2-25-92).
60
-------
Run
Number
1035
1036
1037
1044
1045
1046
1047
1048
1049
1050
1051
Mean
SD
RSD (%)
Benzene
10850
10650
11775
12935
13946
12547
1 1 903
11721
10825
11003
11588
11796
972
8.2
Area (//V X
Toluene
9826
10173
11940
11559
12768
11828
10987
12345
8980
9194
10337
10903
1227
11.3
sec / 8)
Ethylbenzene
9717
10195
11906
11596
12416
11747
10790
12120
8094
8186
9271
10549
1491
14.1
o-Xylene
10126
10752
13165
11513
12588
12001
12197
13023
8365
8552
9249
11048
1673
15.1
Method Detection Limits
3*SD
MDL (ppb)
2917
0.25
3861
0.34
4472
0.42
5020
0.45
Table B-10. HP 5890/PID detector - method detection limits in water
spiked at 1 fJQ/L (ppb) (analyzed 3-13-92).
61
-------
Port
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Mean
SD
RSD (%)
Benzene
1 64449
139872
149744
147412
150194
145834
152719
157745
148831
157991
147731
148059
142816
147917
151058
150158
5965
4.0
Area U/V X
Toluene
159916
135061
145185
141334
145009
139233
143833
144081
140996
141619
143255
143060
135900
140173
141917
142711
5415
3.8
sec / 8}
Ethylbenzene
139537
1 1 9468
127832
127001
127682
120861
125823
126124
121540
1 24077
124178
123665
120035
119087
1 20680
1 24507
4968
4.0
o-Xylene
135173
117475
127576
1 24080
128535
1 22467
125822
125652
122972
121615
123363
121287
117943
119240
121075
123618
4392
3.6
Table B-11. HP 5890/PID detector - precision study replicates
analyzed at 20//g/L (analyzed 1-09-92).
62
-------
APPENDIX C
MODIFIED PROCEDURE FOR PREPARATION OF
TRIP AUDIT/PERFORMANCE AUDIT SAMPLES
63
-------
MODIFIED PROCEDURE FOR PREPARATION OF
TRIP AUDIT/PERFORMANCE AUDIT SAMPLES
1. Place 1.1 L of deionized water in a glass bottle and chill in ice to 4°C.
2. Prepare 10 ppm BTX standard as per SOP in the Demonstration Plan.
3. Label one clean 250-mL volumetric flask as the 2.5 ppb standard. Fill flask 9/10 full with
ice-cold deionized water.
4. With a 100-j/L (gas-tight) syringe, transfer 62.5-j/L of the 10 ppm standard into the flask
and bring the volume to the mark with ice-cold deionized water. While pipetting, keep the
syringe tip under water.
5. Quickly stopper the flask. Wrap Parafilm around the stopper seal and thoroughly mix the
contents by repeated inversion. Make sure the solution is well mixed.
6. Transfer the contents to four clean, appropriately labelled* VOA vials and quickly cap each
vial and place on ice in a cooler.
7. For the 25 ppb standard, proceed as in steps 3 through 6 above with the exception that in
step 3 the 250-mL volumetric flask is labelled as the 25 ppb standard and in step 4, 625 //L
of the 10 ppm standard is pipetted into the flask using a 1.0-mL (gas-tight) syringe. While
pipetting, keep the syringe under water.
8. For the 100 ppb standard, proceed as in steps 3 through 6 above except that in step 3 the
250-mL volumetric flask is labelled as the 100 ppb standard and in step 4, 2.5 ml of the
10 ppm standard is pipetted into the flask using a 2-mL graduated glass pipet.
9. Assay two of the sets of three standards (2.5, 25, and 100 ppb) as the first and last runs
of the day in the BTX immunoassay field analysis.
10. Transport the other two sets of standards back to the laboratory in the cooler, along with the
other field samples. Store the sets of standards at 4°C.
11. Within 7 days of the collection date, analyze one set of standards by the BTX immunoassay
in the laboratory and analyze the other set by GC using EPA Method 8020.
* Sample codes for these samples are: W1 for 2.5 ppb sample
W2 for 25 ppb sample
W3 for .100 ppb sample
Run code for the field analysis at the beginning of the day is 01,
and 02 for the field analysis at the end of the day.
64
-------
Additional supplies needed:
5 -- 250-mL volumetric flasks with stoppers.
2 - 2-mL graduated glass pipets.
1 - 100-/A. (gas-tight) syringe.
65
-------
APPENDIX D
FIELD AUDIT REPORT
66
-------
FIELD AUDIT REPORT
ANTOX BTX WATER SCREEN
by
Neal Amick
1. SUMMARY
An on-site evaluation of field operations in support of the Superfund Innovative Technology
Evaluation (SITE) for Antox BTX water screen was conducted on April 16, 1992. Sample
collection procedures, field analytical procedures, and data documentation were carefully observed.
Overall, the field work was of very good quality, and nothing was observed which would adversely
effect the study results. Only minor deviations from the Demonstration Plan were observed, which
are recorded in this report.
2. PERSONNEL AND EQUIPMENT
The personnel involved and the supporting field equipment were more than adequate to fulfill the
requirements of the study. The operation was organized and neat. All supplies required for the
study were present and set up or stored in designated spaces.
3. SAMPLING PROCEDURES
Sampling procedures were performed well and according to the plan. Samples were stored on ice
in coolers. Temperature of the coolers were checked regularly.
4. FIELD PROCEDURES
The field analyst took ultimate care that the procedure was performed precisely the same way
every time. The field kit standard operating procedure was followed to the letter for the analysis of
samples. Minor differences were noted in the procedures for preparing audit samples between
those in the Demonstration Plan and as performed in the field. For example, the plan calls for
preparing audit samples by diluting a stock solution of BTX into 4 mL of distilled water. Instead,
the stock solution was diluted into volumetric flasks. From the plan, one might assume the audit
sample contained benzene, toluene, and xylene (BTX). Only toluene was used for a spike for the
audit samples. The amputated stock solution used for preparing the audit samples was 1000 ppm,
instead of 10 ppm as stated in the plan. This 1000 ppm solution was diluted in two steps (one to
ten dilutions) to obtain the 10 ppm stock solution.
These small changes in procedure should not influence the result of the study.
5. DOCUMENTATION
All facets of the field study were well documented. Chain-of-Custody forms were completed in a
timely manner after sample collection. The field data form as provided in the Demonstration Plan
was not used; however, all information provided by that form was recorded either in the sample
data log sheet, or in a field logbook.
Samples were labeled with identification codes as specified in the Demonstration Plan, except the
standards were labeled as W1, W2, and W3 instead of P1, P2, and P3.
67
-------
6. CONCLUSION
All observations made during this field audit indicate good quality data was obtained.
Figure D-1 is a copy of the audit sheet completed in the field.
68
-------
ON-SITE AUDIT CHECKLIST FORM (F-2)
O)
(O
LOCATION:
Field Forms Filled Out Completely
And Accurately
Field Kit SOP Followed
QC Acceptance Criteria Observed
Sampling And Analysts Plan
Followed
Sample Handling. Labeling, Tracking
And Archiving Done According To
Instructions
Sample Packaging And Shipment
Handled Properly
Safely Observed
Cleanliness. Adherence To
CLP Observed
YES
X
•\J
r**>
V
X
V
X
YES. WITH
NUMEROUS
EXCEPTIONS
YES. WITH
FEW
EXCEPTIONS
X
X
NO
PROBLEM
^uTfcP V*.() ^ lv* f
«OM *™iW fa
Auditor's Signature.
Dale
Figure D-1. Field audit data sheet completed by auditor.
-------
APPENDIX E
PARTICIPATING PERSONNEL
70
-------
PARTICIPATING PERSONNEL
Many individuals from various organizations contributed to the success of the BTX immunoas-
say demonstration and participated in the preparation of this report. Their names, organizations,
and contributions are provided below.
Responsibility
Authors
SITE Matrix Manager
Project Management
Technical Leads
Quality Assurance
Experimental Design
Health and Safety
Logistical Support
Field Sampling/Analysis
Analytical Support
Data Base Management
Statistical Analysis
Name
R. W. Gerlach
R. J. White
N. F. D. 0' Leary
J. M. Van Emon
E. N. Koglin
J. M. Van Emon
R. Piasio
N. F. D. 0' Leary
R. J. White
R. W. Gerlach
V. A. Ecker
L. R. Williams
R. W. Gerlach
R. J. White
G. R. Mongeau
D. Alford
E. Noack
J. W. Curtis
J. H. Zimmerman
M. Brand
R. J. White
P. A. Amick
E. N. Amick
G. I. Martucci
R. W. Gerlach
Affiliation*
Lockheed
Lockheed
Lockheed
U.S. EPA
U.S. EPA
U.S. EPA
ANTOX
Lockheed
Lockheed
Lockheed
Lockheed
U.S. EPA
Lockheed
Lockheed
Lockheed
CECS
CECS
Lockheed
Lockheed
CECS
Lockheed
Lockheed
Lockheed
Lockheed
Lockheed
71
-------
Graphics S. 0. Garcia Lockheed
* ANTOX Antox Inc., South Portland, ME.
CECS Converse Environmental Consultants Southwest, Las Vegas, NV.
U.S. EPA U.S. Environmental Protection Agency, Las Vegas, NV.
72
-------
APPENDIX F
DATA TABLES FOR FIELD IMMUNOASSAY, LABORATORY IMMUNOASSAY,
AND GAS CHROMATOGRAPY ANALYSES
List of Tables
Table page
F-1. Data for field immunoassay analysis 75
F-2. Data for laboratory immunoassay analysis 82
F-3. Data for gas chromatography analysis 87
Abbreviations
* exceeds calibration
B benzene
C co-elution
DIL dilution
DCE c/s-1,2-dichloroethene
FB fluorobenzene
E ethylbenzene
nd not detected
T toluene
X xylene
73
-------
Sample ID Number coding scheme for the BTX SITE demonstration:
Location of Analysis
Analysis Type
Site Identification
Water Type
Replicate Number
F = Field
L = Laboratory
G = GC
I = Immunoassay
A = Site A
B = Site B, etc.
01 = Well 1
02 = Well 2, etc.
P1 = 2.5 ppb Standard
P2 = 25 ppb Standard
P3 = 100 ppb Standard
TB = Trip Blank
TL = Trip Audit Low
TH = Trip Audit High
W1 = 2.5 ppb Audit Sample
W2 = 25 ppb Audit Sample
W3 = 100 ppb Audit Sample
01 = 1st Analysis
02 = 2nd Analysis, etc.
Example = FIA0101
F = Field Location
I = Immunoassay Analysis
A = Site A
01 = Well Number 1
01 = Replicate Number 1
74
-------
I SAMPLING
|l DATE
1/22/92
1/22/92
1/22/92
1/22/92
1 /22/92
1 /22/92
1/22/92
1/22/92
1/22/92
1/22/92
1 /22/92
1 /22/92
1 /22/92
1 /22/92
1 /22/92
1 /22/92
1 /22/92
1/22/92
1/22/92
1 /22/92
1 /22/92
1 /22/92
1/22/92
1/22/92
1/23/92
1/23/92
1/23/92
1 /23/92
1/23/92
1/23/92
1/23/92
1/23/92
1 /23/92
1 /23/92
1 /23/92
1/23/92
1 /23/92
1/23/92
1/23/92
1/23/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
SAMPLE REFERENCE
ID NO. OD
FIA0401
FIA0402
FIA0403
FIA0601
FIA0602
FIA0603
FIA0701
FIA0702
FIA0703
FIA0901
FIA0902
FIA0903
FIA1001
FIA1002
FIA1003
FIA1101
FIA1102
FIA1103
FIAP101
FIAP102
FIAP201
FIAP202
FIAP301
FIAP302
FIA0101
FIA0102
FIA0201
FIA0202
FIA0301
FIA0302
FIA0501
FIA0502
FIA0801
FIA0802
FIAP103
FIAP104
FIAP203
FIAP204
FIAP303
FIAP304
FIA0101
FIA0102
FIA0103
FIA0201
FIA0202
1.746
1.836
1.809
1.746
1.836
1.809
1.408
1.984
1.865
1.408
1.984
1.865
1.746
1.836
1.809
1.408
1.984
1.865
1.506
1.880
1.506
1.880
1.506
1.880
1.766
1.900
1.766
1.900
1.972
1.898
1.766
1.900
1.972
1.898
1.319
1.977
1.319
1.977
1.319
1.977
1.640
1.754
1.817
1.440
1.847
SAMPLE
OD
1.854
1.961
>2
0.517
0.742
0.794
1.600
1.993
1.972
0.539
0.802
1.041
1.947
1.947
1.963
1.467
1.890
1.280
1.387
1.907
1.184
1.849
0.753
1.356
0.383
0.416
1.869
1.905
1.039
0.712
1.912
1.807
>2
1.992
1.451
1.981
1.255
1.809
0.876
1.592
0.434
0.434
0.484
1.580
1.833
S/R RUN ||
RATIO FLAGS DATE ||
1.06
1.07
*
0.30
0.40
0.44
1.14
1.00
1.06
0.38
0.40
0.56
1.12
1.06
1.09
1.04
0.95
0.69
0.92
1.01
0.79
0.98
0.50
0.72
0.22
0.22
1.06
1.00
0.53
0.38
1.08
0.95
*
1.05
1.10
1.00
0.95
0.92
0.66
0.81
0.26
0.25
0.27
1.10
0.99
Table F-1. Data for field immunoassay analysis. (Page 1 of 7)
75
-------
1 SAMPLING
DATE
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/26/92
2/26/92
SAMPLE
ID NO.
FIA0203
FIA0301
FIA0302
FIA0303
FIA0401
FIA0402
FIA0403
FIA0501
FIA0502
FIA0503
FIA0601
FIA0602
FIA0603
FIA0701
FIA0702
FIA0703
FIAP101
FIAP102
FIAP201
FIAP202
FIAP301
FIAP302
FIA0801
FIA0802
FIA0901
FIA0902
FIA1001
FIA1002
FIA1101
FIA1102
FIAP101
FIAP10X
FIAP201
FIAP20X
FIAP301
FIAP30X
FIB0101
FIB0102
FIB0201
FIB0202
FIBP102
FIBP202
FIBP302
FIB0301
FIB0302
Table
REFERENCE
OD
1.838
1.640
1.754
1.817
1.440
1.847
1.838
1.440
1.847
1.838
1.640
1.754
1.817
1.640
1.754
1.817
1.348
1.430
1.348
1.430
1.348
1.430
1.837
1.932
1.837
1.932
1.837
1.932
1.837
1.932
1.799
1.092
1.799
1.092
1.799
1.092
1.730
1.769
1.730
1.769
1.794
1.794
1.794
1.952
1.705
F-1. Continued
76
SAMPLE
OD
1.869
0.940
1.078
0.976
1.378
1.869
1.889
1.715
1.874
1.775
0.925
0.901
0.855
1.570
1.754
1.500
1.531
1.345
1.373
0.991
1.125
0.732
1.940
1.912
0.353
0.410
1.539
1.601
1.707
1.835
1.75
1.14
1.828
1.529
1,510
1.032
1.507
1.814
1.868
1.882
1.777
1.665
1.440
0.496
0.431
. (Page 2 of
SIR RUN II
RATIO FLAGS DATE II
1.02
0.57
0.61
0.54
0.96
1.01
1.03
1.19
1.01
0.97
0.56
0.51
0.47
0.96
1.00
0.83
1.14
0.94
1.02
0.69
0.83
0.51
1.06
0.99
0.19
0.21
0.84
0.83
0.93
0.95
0.97
1 .04 BAD RUN
1.02
1 .40 BAD RUN
0.84
0.95 BAD RUN
0.87
1.03
1.08
1.06
0.93
0.93
0.80
0.25
0.25
7)
-------
(SAMPLING
DATE
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
SAMPLE
ID NO.
FIB0401
FfB0402
FIB0501
FIB0502
FIB0701
FIB0702
FIB0901
FIB0902
FIB1001
FIB1002
FIBP101
FIBP102
FIBP10X
FIBP201
FIBP202
FIBP20X
FIBP301
FIBP302
FIBP30X
FIB1101
FIB1102
FIB1401
FIB 1402
FIB1501
FIB1502
FIB1601
FIB1602
FIB1701
FIB1 702
FIB1801
FIB1 802
FIB1901
FIB1 902
FIB2001
FIB2002
FIB2101
FIB2102
FIBP101
FIBP102
FIBP10X
FIBP201
FIBP202
FIBP20X
FIBP301
FIBP302
Table
REFERENCE
OD
1.844
1.796
1.844
1.796
1.952
1.705
1.952
.705
.952
.705
.882
.486
0.064
1.882
1.486
0.064
1.882
1.486
0.064
1.847
1.964
1.951
1.960
1.951
1.960
1.847
1.964
1.813
1.855
1.813
1.855
1.951
1.960
1.847
1.964
1.813
1.855
1.927
1.835
0.477
1.927
1.835
0.477
1.927
1.835
F-1. Continued
77
SAMPLE
OD
1.904
1.933
1.455
1.138
1.478
1.277
1.790
1.617
1.179
0.809
1.863
1.504
0.100
1.741
1.493
0.270
1.357
1.007
0.344
1.847
1.909
1.935
1.897
1.919
1.840
1.816
1.862
1.928
1.842
1.882
1.867
1.955
1.894
0.116
0.129
1.873
1.856
1.883
1.668
0.554
1.677
1.662
0.784
1.252
1.320
.(Page 3 of
SIR
RATIO
1.03
1.08
0.79
0.63
0.76
0.75
0.92
0.95
0.60
0.47
0.99
1.01
1.56
0.93
1.00
4.22
0.72
0.68
5.38
1.00
0.97
0.99
0.97
0.98
0.94
0.98
0.95
1.06
0.99
1.04
1.01
1.00
0.97
0.06
0.07
1.03
1.00
0.98
0.91
1.16
0.87
0.91
1.64
0.65
0.72
7)
RUN I
FLAGS DATE I
BAD RUN
BAD RUN
BAD RUN
BAD RUN
BAD RUN
-------
» SAMPLING
DATE
2/27/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
SAMPLE F
ID NO.
FIBP30X
FIB0801
FIB0802
FIB1201
FIB1202
FIB1203
FIB1301
FIB1302
FIBP101
FIBP102
FIBP201
FIBP202
FIBP301
FIBP302
FIC1601
FIC1602
FIC1701
FIC1702
FIC1801
FIC1802
FICW101
FICW102
FICW201
FICW202
FICW301
FICW302
FIC0801
FIC0802
FIC1001
FIC1002
FICW101
FICW102
FICW102
FICW201
FICW202
FICW202
FICW301
FICW302
FICW302
FIC0401
FIC0402
FIC0901
FIC0902
FIC0903
FIC2301
Table
REFERENCE S
OD
0.477
1.875
1.882
1.875
1.882
1.820
1.875
1.882
1.559
1.820
1.559
1.820
1.559
1.820
1.696
1.452
1.696
1.452
1.696
1.452
1.666
1.812
1.666
1.812
1.666
1.812
1.856
1.629
1.856
1.629
1.85
1.823
1.896
1.85
1.823
1.896
1.85
1.823
1.896
1.794
1.548
1.794
1.548
1.692
1.794
F-1 . Continued.
78
AMPLE
OD
0.627
0.938
0.895
1.655
1.505
1.391
1.907
1.746
1.809
1.720
1.656
1.578
1.331
1.238
0.099
0.097
0.129
0.068
0.315
0.223
1.708
1.668
1.562
1.522
1.434
1.352
0.247
0.177
0.129
0.142
1.888
1.729
1.855
1.833
1.566
1.75
1.683
1.482
1.394
1.703
1.357
1.68
1.246
1.485
1.273
(Page 4 of
S/R RUN I!
RATIO FLAGS DATE l|
1.31 BAD RUN
0.50
0.48
0.88
0.80
0.76
1.02
0.93
1.16
0.95
1.06
0.87
0.85
0.68
0.06
0.07
0.08
0.05
0.19
0.15
1.03
0.92
0.94
0.84
0.86
0.75
0.13
0.11
0.07
0.09
1.02
0.-95
0.98
0.99
0.86
0.92
0.91
0.81
0.74
0.95
0.88
0.94
0.80
0.88
0.71
7)
-------
1 SAMPLING
DATE
3/11/92
3/11/92
3/1 1 /92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/1 2/92
3/12/92
3/1 2/92
3/12/92
3/12/92
3/1 2/92
3/12/92
3/12/92
3/12/92
3/1 2/92
3/1 2/92
3/12/92
3/12/92
3/1 2/92
3/12/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/1 3/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
4/1 5/92
4/1 5/92
4/1 5/92
4/1 5/92
SAMPLE REFERENCE SAMPLE
ID NO. OD OD
FIC2302
FIC2401
FIC2402
FIC2601
FIC2602
FICW101
FICW102
FICW201
FICW202
FICW301
FICW302
FIC0601
FIC0602
FIC0701
FIC0702
FIC1101
FIC1102
FICW101
FICW102
FICW102
FICW201
FICW202
FICW202
FICW301
FICW302
FICW302
FIC0201
FIC0202
FIC0301
FIC0302
FIC0501
FIC0502
FICW101
FICW102
FICW1 02
FICW201
FICW202
FICW202
FICW301
FICW302
FICW302
FIA0101
FIA0102
FIA0201
FIA0202
Table
1.548
1.794
1.548
1.794
1.548
1.89
1.692
1.89
1.692
1.89
1.692
1.87
1.884
1.87
1.884
1.87
.884
.891
.725
1.88
.891
.725
1.88
1.891
1.725
1.88
1.857
1.937
1.857
1.937
1.857
1.937
.971
.934
.943
.971
.934
.943
1.971
1.934
1.943
1.773
1.968
1.773
1.968
F-1 . Continued.
79
1.014
1.693
1.498
0.1
0.06
1.887
1.756
1.867
1.392
1.615
1.241
0.819
0.817
1.846
1.831
0.154
0.219
1.938
1.823
1.899
1.949
1.772
1.795
1.665
1.363
1.632
0.306
0.48
0.141
0.161
1.129
1.271
1.998
1.948
1.922
1.936
1.835
1.837
1.726
1.573
1.778
0.03
0.065
1.914
1.975
(Page 5
S/R RUN I
RATIO FLAGS DATE |
0.66
0.94
0.97
0.06
0.04
1.00
1.04
0.99
0.82
0.85
0.73
0.44
0.43
0.99
0.97
0.08
0.12
1.02
1.06
1.01
1.03
1.03
0.95
0.88
0.79
0.87
0.16
0.25
0.08
0.08
0.61
0.66
1.01
1.01
0.99
0.98
0.95
0.95
0.88
0.81
0.92
0.02
0.03
1.08
1.00
of 7)
-------
» SAMPLING
DATE
4/15/92
4/1 5/92
4/15/92
4/15/92
4/15/92
4/1 5/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/1 5/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/1 6/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
SAMPLE REFERENCE SAMPLE
ID NO. OD OD
FIA0301
FIA0302
FIA0401
FIA0402
FIA0501
FIA0502
FIAW101
FIAW102
FIAW201
FIAW202
FIAW301
FIAW302
FIA0601
FIA0602
FIA0701
FIA0702
FIA0703
FIA0801
FIA0802
FIA0803
FIA0804
FIA0901
FIA0902
FIA0903
FIA0904
FIA1001
FIA1002
FIA1003
FIA1004
FIA1101
FIA1102
FIA1103
FIAW101
FIAW102
FIAW201
FJAW202
FIAW301
FIAW302
FID1001
FID 1002
FID1101
FID1102
FID1201
FID 1202
FIDW101
Table
1.773
1.968
1.773
1.968
1.773
1.968
1.862
1.777
1.862
1.777
1.862
1.777
1.95
1.935
1.95
1.935
1.944
1.974
0.658
1.877
1.375
1.974
0.658
1.877
1.375
1.974
0.658
1.877
1.375
1.95
1 .935
1.944
1.91
1.993
1.91
1.993
1.91
1.993
1.978
1.913
1.978
1.913
1.978
1.067
1.917
F-1. Continued.
80
0.142
0.19
1.868
1.914
1.81
1.86
1.883
.863
.848
.767
.481
.518
0.542
0.485
>2
>2
1.785
>2
1.425
1.936
1.312
0.164
0.131
0.107
0.057
1.063
0.536
0.818
0.539
>2
1.992
1.836
1.928
1.934
1.831
1.738
1.558
1.557
1.992
1.994
1.312
1.531
1.955
1.409
1.771
(Page 6
S/R
RATIO
0.08
0.10
1.05
0.97
1.02
0.95
1.01
1.05
0.99
0.99
0.80
0.85
0.28
0.25
0.92
2.17
1.03
0.95
0.08
0.20
0.06
0.04
0.54
0.81
0.44
0.39
1.03
0.94
1.01
0.97
0.96
0.87
0.82
0.78
1.01
1.04
0.66
0.80
0.99
1.32
0.92
of 7)
RUN I
FLAGS DATE li
•
*
•
BAD RUN
DIL 1 :2
BAD RUN
DIL 1:2
BAD RUN
DIL 1:2
*
DIL
-------
E SAMPLING
DATE
4/17/92
4/17/92
4/17/92
4/1 7/92
4/17/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
SAMPLE 1
ID NO.
FIDW102
FIDW201
FIDW202
FIDW301
FIDW302
FID0101
FID0101X
FID0 102
FID0201
FID0201X
FID0202
FID0301
FID0301X
FID0302
FID9901
FID9901X
FID9902
FIDW101
FIDW102
FIDW201
FIDW202
FIDW301
FIDW302
FID0501
FID0502
FID0801
FID0802
FID0901
FID0902
FIDW101
FIDW102
FIDW201
FIDW202
FIDW301
FIDW302
FID0401
FID0402
FID0701
FID0702
FIDW101
FIDW102
FIDW201
FIDW202
FIDW301
FIDW302
REFERENCE
OD
1.948
.917
.948
.917
.948
.905
2
.952
.905
2
1.952
1.905
2
1.952
1.905
2
1.952
.907
.985
.907
.985
.907
.985
.975
.825
.975
.825
.975
.825
.938
1.944
1.938
1.944
1.938
1.944
1.792
1.998
1.792
1.998
1.91
1.887
1.91
1.887
1.91
1.887
SAMPLE
OD
1.901
1.713
1.755
1.203
1.553
1.945
1.933
1.865
0.788
0.696
0.911
0.278
0.269
0.29
1.685
1.081
1.673
1.871
1.987
1.73
1.866
1.166
1.722
0.861
0.756
.927
.973
.949
.979
.948
1.92
1.839
1.9
1.368
1.563
0.981
1.263
0.013
0.045
1.817
1.835
1.778
1.803
1.36
1.558
S/R RUN I
RATIO FLAGS DATE |
0.98
0.89
0.90
0.63
0.80
1.02
*
0.96
0.41
*
0.47
0.15
*
0.15
0.88
*
0.86
0.98
1.00
0.91
0.94
0.61
0.87
0.44
0.41
0.98
1.08
0.99
1.08
1.01
0.99
0.95
0.98
0.71
0.80
0.55
0.63
0.01
0.02
0.95
0.97
0.93
0.96
0.71
0.83
Table F-1. Continued. (Page 7 of 7)
81
-------
1 SAMPLING
DATE
1/22/92
1/22/92
1/22/92
1/22/92
1/22/92
1/22/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/23/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/28/92
1/28/92
1/28/92
1/29/92
1/29/92
1/29/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
2/24/92
Table
SAMPLE REFERENCE
ID NO. OD
UA0401
LJA0402
LJA0601
LJA0602
UA0701
UA0901
UA0101
UA0101
UA0102
UA0102
UA0201
UA0201
UA0202
UA0202
UA0301
UA0301
UA0302
UA0302
LJATB01
UATB02
LJATH01
LJATH02
LJATL01
UATL02
UAP101
LIAP102
UAP201
UAP202
UAP301
UAP302
UATH01
LJATH02
UATL01
LJATL02
UAP101
UAP201
UAP301
UAP101
UAP201
UAP301
UA0101
UA0102
UA0201
LJA0202
UA0701
UA0702
UATB01
LJATB02
UATH01
UATH02
F-2. Data for
1.875
1.795
1.875
1.795
1.924
1.924
1.875
1.941
0.505
1.672
1.875
1.941
0.505
1.672
1.875
1.941
0.505
1.672
1.803
1.406
1.803
1.406
1.803
1.406
1.962
1.924
1.962
1.924
1.962
1.924
1.875
1.795
1.875
1.795
1.785
1.785
1.785
1.788
1.788
1.788
0.818
1.738
1.812
1.686
1.812
1.686
0.818
1.738
0.818
1.738
laboratory
SAMPLE S/R
OD RATIO FLAGS
1.900
1.862
0.648
0.810
>2
0.94
1.102
0.492
0.526
0.605
1.946
1.888
1.793
1.897
1.163
0.736
0.795
0.735
1.621
1.258
0.949
1.037
1.479
1.236
1.875
1.890
1.854
1.689
1.579
1.500
1.451
1.515
1.636
1.606
1.787
1.617
1.582
1.452
1.305
0.981
0.238
0.590
1.840
1.824
1.729
1.179
1.110
1.805
0.443
1.168
immunoassay
1.01
1.04
0.35
0.45
*
0.49
0.59
0.25
1.04 BAD RUN
0.36
1.04
0.97
3.55 BAD RUN
1.13
0.62
0.38
1.57 BAD RUN
0.44
0.90
0.89
0.53
0.74
0.82
0.88
0.96
0.98
0.94
0.88
0.80
0.78
0.77
0.84
0.87
0.89
1.00
0.91
0.89
0.81
0.73
0.55
0.29
0.34
1.02
1.08
0.95
0.70
1.36
1.04
0.54
0.67
analysis. (Page 1 of
RUN I
DATE!
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/28/92
1/29/92
1/28/92
1/29/92
1/28/92
1/29/92
1/28/92
1/29/92
1/28/92
1/29/92
1/28/92
1/29/92
1/28/92
1/28/92
1/28/92
1/28/92
1/28/92
1/28/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/27/92
1/28/92
1/28/92
1/28/92
1/29/92
1/29/92
1/29/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
5)
82
-------
! SAMPLING
DATE
2/24/92
2/24/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/25/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
3/3/92
3/3/92
3/3/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/9/92
3/10/92
SAMPLE REFERENCE SAMPLE
ID NO OD OD
LJATLQ1
LJATL02
UA1001
LJA1002
UATB01
UATB02
UATH01
UATH02
UATL01
UATL02
UB0301
UB0302
UB0401
UB0402
UBTB01
UBTB02
UBTH01
UBTH02
LJBTL01
UBTL02
UB2001
UB2002
UBTB01
UBTB02
UBTH01
UBTH02
UBTL01
UBTL02
UB1201
UB1202
UBTB01
UBTB02
UBTH01
UBTH02
UBTL01
UBTL02
UAP101
LIAP201
UAP301
LJC1701
UC1702
UCW101
UCW102
UCW201
UCW202
UCW301
UCW302
UTB101
LJTB102
UC1001
Table
0.818
1.738
1.812
1.686
1.918
1.835
1.918
1.835
1.918
1.835
1.270
1.580
1.593
1.566
1.270
1.580
1.270
1.580
1.270
1.580
1.593
1.566
1.694
1.688
1.694
1.688
1.694
1.688
1.593
1.566
-0.025
0.852
-0.025
0.852
-0.025
0.852
1.220
1.220
1.220
1.593
1.666
1.593
1.666
1.593
1.666
1.593
1.666
1.593
1.666
1.657
0.686
1.733
1.696
1.475
1.870
1.909
1.598
1.551
1.703
1.621
0.406
0.612
1.580
1.778
1.535
1.579
1.071
1.280
1.309
1..467
0.103
0.101
1.656
t.675
1.188
1.358
1.303
1.454
0.932
0.932
0.593
1.051
0.855
0.581
0.392
0.488
1.349
1.246
0.888
0.051
0.054
1.404
1.481
1.333
1.359
1.016
1.018
1.473
1.638
0.038
F-2. Continued. (Page
S/R
RATIO FLAGS
0.84
1.00
0.94
0.87
0.97
1.04
0.83
0.85
0.89
0.88
0.32
0.39
0.99
1.14
1.21
1.00
0.84
0.81
1.03
0.93
0.06
0,06
0.98
0.99
0.70
0.80
0.77
0.86
0.59
0.60
-23.72 BAD TUBE
1.23
-34.20 BAD TUBE
0.68
-15.68 BAD TUBE
0.57
1.11
1.02
0.73
0.03
0.03
0.88
0.89
0.84
0.82
0.64
0.61
0.92
0.98
0.02
2 of 5)
RUN II
DATE II
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/3/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/18/92
83
-------
» SAMPLING
DATE
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/10/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/11/92
3/12/92
3/12/92
3/12/92
3/12/92
3/12/92
3/12/92
3/12/92
3/12/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/13/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
4/15/92
4/15/92
4/15/92
SAMPLE REFERENCE SAMPLE
ID NO. OD OD
LJC1002
UCW101
LJCW102
LJCW201
LJCW202
UCW301
LJCW302
LJTB101
LJTB102
UC0901
UC0902
UC2301
UC2302
UCW201
UCW202
UCW301
UCW302
UCW101
UCW102
UCW201
UCW202
UCW301
UCW302
LJTB101
LJTB102
UC0301
LJC0302
UCW101
UCW102
UCW201
UCW202
UCW301
UCW302
UTB101
UTB102
UCW101
LJCW102
UCW201
UCW202
LJCW301
UCW302
UCW101
UCW102
UCW201
UCW202
UCW301
UCW302
UA0301
UA0302
UA0501
1.689
1.657
1.689
1.657
1.689
1.657
1.689
1.657
1.689
1.439
1.486
1.439
1.486
1.439
1.486
1.439
1.486
1.374
1.701
1.374
1.701
1.374
1.701
1.374
1.701
1.621
1.493
1.621
1.493
1.621
1.493
1.621
1.493
1.621
1.493
1.4
1.529
1.4
1.529
1.4
1.529
1.594
1.635
1.594
1.635
1.594
1.635
1.82
1.827
1.82
0.094
1.514
1.648
1.52
1.463
1.13
1.144
1.478
1.636
1.272
1.232
0.88
0.992
1.114
1.355
0.979
0.893
1.456
1.562
1.164
1.368
0.906
0.946
1.339
1.603
0.063
0.068
1.527
1.478
1.039
1.411
0.958
1.1
1.714
1.615
1.718
1.506
1.271
1.375
1.099
1.226
1.547
1.679
1.377
1.503
1.382
1.538
0.205
0.167
1.715
S/R
RATIO FLAGS
0.06
0.91
0.98
0.92
0.87
0.68
0.68
0.89
0.97
0.88
0.83
0.61
0.67
0.77 ICY
0.91
0.68 ICY
0.60
1.06
0.92
0.85
0.80
0.66
0.56
0.97
0.94
0.04
0.05
0.94
0.99 ICY
0.64
0.95 ICY
0.59
0.74 ICY
1.06
1.08 ICY
1.23
0.98
0.91
0.90
0.79
0.80
0.97
1.03
0.86
0.92
0.87
0.94
0.11
0.09
0.94
RUN II
DATE l|
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/19/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/18/92
3/19/92
3/19/92
3/19/92
3/19/92
^/ / w~
3/19/92
•»/ V j V —
3/19/92
/ f "~™
4/27/92
/ ™" / »•
4/27/92
4/27/92
Table F-2. Continued. (Page 3 of 5)
84
-------
[SAMPLING
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/15/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/16/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
4/17/92
A/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/20/92
4/21/92
4/21/92
SAMPLE REFERENCE SAMPLE
ID NO. OD OD
UA0502
LJDTB01
LJDTB02
UDTB03
UDW101
UDW102
UDW103
UDW201
UDW202
UDW203
UDW301
UDW302
LJDW303
LJA0601
LJA0602
UATB01
UAW101
UAW201
UAW301
UDTB02
UDW101
UDW202
UDW302
UD2301
LJDTB01
UDTB02
UOW101
LJDW102
UDW201
UDW202
UDW302
LJD0301
UD0302
LJD9901
UD9902
UD9903
UDTB01
UDTB02
UDTB03
UDW101
UDW102
UDW103
UDW201
UOW202
UDW203
UDW301
UDW302
UDW303
LJDTB01
UDTB02
Table
1.827
1.678
1.47
1.866
1.678
1.47
1.866
1.678
1.47
1.866
1.678
1.47
1.866
1.836
1.924
1.866
1.866
1.866
1.866
1.849
1.849
1.849
1.849
1.918
1.918
1.907
1.918
1.907
1.918
1.907
1.907
1.82
1.827
1.836
1.924
1.82
1.607
1.948
1.917
1.607
1.948
1.917
1.607
1.948
1.917
1.607
1.948
1.917
1.863
1.799
1.890
1.234
1.502
1.883
1.159
1.458
1.774
1.283
1.251
1.868
1.007
0.905
1.226
0.324
0.362
1.872
1.836
1.743
1.320
1.829
1.714
1.688
1.196
1.237
1.766
1.766
1.906
1.764
1.844
1.643
1.145
0.195
0.255
1.598
1.414
1.458
1.142
1.996
1.847
0.825
1.958
1.883
0.834
1.848
1.791
0.016
1.341
1.342
1.737
1.828
F-2. Continued. (Page
S/R
RATIO FLAGS
1.03
0.74
1.02
1.01
0.69
0.99
0.95
0.76
0.85
1.00
0.60
0.62
0.66
0.18
0.19
1.00
0.98
0.93
0171
0.99
0.93
0.91
0.65
0.64
0.92
0.93
0.99
0.93
0.96
0.86
0.60
0.11
0.14
0.87
0.73
0.80
0.71 BAD TUBE
1.02
0.96
0.51 BAD TUBE
1.01
0.98
0.52 BAD TUBE
0.95
0.93
0.01 BAD TUBE
0.69
0.70
0.93
1.02
i 4 of 5)
RUN II
DATE 1!
4/27/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/27/92
4/27/92
4/23/92
4/23/92
4/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/27/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
A/23/92
85
-------
1 SAMPLING
DATE
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/21/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/22/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
SAMPLE REFERENCE SAMPLE
ID NO. OD OD
LJDTB03
UDW101
UDW102
UDW103
UOW201
UOW202
UDW203
UDW301
UDW302
UDW303
LJD0701
LJD0702
UDTB01
UDTB02
UDW101
UDW102
UDW201
UDW202
UDW301
UDW302
UZTB01
UZTB02
UZW101
UZW102
UZW201
UZW202
LJZW301
UZW302
UZTB01
UZTB02
UZW101
UZW102
UZW201
UZW202
UZW301
UZW302
1.712
1.863
1.799
1.712
1.863
1.799
1.712
1.863
1.799
1.712
1.836
1.924
1.754
1.75
1.754
1.75
1.754
1.75
1.754
1.75
1.961
1.634
1.961
1.634
1.961
1.634
1.961
1.634
1.835
1.916
1.835
1.916
1.835
1.916
1.835
1.916
1.886
1.639
1.471
1.782
1.366
1.361
1.697
0.908
0.966
1.310
0.064
0.012
1.817
1.857
1.614
1.650
1,417
1.494
0.965
1.052
1.858
1.896
1.872
1.652
1.696
1.556
1.360
1.269
1.914
1.851
1.930
1.846
1.587
1.657
1.568
1.233
S/R
RATIO
1.10
0.88
0.82
1.04
0.73
0.76
0.99
0.49
0.54
0.77
0.03
0.01
1.04
1.06
0.92
0.94
0.81
0.85
0.55
0.60
0.95
1.16
0.95
1.01
0.86
0.95
0.69
0.78
1.04
0.97
1.05
0.96
0.86
0.86
0.85
0.64
FLAGS
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
LABSTD
RUN ||
DATE II
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
A/23/92
4/27/92
4/27/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
4/27/92
» /
4/27/92
Table F-2. Continued. (Page 5 of 5)
86
-------
SAMPLING SAMPLE RUN
DATE ID RUN DATE DIL
CONCENTRATION (ug/L)
B T E X
SURROGATE
% RECOVERY
CIS- 1,2
DCE FB
FLAGS
oo
1/22/92 LAGTH01
1/22/92 LAGTL01
1/22/92 LGA0401
1/22/92 LGA0601
1/22/92 LGA0701
1/22/92 LGA0901
1/22/92 LGA1001
1/22/92 LGA1101
1/22/92 LGATB01
1/23/92 LGA0101
1/23/92 LGA0201
1/23/92 LGA0301
1/23/92 LGA0501
1/23/92 LGA0801
1/23/92 LGATB01
1/23/92 LGATH01
1/23/92 LGATL01
2/24/92 LGA0101
2/24/92 LGA0201
2/24/92 LGA0301
2/24/92 LGA0401
2/24/92 LGA0501
2/24/92 LGA0601
2/24/92 LGA0701
2/24/92 LGATB01
2/24/92 LGATH01
2/24/92 LGATL01
2/25/92 LGA0801
2/25/92 LGA0901
2/25/92 LGA1001
2/25/92 LGA1101
2/25/92 LGATB01
266 1/27/92
265 1/27/92
275 1/27/92
277 1/27/92
273 1/27/92
271 1/27/92
268 1/27/92
276 1/27/92
264 1/27/92
289 1/28/92
286 1/28/92
295 1/28/92
287 1/28/92
288 1/28/92
282 1/28/92
284 1/28/92
283 1/28/92
772 3/1/92
727 2/28/92
773 3/1/92
729 2/28/92
726 2/28/92
774 3/1/92
730 2/28/92
728 2/28/92
759 2/29/92
736 2/28/92
731 2/28/92
771 3/1/92
737 2/28/92
732 2/28/92
733 2/28/92
1
1
1
20
1
5
1
1
1
5
1
20
1
1
1
1
1
10
1
50
1
1
50
1
1
1
1
1
10
1
1
1
nd
2.3
1.8
1820.3
nd
231
nd
6.5
2
320.3
nd
1393
nd
nd
nd
nd
nd
455.1
<1
889
<1
<1
983.9
1.2
<1
nd
2.4
1.7
305.4
1.1
1.3
<1
86.30
25.60
<1
220.80
nd
14.00
nd
2.10
nd
112.10
<1
<20
<1
<1
<1
89.70
28.70
41.10
<1
<50
<1
<1
<50
<1
<1
100.80
39.40
<1
22.40
2.50
<1
<1
nd
nd
nd
nd
nd
94.4
nd
1.9
nd
59.4
nd
90.1
nd
nd
nd
nd
nd
93.4
<1
<50
<50
nd
nd
nd
<1
nd
127.1
nd
nd
nd
nd
nd
49.3
nd
nd
nd
26
nd
nd
nd
nd
nd
nd
nd
17.9
nd
nd
nd
nd
nd
nd
nd
nd
nd
<1
54.7
nd
<1
nd
82
85
110
85
95
97
96
91
102
81
97
97
97
101
105
79
84
69
95
91
94
99
91
90
92
155
90
92
94
75
81
92
95
98
97
131
104
96
120
118
93
101
124
133
101
109
117
Table F-3. Data for gas chromatography analysis. (Page 1 of 5)
-------
SAMPLING SAMPLE RUN
DATE ID RUN DATE OIL
CONCENTRATION (ug/L)
B T E X
SURROGATE
% RECOVERY
CIS-1,2
DCE FB
FLAGS
oo
oo
2/25/92
2/25/92
2/25/92
2/25/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/26/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/27/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
2/28/92
3/9/92
LGATH01
LGATL01
LGB0101
LGB0201
LGB0301
LGB0401
LGB0501
LGB0701
LGB0901
LGB1001
LGBTB01
LGBTH01
LGBTL01
LGB1101
LGB1401
LGB1501
LGB1601
LGB1701
LGB1801
LGB1901
LGB2001
LGB2101
LGBTB01
LGBTH01
LGBTL01
LGB0801
LGB1201
LGB1301
LGBTB01
LGBTH01
LGBTL01
LGC1601
760
756
734
735
787
751
754
783
770
778
753
786
769
805
803
806
804
794
789
790
779
791
792
788
793
811
784
809
807
810
808
1073
2/29/92
2/29/92
2/28/92
2/28/92
3/1/92
2/29/92
2/29/92
3/1/92
3/1/92
3/1/92
2/29/92
3/1/92
3/1/92
3/2/92
3/2/92
3/2/92
3/2/92
3/2/92
3/1/92
3/1/92
3/1/92
3/2/92
3/2/92
3/1/92
3/2/92
3/2/92
3/1/92
3/2/92
3/2/92
3/2/92
3/2/92
3/14/92
1
1
1
1
5
1
1
1
1
20
1
2
1
1
1
1
1
1
1
1
20
1
1
2
1
50
1
1
1
2
1
40
1
1
1
1
5
1
1
1
1
20
1
2
1
1
1
1
1
1
1
1
20
1
1
2
1
50
1
1
1
2
1
40
nd
<1
<1
<1
106.9
<1
21.7
<1
28.7
382.4
<1
<1
<1
nd
<1
nd
nd
<1
<1
<1
1067.5
<1
<1
<2
1.2
1113.6
<1
<1
<1
2.1
1
1641.8
95.10
4.00
<1
<1
217.20
<1
5.00
<1
3.40
nd
<1
99.20
32.60
nd
nd
nd
nd
<1
<1
<1
90.10
<1
<1
110.20
42.80
20.00
<1
<1
nd
106.80
38.70
4,312.10
nd
nd
<1
<1
58.9
<1
28.8
<1
<1
<20
nd
nd
nd
nd
nd
nd
nd
<1
nd
nd
1047.6
nd
nd
nd
nd
nd
<1
nd
nd
nd
nd
689
nd
nd
nd
nd
93.8
<1
8.1
2.1
3.4
<20
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
225.5
nd
nd
nd
nd
nd
<1
nd
nd
nd
nd
864.8
82
92
91
92
98
99
101
100
90
78
93
91
94
98
101
98
101
91
92
93
96
92
93
91
89
91
92
98
97
91
97
107
92
97
110
100
125
103
124
110
104
81
97
98
103
99
103
99
103
99
97
99
297
97
98
97
93
91
101
99
97
92
98
295
Table F-3. Data for gas chromatography analysis. (Page 2 of 5)
-------
SAMPLING SAMPLE RUN
DATE ID RUN DATE DIL
CONCENTRATION (ug/L)
B T E X
SURROGATE
% RECOVERY
CIS- 1,2
DCE FB
FLAGS
00
CO
3/9/92 LGC1701
3/9/92 LGC1801
3/9/92 LGCW101
3/9/92 LGCW201
3/9/92 LGCW301
3/9/92 GCTB101
3/10/92 LGC0801
3/10/92 LGC1001
3/11/92 LGC2301
3/10/92 LGCW101
3/10/92 LGCW201
3/10/92 LGCW301
3/10/92 GCTB101
3/11/92 LGC0401
3/11/92 LGC0901
3/11/92 LGC2401
3/11/92 LGC2601
3/11/92 LGCW101
3/11/92 LGCW201
3/11/92LGCW301
3/11/92 GCTB101
3/12/92 LGC0601
3/12/92 LGC0701
3/12/92 LGC1101
3/12/92 LGCW101
3/12/92 LGCW201
3/12/92 LGCW301
3/12/92 GCTB101
3/13/92 LGC0201
3/13/92 LGC0301
3/13/92 LGC0501
3/13/92 LGCW101
1074 3/14/92
1064 3/13/92
1056 3/13/92
1057 3/13/92
1058 3/13/92
1054 3/13/92
1076 3/14/92
1075 3/14/92
1097 3/15/92
1059 3/13/92
1060 3/13/92
1061 3/13/92
1055 3/13/92
1106 3/15/92
1101 3/15/92
1117 3/16/92
1095 3/15/92
1105 3/15/92
1108 3/15/92
1107 3/15/92
1116 3/16/92
1098 3/15/92
1114 3/16/92
1096 3/15/92
1102 3/15/92
1103 3/15/92
1104 3/15/92
1115 3/16/92
1100 3/15/92
1099 3/15/92
1119 3/16/92
1081 3/14/92
Table F-3.
40
20
1
1
1
1
20
200
40
1
1
1
1
1
1
1
40
1
1
1
1
40
1
40
1
1
1
1
200
400
1
1
Data for
5762.2 4,661.60
443.3 228.30
nd 3.30
nd 30.00
nd 113.00
nd nd
395.6 200.40
7252 28.603.00
437.7 202.40
nd 3.10
nd 29.90
<1 113.90
nd nd
14.4 49.40
8.4 4.50
3.1 6.00
2304.4 6,515.80
2.3 9.40
1 32.00
2.3 119.80
nd 3.50
1271.8 313.50
1.5 4.60
5484.4 6,168.20
1.4 7.50
3.2 34.00
2.9 112.70
1.2 3.20
930.8 2,520.20
12359.6 25.060.40
144.6 28.90
< 1 3.80
gas chromatography
1825.8
136.1
nd
nd
nd
nd
68.4
3031
55.6
nd
nd
nd
nd
5.8
<1
nd
1097.6
nd
<1
<1
nd
51.4
<1
1038.2
nd
nd
nd
nd
277
1966.8
9.2
nd
analysis.
1796.6
124
nd
nd
nd
nd
109.4
5254
37.5
nd
nd
<1
nd
11
1.3
1.2
1404.5
<1
1.4
1.4
<1
39.5
<1
1059.4
<1
<1
1.4
<1
481.8
3675.6
8.7
nd
(Page 3 of 5)
106
106
102
106
87
115
98
105
102
106
103
97
114
101
119
107
113
101
92
98
97
105
111
109
102
101
94
105
110
104
102
98
484
132
99
104
87
108
147
379
101
102
102
102
106
108
100
116
511
100
96
106
94
103
104
329
99
100
99
101
118
115
133
97
F*
F*
-------
SAMPLING SAMPLE RUN
DATE ID RUN DATE DIL
CONCENTRATION (ug/L)
B T E X
SURROGATE
% RECOVERY
CIS- 1,2
DCE FB
FLAGS
-------
SAMPLING SAMPLE RUN
DATE ID RUN DATE DIL
CONCENTRATION (ug/L)
B T E X
SURROGATE
% RECOVERY
CIS-1,2
DCE FB
FLAGS
CO
4/20/92 LGD9902
4/20/92 LGDTB01
4/20/92 LGDW101
4/20/92 LGDW201
4/20/92 LGDW301
4/21/92 LGD0501
4/21/92 LGD0801
4/21/92 LGD0901
4/21/92 LGDTB01
4/21/92 LGDW101
4/21/92 LGDW201
4/21/92 LGDW301
4/22/92 LGD0401
4/22/92 LGD0701
4/22/92 LGW101
4/22/92 LGW201
4/22/92 LGW301
1474
1466
1467
1468
1484
1496
1481
1483
1482
1479
1480
1486
4/23/92
4/23/92
4/23/92
A/23/92
4/23/92
4/24/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
4/23/92
1
1
1
1
1
1
1
1
1
1
1
1
nd
nd
nd
nd
3.6
11.6
nd
nd
nd
nd
nd
2.2
<1
nd
2.80
23.60
92.40
2.50
nd
<1
nd
2.50
26.50
91.30
nd
nd
nd
nd
<1
94.4
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
<1
12.6
nd
nd
nd
nd
nd
nd
91
96
95
90
106
121
95
91
92
96
88
96
95
98
95
95
93
157
96
92
93
98
95
94
DC.FC
MISSING
MISSING
MISSING
MISSING
MISSING
Table F-3. Data for gas chromatography analysis. (Page 5 of 5}
------- |