EPA600/B-22/001 | January 2022 | www.epa.gov/research
Guidelines on Validation of
Non-Regulatory Chemical
and Radiochemical Methods
-------
EPA/600/B-22/001
Guidelines on Validation of Non-Regulatory Chemical and
Radiochemical Methods
Prepared by:
EPA Environmental Methods Forum (EMF)
Method Validation Workgroup
January 2022
Workgroup Members:
John Griggs, Chair
Office of Air and Radiation
William Adams
Office of Water
Stephen Blaze
Office of Land and Emergency Management
Margie St. Germain
Region 7
Jennifer Gundersen
Office of Research and Development
Adrian Hanley
Office of Water
Keith McCroan
Office of Air and Radiation
Anand Mudambi
Office of Research and Development
Barry Pepich
Region 10
Yaorong Qian
Office of Chemical Safety and Pollution Prevention
Robin Segall
Office of Air and Radiation
Kent Thomas
Office of Research and Development
Contributors:
Charles Cook
Region 7
Troy Strock
Office of Land and Emergency Management
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Preface
The goal of this document is to provide a consistent general approach for the validation
and communication of newly developed, adopted (e.g., from another agency, voluntary
consensus standard development body such as ASTM International, not previously
validated, etc.), or modified chemical and radiochemical methods for non-regulatory
purposes. Specifically, this document provides collected information on critical areas of method
performance assessment for validation studies. This document also introduces the following new
concepts:
• Lifecycle of a method, which identifies the typical activities a method goes through
during its development and use.
• Validation Design, which is a short descriptor indicating the number of laboratories
and matrices in a validation study.
• Method Validation Summary, which is designed to provide consistency in delivery
of summary method validation results in a concise, easy-to-prepare and share
format.
Use of this guidance will assist the Environmental Protection Agency (EPA) in
both validating methods for non-regulatory purposes and communicating the results in a
consistent manner, allowing them to be used or further developed for other purposes. It will
also serve to assist external parties that develop methods to communicate their method
validation results in a standard format to make comparisons between similar validation studies
easier.
This document was prepared by the Environmental Methods Forum (EMF) Method
Validation Workgroup. The EMF is a cross-Agency forum chartered under the EPA's Laboratory
Enterprise Council (LEC). For more information on the EMF and LEC, please go to:
https://www.epa.gov/labs/national-program-manager-regional-laboratories-activities
EPA extends its appreciation to Stephanie Buehler and Ryan James from Battelle for their
diligent work and support.
11
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Disclaimer
This document is intended to provide internal guidance to U.S. Environmental Protection
Agency (EPA) personnel engaged in method validation activities, and for the understanding of
those who use EPA methods. This document can also be used by external parties for informational
purposes (e.g., private and state laboratories, consensus standard bodies, etc.) engaged in method
validation efforts. This document is not in any way binding and EPA retains the discretion,
however, to adopt, on a case-by-case basis, approaches that differ from this guidance. The guidance
set forth in this document does not create any rights, substantive or procedural, enforceable by law
for a party in litigation with EPA or the United States.
The intent of the document is not to supersede established practices. It does not replace
existing validation practices used by the EPA national program offices for published EPA
methods. Rather, the intent is to collect information from various documents and assemble them
in one place. The use of mandatory language such as "must" and "require" in this guidance manual
reflects sound scientific practice and does not create any legal rights or requirements. The use of
non-mandatory language such as "may," "can" or "should" in this guidance does not connote a
requirement but does indicate EPA's strong preference for validation and peer review of methods
prior to publication for general use. EPA is publicly releasing this document to increase
transparency in agency activities and this document may provide useful information for external
parties engaged in method validation.
References within this document to any specific commercial product, process or service by
trade name, trademark, manufacturer or otherwise does not necessarily imply its endorsement or
recommendation by EPA. Neither EPA nor any of its employees make any warranty, expressed or
implied, nor assume any legal liability of responsibility for any third party's use, or the results of
such use, of any information, apparatus, product, or process disclosed in this manual, nor represent
that its use by such third party would not infringe on privately owned rights.
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Executive Summary
Method validation is an important aspect of establishing chemical and radiochemical
laboratory methods. EPA methods used for regulatory purposes rely on program-specific
assessment criteria and documentation to guide the conduct of method validation studies.
However, EPA also has wide-ranging needs for developing/modifying and validating laboratory
methods for non-regulatory purposes to address measurement gaps for both current and emerging
contaminants of concern.
The guidance in this document is specific to non-regulatory methods. It provides general
explanations and concepts collected from a variety of references from standard setting
organizations on critical areas of method validation performance assessment to provide guidance
on a consistent, general approach. While step-by-step guidance or a requirement for implementing
a method validation is not included here, basic method validation principles and possible areas of
assessment, as drawn from Agency and non-Agency programs and international programs and
guidance bodies, are provided.
These principles and areas of assessment are discussed according to the method lifecycle,
a novel concept that depicts the steps and processes involved with a method, from the method's
beginning by determining its need, to its retirement. A method is developed, depending on needs
and intended uses of data, and then validated, taking into account implementation considerations
such as holding times and cost.
Method validation can be conducted by a single laboratory or multiple laboratories
(interlaboratory) on a specific set of analytes in a defined matrix or matrices. Matrix variations
should also be part of a well-planned validation study. The number of matrices tested and
laboratories participating in a method validation study will vary and are not dictated or defined in
this document.
In order to appropriately validate a method, method performance characteristics are used.
Performance characteristics are a set of parameters that can directly and quantitatively assess the
performance of a method, demonstrating if it is fit for its intended purpose. The following typical
method performance characteristics are discussed in this document:
• Bias/Trueness
• Detection and Quantification Capability
• Instrument Calibration/Verification
• Measurement Uncertainty
• Precision
• Range
• Ruggedness
• Selectivity
iv
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Each of these performance characteristics are defined along with a discussion of their use
and implementation, as well as a list of relevant resources for more details and information.
Once completed, consistent and concise communication of method validation studies is
important to ensure an accurate and thorough understanding of the method's performance and
application. To this end, this document proposes the use of two new communication tools for
reporting method validation results: the Validation Design and the Method Validation Summary
to convey both the level of validation performed and pertinent information regarding the
validation:
• The Validation Design describes the validation in a succinct descriptor presented as
[aL, bM], where aL is the number (a) of laboratories (L) that participated in the method
validation, and bM is the number (b) of different matrices (M) that were used in the
method validation. It provides a standardized, easily reported, and easily understood
format to convey the level of validation performed based on the number of laboratories
that participated and the number of different matrices evaluated.
• The Method Validation Summary is a stand-alone table that provides a brief synopsis
of the method validation process and results. The Method Validation Summary is
intended to be placed at the front of a method validation report to provide the reader
with easy access to pertinent information concerning the method validation in a format
that is consistent across all reports. In this way, the Method Validation Summary will
allow for easier sharing of method validation results across the Agency and provide
consistency across all documents and offices.
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Table of Contents
Preface ii
Disclaimer iii
Executive Summary iv
Table of Contents vi
List of Acronyms viii
1 Introduction 1
1.1 Purpose 2
1.2 Intended Audience 2
1.3 Scope of Guidance 2
2 Activities and Processes Preceding Method Validation 4
3 Interlaboratory and Single Laboratory Method Validation Studies 5
3.1 Interlaboratory and Single Laboratory Considerations 5
3.2 Matrix Definition and Variability 6
4 Method Performance Characteristics 8
4.1 Bias/Trueness 8
4.2 Detection Capability and Quantification Capability 11
4.3 Instrument Calibration 17
4.4 Measurement Uncertainty 19
4.5 Precision 21
4.6 Range 23
4.7 Ruggedness 24
4.8 Selectivity in the Presence of Interferences 27
5 Statistical Evaluation and References 29
6 Method Validation Report 30
7 Consistently and Concisely Communicating Method Validation Studies 31
7.1 Validation Design 31
7.2 Method Validation Summary 32
8 References 34
vi
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Tables
Table 1. Cover template, with structure guidelines, for the Method Validation Summary 33
Figures
Figure 1. Diagram of the typical lifecycle of a method 3
Appendices
Appendix A: Method Validation Matrix Considerations for Individual EPA Offices
Appendix B: Analyte Detection and Quantitation
Appendix C: Detection and Quantitation Limit Definitions
Appendix D: Example Method Validation Summary and Associated Full Method Validation
Report
Appendix E: EPA Offices Method Validation References
Appendix F: Non-EPA Method Validation References
vii
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
List of Acronyms
ASME
American Society of Mechanical Engineers
CDC
Centers for Disease Control and Prevention
CFR
Code of Federal Regulations
CITAC
Co-operation on International Traceability in Analytical Chemistry
COA
certificate of analysis
CRM
certified reference material
CWA
Clean Water Act
EMF
Environmental Methods Forum
EPA
Environmental Protection Agency
EU
European Union
FEM
Forum on Environmental Measurements
GC/HSD
gas chromatography/halogen-specific detector
GC/MS
gas chromatography/mass spectrometry
GUM
Guide to the Expression of Uncertainty in Measurement
IDL
instrument detection limit
IEC
International Electrochemical Commission
ILS
interlaboratory study
ISO
International Organization for Standardization
IUPAC
International Union of Pure and Applied Chemistry
JCGM
Joint Committee for Guides in Metrology
LCMRL
lowest concentration minimum reporting level
LEC
Laboratory Enterprise Council
LLOQ
Lower Limit of Quantitation
LOD
limit of detection
LOQ
limit of quantification
MARLAP
Multi-Agency Radiological Laboratory Analytical Protocols
MDC
minimum detectable concentration
MDL
method detection limit
ML
minimum level
MRL
minimum reporting level
NIST
National Institute of Standards and Technology
NPDES
National Pollutant Discharge Elimination System
PQL
practical quantitation limit
viii
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
QA
quality assurance
ReMAP
Reference Method Accuracy and Precision
RM
reference material
SI
International System
SOP
standard operating procedure
UC MR
Unregulated Contaminant Monitoring Rule
UTL
Upper tolerance limit
IX
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
1 Introduction
Validation is defined as confirmation by examination and provision of objective evidence
that the requirements for a specific intended use or application have been fulfilled (7, 2). Method
validation applies the concept of validation to laboratory chemical and/or radiochemical methods.
Thus, per the International Organization for Standardization (ISO) definition, method validation
is the confirmation (verification and demonstration are considered as alternative terms here),
through the provision of objective evidence, that the requirements for a specific intended use or
application of a method have been fulfilled (7).
Different organizations have various definitions of method validation. One such definition
provides that method validation is basically the process of defining an analytical requirement and
confirming that the method under consideration has capabilities consistent with what the
application requires (3). Method validation is further defined as a process of demonstrating that
the method meets the required performance capabilities (4). That is, it makes use of a set of tests
that verify any assumptions on which the analytical method is based and establishes and documents
the performance characteristics of a method, thereby demonstrating whether the method is fit for
a particular analytical purpose (5). United States Environmental Protection Agency (EPA or 'the
Agency') and other documents have previously described method validation as the process of
demonstrating that an analytical method is suitable for its use and involves conducting a variety of
studies to evaluate method performance under defined conditions (3, 6).
EPA methods used for regulatory purposes cover a wide range of matrices, methodological
approaches, and objectives. Program-specific documentation guides the conduct of method
validation studies and development of assessment criteria for such studies. However, EPA also has
wide-ranging needs for laboratory methods (e.g., radiochemical and chemical) for non-regulatory
purposes. This includes a significant need for internal agency method development and method
validation from EPA Offices, Regions, and Programs, which are most aware of the Agency's
evolving needs to define new methods and modify existing ones to equip the EPA to best measure
contaminants of concern. Efforts by parties outside of EPA (e.g., private and state laboratories,
voluntary consensus standard bodies, etc.) may need to be leveraged by the Agency to help fill
gaps in method development and method validation to effectively and efficiently address both
current and emerging contaminants. Consistent processes and approaches for validating these new
non-regulatory laboratory methods are critical to ensuring that efforts by external parties are in
harmony with internal agency approaches and criteria, and that methods developed for non-
regulatory purposes can potentially be adapted for regulatory purposes as needed. This document
presents basic method validation principles and areas of assessment to consider or address when
validating laboratory methods intended for non-regulatory purposes. This information is based on
approaches and guidelines set forth in documents gathered across Agency programs, international
programs and guidance bodies, and non-Agency programs, such as state laboratories and non-EPA
federal agencies.
1
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
1.1 Purpose
The purpose of this document is to provide a consistent general approach for the validation
of newly developed, adopted (e.g., from another agency, voluntary consensus standard
development body such as ASTM International, not previously validated, etc.), or modified1
chemical and radiochemical methods for non-regulatory uses. Specifically, this document
provides collected information on critical areas of method performance assessment for validation
studies. This document introduces the following new concepts:
• Lifecycle of a method, which identifies the typical activities a method goes through
during its development and use.
• Validation Design, which is a short descriptor indicating the number of laboratories
and matrices in a validation study.
• Method Validation Summary, which is designed to provide consistency in delivery of
summary method validation results in a concise, easy-to-prepare and share format.
This document is not meant to provide prescriptive or step-by-step guidance on conducting
method validation studies. Instead, the intent of this document is to provide an overview of the
general principles and important areas of consideration for method validation and to provide lists
and, in some cases, links to more detailed resources (e.g., guides and standards, some of which
may need to be purchased, and Agency documents) to assist the user in conducting a method
validation study.
1.2 Intended Audience
This document is intended for use by internal EPA personnel engaged in method validation
activities, and for the understanding of those who use EPA methods. This document can also be
used by external bodies (e.g., private and state laboratories, voluntary consensus standard bodies,
etc.) engaged in method validation efforts.
1.3 Scope of Guidance
The lifecycle of a method starts with identifying a need for a method followed by several
activities, including defining the method purpose and the intended use of the data, method
development, implementation considerations, method validation, release/adoption, use, method
review/revision, and method retirement (see Figure 1). It should be noted that many activities in
the method lifecycle overlap and are interrelated. This document focuses on the validation of
laboratory methods for non-regulatory purposes. It is intended for use across different types of
chemical and radiochemical methods and provides a general approach for the validation of a newly
developed, adopted, or modified method as well as guidelines for communicating the validation
1 Method modifications that are within the accepted flexibilities of the applied method do not require
additional method validation; however, if modifications are made outside the accepted flexibilities of the
applied method, then users are responsible for ensuring that the modifications are documented and are
supported by a validation study that addresses those modifications.
2
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
performed in a consistent manner. The other activities of the method lifecycle are outside the scope
of this document, but some are briefly described in Section 2.0. In addition, sampling methods are
outside the scope of this document.
Figure 1. Diagram of the typical lifecycle of a method
(*) NOTE: Validation is not needed when non-technical edits, clarifications, or grammatical edits result in
the release of a method revision that does not change any technical aspects of the method protocol.
3
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
2 Activities and Processes Preceding Method Validation
Several activities of the method lifecycle precede method validation and some of these are
briefly described in this section. One of the first steps in developing a method involves defining
the reason for, or purpose of, the method. A method may be developed to identify and measure a
new or emerging analyte, achieve a lower detection capability, meet stricter quality control
objectives, or address any number of reasons or needs. The initial use of the data and, if applicable,
fitness for purpose criteria should be considered prior to undertaking any method development or
validation activities.
Method development and method validation are generally not viewed as completely
separate processes but rather are considered to be significantly interrelated. Method development
is often a complex iterative process. A detailed discussion of method development is outside the
scope of this document. However, there are several similarities between the types of experiments
conducted during method development and the tests performed during method validation. For
example, many of the same performance characteristics assessed in method validation (see Section
4) are evaluated to some extent during method development (3, 6).
Before method validation, several aspects related to method implementation are often
considered. These include, but are not limited to, holding times, sample preservation, cost, and
waste generation. These and other implementation aspects relate to how the method will be used
in a laboratory and could have implications for its use and applicability. A detailed discussion of
these and other method implementation issues is outside the scope of this document. However,
consideration should be given to including information on at least the implementation aspects
mentioned in this paragraph and in the Method Validation Summary (see Section 10).
All laboratory activities related to the development of a method including method
development and method validation activities should be conducted in accordance with the
laboratory's quality system. In general, this means compliance with an applicable Quality
Assurance Plan (e.g., https://www.epa.gov/qualitv/epa-qar-5~epa-requirements-qualitv-assurance-
proiect-plans) and any applicable standard operating procedures (SOPs).
4
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
3 Interlaboratory and Single Laboratory Method Validation Studies
3.1 Interlaboratory and Single Laboratory Considerations
An interlaboratory study, as defined by ASTM E691-19el, Standard Practice for
Conducting an Interlaboratory Study to Determine the Precision of a Test Method (7), measures
the variability of results when a test method is applied many times in multiple laboratories. Thus,
interlaboratory validation studies are designed to evaluate the performance of a method, in
particular its precision (e.g., reproducibility and repeatability), across multiple laboratories, when
the method is performed as written (6-8).
Laboratories that participate in an interlaboratory validation study should be representative
of the kind of laboratory expected to use the method and should be considered qualified to perform
the method (e.g., have qualified staff, correct instrumentation, and appropriate facilities to support
the conduct of the method) (7-9). Laboratories selected to participate in the interlaboratory
validation should not be limited to those that are exceptionally well qualified or equipped to
perform the method (7). Participating laboratories should be expected to conduct the method as
written, without deviations, but pay close attention to the flexibilities allowed within the method.
This is important to assessing the performance of the method in different settings with different
operators and in providing appropriate results for statistical analyses. If deviations from the
provided interlaboratory study protocol are made, they should be documented (6) and considered
when evaluating the study results.
Appropriate test materials should be selected for the interlaboratory validation study. The
matrices selected should be within the scope of the method and of the type expected to be
encountered when using the method. ASTM E1601-19, Standard Practice for Conducting an
Interlaboratory Study to Evaluate the Performance of an Analytical Method (9), recommends that
all test materials included in the expected scope of the method be included in the interlaboratory
validation. Samples of a certain matrix and concentration should be as homogenous as possible
prior to aliquoting into individual test samples and distributing to individual laboratories (7). The
AOAC International Official Methods of Analysis notes that any heterogeneity between test
samples generated from a single test material must be negligible as compared to any analytical
variability, so as not to be a factor in the test results (8). Samples should cover the range of
concentrations applicable to the method, and, if possible, should include levels near the upper and
lower limits of the established concentration range of the compound(s) being evaluated (9).
The number of laboratories that participate in an interlaboratory validation and the number
of samples that are used in each validation study may vary. The following discussion on number
and type of materials used in an interlaboratory study comes from ASTM E691-19el (7):
The number and type of materials to be included in an interlaboratory study (ILS)
will depend on the range of the levels in the class of materials to be tested and likely
relation of precision to level over that range, the number of different types of
materials to which the test method is to be applied, the difficulty and expense
5
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
involved in obtaining, processing, and distributing samples, the difficulty of, length
of time required for, and expense ofperforming the test, the commercial or legal
need for obtaining a reliable and comprehensive estimate of precision, and the
uncertainty ofprior information on any of these points.
Sometimes a formal collaborative study is not practical (4) and a single laboratory
validation is performed. This can be considered a special case of an interlaboratory validation
where only one laboratory conducts the validation activities. Most if not all of the same laboratory
tests performed in interlaboratory studies are performed during a single laboratory validation
study. However, single laboratory validations may need to consider potential limitations for some
validation activities. The International Union of Pure and Applied Chemistry (IUPAC) Technical
Report, Harmonized Guidelines for Single-Laboratory Validation of Methods of Analysis (5),
notes:
It is critically important in single-laboratory method validation to take account of
method bias and the laboratory effect. There are a few laboratories with special
facilities where these biases can be regarded as negligible, but that circumstance
is wholly exceptional. (However, if there is only one laboratory carrying out a
particular analysis, then method bias and laboratory bias take on a different
perspective.) Normally, method and laboratory effects have to be included in the
uncertainty budget, but often they are more difficult to address than repeatability
error and the run effect. In general, to assess the respective uncertainties it is
necessary to use information gathered independently of the laboratory. The most
generally useful sources of such information are (i) statistics from collaborative
trials (not available in many situations of single-laboratory method validation), (ii)
statistics from proficiency tests, and (Hi) results from the analysis of certified
reference materials.
When available, certified reference materials (CRMs) can be used in single laboratory
validations to assess laboratory and method bias in combination (5). If appropriate CRMs are not
available, a laboratory could alternatively use spiked samples (4, 5).
In addition, results from a single laboratory validation could be compared to published
results or statistics for the method (4) to further confirm the method performance. Also, to better
ascertain method bias, an internal round robin could be performed within the laboratory using
multiple qualified analysts following general interlaboratory validation guidelines (4).
3.2 Matrix Definition and Variability
Method validation is conducted for a specific analyte or set of analytes in a defined matrix
or matrices. When planning and conducting a method validation study, the matrix or matrix
variations that will be included in the validation need to be defined. The determination of what
matrices and associated variables to include in the method validation study depends on the
6
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
intended use and application of the final method and should be contemplated during method
development.
Matrix variability can have significant implications for method performance. The matrix
will have an impact on the sample from collection through analysis. Matrix variation has been
cited as one of the most important but least acknowledged sources of error in analytical
measurements (5). Variations in matrix can impact detection limits and introduce bias into
measurements, and thus are an important consideration of any method validation study. Matrix
variability should be an important factor in method ruggedness testing (see Section 7.7). As the
Multi-Agency Radiological Laboratory Analytical Protocols (MARLAP) Manual notes, samples
collected from different geographical regions or different processes may have differing
characteristics that can impact the performance of a method (10). Thus, the ruggedness of a
technique for handling variations of a matrix should be investigated, and matrix variations should
be a part of any well-planned interlaboratory collaborative study or single laboratory study (10).
Additional method application criteria or quality assurance measures (e.g., matrix or surrogate
spikes; internal and external calibration techniques; performance reference compounds) may be
needed to account for matrix impacts.
No universal guidelines or algorithms apply to all matrices that define how many matrices
or types of a matrix (e.g., different variations of soil) should be included in a method validation
study. However, individual EPA offices may have specific guidelines on matrix variability. Matrix
selection will be method-specific and should be based on the scope and method application. For
example, when validating a method for use nationally, one may need to include a larger number
of matrix variations to cover the range of anticipated method applications, while validation of a
method developed for samples from a specific or local site or region may need a much more limited
range of matrices for method validation. These aspects of the method should be considered when
determining the matrix variations to test during a method validation study and be included in the
method validation report. In addition, the use of varied matrices should be considered in
conducting different method performance characteristic evaluations (see Section 7). Appendix A
provides examples of method validation matrix considerations used by some EPA offices.
7
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
4 Method Performance Characteristics
The purpose, scope, and applicability of a method determine the method performance
characteristics necessary to properly validate the method. Performance characteristics offer a
defined and quantitative set of parameters against which a method can be validated, directly
assessing the method and demonstrating that it is fit for its intended purpose.
Typical method performance characteristics evaluated during a method validation study
include:
• Bias/Trueness
• Detection and Quantification Capability
• Instrument Calibration
• Measurement Uncertainty
• Precision
• Range
• Ruggedness
• Selectivity
The following subsections on each of the performance characteristics provide a definition
of the performance characteristic, a discussion of the use and implementation of the performance
characteristic, and a list of relevant resources for more details and information. The performance
characteristics are provided in alphabetical order which is not meant to imply order of importance.
The performance characteristics listed here and discussed in the following subsections are
not an exhaustive list but rather those that are typically used and found in various method validation
guidelines (3-5). Other performance characteristics may be applicable or specific to certain
methods and should be considered for evaluation as part of those method validation studies.
4.1 Bias/Trueness
Definition
The following definition for bias, specifically for methods, is from ASTM E177-20,
Standard Practice for Use of the Terms Precision and Bias in ASTM Test Methods (11).
Bias is the difference between the expectation of the test result, and an accepted
reference value.
ASTM Practice El77 also defines an alternative term for bias - trueness (11). The standard
notes that trueness has a more positive connotation. In some EPA programs, accuracy is used as a
synonym for bias, but this will not be how it is used in this document. More information about the
definition and use of these terms can be found using the following link:
https://www.itl.nist.eov/div898/handbook/mpc/sectionl/mi itm#:~: text=In %2 Oparti cut ar%
2C%20for%20a%20measurement.on%20the%20same%20test%20item.
8
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Discussion
The following general overview of the determination of bias for a method is taken from
the Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide to
Method Evaluation and Related Topics (3).
A practical determination of bias relies on comparison of the mean of the results
(x) from the candidate method with a suitable reference value (xret). The reference
value is sometimes referred as a 'true value' or a 'conventional true value '. Three
general approaches are available: a) analysis of reference materials; b) recovery
experiments using spiked samples, and c) comparison with results obtained with
another method. Bias studies should cover the method scope and may therefore
require the analysis of different sample types and/or different analyte levels. To
achieve this, a combination of these different approaches may be required.
The bias can be expressed in absolute terms
b = x — xref (Eq.l)
or relative in percent
b(%) = X 100 (Eq.2)
Xref
or as a relative spike recovery
R'(%) = X 100 (Eq.3)
xspike
where x' is the mean value of the spiked sample and xspike is the added
concentration.
However, in some sectors of analytical measurement, the relative recovery
('apparent recovery ) in percent is also used.
R(%) = — X 100 (Eq. 4)
xref
To determine the bias using reference material (RM), the mean and standard
deviation of a series of replicate measurements are determined and the results
compared with the assigned property value of the RM. The ideal RM is a certified
matrix reference material [CRM] with property values close to those of the test
samples of interest. CRMs are generally accepted as providing traceable values. It
is also important to remember that a RM should only be used for one purpose
during a validation study. For example, an RM usedfor calibration shall not also
be used to evaluate bias.
Compared to the wide range of sample types and analytes encountered by
laboratories, the availability of RM is limited, but it is also important that the
9
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
chosen material is appropriate to the use. It may be necessary to consider how the
RM was characterized (for example, if the sample preparation procedure used
during characterization of the material is not intended to give the total analyte
concentration, but the amount extracted under certain conditions). For regulatory
work, a relevant certified material (ideally matrix matched if available) should be
used. For methods used for long-term, in-house work, a stable in-house material
can be used to monitor bias, but a CRM should be used in the initial assessment.
In the absence of suitable RMs, recovery studies (spiking experiments) may be used
to give an indication of the likely level of bias. Analytes may be present in a variety
of forms in the sample, and sometimes only certain forms are of interest to the
analyst. The method may thus be deliberately designed to determine only a form of
the analyte. A failure to determine part of, or all the analyte present may reflect an
inherent problem with the method. Hence, it is necessary to assess the efficiency of
the methodfor detecting all the analyte present.
Because it is not usually known how much of a particular analyte is present in a
test portion, it is difficult to be certain how successful the method has been at
extracting it from the sample matrix. One way to determine the efficiency of
extraction is to spike test portions with the analyte at various concentrations, then
extract the spiked test portions and measure the analyte concentration. The
inherent problem with this is that an analyte introduced in such a way will probably
not be bound as strongly as that which is naturally present in the test portion matrix
and so the technique will give an unrealistically high impression of the extraction
efficiency.
It may be possible to assess bias by comparing results from the candidate method
with those obtainedfrom an alternative method.
There are challenges associated with evaluating bias. The following discussion of some of
the challenges, which parallels the discussion on the three approaches to determine bias in the
Eurachem Guide (3), is taken from Validation and Peer Review of U.S. Environmental Protection
Agency Chemical Methods of Analysis (6):
The reference material approach introduces the important uncertainty of matrix
matching, and the reliability of a bias estimate depends upon the relationship
between the composition of the reference material and the samples.
When using the alternate method approach, the reliability of a bias estimate is
dependent on how much is known about the performance characteristics of the
alternate method.
10
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
The matrix spiking approach introduces uncertainty regarding the behavior of
spiked materials, compared to materials containing native analyte. This may be
particularly problematic for solid materials.
RMs are a critical tool in method validation, particularly in regard to assessing method
bias. RMs are also important in instrument calibration and are described in more detail for this
application in Section 4.3.
Useful Resources
The following resources can provide further details and information on bias/trueness.
• ASTM E177-20, Standard Practice for Use of the Terms Precision and Bias in ASTM
Test Methods (11)
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics (3)
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analysis (5)
• ASTM E1601-19, Standard Practice for Conducting an Interlaboratory Study to
Evaluate the Performance of an Analytical Method (9)
4.2 Detection Capability and Quantification Capability
Definition
For both analyte detection capability and quantification capability, a variety of definitions
and terms are used by EPA, as well as the national and international analytical/metrology
community. Regarding detection capability, the IUPAC Technical Report, Harmonized Guidelines
for Single-Laboratory Validation of Methods of Analysis (5), notes:
There are several possible conceptual approaches to the subject, each providing a
somewhat different definition of the limit. Attempts to clarify the issue seem ever
more confusing.
If a method is being validated for use in a particular EPA program area, that program area's
definitions, terms, and calculational procedures related to detection capability and quantification
capability should be incorporated into the method validation process. If appropriate, the method-
specific detector and related systems should be identified.
11
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Discussion
The following general overview of the concepts of detection capability and quantification
capability is from the Eurachem Guide (3).
Where measurements are made at low concentrations, there are three general
concepts to consider. First, it may be necessary to establish a value of the result
which is considered to indicate an analyte level that is significantly different from
zero. Often some action is required at this level, such as declaring a material
contaminated. This level is known as the 'critical value', 'decision limit', or in
[European Union] EU directives, CCa.
Second, it is important to know the lowest concentration of the analyte that can be
detected by the method at a specific level of confidence. That is, at what true
concentration will we confidently exceed the critical value described above? Terms
such as 'limit of detection' (LOD), 'minimum detectable value ', 'detection limit',
or, in EU directives, CCb are used for this concept.
Third, it is also important to establish the lowest level at which the performance is
acceptable for a typical application. This third concept is usually referred to as the
limit of quantification (LOQ).
More detailed discussions on the concepts of detection capability and quantification
capability are provided in Appendix B.
While it is beyond the scope of this document to include a detailed discussion of the
different definitions and terms for detection capability and quantification capability used by EPA
and the analytical/metrology community, a few examples have been included to provide greater
context for this discussion.
Method Detection Limit (MDL). The U.S. EPA Office of Water MDL procedure is worth
noting because it is codified in federal regulations (40 Code of Federal Regulations [CFR] Part
136, Appendix B) (12). The MDL is required for most chemical analyses that support National
Pollutant Discharge Elimination System (NPDES) permits. It would be appropriate to use the
MDL when validating most methods that are applicable to wastewater or Clean Water Act (CWA)
compliance monitoring. Types of methods to which the MDL does not apply are detailed in the
Scope and Application section of the MDL procedure (72):
The MDL procedure is not applicable to methods that do not produce results with
a continuous distribution, such as, but not limited to, methods for whole effluent
toxicity, presence/absence methods, and microbiological methods that involve
counting colonies. The MDL procedure also is not applicable to measurements
such as, but not limited to, biochemical oxygen demand, color, pH, specific
conductance, many titration methods, and any method where low-level spiked
samples cannot be prepared. Except as described in the addendum, for the purposes
12
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
of this procedure, "spiked samples " are prepared from a clean reference matrix,
such as reagent water, spiked with a known and consistent quantity of the analyte.
MDL determinations using spiked samples may not be appropriate for all
gravimetric methods (e.g., residue or total suspended solids), but an MDL based
on method blanks can be determined in such instances.
The EPA MDL procedure uses low-level spikes of a clean reference matrix and method
blanks to calculate a detection limit. The procedure defines the MDL as (12):
the minimum measured concentration of a substance that can be reported with 99%
confidence that the measured concentration is distinguishable from method blank
results.
It is important to note that the MDL incorporates every step of the analytical method,
including extractions and any mandatory cleanups, not just the final instrumental analysis.
The MDL procedure has two subcategories: an initial MDL and an ongoing MDL. The
initial MDL is used when a laboratory is first implementing a method, or if the laboratory does not
already have data available to calculate an MDL. The ongoing MDL is used for analyses that are
run routinely, using low-level spike data that are collected quarterly along with routinely collected
method blank data. The MDL(s) is calculated using the spike data. The MDL(b) is calculated using
the blank data. The final MDL is higher of the two MDL(s) and MDL(b) calculations. The
procedure is relatively short, seven pages, and is available to the public at 40 CFR Part 136 in the
eCFR at
https://www.ecfr.eov/cei-bin/text-
d!\ h U ? '<0. I m i 101> > \ _ * _ * ••'. An MDL frequent questions webpage is also
available at https://www.epa.eov/cwa-methods/method-detection-limit-firequent-questions.
In the EPA Office of Air and Radiation's ambient air monitoring program, "detection
limit" is defined in the quality assurance guidance handbook as "[t]he lowest concentration or
amount of the target analyte that can be determined to be different from zero by a single
measurement at a stated level of probability" (13). In addition, the MDL procedure (12) is used
in the ambient air monitoring program Photochemical Assessment Monitoring (14). The ambient
air monitoring data collected by the states is submitted as measured, even if below the MDL.
In the EPA Office of Air and Radiation's stationary source regulatory program which
includes the New Source Performance Standards of 40 CFR Part 60 and the National Emissions
Standards for Hazardous Air Pollutants of 40 CFR Parts 61 and 63, the "limit of detection" is
defined in Method 301, the method validation protocol, as "the minimum concentration of a
substance that can be measured and reported with 99 percent confidence that the analyte is greater
than zero" (75). The Part 136, Appendix B MDL procedure (12) is also specifically referenced by
several of the standards as well as_Method 301 (75). Neither the ambient air monitoring program
nor the station source program utilizes the LOQ in data reporting.
13
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Limit of Detection (LOD) is used in some EPA methods and is recognized by several
standards organizations such as ISO and ASTM International as a means to express the detection
capability of a method. LOD is defined as the lowest concentration of the analyte that can be
detected by the method at a specified level of confidence. ISO standards note that the LOD should
be estimated taking both type I (alpha [a]) and type II (beta [|3]) errors into account (see Appendix
B). IUPAC recommends default values for alpha error and beta error equal to 0.05 (5). Other
approaches to estimating the LOD involve calculating the standard deviation of replicate
measurements of blank samples or replicate measurements of test samples with low concentrations
of the analyte in which the LOD equals some multiple (e.g., 3 times) of the standard deviation.
Given that the definition of the LOD is somewhat general and that there are different means of
calculating the LOD, it is critical that a precise definition of LOD be clearly stated during a method
validation effort and the means of calculating the LOD be fully documented.
Minimum Detectable Concentration (MDC) is used in most EPA radiochemical
methods (drinking water methods are a notable exception) and is used extensively in the
environmental radiochemistry community as a means to express the detection capability of a
method. The following definition of the MDC is taken from the MARLAP Manual (10):
The minimum analyte concentration that must be present in a sample to give a
specified probability, l-p, of measuring a response greater that the critical value,
leading one to conclude correctly that there is analyte in the sample.
The value of P that appears in the definition above, like a, is usually chosen to be 0.05 or
is assumed to be 0.05 by default if no value is specified.
Lowest Concentration Minimum Reporting Level (LCMRL) and Multi-Laboratory
Minimum Reporting Level (MRL) are used in the EPA Unregulated Contaminant Monitoring
Rule (UCMR) Program, a national occurrence study of unregulated contaminants in drinking
water. The LCMRL is defined in Statistical Procedures for Determination and Verification of
Minimum Reporting Levels for Drinking Water Methods (16) as:
A single laboratory Lowest Concentration Minimum Reporting Level (LCMRL) is the
lowest true concentration for which the future recovery is predicted to fall between 50% to
150% with 99% confidence.
It is a statistically-based quantitation procedure that accounts for both precision and
accuracy. The determined concentration is based on a statistical calculation by a single laboratory
where multiple concentration replicates are processed through the entire analytical method and the
data are plotted as measured sample concentration (y-axis) versus true concentration (x-axis). If
the data support an assumption of constant variance over the concentration range, an ordinary
least-squares regression line is drawn; otherwise, a variance-weighted least-squares regression is
used. Prediction interval lines of 99% confidence are drawn about the regression. At the points
14
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
where the prediction interval lines intersect with data quality objective lines of 50% and 150%
recovery, lines are dropped to the x-axis. The higher of the two values is the LCMRL.2
The multi-lab oratory MRL is a statistical calculation based on the incorporation of LCMRL
data collected from multiple laboratories into a 95% one-sided confidence interval on the 75th
percentile of the predicted distribution referred to as the 95-75 upper tolerance limit (UTL). This
effectively means that 75% of participating laboratories will be able to meet a set MRL. with a 95%
confidence interval. The statistical parameters of the multi-laboratory MRL are based on the EPA
established practical quantitation limit (PQL) determination, where the PQL is set at the
concentration that 75% of participating laboratories are predicted to meet an analyte's acceptance
criteria using a specified regression procedure. The calculation itself is defined in Technical Basis
for the LCMRL Calculator (77):
The MRL is calculated in three steps whenever there are three or more laboratories
providing data with valid LCMRLs or calculated LCMRLs that are below the lowest non-
zero spiking level. In the first step, 200 BB LCMRL replicates are calculated for each
laboratory data set. In the second step a predicted distribution of some unknown and yet
to be observed laboratory is built from the population of replicate laboratory LCMRLs
using a random effects model. In the third and last step the MRL is taken to be the upper
95% one-sided confidence interval on the 75th percentile of the predicted distribution
referred to as the 95-75 upper tolerance limit (95-75 UTL).
Minimum Level (ML) refers to either the sample concentration equivalent to the lowest
calibration point in a method or a multiple of the MDL, whichever is higher. Minimum levels can
be obtained in several ways: they may be published in a method; they may be based on the lowest
acceptable calibration point used by a laboratory; or they may be calculated by multiplying the
MDL in a method, or the MDL determined by a laboratory, by a factor of three. For the purposes
of NPDES compliance monitoring (under the CWA), EPA considers the following terms to be
synonymous: quantitation limit, reporting limit, and minimum level (18-20).
Lower Limit of Quantitation (LLOQ). EPA's Office of Resource Conservation and
Recovery (part of Office of Land and Emergency Management) publishes the SW-846
Compendium, which uses the LLOQ concept for quantitative analysis (21). The SW-846
Compendium defines the LLOQ as the lowest concentration at which the laboratory has
demonstrated that a target analyte can be reliably measured and reported with a certain degree of
confidence (21). The LLOQ must be greater than or equal to the lowest initial calibration standard
concentration, and each laboratory is required to establish and periodically verify LLOQs at
concentrations at which both qualitative and quantitative requirements can routinely be met using
the instrumentation, equipment, reagents, supplies, and personnel specific to that laboratory.
2 EPA provides software that will calculate LCMRL values. Instructions regarding the download
and use of the software can be found at the following URL:
https://www.epa.gov/dwanalvticalmethods/lowest-concentration-minimum-reporting-level-
Icmrl-calculator
15
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
LLOQs are established or verified with spiked blanks or representative sample matrices prepared
at or near expected target analyte LLOQs, and these quality control samples are processed through
all sample preparation and analysis steps used for field samples. SW-846 methods provide default
acceptance criteria for establishing or verifying LLOQs, and laboratories are encouraged to use
statistically-based limits once they have acquired sufficient data (21). SW-846 methods also
recommend including LLOQ verifications on a project-specific basis, as needed (e.g., to evaluate
the potential for measurement bias when decision limits are near established LLOQs), and SW-
846 methods defer to project planning documents regarding whether and how to report
concentrations below the LLOQ in field samples (21).
Appendix C is a compilation of most of the terms used in EPA methods for analyte
detection capability and quantification capability.
Useful Resources
The following resources can provide further details and information on detection capability
and quantification capability.
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analyses (5)
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics (3)
• 40 CFRPart 136, Appendix B (12)
• Multi-Agency Radiological Laboratory Analytical Protocols (MARLAP) Manual (10)
• Currie, L.A. Limits for Qualitative Detection and Quantitative Determination:
Application to Radiochemistry (22)
• Currie, L.A., Detection: Overview of Historical, Societal, and Technical Issues, in
Detection and Analytical Chemistry (23)
• Currie, L.A., Presentation of the Results of Chemical Analysis in IUPAC Compendium
of Analytical Nomenclature (24)
• Currie, L.A. Quality Assurance of Analytical Processes, in IUPAC Compendium of
Analytical Nomenclature (25)
• Lanier, S. W., Hendrix, C. D. Reference Method Accuracy and Precision (ReMAP):
Phase 1 (26)
16
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
4.3 Instrument Calibration
Definition
Instrument calibration refers to the procedures used for correlating instrument response to
an amount of analyte (concentration or other quantity) using measurements of suitable RMs.
Discussion
An instrument calibration approach is established for a particular application during
method development and is confirmed during single laboratory and/or multi-laboratory method
validation. There are two major components of instrument calibration: calibration approach and
RMs, both of which are described below.
Calibration Approach. The calibration approach often includes the selection and
application of a calibration model. The following excerpt on calibration model is from Validation
and Peer Review of U.S. Environmental Protection Agency Chemical Methods of Analysis (6):
The characteristics of a calibration function and justification for a selected
calibration model should be demonstrated during an intra-laboratory method
validation study. The performance of a calibration technique and the choice of
calibration model (e.g., first-order linear, curvilinear, or nonlinear) are criticalfor
minimizing sources of instrument bias and optimizing precision. A calibration
model is a mathematical function that relates composition to instrument response.
The parameters of the model are usually estimated from the responses of known,
pure analytes. Calibration errors can result from failure to identify the best
calibration model; inaccurate estimates of the parameters of the model; or
inadequately studied, systematic effects from matrix components.
The following excerpt on calibration is from ASTM E2857-11, Standard Guide for
Validating Analytical Methods (4):
Methods require calibration using measurements of suitable reference materials
and mathematical fitting of the measured responses to an algorithm, that is, an
equation thought to describe adequately the relationship between the amount of
analyte and the measured response. Algorithms are almost always an
approximation of the real world, and as such, their ability to fit the data has limits
that can be tested by a variety of means.
The process of method validation includes evaluation of the mathematical model or models
used for instrument calibration. However, it is beyond the scope of this document to describe all
possible models or algorithms that might be used for calibration, or the approaches that might be
used for their evaluations.
17
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Reference Materials (RMs). RMs are often used to verify instrument calibration. In ISO/
International Electrochemical Commission (IEC) 17025, General Requirements for the
Competence of Testing and Calibration Laboratories (2), the standard notes that RMs shall, where
possible, be traceable to SI units through a National Metrology Institute (e.g., National Institute of
Standards and Technology [NIST] in the United States) or, if that is not possible, then traceable to
CRMs. It also states that internal RMs shall be checked as far as technically and economically
practicable.
The following excerpt on RMs and CRMs is from the Eurachem Guide (3):
RMs can be virtually any material used as a basis for reference, and could include
laboratory reagents of known purity, industrial chemicals, or other artefacts. The
property or analyte of interest needs to be stable and homogenous, but the material
does not need to have the high degree of characterization, metrological
traceability, uncertainty and documentation associated with CRMs.
The characterization of the parameter of interest in a CRM is generally more
strictly controlled than for an RM, and in addition the characterized value is
certified with a documented metrological traceability and uncertainty.
There are generally three options for suitable RMs for instrument calibration: 1) CRMs; 2)
RMs with traceability to CRMs; and 3) RMs from other sources. For chemical analysis, it is often
the case that CRMs are not available, but it may be possible to obtain RMs with traceability to
CRMs from a manufacturer or the laboratory conducting the method validation may prepare RMs
with traceability to CRMs. In many cases, it may not be possible to obtain or produce RMs with
traceability to CRMs. In those instances, the laboratory conducting the method validation may
prepare internal RMs from other sources or obtain them from a manufacturer. These RMs, prepared
from other sources should have a supporting certificate of analysis (COA) and should meet certain
purity acceptance criteria based on the intended use of the method.
Given that there are significantly fewer radiochemical analytes than chemical analytes,
CRMs are much more likely to be available for the calibration of radiation instruments for a
radiochemical method validation, though CRM availability for chemical analytes may also be
related to difficulties in synthesizing or purifying a chemical of interest. The radiochemical CRMs
can be used directly for instrument calibration, or as is most often the case, RMs with traceability
to CRMs can be used for calibration.
Useful Resources
The following resources can provide further details and information on instrument
calibration.
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics (3)
18
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analysis (5)
• ISO/IEC 17025 General Requirements for the Competence of Testing and Calibration
Laboratories (2)
• ASTM E2857-11, Standard Guide for Validating Analytical Methods (4)
• Validation and Peer Review of U.S. Environmental Protection Agency Chemical
Methods of Analysis (6)
4.4 Measurement Uncertainty
Definition
The Joint Committee for Guides in Metrology (JCGM) Guide to the Expression of Uncer-
tainty in Measurement (27), often abbreviated as GUM, defines measurement uncertainty as
follows:
a parameter, associated with the result of a measurement, that characterizes the
dispersion of the values that could reasonably be attributed to the measurand
NOTE 1 The parameter may be, for example, a standard deviation (or a given
multiple of it), or the half-width of an interval having a stated level of confidence.
NOTE 2 Uncertainty of measurement comprises, in general, many components.
Some of these components may be evaluatedfrom the statistical distribution of the
results of series of measurements and can be characterized by experimental
standard deviations. The other components, which also can be characterized by
standard deviations, are evaluated from assumed probability distributions based
on experience or other information.
NOTE 3 It is understood that the result of the measurement is the best estimate of
the value of the measurand, and that all components of uncertainty, including those
arising from systematic effects, such as components associated with corrections
and reference standards, contribute to the dispersion.
Discussion
The result of a measurement is never exactly equal to the true value of the measurand (the
particular quantity subject to measurement). The difference between the result and the true value
is called the error of the measurement. Since the true value is always unknown, so is the error;
however, a properly determined measurement uncertainty allows one to put bounds on the likely
magnitude of the error. In conjunction with the measurement result, the uncertainty allows one to
find bounds for the most likely values of the measurand. In this regard, it is conceptually similar
to the "margin of error" that is commonly reported with statistical polling results.
19
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
When one follows the guidance of the GUM (27), the uncertainty of a measurement is
expressed first as a standard deviation, called the combined standard uncertainty. The combined
standard uncertainty may be multiplied by a coverage factor, k, to obtain an expanded uncertainty,
which describes an interval about the measurement result that is believed to contain the true value
with high confidence. The most commonly used coverage factor is k = 2, which is assumed to
provide approximately 95 % confidence.
The uncertainty of a measurement is intended to describe the quality of that measurement
and should be considered when making decisions about the true value of the quantity being
measured (for example, when comparing a single measurement result to an action level). It can
also be useful when assessing whether a measurement process is producing results of the expected
or required quality. Measurement uncertainties may or may not be considered when using
statistical tests to make decisions about sampled populations.
All or nearly all methods of radiochemical analysis used in the US include procedures for
calculating measurement uncertainties. It has become common for laboratories to specify
requirements for radioactive standards, calibrations, and calibration verifications in terms of their
measurement uncertainties and also to include measurement uncertainties in the evaluation of
radiochemical method quality control parameters.
The Stationary Source and Ambient Air Monitoring Programs in EPA's Office of Air and
Radiation, Office of Air Quality Planning & Standards utilize uncertainty in qualifying RMs used
in ambient air and pollutant emissions measurements. The regulatory programs determine the
quality (uncertainty) of the gas needed for a particular application, and the EPA Traceability
Protocol for Assay and Certification of Gaseous Calibration Standards (28) sets forth the
procedures to be followed in certifying the uncertainty of the gases manufactured for use in
meeting the program requirements.
Measurement uncertainty is most properly understood as a property of a measurement, not
a measurement method or even a measurement process. However, an analytical method can still
be evaluated in terms of the measurement uncertainty that it is expected to be achieved when used
for analysis of samples at specified analyte levels under specified measurement conditions. Note
that the predicted uncertainty may vary with the analyte level and may depend on other factors,
including interferences. Predicting the uncertainty for a hypothetical measurement requires making
assumptions about all such factors.
Measurement uncertainty accounts for the effects of both random and systematic
measurement errors, that is, for both imprecision and bias in the measurement process. (See
Sections 4.5 and 4.1, which describe precision and bias for chemical and radiochemical methods.)
When an analytical method provides estimates of measurement uncertainty, the validation process
for the method can test the plausibility of the uncertainty estimates by comparing them to either
the estimated precision of the method or its estimated root-mean-squared error. More extensive
and rigorous testing of the uncertainty can be based on a large number of repeated analyses of the
20
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
same material, accounting not only for the uncertainty of each result but also for estimated
measurement correlations among the results.
Useful Resources
The following resources can provide further details and information on measurement
uncertainty.
• JCGM 100:2008 Evaluation of Measurement Data: Guide to the Expression of
Uncertainty in Measurement (27)
• JCGM 101:2008 Evaluation of Measurement Data: Supplement 1 to the Guide to the
Expression of Uncertainty in Measurement — Propagation of Distributions Using a
Monte Carlo Method (29)
• JCGM 102:2011 Evaluation of Measurement Data: Supplement 2 to the Guide to the
Expression of Uncertainty in Measurement — Extension to Any Number of Quantities
(30)
• Eurachem/Co-Operation on International Traceability in Analytical Chemistry
(CITAC) Guide, Quantifying Uncertainty in Analytical Measurement, CG 4, Third
edition (37)
• ASTM D8293-19, Guide for Evaluating and Expressing the Uncertainty of
Radiochemical Measurements (32)
• Multi-Agency Radiological Laboratory Analytical Protocols (MARLAP) Manual (10)
• NIST "Uncertainty of Measurement Results" (33)
• The NIST Traceable Reference Material Program for Gas Standards (34)
4.5 Precision
Definition
Precision, specifically for methods, is defined by ASTM E177-20 (11) as:
Precision is the closeness of agreement between independent test results under
stipulated conditions.
ASTM Practice E177 notes that quantitative measures depend on the stipulated conditions
and that independent test results mean that results are obtained in a manner not influenced by any
previous result on the same or similar test object (11).
Discussion
The following discussion of precision as a method performance characteristic is taken from
the Eurachem Guide (3):
21
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Precision (measurement precision) is a measure of how close results are to one
another. It is usually expressed by statistical parameters which describe the spread
of results, typically the standard deviation (or relative standard deviation),
calculated from results obtained by carrying out replicate measurements on a
suitable material under specified conditions. Deciding on the 'specified conditions'
is an important aspect of evaluating measurement precision - the conditions
determine the type of precision estimate obtained.
'Measurement repeatability' and 'measurement reproducibility' represent the two
extreme measures ofprecision which can be obtained. Documentation of standard
methods (e.g. from ISO) will normally include both repeatability and
reproducibility data where applicable.
Repeatability, expected to give the smallest variation in results, is a measure of the
variability in results when a measurement is performed by a single analyst using
the same equipment over a short timescale. Repeatability is sometimes referred to
as 'within-run', 'within-batch' or 'intra-assay 'precision.
Reproducibility, expected to give the largest variation in results, is a measure of
the variability in results between laboratories. In validation reproducibility refers
to the variation between laboratories using the same method.
Between these two extremes, 'intermediate (measurement) precision' gives an
estimate of the variation in results when measurements are made in a single
laboratory but under conditions that are more variable than repeatability.
The exact conditions used should be stated in each case. The aim is to obtain a
precision estimate that reflects all sources of variation that will occur in a single
laboratory under routine conditions (different analysts, extended timescale,
different pieces of equipment etc.). 'Intermediate precision is sometimes referred
to as 'within-laboratory reproducibility', 'between-run variation', 'between
batches variation' or 'inter-assay variation'.
ASTM E177-20 (77) has similar definitions for repeatability and reproducibility as they
pertain to replicate measurements using a method. The Eurachem Guide (3) provides a spectrum
(measurement repeatability, intermediate measurement precision, reproducibility) for the range of
experiments involving replicate analyses designed to take into account variations in operational
conditions, which can be expected during routine use of a method. The extent of replicate analysis
should be based on the intended use and application of the method and the need to adequately
demonstrate that a method is fit for its intended purpose.
Useful Resources
The following resources can provide further details and information on precision.
22
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
• ASTM E177-20, Standard Practice for Use of the Terms Precision and Bias in ASTM
Test Methods (11)
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics, Second Edition (3)
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analysis (5)
• ASTM E1601-19, Standard Practice for Conducting an Interlaboratory Study to
Evaluate the Performance of an Analytical Method (9)
• ASTM E691-19el, Standard Practice for Conducting an Interlaboratory Study to
Evaluate the Precision of a Test Method (7)
4.6 Range
Definition
The range is defined as the interval of analyte concentrations for which there is a
meaningful response from the analytical system (in other words, from the characterized detection
level e.g., LOD at the low end to the below saturation at the high end). The following types of
ranges are normally confirmed during method validation.
The calibration range is defined by the lowest and highest standards used for calibration
that meet calibration performance criteria (e.g., linearity check, precision).
The quantitation range is the range of analyte concentrations for which acceptable
quantitative measurement results are obtained and reported. It spans from the characterized lower
quantitation level (e.g., LOQ), through the level of interest (e.g., action levels) to an upper
quantitation level (e.g., the highest calibration standard). This range may be extended through
sample preparation techniques such as sample dilution or concentration.
Discussion
Where applicable, method validation should start by determining if the level(s) of interest
(e.g., action level, risk limit, target level) can be reliably detected and quantitated on a candidate
instrument, followed by determining the range of detections centered on the levels of interest. This
defines the range and acceptability of the method for the desired sample results.
The characterized detection level (e.g., MDL or LOD) should be well below the level(s) of
interest, if possible. The upper limit should be well above any expected levels and is bound by
analytical system constraints (e.g., detector saturation). The inclusion of additional factors, such
as sample size, dilutions, or concentrations, help define the full range of the method. The
determination of the range accounts for all steps of sample preparation and instrument conditions.
23
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
The quantitation range defines the range that meets known performance criteria and is
unique to each specific combination of sample preparation and analysis procedures. The
quantitation range produces acceptable calibration (linear or other) and is limited by the
method/instrument's range of concentrations. This range begins at the characterized lower
quantitation level and ends at the upper concentration of the calibration curve, which does not
saturate the detector. The low point of the quantitation range can be between the characterized
detection level and the lower characterized quantitation level. This quantitation range must be
validated for all methods (except for single point calibrations or presence/absence tests). This
range must be confirmed each time the method is applied after the validation. This quantitation
range must also be confirmed if the method is applied using different instrumentation (within the
same laboratory) or in a different laboratory (with the same or different instrumentation). This
range may be extended through sample preparation techniques such as sample dilution or
concentration.
Note that the characterized detection limit may not meet the performance criteria of the
calibration curve. The lower characterized quantitation limit is a concentration above the
characterized detection limit and has defined performance levels. When the detectors become
saturated, the response plateaus, and it is difficult to differentiate concentrations in the upper part
of the range. It is important not to include any plateauing part of the range in the quantitation range.
Useful Resources
The following resources can provide further details and information on range.
• ASTM E2857-11, Standard Guide for Validating Analytical Methods (4)
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analyses (5)
• Validation and Peer Review of U.S. Environmental Protection Agency Chemical
Methods of Analysis (6)
4.7 Ruggedness
Definition
The following definition of ruggedness comes from the EPA Forum on Environmental
Measurements (FEM) Validation and Peer Review of U.S. Environmental Protection Agency
Chemical Methods of Analysis (6):
the extent to which an analytical method remains unaffected by minor variations in
operating conditions.
The document further describes that:
24
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Ruggedness testing involves experimental designs for examining method
performance when minor changes are made in operating or environmental
conditions. The changes should reflect expected, reasonable variations that are
likely to be encountered in different laboratories.
Discussion
The following discussion of ruggedness testing comes from ASTM El 169-18, Standard
Practice for Conducting Ruggedness Tests (35):
A ruggedness test is a special application of a statistically designed experiment. It
is generally carried out when it is desirable to examine a large number of possible
factors to determine which of these factors might have the greatest effect on the
outcome of the test method. Statistical design enables more efficient and cost
effective determination of the factor effects that would be achieved if separate
experiments were carried out for each factor.
Ruggedness testing approaches can be univariate or mutlivariate in nature. In a univariate
approach, one factor is changed at a time and the method performance assessed. Multivariate
testing involves changing more than one factor at a time and is a more efficient way to assess
ruggedness of a method. Two commonly used multivariate approaches are fractional factorial
design (6, 35, 36) and Plackett-Burman design (3, 35, 37). These two designs are used to identify
a smaller number of important factors from a list of many potential ones. In a fractional factorial
design, only an adequately chosen fraction of the treatment combinations required for the complete
factorial experiment is selected to be run (35). Appropriately chosen fractional factorial designs
for two-level experiments have the desirable properties of being both balanced and orthogonal.
Plackett-Burman designs are very economical designs with a run number that is a multiple of four
(rather than a power of 2 for a complete factorial design) (37). Plackett-Burman designs are very
useful for economically detecting large main effects, assuming all interactions are negligible.
Additional information on the Plackett-Burman design can be found in the original paper
describing this approach (37).
An appropriate statistical design for conducting the ruggedness testing experiments should
be completed prior to the start of any testing. The statistical design used may depend on the
ruggedness testing needs that are determined for an individual method.
Ruggedness tests should be designed to evaluate a range of possible variables that may
impact the method. The IUPAC Technical Report (5) provides the following examples of variables
or factors to consider including in a ruggedness test: changes in the instrument operator, changes
in brand of reagent, concentration of a reagent, pH of a solution, temperature of a reaction, time
allowed for completion of the process, etc. Other variables or factors that could be considered may
include sample preparation variations, instrument settings, instrument conditions such as
temperatures and flows, sample and/or extract holding times, or sample and/or extract additives.
Matrix variability, as discussed in Section 3, may also be a factor that is incorporated into
25
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
ruggedness testing. Different variables may be more important to certain methods and not as
important to others. Decisions surrounding which variables to test should be made in consideration
of the method's operational needs and critical factors that may be most influential on method
performance. A skilled method operator as well as someone with experience in designing
ruggedness testing can play an important role in effectively determining which variables to test.
The following note about the variables or factors chosen for ruggedness testing comes from ASTM
El 169-18 (35):
The factors chosen for ruggedness testing are those believed to have the potential
to affect the results. However, since no limits may be provided in the standard for
these factors, ruggedness testing is intended to evaluate this potential.
ASTM El 169-18 (35) recommends testing two levels for each factor evaluated. The
following description of these factors comes from ASTM El 169-18 (35):
In ruggedness testing, the two levels for each factor are chosen to use moderate
separation between the high and low setting. In general, the size effects, and the
likelihood of interactions between the factors, will increase with increased
separations between the high and low settings of the factors.
Experimental runs of each factor of interest should be conducted in random order. Once
test results are obtained, statistical analysis should be used to determine the effects of factors on
the test method (35). In evaluating the results, ASTM El 169-18 notes that statistical significance
is not the same as practical significance (35). There may be practical significance in differences
smaller than those determined by the ruggedness tests and statistical evaluations that may need
consideration and additional experimentation (35).
Ruggedness testing can be conducted towards the end of the method development effort.
The following overview of the rationale behind this approach is taken from ASTM El 169-18 (35):
Ruggedness testing is usually done within a single laboratory on uniform material,
so the effects of changing only the factors are measured. The results may then be
used to assist in determining the degree of control required offactors described in
the test method.
Any operational aspects determined to have a critical impact on the method should be
discussed in the method distributed for interlaboratory evaluation. Such aspects will be important
for collaborating laboratories to take note of and document during validation efforts to help in
assessing overall method performance across different laboratories.
Useful Resources
The following resources can provide further details and information on ruggedness.
26
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
• Validation and Peer Review of U.S. Environmental Protection Agency Chemical
Methods of Analysis (6)
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics (3)
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analysis (5)
• ASTM Ell 69-18, Standard Practice for Conducting Ruggedness Tests (35)
• Youden, W.J. Statistical techniques for collaborative tests. In: Statistical Manual of the
AOAC (36)
• Plackett, R.L. et al. The Design of Optimum Multifactorial Experiments (37)
4.8 Selectivity in the Presence of Interferences
Definition
The following definition for selectivity is from ASTM E2857-11 (4):
The selectivity of a method is its ability to produce a result that is not subject to
change in the presence of interfering constituents.
Discussion
At a minimum, a qualitative assessment of selectivity should be conducted during method
validation. This section provides a general description of the qualitative assessment of selectivity.
The following overview of evaluating selectivity is taken from the IUPAC Technical Report (5):
Ideally, selectivity should be evaluated for any important interferent likely to be
present. It is particularly important to check interferents that are likely, on
chemical principles, to respond to the test. For example, colorimetric tests for
ammonia might reasonably be expected to respond to primary aliphatic amines. It
may be impracticable to consider or test every potential interferent; where that is
the case, it is recommended that the likely worst cases are checked. As a general
principle, selectivity should be sufficiently good for any interferences to be ignored.
The following overview of the assessment of selectivity is taken from the Eurachem Guide
(3):
The selectivity of a method is usually investigated by studying its ability to measure
the analyte of interest in samples to which specific interferences have been
deliberately introduced (those thought likely to be present in samples). Where it is
unclear whether interferences are already present, the selectivity of the method can
27
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
be investigated by studying its ability to measure the analyte compared to other
independent methods.
Confirmatory techniques can be useful as a means of verifying identities. The more
evidence one can gather, the better. Inevitably there is a trade-off between costs
and time taken for analyte identification, and the confidence with which one can
decide if the identification has been made correctly.
Whereas evaluation of repeatability requires the measurement to be repeated
several times by one technique, confirmation of analyte identity requires the
measurement to be performed by several, preferably independent, techniques.
Confirmation increases confidence in the technique under examination and is
especially useful when the confirmatory techniques operate on significantly
different principles. In some applications, for example, the analysis of unknown
organics by gas chromatography, the use of confirmatory techniques is essential.
When the measurement method being evaluated is highly selective, the use of other
confirmatory techniques may not be necessary.
An important aspect of selectivity which must be considered is where an analyte
may exist in the sample in more than one form such as: bound or unbound;
inorganic or organometallic; or different oxidation states. The definition of the
measurand is hence critical to avoid confusion.
Typically, selectivity is expressed qualitatively. As with any method performance
characteristic that is expressed qualitatively, it is critical that the conditions under which the testing
was performed be thoroughly described and documented. The following discussion of a
"qualitative selectivity statement" is taken from Validation and Peer Review of U.S.
Environmental Protection Agency Chemical Methods of Analyses (6):
A qualitative selectivity statement includes a description of known interferences,
interference effects, and the nature of the analytical data and information that
substantiates the identity of the analyte (s) in the matrix of concern (e.g., elemental
or molecular structure data, retention times from chromatographic separations,
selective reaction chemistry, and results from reference standards, reference
materials, matrix blanks, other blanks, or matrix fortifications). Quantitative
measures of selectivity may also be used, although there is no generally accepted
approach for the quantitative treatment of selectivity data. Therefore, the basis for
quantitative selectivity measures should be thoroughly described.
Useful Resources
The following resources can provide further details and information on selectivity.
• ASTM E2857-11, Standard Guide for Validating Analytical Methods (4)
28
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
• IUPAC Technical Report, Harmonized Guidelines for Single-Laboratory Validation of
Methods of Analyses (5)
• Validation and Peer Review of U.S. Environmental Protection Agency Chemical
Methods of Analysis (6)
• Eurachem Guide: The Fitness for Purpose of Analytical Methods, A Laboratory Guide
to Method Validation and Related Topics (3)
5 Statistical Evaluation and References
A method validation study seeks to demonstrate that a method is fit for a purpose through
the generation of results from method performance characteristic testing. These results should be
appropriately evaluated and assessed to determine the validity and acceptability of the method.
This assessment will generally rely on statistical evaluations of the generated data. However,
ASTM E1488-12, Standard Guide for Statistical Procedures to Use in Developing and Applying
Test Methods (38) notes that:
Statistical procedures often result in interpretations that are not absolutes.
Sometimes the information obtained may be inadequate or incomplete, which may
lead to additional questions and the needfor further experimentation.
Statistical evaluations need proper test planning and data collection to ensure the
generation of reliable results. For example, ruggedness testing entails the consideration of the best
statistical design for generating results necessary to appropriately evaluate the method of interest,
as discussed in Section 4.7. Any statistical evaluations of method validation study data should be
conducted by someone knowledgeable (from your office or organization) of the statistical methods
and theories being used.
A variety of statistical approaches are available, but there are often some that are more
widely accepted or suggested for use (38). In addition, different method performance
characteristics may involve implementing different statistical methods to appropriately analyze the
results. It is a good idea to review existing methods or relevant literature to evaluate statistical
methods used. The goal of this document is not to provide detailed, prescriptive guidelines on
statistical analyses and procedures to be used for evaluating method validation study data. Rather,
this section provides a list of suggested resources for use in understanding and implementing the
necessary statistical assessment of generated study data. Any statistical approaches or analysis
guidelines or criteria that are particular to an agency/program or its method development strategy
should also be referred to and used as appropriate in analyzing method validation data. Review of
existing agency method validation reports or scientific literature reporting method validation
results could also be helpful in determining what statistical approaches might be applicable for
use.
The following references provide further information and details on general statistical
evaluations of method validation study data. This is not an exhaustive list.
29
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
• ASTME1488-12, Standard Guide for Statistical Procedures to Use in Developing and
Applying Test Methods (38)
• Belouafa, S., et al. Statistical Tools and Approaches to Validate Analytical Methods:
Methodology and Practical Examples (39)
• Lynch, J.M. Use of AOAC International Method Performance Statistics in the
Laboratory (40)
• Ravisankar, P., et al .A Review on Step-by-Step Analytical Method Validation (41)
• Wernimont, G. Use of Statistics to Develop and Evaluate Analytical Methods (42)
• Guidance for Data Quality Assessment: Practical Methods for Data Analysis (43)
• Data Quality Assessment: Statistical Method for Practitioners (44)
• ASTM E1601-19, Standard Practice for Conducting an Interlaboratory Study to
Evaluate the Performance of an Analytical Method (9)
Other references noted throughout this document may also have sections describing the
application of statistical procedures for more specific data evaluation efforts and should be
considered as well. In addition, specific EPA program offices may have documents that contain
program-specific guidance on statistical procedures that are recommended/should be used in
validating related methods. For example, the Emissions Measurement Center of the Office of Air
Quality Planning and Standards provides guidance on statistical calculations and comparisons to
be made to data in Method 301 (75).
6 Method Validation Report
A method validation report should be prepared and structured in accordance with the
expectations and guidelines/protocols of individual offices and/or programs. This report should
address the scope and purpose of the method as well as detail the results from all method
performance and application characteristic validations performed, as described previously. The
method validation report may also include the following information, as indicated in the Validation
and Peer Review of U.S. Environmental Protection Agency Chemical Methods of Analysis (6):
• background information on method development
• details on the method validation techniques employed
• changes made to the method as a result of the method validation studies, and
• any recommendations for future work.
This is not a definitive list. Information and results to be included in the method validation
report are at the discretion of each office and/or agency developing the method. Review and release
of a method validation report should be conducted in accordance with applicable office and/or
agency protocols.
30
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
7 Consistently and Concisely Communicating Method Validation Studies
Effective communication of method validation study results is an important part of the
release/adoption step of the method lifecycle (see Figure 1). Consistent and concise
communication practices can help to advance agency-wide awareness and understanding of
validated methods. To aid in the communication of method validation study results, this document
includes two newly developed tools:
• Validation Design
• Method Validation Summary
The Validation Design is a very short alphanumeric descriptor meant to concisely describe
important aspects of the method validation study design. The Method Validation Summary is a
concise synopsis of the method validation results that can be easily prepared and included in any
method validation report. Both provide mechanisms for consistent sharing of method validation
results. The following sections explain these two new tools in more detail.
7.1 Validation Design
The Validation Design provides a standardized format to convey the extent of validation
performed for the method based on the number of laboratories that participated in the validation
and the number of different matrices evaluated by each participating laboratory. By using the
Validation Design in communicating method validation results, readers and users of the method
will be easily able to understand how the validation was performed and compare validations across
similar methods. The Validation Design represents the results of the validation in a succinct
descriptor and is presented as [aL, bM] where aL is the number (a) of laboratories (L) that
participated in the method validation, and bM is the number (b) of different matrices (M) that were
used in the method validation. For example, a multi-laboratory method validation that used four
laboratories, where each laboratory evaluated three different matrices would be reported as [4L,
3M], The number of laboratories used in any given method validation may vary from a single
laboratory to multi-laboratory validation, depending on what was determined to be appropriate in
activities preceding method validation (see Section 2.0). The number of different representative
matrices used for validation samples can also vary. Each matrix should be different and, as
previously described, be typical of matrix types applicable to the method under validation.
Pertinent references are available to aid in the design and conduct of an interlaboratory or
single laboratory validation study (7-9). Results from the interlaboratory or single laboratory study
should be statistically evaluated to determine if the performance of the method is acceptable. These
same references provide an excellent resource for performing and evaluating these calculations.
It is not the intent of this document to provide guidelines for the number of laboratories
that should participate in a validation study or the number of samples or different types of matrices
that should be used in each study. Useful references are available that define expectations for
numbers of participating laboratories and matrices used for non-government organizations and
provide reasoning behind these recommendations or requirements (7-9). In addition, the purpose
31
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
and future use of the method may guide the interlaboratory or single laboratory validation study
needs. Rather, this document presents the Validation Design that communicates the characteristics
of an interlaboratory or single laboratory method validation in an easily reported and understood
format.
7.2 Method Validation Summary
It is the recommendation of this document that each method validation report that is
developed include a Method Validation Summary to serve as a brief synopsis of the method
validation process and results. Its intent is to provide a brief overview of the validation at the front
of the method validation report to provide the reader with easy access to pertinent and important
information. Furthermore, final Method Validation Summary documents can be easily shared
across the Agency, and then be linked or referred to the originating office for details to allow for
additional collaboration and discussions, especially under emergency situations.
In order to facilitate sharing of these summaries across the EPA, the format of the Method
Validation Summary should be the same across all methods, such that each Method Validation
Summary contains the same types of summary information and details for a method in the same
structure. This will allow the reader to access needed information quickly across summaries. A
Method Validation Summary should be included in all method validation reports prepared across
the Agency. In this way, all method validation reports that are developed should contain a Method
Validation Summary that conveys the same type of information for a given method, providing
consistency across all documents and offices. Table 1 provides the structure, format, and content
of the Method Validation Summary.
The Method Validation Summary should be brief, e.g., two pages in length and serve as a
stand-alone document. The summary format is designed to be easy to complete and quickly capture
all of the pertinent information from the Method Validation Report with minimal additional effort.
Thus, method development and validation efforts should produce two documents: a Method
Validation Report and Method Validation Summary. It is anticipated that both the report and the
summary for a given method would reside in the records storage of the originating office. EPA
will internally publish Method Validation Summaries separately for Agency staff to access in order
to inform other method development activities across the Agency.
Appendix D provides an example of a completed Method Validation Summary.
32
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
Table 1. Cover template, with structure guidelines, for the Method Validation Summary
A
Validation Design
Description
1
Number of Laboratories
2
Number of Matrices
3
Types of Matrices Tested
(water, soil, sediment, etc.)
B
Method Validation
Overview
Description
1
Method title
2
Author(s) list
3
Date
4
Purpose
5
Qualitative or Quantitative
6
Target
Analytes/Parameters
NOTES
C
Method Development
Considerations
Description and/or Results
1
Sample Cost
2
Sample Holding Times
3
Sample Preservation
4
Waste Generation
NOTES
D
Method Performance
Characteristic
Description and/or Results
1
Bias/Trueness
2
Detection Capability and
Quantification Capability
3
Instrument Calibration
4
Measurement Uncertainty
5
Precision
6
Range
7
Ruggedness
8
Selectivity in the Presence
of Interferences
NOTES
The Method Validation Summary should contain all categories listed here.
33
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
8 References
1. International Organization for Standardization. (2015). ISO 9000:2015, Quality
Management Systems, Fundamentals and Vocabulary.
2. International Organization for Standardization and International Electrotechnical
Commission. (2017). ISO/IEC 17025:2017, General Requirements for the Competence of
Testing and Calibration Laboratories.
3. Magnusson, B., Ornemark, U. (eds.). (2014) Eurachem Guide: The Fitness for Purpose of
Analytical Methods: A Laboratory Guide to Method Validation and Related Topics. (ISBN
978-91-87461-59-0). Available from www.eurachem.org.
4. ASTM International. (2016). ASTM E2857-11(2016), Standard Guide for Validating
Analytical Methods.
5. Thompson, M.; Ellison, S.L.R.; Wood, R. (2002). Harmonized Guidelines for Single-
Laboratory Validation of Methods of Analysis, IUPAC Technical Report. Pure Appl.
Chem., 74 (5), 835-855.
6. U.S. Environmental Protection Agency (EPA) Forum on Environmental Measurements
(FEM) Method Validation Team; Principal Authors: Mishalanie, E. A., Lesnik, B., Araki,
R., and Segal 1, R. (2005). Validation and Peer Review of US Environmental Protection
Agency Chemical Methods of Analysis. FEM Document Number 2005-01.
https://www.epa.gov/sites/production/files/2016-
02/documents/chemicat method guide i c\ ised 020 '< I |;df
7. ASTM International. (2019). ASTM E691-19el, Standard Practice for Conducting an
Interlaboratory Study to Determine the Precision of a Test Method.
8. AO AC International. (2005). Appendix D: Guidelines for Collaborative Study Procedures
to Validate Characteristics of a Method of Analysis. Official Methods of Analysis of A OA ('
International.
9. ASTM International. (2019). ASTM E1601-19, Standard Practice for Conducting an
Interlaboratory Study to Evaluate the Performance of an Analytical Method.
10. U.S. Environmental Protection Agency, U.S. Department of Defense, U.S. Department of
Energy, U.S. Department of Homeland Security, U.S. Nuclear Regulatory Commission,
U.S. Food and Drug Administration, U.S. Geological Survey, and National Institute of
Standards and Technology. (2004). Multi-Agency Radiological Laboratory Analytical
Protocols Manual (MARLAP). NUREG-1576. EPA 402-B-04-001C, NTIS PB2004-
105421. https://www.epa.gov/radiation/marlap-manual-and-supporting-documents
11. ASTM International. (2020). ASTM E177-20, Standard Practice for Use of the Terms
Precision and Bias in ASTM Test Methods.
12. Definition and Procedure for the Determination of the Method Detection Limit. (2017). 40
CFR Part 136, Appendix B. https://www.ecfr.gov/cgi-bin/text-
ic^ * M i 0:s m 11014 L'%7c5f?cd528d526a&MC=tme&no ^ M _ s s U.fg
n=div9
34
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
13. U.S. Environmental Protection Agency. (2017). Quality Assurance Handbook for Air
Pollution Measurement Systems, Volume 11, Ambient Air Quality Monitoring Program.
EPA-454/B-17-001. https://www.epa.gov/sites/production/files/2Q2Q-
1 Q/documents/fmal handbook document 1 17.pdf
14. U.S. Environmental Protection Agency. (2019). Technical Assistance Document for
Sampling and Analysis of Ozone Precursors for the Photochemical Assessment Monitoring
Stations Program, Revision 2, April 2019. EPA 454/B-10-004
¦https://www.epa.gov/sites/production/files/2Q19-
11/documents/pams technical assistance document revision 2 april 2019.pdf
15. Method 301 - Field Validation of Pollutant Measurement Methods from Various Waste
Media. (DATE?). 40 CFR Part 63, Appendix A. https://www.epa.gov/emc/method-301-
field-validation-pollutant-measurement-methods-various-waste-media
16. Winslow, SD; Pepich, BV; Martin, JJ; Hallberg, GR.; Munch, DJ; Frebis, CP; Hedrick, EJ;
Krop, RA. (2006). Statistical procedures for determination and verification of minimum
reporting levels for drinking water methods. Environ Sci Technol. 40 (1), 281-8.
17. U.S. Environmental Protection Agency. Technical Basis for the Lowest Concentration
Minimum Reporting Level (LCMRL) Calculator. (2010). EPA-815-R-11-001.
https://nepis.epa. gov/Exe/ZyPDF.cgi?Dockev=Pl 00J7CA.txt
18. Method 608.3: Organochlorine Pesticides and PCBs by GC/HSD. EPA 821-R-16-009.
(2016). 40 CFR Part 136, Appendix A. https://www.epa.gov/sites/default/files/
08/docum ents/m ethod 601 idf
19. Method 624.1: Purgeables by GC/MS. EPA 821-R-16_008. (2016). 40 CFR Part 136,
Appendix A. https://www.epa.gov/sites/default/files/2017-08/documents/method 624-
1 2016.pdf
20. Method 625.1: Base/Neutrals and Acids by GC/MS. EPA 821-R-16-007. (2016). 40 CFR
Part 136, Appendix A. https://www. epa.gov/ sites/default/files/
08/docum ents/m eth od 62. idf
21. U.S. Environmental Protection Agency. (2015). Test Methods for Evaluating Solid Waste,
Physical/Chemical Methods, EPA publication SW-846, Third Edition, Final Updates I
(1993), II (1995), IIA (1994), IIB (1995), 111 (1997), 111 A (1999), II IB (2005), IV (2008),
and V (2015). https://www.epa.eov/hw-sw846/sw-846-compendium
22. Currie, L.A. (1968). Limits for Qualitative Detection and Quantitative Determination:
Application to Radiochemistry. Anal. Chem. 40, 586-593.
23. Currie, L.A. (1988). Detection: Overview of Historical, Societal, and Technical Issues. In
L.A. Currie (Ed.), Detection and Analytical Chemistry (ACS Symp. Ser. 361) (pp. 1-62).
American Chemical Society, Washington, DC.
24. Currie, L.A. (1998). Presentation of the Results of Chemical Analysis. In J. Inczedy, T.
Lengyel, and A.M. Ure, (Eds.), IUPAC Compendium of Analytical Nomenclature (Chapter
2). Blackwell Science, Oxford.
35
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
25. Currie, L.A. (1998). Quality Assurance of Analytical Processes. In J. Inczedy, T. Lengyel,
and A.M. Ure, (Eds.), IUPAC Compendium of Analytical Nomenclature (Chapter 18).
Blackwell Science, Oxford.
26. Lanier, W.S. and Hendrix, C. D, (2001). Reference Method Accuracy and Precision
(ReMAP): Phase 1 - Precision of Manual Stack Emission Measurements. American
Society of Mechanical Engineers (ASME) Report. CRTD Vol. 60.
27. Joint Committee for Guides in Metrology. (2008). JCGM 100:2008, Evaluation of
Measurement Data: Guide to the Expression of Uncertainty in Measurement.
https://www.bipm.ore/iitils/common/documents/icgm/JCG s08 E.pdf
28. U.S. Environmental Protection Agency. (2012). Traceability Protocol for Assay and
Certification of Gaseous Calibration Standards. EPA/600/R-12/53.
https://www.epa.gov/air-research/epa-traceabilitv-protocol-assav-and-cert.ification-
gaseous-calibration-standards
29. Joint Committee for Guides in Metrology. (2008). JCGM 101:2008, Evaluation of
Measurement Data: Supplement 1 to the "Guide to the Expression of Uncertainty in
Measurement," Propagation of Distributions Using a Monte Carlo Method.
https://www.bipm.org/utils/common/documents/icgm/JCGM 101 JOOS E.pdf
30. Joint Committee for Guides in Metrology. (2011). JCGM 102:2011 Evaluation of
Measurement Data: Supplement 2 to the "Guide to the Expression of Uncertainty in
Measurement," Extension to Any Number of Quantities.
https://www.bipm.org/utils/common/documents/icgm/JCG df
31. Ellison, S. L. R., Williams, A. (eds.). (2012). Eurachem/Co-Operation on International
Traceability in Analytical Chemistry (CITAC) Guide CG 4 Quantifying uncertainty in
analytical measurement. Third Edition. (QUAM:2012.P1). Available from
www.eurachem .org
32. ASTM International. (2019). ASTM D8293-19, Guide for Evaluating and Expressing the
Uncertainty of Radiochemical Measurements.
33. National Institute of Standards and Technology (NIST). Uncertainty of Measurement
Results. Retrieved March 31, 2020, from
https://phvsics.nist.gov/ciiu/Uncertaintv/index.html
34. The NIST Traceable Reference Material Program for Gas Standards. (2013). NIST Special
Publication 260-126, available at
https://nvlpiibs.nist.gOv/nistpiibs/SpecialPublications/.NIST.SP.260-126rev2013.pdf.
35. ASTM International. (2018). ASTM El 169-18, Standard Practice for Conducting
Ruggedness Tests.
36. Youden, W.J. (1975). Statistical Techniques for Collaborative Tests. In: Statistical Manual
of the AOAC, AO AC International, Washington, DC, pp. 33-36.
37. Plackett, R.L. and Burman, J.P. (1946). The Design of Optimum Multifactorial
Experiments. Biometrika, 33(4), pp.305-325.
36
-------
Guidelines on Validation for Non-Regulatory Chemical and Radiochemical Methods January 2022
38. ASTM International. (2012). ASTM E1488-12, Standard Guide for Statistical Procedures
to Use in Developing and Applying Test Methods.
39. Belouafa, S., Habti, F., Benhar, S., Belafkih, B., Tayane, S., Hamdouch, S., ... &
Abourriche, A. (2017). Statistical Tools and Approaches to Validate Analytical Methods:
Methodology and Practical Examples. International Journal of Metrology and Quality
Engineering, 8, 9.
40. Lynch, J. M. (1998). Use of AO AC International Method Performance Statistics in the
Laboratory. Journal of AO AC International, 57(3), 679-684.
41. Ravisankar, P., Navva, C. N., Pravallika, D., & Sri, D. N. (2015). A Review on Step-by-
Step Analytical Method Validation. IO SR Journal of Pharmacy, 5( 10), 7-19.
42. Wernimont, G. Use of Statistics to Develop and Evaluate Analytical Methods; AOAC:
Arlington, VA, 1985. (Fourth printing 1993; ISBN 0-935584-31-5)
43. U.S. EPA. (2000). Guidance for Data Quality Assessment: Practical Methods for Data
Analysis. (EPA QA/G-9; QA00 Update) (EPA/600/R-96/084). Washington, DC; US EPA,
Office of Environmental Information, https://www.epa.gov/sites/production/files/2015-
06/docum ents/g9-final. pdf
44. U. S. EPA. (2006). Data Quality Assessment: Statistical Methodfor Practitioners. (EPA QA/G-
9S) (EPA/240/B-06/003). Washington, DC; US EPA, Office of Environmental Information.
https://www.epa.gOv/sites/production/files/2 /documents/g9s-final.pdf
37
-------
Appendix A
Method Validation Matrix Considerations for Individual EPA Offices
A-l
-------
U.S. EPA Office of Water, Office of Groundwater and Drinking Water
Drinking Water Method Development and Validation
General validation guidelines for types of matrices and number of laboratories used are provided
below.
Note: Method validation is incorporated as part of the method development process.
Matrix Testing (three, including reagent water)
• Reagent water or laboratory water is used as a matrix baseline (no expected matrix
effects).
• Finished drinking water matrices generally fall into two categories: a surface source
tap water (high organic carbon content) and a ground source tap water (high hardness).
Usually, at least one of each is used as a test matrix.
Note: Some methods analyze for target analytes that may only appear in one type of
matrix, e.g., cyanotoxins that would only exist in finished water from surface water
sources. In those cases, the type of matrix used in method validation may only include
reagent water and finished drinking water from different sources of the same type.
Similarly, if specific matrices are expected to have significant effects on method
measurements, more of those types of test matrices may be evaluated, e.g., the effect
of total organic carbon from surface water sources on early eluting organic analytes in
liquid chromatography.
Multi-laboratory Study (three, including in-house method performance)
• At a minimum for statistical purposes, three laboratories are used for drinking water
method validation. More may be used but it depends on availability and the resources
of the laboratories since the participation of laboratories is voluntary.
A-2
-------
U.S. EPA Office of Water, Office of Science and Technology
Wastewater Matrix Types Recommended for Multiple Matrix Type Validation Studies
1. Effluent from a publicly owned treatment works (POTW)
2. ASTM D5905-98 (Reapproved 2018), Standard Specification for Substitute Wastewater
3. Sewage sludge, if sludge will be in the permit
4. ASTM D1141 - 98 (Reapproved 2021), Standard Specification for Substitute Ocean Water,
if ocean water will be in the permit
5. Untreated and treated wastewaters up to a total of nine matrix types (see
https://www.epa.gov/eg/industrial-effluent-guidelines for a list of industrial categories
with existing effluent guidelines)
At least one of the above wastewater matrix types should have at least one of the following
characteristics:
• Total suspended solids (TSS) greater than 40 mg/L
• Total dissolved solids (TDS) greater than 100 mg/L
• Oil and grease greater than 20 mg/L
• NaCl greater than 120 mg/L
• CaC03 greater than 140 mg/L
A-3
-------
U.S. EPA Office of Air and Radiation, Office of Air Quality Planning and Standards
Sample Matrix Considerations for Validation of Emission Test Methods
Section 17.1.1 of Method 301 (40 CFR 63, Appendix A), a protocol for validation of
stationary source emission test methods, recognizes that validation of a method at a 'similar
source' with a similar emission matrix may be adequate to justify application of a candidate
method to other similar sources. Because of the wide range of sample matrices that may be
encountered from emission sources, there are no formal guidelines in place for addressing matrix
differences. However, there is general agreement among those in the emission measurement
community who have developed and validated methods that there are a number of sample matrix
constituents that must be considered/assessed in determining the breadth of applicability of a
particular candidate method. These include:
• Acid Gases including NOx, SO2, HC1, HF, and H2SO4
• Other Reactive Gases including NH3, sulfur compounds
• Particulate Matter including carbon, metals, and salts
• Organic compounds
• High moisture
If it is possible and resources allow, it is preferable to evaluate the performance of the candidate
method on samples from a number of types of emission sources to include various combinations
of these matrix constituents and thus challenge the method capability. However, the cost of
mobilization and sampling for collection of emission samples for validation purposes can often
exceed $100,000, so this is often not a possibility. Thus, Section 14.0 of Method 301 suggests use
of ruggedness testing1 as a potential tool to collect data to support a broader application of the
candidate method. Sampling for ruggedness testing of emission test methods can often be
conducted in the laboratory and multiple variables evaluated using a set of nine test runs, which
can be a significant cost savings.
1 Youden, W.J. Statistical techniques for collaborative tests. In: Statistical Manual of the Association of Official
Analytical Chemists, Association of Official Analytical Chemists, Washington, DC, 1975, pp. 33-36.
A-4
-------
U.S. EPA Office of Resource Conservation and Recovery
Matrix Types for SW-846 Method Validation:
• Sample matrices should be selected to represent those regulated under the Resource
Conservation and Recovery Act (RCRA) (e.g. soil, oily waste, wastewater).
• Developers should analyze different types of matrices included in the scope of the
method. Matrix types refer to different matrices within a particular medium, e.g.
water, soil and ash. Appropriate RCRA matrix types might include:
o Aqueous: groundwater, toxicity characteristic leaching procedure (TCLP)
leachate and wastewater
o Soil: sand, loam and clay soils
o Ash: bottom ash, fly ash and/or combined ash
• Samples should be well-characterized reference materials and/or spiked matrices
containing known amounts of target analytes.
• Bulk samples should be carefully homogenized where appropriate to reduce sub-
sampling errors
Summarized from: https://www.epa.eov/sites/prodiiction/files/2015-10/documents/methdev.pdf
A-5
-------
Appendix B
Analyte Detection and Quantitation
B-l
-------
The purpose of analyzing an environmental sample may be either (1) to make a qualitative
decision about the presence or absence of a particular analyte in the sample or (2) to quantify the
amount of analyte that is present. The first of these goals is referred to as detection, while the
second is called quantitation (or quantification). One analysis can often serve both purposes
simultaneously.
Detection
When analyte detection is of interest, two relevant aspects of the analytical method are its
detection rule and its detection capability.
The detection rule is the statistical test that is used to determine whether the measurement
data justify a decision that the analyte is present in a sample. Typically, the test is designed to limit
the probability of a false detection in a truly analyte-free sample to a specified small value such as
1 % or 5 %. This probability of false detection is the significance level of the test, often denoted
by a.
The detection rule is usually implemented as a straightforward comparison of the measured
result to a calculated threshold value. This detection threshold goes by various names, including
critical level, critical value, decision threshold, and decision level to name a few. The method
detection limit (MDL) defined in 40 Code of Federal Regulations (CFR) 136 Appendix B is also
an example. The procedures for calculating the detection threshold may be program-specific.
The detection capability of a measurement process—its ability to detect the analyte—is
often described in terms of the minimum detectable value, which is defined as the true value of the
analyte that must be present in a sample to ensure a specified high probability of detection using
the given detection rule. This probability is often denoted by 1 -/?, where p denotes the probability
of a "false negative" result (non-detection). Assuming the detection rule involves comparison to a
detection threshold, the minimum detectable value is the smallest true value of the analyte needed
to ensure a specified probability 1 - p of observing a result greater than the detection threshold.
Detection capability may be relevant even when the question to be answered is not whether
the analyte is present but whether it exceeds a specified action level, as long as the expected
background level of the analyte in typical samples is very low relative to the action level. In this
case, to ensure adequate measurement capability at the action level, it may suffice to ensure
adequate detection capability.
Note—The term detection limit, depending on the field of measurement, may be used to
mean either a detection threshold or a minimum detectable value, creating opportunities
for miscommunication between workers in different fields. The same term is also defined
in 40 CFR 141.25 (c) with a somewhat different definition in the context of measuring
radionuclides in drinking water.
B-2
-------
Assuming the measurement process produces normally distributed results with negligible
bias and a well-characterized standard deviation, the relationship between the detection threshold
and the minimum detectable value is illustrated by Figure Bl.
Analyte concentration
Figure Bl - Detection threshold, xc, and minimum detectable value, xd
The curve centered at x = 0 represents the distribution of measurement results that can be
expected when analyte-free samples are analyzed. A specified percentile of this distribution is
identified as the detection threshold, xc. The analyte is considered to be detected if the measured
result exceeds the detection threshold. So, for example, if the threshold is set at the 99th percentile
of the distribution (as shown in Fig. Bl), then 99 % of all measured results for analyte-free samples
should be less than the detection threshold, and approximately 1 % should exceed it, producing
false detections.
The second curve, on the right, represents the distribution of results that can be expected
when the true level of the analyte in a sample equals the minimum detectable value, xd. For such
a sample, there is a specified high probability, 1 - /?, of obtaining a result greater than the detection
threshold. For example, if the minimum detectable value is defined as the analyte level at which
the probability of detection (1 -/?) equals 95 %, then the area under the curve to the right of the
detection threshold is 95 %, and the area to the left (/?) is 5 % (as shown in Fig. Bl).
When the measurement standard deviation is not well known and must be estimated—by
replicate measurements, for example—Fig. Bl is not completely valid but may still be illustrative.
Quantitation (Quantification)
When quantitation is of interest, the most relevant characteristic of the measurement
process is its quantification capability, which is typically defined as the smallest true value of the
analyte that ensures a specified acceptable level of relative measurement precision or uncertainty.
This value of the analyte might be called by various names, including limit of quantitation (LOQ),
quantitation limit, quantification limit, or minimum quantifiable value. The procedures for
estimating quantification capabilities may be program-specific.
Note—As used here, the terms quantitation and quantification have identical meanings.
Each has been used extensively in the literature—for example in the terms listed above.
B-3
-------
Quantitation is typically of interest whenever decisions are being made about the average
analyte concentration in a sampled population rather than an individual environmental sample or
specimen. It may also be of interest if the action level for decision-making about an individual
sample does not greatly exceed the expected background level of the analyte in the sample matrix.
B-4
-------
Appendix C
Detection and Quantitation Limit Definitions
C-l
-------
Appendix C includes some of the detection and quantitation limit definitions that are used
by EPA programs and appear in the literature. This is not a comprehensive list.
C-2
-------
I'vpc Name Kxpliiiiiilion
Notes
AML
Alternate Minimum Level
A regression approach that provides for the
case of nonconstant variance throughout the
instrument calibration working range;
calculated Currie-type Lc, Ld, and Lq; Lq is
based on the standard deviation only (CI).
IDL
Instalment Detection
Limit
The concentration equivalent to a signal, due to
the analyte of interest, which is the smallest
signal that can be distinguished from
background noise by a particular instrument.
The IDL should always be below the MDL,
and is not used for compliance data reporting,
but may be used for statistical data analysis and
comparing the attributes of different
instruments. The IDL is similar to the "critical
level" and "criterion of detection" as defined in
the literature (C2).
Similar to an MDL but is only
intended as an instrumental
measurement. Samples are not
processed through entire method.
IQEZ%
Interlab oratory
Quantitation Estimate
Regression approach that provides for
nonconstant variance throughout the working
range; an interlaboratory quantitation level is
determined on the basis of the use of the
standard deviation only (C3).
Lc, Lu, Lq
Critical Level, Detection
Level, Quantitation Level
Lc, critical level (low false positive error);
Ld, detection level (low false negative error);
Lq, quantitation level, defined as a multiple
(default 10 times) of the standard deviation; the
standard deviation is determined from method
blank replicates (C4, C5).
LCMRL
Lowest Concentration
Minimum Reporting Level
The lowest true concentration for which the
future recovery is predicted to fall between
50% and 150% with 99% confidence (C6).
Typically used by EPA internally
(UCMR Program). It takes into
account both precision and accuracy.
C-3
-------
Typo
Nil inc
Kxpliiiiiilion
Notes
LLOQ
Lower Limit of
Quantitation
The LLOQ is the lowest concentration at which
the laboratory has demonstrated target analytes
can be reliably measured and reported with a
certain degree of confidence, which must be >
the lowest point in the calibration curve.
LOD
Limit of Detection
The lowest concentration level that can be
determined to be statistically different from a
blank (99% confidence). The LOD is typically
determined to be in the region where the signal
to noise ratio is greater than 5. Limits of
detection are matrix, method, and analyte
specific (C2).
5:1 signal to noise
LOQ/MQL
Limit or Level of
Quantitation/Minimum
Quantitation Level
The level above which quantitative results may
be obtained with a specified degree of
confidence. The LOQ is mathematically
defined as equal to 10 times the standard
deviation of the results for a series of replicates
used to determine a justifiable LOD. Limits of
quantitation are matrix, method, and analyte
specific (C2).
Recommended LOQ = lOo, where o is the
standard deviation of the samples (C7).
Typically, it is the concentration that produces
a signal 10 o above the reagent water blank
signal, and should have a defined precision and
bias at that level (C8).
10 times the standard deviation or
10:1 signal to noise.
LT-MDL
and LRL
Long-Term MDL,
Laboratory Reporting
Level
LT-MDL is calculated as the MDL of Glaser et
al.; additional variance is included from
multiple instruments, different matrices, and
over time; LRL = 2(LT-MDL) (C9).
C-4
-------
Typo
Nil inc
Kxpliiiiiilion
Notes
MDL
Method Detection Limit
The MDL is defined as the minimum measured
concentration of a substance that can be
reported with 99% confidence that the
measured concentration is distinguishable from
method blank results. Reference CI 1 contains
the necessary equations for calculating MDLs
(CIO, Cll).
Does not take into account accuracy.
Resulting concentration may
significantly deviate from the true
concentration.
ML
Minimum Level of
Quantitation
The term "minimum level" refers to either the
sample concentration equivalent to the lowest
calibration point in a method or a multiple of
the MDL, whichever is higher. Minimum levels
can be obtained in several ways: they may be
published in a method; they may be based on
the lowest acceptable calibration point used by
a laboratory; or they may be calculated by
multiplying the MDL in a method, or the MDL
determined by a laboratory, by a factor of 3.
For the purposes of NPDES compliance
monitoring, EPA considers the following terms
to be synonymous: "quantitation limit,"
"reporting limit," and "minimum level" (CI2).
C-5
-------
Typo
Nil inc
Kxpliiiiiilion
Notes
MRL
Minimum Reporting
Level/Limit
The minimum concentration that can be
reported by a laboratory as a quantified value
for the method analyte in a sample following
analysis. This concentration must meet MRL
confirmation criteria in the method and must be
no lower than the concentration of the lowest
calibration standard for each method analyte
(CI 3).
Data quality objectives are specified for
accuracy (70-130% recovery) and precision
(10% RSD), though independently applied
(C14).
MRL = 3 x MDL
Where MDL = t* s and Hs a Student's t value
and 5 is the estimated standard deviation (CI 5).
For EPA MRL, the concentration is
assigned by EPA but based on a
multi-laboratory study using the
LCMRL (precision and accuracy).
Current EPA DW methods require an
MRL confirmation, which also takes
into account precision and accuracy.
PQL
Practical Quantitation
Limit
A quantitation limit that represents a practical
and routinely achievable quantitation limit with
a high degree of certainty (>99.9% confidence)
in the results. The PQL appears in older
Department of Natural Resources literature and
in some current EPA methods, however its use
is being phased out by the DNR (C2).
The PQL, which is about three to five times
larger than the MDL, is a practical and
routinely achievable detection level with a
relatively good certainty that any reported
value is reliable (C16).
C-6
-------
Typo
N il mo
Kxpliiiiiilion
Notes
PQLs
Practical Quantitation
Levels
Criteria of quantitation level set for accuracy at
±40% and for precision at less than 20% RSD
(CI 7).
QLs and
RDL
Quantitation Levels and
Reliable Detection Level
Interlaboratory quantitation level: the median
interlaboratory MDL is multiplied by a variable
determined from laboratory performance data
(values -4-7); RDL = 2(MDL) (CIS).
yc, xD
Hubaux-Vos Detection
Limits
yc is the decision limit corresponding to
Currie's Zc; xd is a detection limit
corresponding to Currie's Ld; calibration
design for detection limits; based on the
standard deviation only (C19).
C-7
-------
CI Gibbons, R. D., Coleman, D. E., Maddalone, R. F. (1997). An Alternative Minimum Level
Definition for Analytical Quantification. Environmental Science & Technology, 31(7),
2071-2077. doi: 10.1021/es960899d.
C2 Ripp, J. (1996). Analytical Detection Limit Guidance & Laboratory Guide for Determining
Method Detection Limits. (PUBL-TS-056-96). Madison, WI: Wisconsin Department of
Natural Resources.
C3 ASTM International. (2014). ASTM D6512-07, Standard Practice for Interlaboratory
Quantitation Estimate. West Conshohocken, PA: ASTM International.
C4 Currie, L. A. (1968). Limits for qualitative detection and quantitative determination.
Application to radiochemistry. Analytical Chemistry, 40(3), 586-593.
doi:10.1021/ac60259a007.
C5 Currie, L. A. (1999). Detection and quantification limits: origins and historical overview.
Analytica Chimica Acta, 391(2), 127-134. doi:https://doi.org/10.1016/S0003-
2670(99)00105-1.
C6 Winslow, S. D., Pepich, B. V., Martin, J. J., Hallberg, G. R., Munch, D. J., Frebis, C. P.,
Krop, R. A. (2006). Statistical Procedures for Determination and Verification of Minimum
Reporting Levels for Drinking Water Methods. Environmental Science & Technology,
40(1), 281-288. doi:10.1021/es051069f.
C7 Keith, L. H., Crummett, W., Deegan, J., Libby, R. A., Taylor, J. K., Wentler, G. (1983).
Principles of environmental analysis. Analytical Chemistry, 55(14), 2210-2218.
doi:10.1021/ac00264a003.
C8 American Public Health Association (APHA), American Water Works Asociation
(AWW A), Water Environment Federation (WEF). (2017). 1010 INTRODUCTION.
Standard Methods For the Examination of Water and Wastewater. Retrieved from
https://www.standardmethods.ore/doi/abs/10.2105/SMWW.2882.004.
C9 Childress, C. J. O., Foreman, W. T., Connor, B. F., Maloney, T. J. (1999). New reporting
procedures based on long-term method detection levels and some considerations for
interpretations of water-quality data provided by the US Geological Survey National Water
Quality Laboratory. US Geological Survey Open-File Report, 193, 1.
C10 Glaser, J. A., Foerst, D. L., McKee, G. D., Quave, S. A., Budde, W. L. (1981). Trace
analyses for wastewaters. Environmental Science & Technology, 75(12), 1426-1435.
doi: 10.1021/es00094a002.
Cll U.S. Environmental Protection Agency. (2016). Definition and Procedure for the
Determination of the Method Detection Limit, Revision 2. (EPA 821-R-16-006).
Washington, D.C.: U.S. Environmental Protection Agency, OW/OST/EAD.
C12 U.S. Environmental Protection Agency. (2016). Method608.3: Organochlorine Pesticides
and PCBs by GC/HSD. (EPA Document No. 821-R-16-009). Washington, DC: US
Environmental Protection Agency OW/OST/EAD.
C-8
-------
C13 Adams, W. A., Wendelken, S. C. (2015). EPA Method 545: Determination of
cylindrospermopsin and anatoxin-a in drinking water by liquid chromatography
electrospray ionization tandem mass spectrometry (LC/ESI-MS/MS). (EPA Document No.
815-R-15-009). Cincinnati, OH: U.S. Environmental Protection Agency,
OGWDW/SRMD/TSC.
C14 Hertz, C., Brodovsky, J., Marrollo, L., Harper, R. (1992). Minimum Reporting Levels
Based on Precision and Accuracy for Inorganic Parameters in Water. Proc. WQTC,
Toronto.
C15 Urbansky, E. T. (2000). Perchlorate in the Environment (Vol. 440): Springer.
C16 APHA, AWWA, WEF. (2017). 1030 DATA QUALITY. Standard Methods For the
Examination of Water and Wastewater. Retrieved from
https://www.standardmethods.org/doi/abs/10.2105/SMWW.2882.0Q6.
C17 Oxenford, J. L., McGeorge, L. J., Jenniss, S. W. (1989). Determination of Practical
Quantitation Levels for Organic Compounds in Drinking Water. Journal -AWWA, 57(4),
149-154. doi:10.1002/j. 1551-8833.1989.tb03193.x
C18 Sanders, P. F., Lippincott, R. L., Eaton, A. (1996). Determining quantitation levels for
regulatory purposes. Journal - AWWA, 55(3), 104-114. doi: 10.1002/j. 1551-
8833.1996.tb06523.x.
C19 Hubaux, A., Vos, G. (1970). Decision and detection limits for calibration curves. Analytical
Chemistry, 42(8), 849-855. doi:10.1021/ac60290a013.
C-9
-------
Appendix D
Example Method Validation Summary and Associated Full Method Validation Report
D-l
-------
Appendix D provides an example of a Method Validation Summary developed from a
Method Validation Report previously prepared by EPA Region 7. This is meant to provide
guidance on how a Method Validation Summary is constructed and derived based on Table 1 (in
Section 7.2 of the document), the information provided in Section 7.2, and the full Method
Validation Report.
Below is the example Method Validation Summary. Though information for an individual
Method Validation Summary will change based on the method it is describing, the same table
should be used to prepare all Method Validation Summaries.
Note that the Method Validation Report provided following the Method Validation
Summary serves to provide background for the construction of the Method Validation Summary
and is not meant to dictate how Method Validation Reports should be prepared. The format of
such reports will vary and should be based on program requirements.
Example Method Validation Summary
EPA Region 7 "Stir Bar Sorptive Extraction (SBSE or Twister™)
Collaborative Research Project
A
Validation Design
Description
1
Number of Laboratories
1
2
Number of Matrices
1 (surface water)
3
Types of Matrices Tested
(water, soil, sediment, etc.)
Surface water in the Kansas City Urban area; Tested three
locations on 12 different streams in the Kansas City area.
Noted that 1 -2 streams had high chlorine which impacted IS
results.
B
Method Validation
Overview
Description
1
Method title
Stir Bar Sorptive Extraction (SBSE or Twister)
2
Author(s) list
Lorraine Iverson (Kimball), EPA Region 7 Science and
Technology Center
3
Date
January 8, 2010 (Final Internal Report with Attachments—
Region 7)
4
Purpose
Test new sorptive extraction technique that reduces the use of
methylene chloride while providing better sample results.
Develop alternative test procedure for polycyclic aromatic
hydrocarbons.
5
Qualitative or Quantitative
Quantitative
6
Target
Analytes/Parameters
66 (45) semi-volatile organic compounds including PAHs
(18), 17 (14) pesticides, 4 pharmaceutical and personal care
products, 5 brominated flame retardants
NOTES
Have expanded the list to include selected herbicides. Evaluating in-situ sample collection.
C
Method Development
Considerations
Description and/or Results
D-2
-------
1
Sample Cost
Significant reduction in costs for sample shipment, waste
disposal, and solvent purchases; Annualized savings over
traditional techniques of up to $2162 in solvent and
glassware costs and 75% reduction in shipping costs
2
Sample Holding Times
Tested for holding time—results good for 14 days without
preservation
3
Sample Preservation
Tested for holding time—results good for 14 days without
preservation
4
Waste Generation
Significant reduction in solvent usage and corresponding
waste disposal; Annualized savings of up to 32 gallons of
solvent, hundreds of glassware
NOTES
D
Method Performance
Characteristic
Description and/or Results
1
Bias/Trueness
Met SW846 8270 and EPA 625 criteria
2
Detection Capability and
Quantification Capability
Detection limit is 10-100 times lower than SW-846 8270 and
EPA 625, pesticide results are comparable to 608 by gas
chromatography/electron capture detection
3
Instrument Calibration
For polycyclic aromatic hydrocarbons: Linearity of the
calibration curves was excellent for the range of 0.2 ug/L to 8
ug/L - a factor of 40.
Overall Summary: Linear range varied from 40-fold to only
4-fold
4
Measurement Uncertainty
Excellent internal standard area reproducibility, at <10%
with no interferents
5
Precision
Met SW846 8270 and EPA 625 criteria
6
Range
0.1-20 (ig/L
7
Ruggedness
Eight extraction parameters were tested: liners, split flow
rates, range of sample volumes and stir times, temperature
for desorption, extraction additives (methanol or salt),
immediate removal or wait time, reanalysis of stir bar for
removal rates
8
Selectivity in the Presence
of Interferences
Consistent with traditional semi-volatile organic compound
and pesticide methods on gas chromatography/mass
spectrometry
NOTES
This method has also been tested on three water sources as part of a multi-laboratory study and is one of
the accepted solid phase extraction techniques in the updated EPA Method 625.
D-3
-------
The following is the Method Validation Report, previously prepared by EPA Region 7,
that was used to construct the Method Validation Summary presented above. This Report is being
used to illustrate the development of a Method Validation Summary and is not indicative of how
a Method Validation Report must be prepared as such reports should be structured based on
program-specific requirements.
D-4
-------
Stir Bar Sorptive Extraction (SBSE or Twister™)
Collaborative Research Project
July 2009 to September 2009
Final Internal Report with Attachments
Margie St. Germain
Lorraine Iverson
ENSV/CARB/ORCS
Jeff Robichaud
Laura Webb
ENSV/EAMB
January 8, 2010
D-5
-------
Executive Summary
EPA Region 7 completed a 90 day research project to test the feasibility of using a new and
novel technology: Stir Bar Sorptive Extraction (SBSE or Twister™) for aqueous environmental
samples. We accomplished the tasks that we planned and generated data to show the limits of
feasibility using surface water samples. Twisters™ work well for neutral compounds that have
limited solubility in water. Because of the benefits of this technology, we would like to purchase
the equipment, implement sample analysis for PAHs while continuing work with other
compounds.
Accompli shments
• Analyzed 66 Semi-volatile compounds, meeting method requirements for 42
compounds, including all 18 polyaromatic hydrocarbons.
• Analyzed 17 UAA pesticides, meeting method requirements for 14 compounds.
• Analyzed 4 pharmaceutical and personal care products with mixed success because of
their solubility in water.
• Analyzed 5 brominated flame retardants with an indication of good success.
• Completed the documentation necessary for an Alternate Test Procedure for
polyaromatic hydrocarbons.
• Drafted a Standard Operating Procedure for future use.
Benefits of Technique
• Significant reduction in solvent usage and corresponding waste disposal.
• Significant reduction in staff time to perform extraction and analysis.
• Significant reduction in costs for sample shipment, waste disposal, and solvent
purchases.
• Significant reduction in staff exposure to repetitive movements and toxic solvents.
• Faster turn-around time for emergency response and screening unknowns.
• Field portability of the extraction process.
• Ability to analyze semi-volatile compounds, pesticides, and brominated flame
retardants in a single aqueous sample.
Possible Future Direction
• Re-evaluate 12 semi-volatile compounds at a lower concentration to meet method
detection criteria.
• Evaluate remaining pesticides, and gather data for an ATP for pesticides.
• Select an appropriate list of compounds and perform the analyses needed for the ATP
for the brominated flame retardants.
• Identify other environmental contaminants that would be amenable to this novel
technology.
• Investigate LC/MS methods and instrumentation in order to analyze samples for the
water soluble contaminants that are not amenable to Twister™.
D-6
-------
Outline
Executive Summary 2
I. Introduction 4
II. Background 4
III. Detailed Overview
A. Field Procedures 7
B. Overview of Conditions and Compounds Evaluated 7
C. Sample Preparation Procedures 11
D. Analysis Procedures 11
E. Analytes Tested and Overall Results
1. Semi-volatile Organic Compounds 13
2. Polyaromatic Compounds 17
3. UAA Pesticides 19
4. Pharmaceutical and Personal Care Products 20
5. Brominated Flame Retardants 21
F. Holding Time Study for Semi-volatile Organic Compounds 21
G. Comparability Study 24
H. Matrix Interferences 25
IV. Cost Benefit Analysis 25
V. Conclusions 27
VI. Future Work and Development 30
VII. Acknowledgement and Disclaimer 31
VIII. References 32
IX. Attachments
A. Overview of Project Activities A-l
B. SVOC Data B-l
C. PAH Data C-l
D. UAA Pesticide Data D-l
E. PPCP and BFR Data E-l
F. Holding Time Study Data F-l
G. Comparability Study Data G-l
H. ATP for PAHs and selected SVOCs H-1
I. Draft SOP for Twister 1-1
D-7
-------
I. Introduction
On July 6, 2009, EPA Region 7 entered into a loan agreement with Gerstel, Inc. for a period of
60 days which was later extended to 90 days. During this period we had free use of the Gerstel
equipment to perform studies related to environmental analysis of surface waters. The Gerstel
equipment included the autosampler, thermal desorption unit (TDU), cooled injection system
(CIS), 15 position stir plate, and 10 Twisters™. We purchased an additional 40 Twisters™ for
this project. The initial focus of the study was on the semi-volatile compound list, hoping to
obtain satisfactory recovery and precision data to warrant an Alternate Test Procedure for the
analysis in semi-volatile compounds in water. When the loan agreement was extended, we
shifted our focus to add several compounds of concern for urban stream studies. These
compound classes included traditional pesticides, polyaromatic hydrocarbons (PAHs), selected
pharmaceutical and personal care products (PPCPs), and brominated flame retardants (BFRs).
This report summarizes the work and the findings of this 90 day project, and provides our
recommendations for future work and directions. The report details are summarized in the
attachment while the body of this report provides a detailed overview. The attachments are
scientific reports of the actual work, and should be able to stand on their own for the specific
topic.
II. Background
Traditional extraction techniques suffer from several problems and limitations.
First, the technique uses dichloromethane, a toxic solvent. The preferred technique is a
liquid/liquid extraction using a mechanical shaking motion. This technique depends on
comparatively large volumes of dichloromethane as the solvent (approximately 500 milliliters).
Most laboratories are moving towards "greener" techniques including liquid-solid extraction (C-
18 impregnated Teflon discs or columns), micro-extraction, and solid phase micro-extraction
(SPME). Under executive order 13423, section 2. (e.) (i.), all EPA laboratories are charged with
reducing the use of 15 toxic chemicals, including dichloromethane.
Second, the liquid/liquid extraction technique is not efficient in either time or labor. This is a
relatively labor intensive process, taking roughly 8 hours to complete an extraction of a set of 12
samples. Shaking 1 liter (L) of sample with solvent is physically demanding, and is multiplied by
the number of samples extracted. Additional time is needed to perform clean up steps, if needed,
and to concentrate the extract to 1 milliliter (mL), resulting in staff time of 2-4 days for a set of
12 samples.
Third, the sensitivity is moderate, and occasionally is barely sufficient for the lower action
levels. One liter of sample is extracted and concentrated to 1 mL of extract. Then, only 1
microliter (|iL) of the final extract is injected for analysis. This translates to essentially extracting
and analyzing only 1 milliliter of sample. Our reporting limits for most analytes are 2
micrograms per liter (|ig/L).
Fourth, our traditional extraction technique is not portable. It cannot be done anywhere except
within a permanent laboratory. Therefore, field samplers must collect the samples in one gallon
D-8
-------
glass containers and transport them back to the laboratory. These samples are very heavy, and if
they need to be shipped, the cost of shipping is high. The field samplers must also be cognizant
of the seven day holding time from the time of collection to the time of extraction. The field
samplers do not have the luxury of keeping the samples with them for several days until it is
convenient to return to the laboratory.
Stir bar sorptive extraction (SBSE or Twister™) is a relatively new technique to extract organic
compounds from aqueous or other liquid samples for analysis by instrumentation such as gas
chromatography/mass spectrometry (GC/MS). Twister™ employs an adsorptive coating on a
magnetic stir bar. The technique is similar to solid phase micro-extraction (SPME) in theory, but
in practice is considerably different due to the difference in physical design. The Twister™
technique provides enhanced sensitivity over the traditional extraction techniques such as
liquid/liquid because the complete extracted fraction is quantitatively introduced into the GC/MS
system by thermal desorption. While stirring the sample solution, the Twister™ efficiently
extracts organic compounds from aqueous or other liquid samples such as milk. After extracting
analytes from the samples, each Twister™ is placed into a sealed Twister™ desorption liner,
placed in a TDU and introduced directly into the GC/MS. Additionally, Twister™ has the
advantage over SPME due to the larger amount of sorptive material (polydimethylsiloxane) in
contact with the sample on a physically stable form.
Currently, this technique is used in the food manufacturing industry for quality control purposes.
There has not been an extensive evaluation of this technique in the environmental chemistry
industry in the United States. However, there are 3 articles published from European sources
which show that Twister™ is promising in the area of environmental chemistry. EPA action
limits have been traditionally set at the analytical method detection limits, parts per billion
(ppm). Recently, EPA is shifting towards human risk values as action levels requiring that the
analysis be pushed to lower concentrations, parts per trillion (ppt). In a publication about
research in Germany using the Twister™ [1], the method detection limits for PAHs are in the
low nanograms per liter (ng/L) range instead of the traditional low ug/L range. These lower
detection limits approach the concentrations related to risk factors for humans. Currently,
pesticides and polychlorobiphenyls (PCBs) are analyzed by a non-specific method of gas
chromatography (GC) to reach the medium ng/L concentrations required in the regulations. A
second publication from Spain [2] shows data that the method detection limits from Twister™
and gas chromatography/mass spectrometry (GC/MS) can reach low ng/L using full scan and
acquiring confirmatory mass spectra for each compound. A third publication also from Spain
shows that three classes of compounds can be detected on the Twister™ in a single analysis, thus
eliminating multiple extractions. Again, the method detection limits reach the low ng/L or parts
per trillion range. Frequently, when EPA analyzes surface water samples for environmental
contaminants, EPA looks for a variety of compounds requiring multiple extractions and analyses.
These three articles indicate that the use of Twister™ for environmental sample testing would be
feasible.
The under-appreciated attribute of Twister™ is the potential application for "unknown" analysis.
For example, working with the water emergency response group, a water sample could be
extracted and analyzed within one or two hours and produce a tentatively identified compound
report where the unknown spectra are compared with a spectral library. Any number of organic
D-9
-------
contaminants could be quickly detected and identified with this approach at parts per trillion
(ppt) levels.
Staff members from the EPA Region 7 branches (ENSV/EAMB and ENSV/CARB) worked
jointly to plan and implement the research study to evaluate the Twister™ technique. Laura
Webb led the field activities including the sample extraction in the field. Lorraine Iverson led the
laboratory evaluation. Margie St. Germain coordinated the project and assisted where needed.
During the planning meeting, the team of chemists developed a list of questions that needed to be
answered in determining the merits of the Gerstel system. Those questions were
• What are the optimum conditions for extraction?
• What are the optimum conditions of the instrument for analysis?
• What is the precision and accuracy of the Gerstel technique?
• What are the method detection limits for semi-volatile compounds?
• Can traditional method criteria be met using this technique?
• What is the holding time of the Twisters™ once the samples were extracted?
• Can all semi-volatile compounds be extracted by this technique?
• What other compounds would be amenable to this technique?
• Does the matrix interfere with this technique?
• Would it be possible to gather enough data for an Alternate Test Procedure for the
target analytes in the Kansas Urban Stream Studies?
The ENSV/EAMB monitoring team, led by Laura Webb, collected surface water samples from
twelve different surface waters in the Kansas Urban Stream study. The water monitoring team
collected the normal two gallons of sample to be sent back to the laboratory for the semi-
volatiles and pesticides analysis in water using the traditional liquid/liquid extraction technique.
In addition, the team exposed a separate much smaller portion of the stream sample (10
milliliters initially) to the Twister™ in the field at the time of sample collection.
The ENSV/CARB team, led by Lorraine Iverson, analyzed the samples, both for the traditional
method and for the Gerstel method. The gallon water sample was extracted and analyzed in the
laboratory by following existing methodology. The field extracted Twisters™ were brought back
to the laboratory also and analyzed with the GC/MS system. Two additional samples were
laboratory extracted Twisters™ using a small aliquot from the gallon samples.
The quality assurance requirements for this activity are described in the corresponding Quality
Assurance Project Plan (QAPP) for the Kansas Urban Streams Study. The same quality control
procedures were followed for both techniques. Method blanks, lab control samples, and matrix
spikes were prepared and extracted for the Twister™ technique in the field. Additional quality
control samples were also prepared in the laboratory. Finally, a method detection limit study was
performed for the Twister™ method concurrently with the Kansas Urban Stream sampling
events.
D-10
-------
III. Detailed Overview
Because of the sheer volume of data that was generated by this project, discussion of each data
point would be extensive. This detailed overview will summarize the results in sufficient detail
to provide an understanding of the project results. More detailed discussions are provided in the
subsequent attachments.
1. Field Procedures
The Kansas Urban Streams project was chosen to initially test the Gerstel technique. This is an
on-going project of the EAMB branch with the purpose of monitoring urban streams throughout
the Kansas City Metro area. There are a series of 12 streams, with three sampling sites on each
stream, which have been sampled for chemical, physical, and biological parameters for a period
of 4 years [4], These streams were scheduled for sampling at the same time as the equipment
loan. These surface water samples provided a perfect media for testing the durability of the
equipment and feasibility of the project. Specifically, samples provided an environmental media
where there is potential for typical urban analytes, such as pesticides, pharmaceuticals, personal
care products, and other industrial chemicals. The water sampling was done with the water
monitoring mobile laboratory, thus offering a good platform for the field portion of the study.
The samples were collected beginning in mid-July through late August. Each sample was
collected in two 4-L amber bottles for the traditional laboratory procedures, and in a 40 mL VOA
vial for the Twister™ study. The volatile organic analysis (VOA) vials were immediately placed
in a dark cooler, and later stored in the refrigerator in the mobile lab. With one exception,
samples were extracted via the Twister™ technique on the same day as sampling in the mobile
lab. The final set of samples was extracted in the regional laboratory on the following day. The
techniques used in the mobile lab were identical to those used in the regional laboratory, with the
exception that the field sample vials were pre-loaded with modifier in the regional laboratory
prior to field sampling to reduce the material needed in the mobile lab.
B. Overview of Conditions and Compounds Evaluated
Once the instrument was set-up, preliminary conditions were determined and tested. With each
series of tests, the conditions were further optimized. Five problems were identified and resolved
during this project. Finally, additional environmental contaminants were evaluated. A timeline of
events and general data results are presented in Attachment A.
1. Conditions and Parameters Tested
Eight parameters and conditions were tested; ensuring the analysis of water samples would
provide accurate and reproducible data.
Different liners were available for the injector. These liners were open tubular, beveled tubular,
glass bead filled, glass wool filled, tenax filled, and quartz wool filled. Initial analysis used the
default tenax filled liner resulting in many of the compounds not being transferred to the GC/MS.
Next we used the beveled tubular and the open tubular. However, there was not enough surface
area to capture the high boiling compounds in the semi-volatile list. Next we tried the glass filled
liner, and the very reactive compounds, such as 2,4-dinitrophenol and 4,6-dinitro-2-
methylphenol, were being captured or destroyed on the liner. Finally, we tried the quartz wool
D-ll
-------
liner which provided a large surface area for capturing the compounds including the reactive
compounds.
We tried several variations of the split flow and splitless flow at the injector. We determined that
for parts per billion concentrations, that split flows at 80 ml/min and a 1:5 split were optimum for
full scan quantitation. However, as the concentrations dropped to the parts per trillion, then
splitless flow was optimal. When using the splitless option, the standards had to be less
concentrated to obtain a linear range for quantitation.
We evaluated a range of sample volumes and stir times. We tried 5 ml, 10 ml, and 100 ml sample
volumes. When evaluating samples for semi-volatile compounds, we found that 10 ml samples
did not significantly improve the results, so the samples were analyzed with 5 ml aliquots. With
this volume, we were able to use 60 minutes (min) as the optimum stir time. We also tried 90
min and 120 min stir times, and again did not observe significant improvements for these times.
Any stir times greater than 120 min would not be amenable to field extractions. When we began
testing the brominated flame retardants, we used 100 ml samples with a 24 hour stir time
allowing us to reach parts per trillion concentrations. Therefore, for routine analyses, 5 ml
sample aliquots stirred for 60 minutes would be the optimum time and volume, while for
extremely low concentrated analyses, 100 ml sample aliquots stirred for 24 hours would be
optimum.
We used a stir rate of 1200 rpm for all tests. We did not evaluate stir rates for this project.
We evaluated the temperatures during desorption. We tried -120°C and -70°C for the low
temperature on the CIS. When we saw no difference between these two temperatures, we used
the -70°C for all analyses. For the high temperature, we began at 300°C. When the high
molecular weight compounds, high boiling PAHs, were not efficiently removed from the
Twister™, the temperature was raised to 310°C and held for 15 min. This solved the carryover
problem of the late eluting compounds in the SVOC standards.
We evaluated the need for additives to the samples to enhance the adsorption onto the Twister™.
The additives we considered were methanol, salt, acid, and base. All tests with base eliminated
all the acidic compounds; therefore, base was not used with the project. Methanol improved the
adsorption of the PAHs onto the Twister™ by preventing adsorption onto the glassware. We
tested 20% methanol and 50% methanol as an additive. The 50% methanol did not significantly
improve the recovery of the compounds, so 20% methanol was used for all the tests. Salt
improved the adsorption of the compounds with a low octanol-water partition co-efficient
(Ko/w) or the compounds that were water soluble. We tested 30% salt and 50% salt, discovering
that the 50% salt additive did not significantly improve the recovery of compounds. Therefore,
30%) salt was used for all tests. Finally, we tested the addition of acid in order to enhance the
recovery of acidic compounds. Acid did improve the recovery of acidic compounds more so
when added to the salt portion rather than the methanol portion. When we added the acid first,
we lost all basic compounds. Therefore, when evaluating acidic compounds in a mixture, we first
added salt and stirred for 60 min, then added the acid and stirred for an additional 60 min. This
technique provided the best recovery of salt and acid combinations.
D-12
-------
We wanted to know if removal of the Twister™ from the sample was critical for reproducibility.
We tested replicate samples, some where the Twister™ was removed immediately and some
where the Twister™ was allowed to sit for an additional 60 min without stirring. We observed no
significant differences between the replicate sets.
We were also curious if all the compounds were removed from the sample with the first
Twister™ run. We set up a series of analyses where the samples had one Twister™ spun. Upon
completion, the first Twister™ was removed from the sample, and a second was added and the
sample stirred for an additional 60 minutes. We learned that the basic compounds and the neutral
compounds generally had 90% of the compounds adsorbed onto the first Twister™ and less than
10% adsorbed onto the second Twister™. The acids and the water soluble compounds were
extracted from the sample with more difficulty and ranged from 10% to 50% onto the first
Twister™ with something less onto the second Twister™.
Finally, we were concerned with the holding time after a Twister™ has been used to extract a
sample. Discussion of the results is presented Section F of this report. Generally, the Twisters™
reproduced the results for a 14 day period without any loss.
2. Problems Resolved
We identified and resolved five problems with varying degrees of success. The problems
included phthalate contamination, carryover of the high boiling compounds, poor recoveries of
the basic compounds, poor recoveries of several acidic compounds, and poor precision of the
internal standards.
We observed sporadic phthalate contamination during the first six weeks of the project. Initially,
some of the Twisters™ had not been properly cleaned and contained traces of phthalate, a
common contaminant from plastics. The phthalate contamination was reduced by changing the
cleaning technique but still observable. Next, we identified possible sources of contamination in
the laboratory. By modifying our handling protocols for the Twisters™ and their containers, the
phthalate concentration was further reduced but not eliminated. In discussions with Gerstel
representatives, we agreed that the new, 1 mL vials could have contamination that would transfer
to the Twisters™ while being stored. Future work will ensure that the Twisters™ are stored in
the transfer tube and adapter which have been cleaned of phthalates. These practices should
eliminate any laboratory and handling sources of contamination.
During the initial testing we observed non-reproducibility of the high boiling compounds and the
high molecular weight, aromatic compounds. We determined that the CIS temperature of 300°C
held for 1 min was not sufficient. Increasing the temperature to 310°C and holding for 15
minutes improved the results; however, the precision was still not sufficient. Late in the study,
the transfer temperature between the TDU and CIS inlet was raised from 275°C to 325°C, which
improved the results further.
In the semi-volatile list there are five very basic compounds which never performed well. The
compounds were N-nitroso-di-n-propylamine, 4-chloroaniline, 2-nitroaniline, 3-nitroaniline, and
4-nitroaniline. These compounds are all water soluble. N-nitroso-di-n-propylamine is easily
D-13
-------
degraded at high temperatures. A different sorbent or a different analytical technique would need
to be employed to obtain acceptable data for these compounds.
In the semi-volatile list there are 17 acidic compounds. We were able to improve the recoveries
of 11 compounds by using a two step extraction with Twisters™, first with 30% salt followed by
acid addition to pH<2. Six acidic compounds are also water soluble and good recoveries were
never obtained for these compounds (phenol, benzyl alcohol, benzoic acid, 2-methylphenol, 4-
methylphenol, 2,4-dinitrophenol).
Throughout the first eight weeks, we observed that the internal standard areas were not very
precise as with standard methods, specifically l,4-dichlorobenzene-D4 and perylene-D12. We
had been spiking the samples with the method prescribed concentrations using the traditional
methods. We also noticed the normal high concentrations of the calibration curves were not
consistent with the traditional methods. When we finally reduced the concentration by a factor of
ten, and used lower concentrations for the calibration curves, we were able to obtain precise
internal standard areas for the calibration curves and for subsequent samples. We were able to
see standard deviations less than 10% for the areas of all six internal standards. Using the
traditional high level standard, we calculated that we were loading a single Twister with 75,000
ng of contaminants and internal standards. We determined that we had saturated the Twisters
with the concentration of possible compounds. Therefore, for future work, the calibration curves
and the internal standard areas should be reduced in order to properly function.
3. Rationale for the Change in Scope
When we were nearing the original deadline, we realized that we would not be able to resolve all
the issues for the complete semi-volatile list of 66 compounds. At that point, we were routinely
getting acceptable results for 45 compounds out of 66 compounds. Our ultimate goal was to
develop an Alternate Test Procedure (ATP) document for one list of compounds. We realized
that the list of Kansas Urban Stream analytes contained mostly neutral compounds, including
pesticides and emerging contaminants. Therefore, we were able to receive an additional 30 days
on the loan of the equipment. During this time we focused on completing the components of the
ATP for PAHs, and determining the feasibility of adding phthalates, pesticides and other
emerging contaminants such as pharmaceutical and personal care products (PPCP) and
brominated flame retardants (BFR). By determining the feasibility of these additional
compounds, we would be able to define the method development for an additional ATP. Detailed
discussions are provided in Section E on each of the categories of compounds investigated.
C. Sample Preparation Procedures
The samples were measured into extraction vials (either 20 mL VOA vials if using 5 or 10 mL
sample sizes, or larger bottles when using 100 mL sample size). Four reagent water aliquots were
prepared, two for the method blank and two for the laboratory control sample (LCS). One of the
samples was chosen for matrix spike (MS) and matrix spike duplicate (MSD); four additional
aliquots of this sample were also prepared. To each aliquot, 10 |iL of the internal
standard/surrogate solution was added. To the LCS, MS, and MSD 10|iL of spiking solution was
added. The Kansas Urban Stream samples were all analyzed using two sample aliquots - one
with 20% methanol as the modifier, and one with 3g of salt and concentrated sulfuric acid to pH
< 2. Initially, this salt/acid combination was performed at one time and the sample stirred for the
D-14
-------
same time period as the methanol portion at 90 minutes. However, during the project it was
determined that improved recoveries of basic compounds were obtained if the salt were added
first, the sample stirred for a period of time, then the acid added and the sample stirred again for
a total time of 120 min. The final sets of samples used this procedure: 5 mL sample with 20%
methanol for 1 hour, 5 mL sample with 3g salt for 1 hour, then several drops of acid added and
this portion stirred for an additional hour. The methanol portion was held until the other portion
was finished, and then both stir bars were removed, rinsed with de-ionized (DI) water, and
placed in a TDU tube. It was determined that the stir bars would be placed in the tubes in the
same position for all samples and standards, thus the methanol stir bar was inserted first, then
followed by the salt/acid stir bar. The TDU tube was then sealed with a transport adaptor and
placed in the sampler tray.
D. Analysis Procedures
The final analytical conditions are listed below. We evaluated several conditions, such as transfer
temperature, split and splitless desorption, and CIS initial temperature. These evaluations are
summarized Section B.
a. TDU
• Initial Temperature: 40°C
• Delay Time: 0.5 min
• Initial Time: 0.01 min
• Temperature Program: 60°C/min to 310°C, hold 5 min
• Transfer Temperature: 325°C, fixed
• Splitless Desorption
b. CIS
• Initial Temperature: -70°C
• Equilibrium Time: 0.5 min
• Initial Time: 0.2 min
• Temp Program: 12°C/s to 310°C, hold 15 min
c. GC
• Column: HP 5 MS, 30m x ,25mm x .25 |im
• Initial Column Temperature Hold: 40°C for 2 min
• Column Temp Program: 40-284 @ 8°/min, 15°/minto 310°
• Final Column Temperature Hold: 310° C for 2.9 min
• Transfer line Temperature 280°C
• Carrier Gas: Helium at lmL/min constant flow
d. MS
• Source Temperature 230°C
• Electron Energy: 70 V (nominal)
• Mass Range: 35-450 amu (except for Brominated Flame Retardants
which require mass range 100 - 660 amu)
• Scan Time: 1.81 scans/s (1.36 scans/s for BFR)
e. Selected Ion Monitoring for Brominated Flame Retardants (SIM, Table 1)
f. Selected Ion Monitoring for Pharmaceutical and Personal Care Products (SIM,
Table 2)
D-15
-------
Table 1.
Group
SIM Conditio
Start Time
ns for Brominated Flame
Compound
ietardants
Quant Ion
Other Ions
1
10.00
Naphthalene-d8 (IS)
136.2
107.9
2-Fluorobiphenyl (s)
172.0
171.1
2
16.00
Acenaphthalene-DIO (IS)
164.2
162.2
Phenanthrene-dlO (IS)
188.2
184.2
Terphenyl-dl4 (s)
244.2
3
28.00
BDE-47
138.1
Chrysene-dl2 (IS)
240.2
236.2
BDE-47
325.9
485.8
BDE-100/BDE-99
403.7
405.7, 563.9
4
33.00
Perylene-dl2 (IS)
264.2
132.2
BDE-153
483.8
140.9, 643.4
IS = inte
rnal standard, s
Hexabromobiphenyl
= surrogate, BDE - bromin
307.8
ated dipheny
467.8, 627.4
ether
Table 2. SIM Conditions for Pharmaceuticals and Personal Care Products (PPCP)
Group
Start Time
Compound
Quant Ion
Other Ions
1
10.00
Naphthalene-d8 (IS)
136.2
107.9
2-Fluorobiphenyl (s)
172.0
171.1
2
17.00
Bisphenol A
213.2
119.1, 228.1
Phenanthrene-dlO (IS)
188.2
184.2
Terphenyl-dl4 (s)
Y
244.2
Triclosan
217.9
288.0, 289.9
3
28.00
Ethinyl Estradiol
213.1
160.0, 296.1
Estrone
270.2
146.0, 185.1
Chrysene-dl2 (IS)
240.2
236.2
IS = internal standard, s = surrogate
E. Analytes Tested and Overall Results
We fully tested the semi-volatile compounds listed in CARB Method 3230. 2 and EPA Method
625 against method performance criteria. The method performance criteria were calibration
curve linearity, method detection limit determinations, precision of four replications (initial
demonstration of proficiency-IDP), and matrix spike/matrix spike duplicate recovery and bias.
The semi-volatile list contains acidic compounds, basic compounds, and neutral compounds.
Within the list of neutral compounds, there is a shorter list of polyaromatic hydrocarbons
(PAHs). We tested the feasibility of additional compounds, including pesticides listed in EPA
Method 608, pharmaceutical and personal care products (PPCP), and brominated flame
retardants (BFRs) not previously analyzed by EPA Region 7. This section of the report
summarizes the specific compounds within each of these lists, and the successes and failures.
D-16
-------
1. Semi-volatile Organic Compounds (SVOCs)
We tested the semi-volatile compounds, not including the PAHs, summarized in Table 3 and
detailed in Attachment B. We were able to show Twister™ method comparability with EPA 625
for 23 compounds which are listed in Table 4. The extraction procedure was 20% methanol, 30%
salt, or 4 drops 1:1 sulfuric acid to a 5 mL sample size. These compounds met four method
performance criteria listed above.
Table 5 lists the 12 additional compounds that met three of four method performance criteria and
should be considered acceptable once the additional criteria are met. Generally, either the
precision was not within the strict ranges set by the method or the method detection limit
determinations needed to be performed at a different concentration.
For the compounds in Tables 4 and 5, the linear range varied from 40 fold to only 4 fold. The
internal standard areas were generally stable; however, the salt and acid extractions proved to be
susceptible to matrix effects. The base/neutral surrogate recoveries were generally comparable
with what one might obtain from CARB Method 3230.2F; however, the acid surrogates often
proved difficult due to lack of sensitivity or due to matrix effects. If these compounds are needed
by a client for a certain site, they would prove to be more difficult because multiple extractions
and analyses would be required. The time spent would likely still be less than with the traditional
technique, because extraction could be done all three ways (methanol, salt, and acid)
simultaneously on the same stir plate. The analysis time would be tripled, but this problem is
minimal if an auto sampler is used.
Finally, the analytes listed in Table 6 were unsuccessful. The majority of the unsuccessful
compounds fall into the acidic or basic category. Two unsuccessful compounds (benzyl alcohol
and benzoic acid) are not listed in the Clean Water Act for regulatory purposes, but are listed in
EPA Method 625. N-nitroso-di-n-propylamine can be thermally reactive, and may never do well
with the Twisters™. Some of the unsuccessful analytes are difficult at best when using the
traditional extraction and analysis method. Typically these compounds have low sensitivity
and/or poor chromatography. Some of the unsuccessful analytes are much more soluble in water
and may be best analyzed by an alternate method such as LC/MS. As different sorbents are
added to the Twister™ technology, these compounds could be re-evaluated
Table 3. Semi-volatile Compounds (not including PAHs)
Analyte Name
CAS#
Log Ko/w
Type
Phenol
108-95-2
1.46
Acid
bis-(2-Chloroethyl)ether
111-44-4
1.29
Neutral
1,3 -Di chl orob enzene
541-73-1
3.53
Neutral
2-Chlorophenol
95-57-8
2.15
Acid
1,4-Dichlorobenzene
106-46-7
3.44
Neutral
1,2-Dichlorobenzene
95-50-1
3.43
Neutral
Benzyl alcohol
100-51-6
1.10
Acid
2-Methylphenol
95-48-7
1.95
Acid
bi s(2-Chloroi sopropyl)ether
108-60-1
2.48
Neutral
Hexachl oroethane
67-72-1
4.14
Neutral
D-17
-------
Table 3 (continued). Semi-volatile Compounds (not including PAHs)
Analyte Name
CAS#
Log Ko/w
Type
4-Methylphenol
106-44-5
1.94
Acid
N-nitroso-di-n-propylamine
621-64-7
1.36
Base
Nitrobenzene
98-95-3
1.85
Base
Isophorone
78-59-1
1.70
Neutral
2-Nitrophenol
88-75-5
1.79
Acid
2,4-Dimethylphenol
105-67-9
2.30
Acid
bis(2-Chloroethoxy)methane
111-91-1
1.30
Neutral
2,4-Dichlorophenol
120-83-2
3.06
Acid
Benzoic acid
65-85-0
1.87
Acid
1,2,4-Trichlorobenzene
120-82-1
4.02
Neutral
4-Chloroaniline
106-47-8
1.83
Base
Hexachl orobutadi ene
87-68-3
4.78
Neutral
4-Chloro-3-methylphenol
59-50-7
2.70
Acid
Hexachl orocy cl opentadi ene
77-47-4
5.04
Neutral
2,4,6-Trichlorophenol
88-06-2
3.69
Acid
2,4,5-Trichlorophenol
95-95-4
3.72
Acid
2-Chloronaphthalene
91-58-7
3.90
Neutral
2-Nitroaniline
88-74-4
1.85
Base
Dimethylphthalate
131-11-3
1.60
Neutral
2,6-Dinitrotoluene
606-20-2
2.10
Base
3-Nitroaniline
99-09-2
1.37
Base
2,4-Dinitrophenol
51-28-5
1.67
Acid
Dibenzofuran
132-64-9
4.12
Neutral
4-Nitrophenol
100-02-7
1.91
Acid
2,4-Dinitrotoluene
121-14-2
1.98
Base
Diethylphthalate
84-66-2
2.42
Neutral
4-Chlorophenyl-phenyl ether
7005-72-3
4.70
Neutral
4-Nitroaniline
100-01-6
1.39
Base
4,6-Dinitro-2-methylphenol
534-52-1
2.13
Acid
N-nitrosodiphenylamine
86-30-6
3.13
Base
4-Bromophenyl-phenyl ether
101-55-3
4.94
Neutral
Hexachl orob enzene
118-74-1
5.73
Neutral
Pentachl orophenol
87-86-5
5.12
Acid
Carbazole
86-74-8
3.72
Neutral
Di -n-buty lphthal ate
84-74-2
4.50
Neutral
Buty lb enzylphthal ate
85-68-7
4.73
Neutral
3,3'-Dichlorobenzidine
91-94-1
3.51
Basic
bis(2-Ethylhexyl)phthalate
117-81-7
7.60
Neutral
Di -n-octy 1 phthal ate
117-84-0
8.10
Neutral
D-18
-------
Table 4. Summary of Successful Semi-volatile Compounds
CARB
Type of
Twister™
EPA
Method
extraction
MDL,
Method 625
3230.2F
Analyte
ug/L
MDL, ug/L
MDL, ug/L
bis(2-Chloroethyl) ether
0.23
5.7
1
Salt
1,3 -Di chl orob enzene
0.14
1.9
1
Methanol
1,4-Dichlorobenzene
0.15
4.4
1
Methanol
1,2-Dichlorobenzene
0.18
1.9
1
Methanol
Nitrobenzene
0.50
1.9
1
Methanol
bis(2-Chloroethoxy)
methane
0.53
5.3
1
Methanol
1,2,4-Trichlorobenzene
0.14
1.9
1
Methanol
Hexachl orocy cl opentadi ene
0.40
1
Methanol
2,4,6-Trichlorophenol
0.36
2.7
1.6
Acid
2,6-Dinitrotoluene
0.19
1.9
1
Salt
Dibenzofuran
0.08
1
Methanol
2,4-Dinitrotoluene
0.51
5.7
1
Methanol
4-Chlorophenyl-phenyl
ether
0.07
4.2
1
Methanol
4,6-Dinitro-2-methylphenol
0.21
24
1.8
Salt
N-nitrosodiphenylamine
0.32
1.9
1
Methanol
4-Bromophenyl-phenyl
ether
0.08
1.9
1
Methanol
Hexachl orob enzene
0.13
1.9
0.8
Methanol
Pentachl orophenol
0.25
3.6
2
Acid
Carbazole
0.36
2.5
Methanol
Di -n-buty lphthal ate
0.09
2.5
1.9
Methanol
Buty lb enzylphthal ate
0.06
2.5
1.6
Methanol
bis(2-Ethylhexyl)phthalate
0.79
2.5
1.1
Salt
Di -n-octy 1 phthal ate
0.29
2.5
1.1
Methanol
D-19
-------
Table 5. Summary of Compounds that May be Successful
Analyte
Log Ko/w
Type
Comment
2-Chlorophenol
2.15
Acid
Low sensitivity; Repeat MDL with
higher spike level
bis(2-
Chl oroi sopropy l)ether
2.48
Neutral
Repeat MDL with higher spike level
Hexachl oroethane
4.14
Neutral
Good performance; Recovery higher
than method upper limit (100%)
Isophorone
1.70
Neutral
Low sensitivity; Spike levels need to be
increased
2-Nitrophenol
1.79
Acid
Low sensitivity; Repeat MDL with
higher spike level
2,4-Dichlorophenol
3.06
Acid
Low sensitivity; Repeat MDL with
higher spike level
Neutral
Poor sensitivity and linearity;
Recoveries were above the method
Hexachl orobutadi ene
4.78
upper limit
4-Chloro-3-
methylphenol
2.70
Acid
Low sensitivity
Neutral
Poor sensitivity and linearity;
Recoveries were above the method
Dimethylphthalate
1.60
upper limit
Neutral
Poor sensitivity and linearity;
Recoveries were above the method
Diethylphthalate
2.42
upper limit
3,3'-Dichlorobenzidine
3.51
Base
Low sensitivity; Repeat MDL with
higher spike level
Table 6. Summary of Unsuccessful Compounds
Analyte
Log Ko/w
Type
Comment
Phenol
1.46
Acid
Poor chromatography; Poor sensitivity
Benzyl alcohol
1.10
Acid
Not regulated; Poor sensitivity
2-Methylphenol
1.95
Acid
Water soluble; Poor sensitivity
4-Methylphenol
1.94
Acid
Water soluble; Poor sensitivity
N-nitroso-di-n-
propylamine
1.36
Base
Thermally reactive; Poor sensitivity
Benzoic acid
1.87
Acid
Not regulated; Poor sensitivity
4-Chloroaniline
1.83
Base
Poor sensitivity
2,4,5-Trichlorophenol
3.72
Acid
Easily complexes with phosphates; Poor
sensitivity
2-Nitroaniline
1.85
Base
Water soluble; Poor sensitivity
3-Nitroaniline
1.37
Base
Water soluble; Poor sensitivity
2,4-Dinitrophenol
1.67
Acid
Poor sensitivity
4-Nitrophenol
1.91
Acid
Poor sensitivity
4-Nitroaniline
1.39
Base
Water soluble; Poor sensitivity
D-20
-------
2. Polycyclic Aromatic Hydrocarbons (PAHs)
We tested the polycyclic aromatic hydrocarbons, detailed summary is provided in Attachment C.
The PAH compounds (Table 7) are neutral compounds that can be extracted with a single stir
bar, 20% methanol as the matrix modifier, and 5 mL sample size. The sensitivity of the PAH
compounds was excellent, and reporting limits of 0.2 to 0.4 ug/L were obtained with this small
sample size. Table 8 shows the PAH compounds with the lowest MDL obtained by the
Twister™, and compared to the traditional method. The MDLs by the Twister™ method were a
factor of 10 or more less than EPA Method 625 and CARB Method 3230.2F.
Once all the system temperatures were optimized, all method performance criteria were met for
all PAH compounds attempted. Linearity of the calibration curves was excellent for the range of
0.2 ug/L to 8 ug/L - a factor of 40. Reproducibility of the internal standard areas was excellent,
and was typically less than 10% if interferences such as isooctane were not present. Laboratory
control samples and matrix spikes typically gave comparable recovery and precision to Methods
625 and 3230.2F. Surrogate recoveries (base/neutral surrogates only) were similar to those
expected in EPA Method 625 and CARB Method 3230.2F. Since the Twister™ technique is a
procedural calibration, the recoveries seen were slightly higher than those obtained by the
reference method, EPA Method 624.
Table 7. Polyaromatic Compounds
Analyte Name
CAS#
Log Ko/w
Naphthalene
91-20-3
3.30
2-Methylnaphthalene
91-57-6
3.86
2-Chloronaphthalene
91-58-7
3.90
Acenaphthylene
208-96-8
3.94
Acenaphthene
83-32-9
3.92
Fluorene
86-73-7
4.18
Phenanthrene
85-01-8
4.46
Anthracene
120-12-7
4.45
Fluoranthene
206-44-0
5.16
Pyrene
129-00-0
4.88
Benzo(a)anthracene
56-55-3
5.76
Chrysene
218-01-9
5.81
Benzo(a)pyrene
50-32-8
6.13
Benzo(b)fluoranthene
205-99-2
5.78
Benzo(k)fluoranthene
207-08-9
6.11
Indeno( 1,2,3 -cd)pyrene
193-39-5
6.70
Dibenz(a,h)anthracene
53-70-3
6.75
Benzo(g,h,i)perylene
191-24-2
6.63
D-21
-------
Table 8. Method Detection Limits for PAH Com
EPA Method
CARB Method
Twister™ MDL,
625—MDL,
3230.2F—MDL,
Analyte
ug/L
ug/L
ug/L
Naphthalene
0.07
1.6
1.00
2-Methylnaphthalene
0.07
1.00
2-Chloronaphthalene
0.06
1.9
1.00
Acenaphthylene
0.06
3.5
1.00
Acenaphthene
0.04
1.9
1.00
Fluorene
0.04
1.9
1.00
Phenanthrene
0.04
5.4
1.00
Anthracene
0.04
1.9
1.00
Fluoranthene
0.05
2.2
0.98
Pyrene
0.05
1.9
0.96
Benzo(a)anthracene
0.06
7.8
0.93
Chrysene
0.05
2.5
0.99
Benzo(b)fluoranthene
0.05
4.8
0.82
Benzo(k)fluoranthene
0.05
2.5
0.93
Benzo(a)pyrene
0.07
2.5
0.84
Indeno( 1,2,3 -cd)pyrene
0.28
3.7
1.00
Dibenz(a,h)anthracene
0.25
2.5
0.76
Benzo(g,h,i)perylene
0.20
4.1
0.92
Dounds
3. UAA Pesticides
We spent much less time with the pesticide list compounds, and thus have much less data with
which to judge success or failure. Table 9 summarizes the pesticides, and Attachment D provides
details. The log Ko/w values are promising for the majority of the analytes. As a reminder,
pesticides are analyzed by dual column GC/ECD which has a typical MDL of 100-1000 times
lower than full scan GC/MS. Several compounds compare quite favorably to the traditional
techniques.
Table 10 summarizes all but three pesticides which gave good results and the MDLs obtained
compared to the traditional technique. The successful compounds would need to be extracted two
different ways (methanol and salt) and analyzed separately. Time needed for the extraction and
analysis would likely be less than the traditional methods because the two extractions can be
done simultaneously, and an auto sampler minimizes the chemist's effort in analysis. However, it
should be noted that we performed the UAA pesticide list tests concurrently with the semi-
volatile list. Thus, pesticides and semi-volatiles that are amenable to the methanol extraction can
be combined into one extraction/analysis. The same is true for the analytes that are amenable to
the salt extraction.
Tebuthiuron, Chlorothalonil and Pyrethrins were unsuccessful. The first two gave poor response;
and Pyrethrins had problems with linearity. It should be noted, however, that Pyrethrins have the
same problems with linearity on GC/ECD.
D-22
-------
Table 9. Summary of Tested Pesticides
Analyte
CAS#
Log Ko/w
Diethyltoluamide
(DEET)
134-62-3
2.20
Propachlor
1918-16-7
2.20
Trifluralin
1582-09-8
5.30
Simazine
122-34-9
2.20
Atrazine
1912-24-9
2.61
Diazinon
333-41-5
3.80
Alachlor
15972-60-8
3.50
Malathion
121-75-5
2.30
Metolachlor
51218-45-2
3.10
Chlorpyrifos
2921-88-2
4.70
Pendimethalin
40487-42-1
4.80
Bromacil
314-40-9
1.70
Bifenthrin
82657-04-3
8.10
Permethrin
52645-53-1
6.50
Chlorothalonil
1897-45-6
3.10
Pyrethrins
8003-34-7
5.00
Tebuthiuron
34014-18-1
1.78
"able 10. Summary
of Successfu
Pesticides
Twister
TM
CARB Method
CARB Method
Best
MDL,
3240.2 - MDL,
3250.4 - MDL,
Compound
technique
ug/L
ug/L
ug/L
Diethyltoluamide
(DEET)
Salt
0.09
Propachlor
Salt
0.09
0.1
Trifluralin
Methanol
0.02
0.025
Simazine
Salt
0.17
0.065
Atrazine
Salt
0.85
1.5
0.056
Diazinon
Methanol
0.22
0.5
Alachlor
Salt
0.05
0.065
0.047
Malathion
Salt
0.17
Metolachlor
Salt
0.21
0.25
0.058
Chlorpyrifos
Methanol
0.03
Pendimethalin
Methanol
0.24
Bromacil
Methanol
0.41
Bifenthrin
Methanol
0.95
Permethrin
Methanol
4.11
4. Pharmaceutical and Personal Care Products
There were mixed results for the pharmaceutical and personal care product (PPCP) list. The
PPCP list includes a wide variety of compounds, such as ordinary daily use compounds like
D-23
-------
caffeine, over-the-counter pharmaceuticals like ibuprofen, wastes from prescription drugs like
hormone supplement, and ordinary contaminants like phthalates from plastic materials. This
wide variety of compounds creates the potential need for various analytical techniques. For
example, many pharmaceuticals are best analyzed with liquid chromatography/mass
spectrometry because of the solubility in water. Compounds that have a low Log Ko/w and are
readily soluble in water, like caffeine, do not perform well using the Twister™ technique. To get
a sense of the feasibility of using the Twister™ technique, we selected six compounds from the
extensive PPCP list to evaluate (Table 11). Details are provided in Attachment E.
Table 11. Summary of PPCP Considered
Compound
CAS#
Log Ko/w
Reporting Limit-SIM
Comment
Caffeine
Water soluble
Ibuprofen
Low sensitivity
Triclosan
3380-34-5
4.66
0.02 ug/L
Bisphenol A
80-05-7
3.64
0.02 ug/L
Estrone
53-16-7
3.43
0.05 ug/L
Ethinyl
Estradiol
57-63-6
4.12
0.05 ug/L
Three compounds performed well. The two estrogen compounds, estrone and ethinyl estradiol,
were chosen from a larger list of hormones. Bisphenol A was chosen from commonly used
plastics. These compounds showed good linearity using full scan GC/MS and using selected ion
monitoring (SIM) GC/MS. The estimated reporting limits are listed in Table 11. The compounds
were not detected in any samples using the standard 5 mL aliquot. The concentrations of these
compounds may be much lower than anticipated.
Triclosan appears to extract in any modifier and behaves well in SIM and scan, with small
volumes or large. Using the salt additive, linearity was acceptable; only one MS/MSD pair was
analyzed, and the recovery was not reproducible. Unfortunately, this compound was not
detected in any samples, and may be present at lower concentrations.
Three compounds did not perform well. Ibuprofen was inconsistent and showed poor sensitivity.
Caffeine was water soluble and was not extractable with Twister™. These compounds may need
to be analyzed by LC/MS.
5. Brominated Flame Retardants
Brominated flame retardants (BFR) are composed of polybrominated biphenyl ethers (BDE)
which are used to treat clothing for infants. For convenience, a commercially available mix of
several polybrominated biphenyl ethers was chosen to evaluate the method. The five compounds
that were evaluated were linear, especially those eluting early, and showed decent sensitivity
using either a larger sample volume or SIM. The proposed regulatory detection limits would
require that these compounds be analyzed using 100 mL samples or SIM. The compounds were
evaluated on only a few samples, and not detected. The one set of MS/MSD run showed
promising results using SIM. There was also one sample/MS run using 100 mL, and acceptable
recovery was obtained. Estimated reporting limits for the BFRs are summarized in Table 12, and
details are provided in Attachment E.
D-24
-------
Table 12. Summary of Brominated Flame Retardants Tested
Compound
CAS#
Log Ko/w
Reporting Limit-
SIM (ug/L)
BDE-47
5436-43-1
6.77
0.01
BDE-100
189084-64-8
7.66
0.01
BDE-99
60348-60-9
7.66
0.01
Hexabromobiphenyl
59080-40-9
9.10
0.05
BDE-153
68631-49-2
8.55
0.01
F. Holding Time Study for Semi-volatile Organic Compounds
We began a holding time study on July 22nd by preparing, extracting, and storing replicates of
samples at 4°C that had been spiked with a known amount of each compound from the semi-
volatile list. We wanted to assess whether it would be acceptable to extract samples with
Twister™ and then store them at 4 0 C before analysis. We analyzed the stored Twisters on days
0, 1, 2, 5, 8, 14, 21, and 28. The analytes, once captured on the Twister™ and stored, are stable
for at least 14 days. Benzoic Acid, 4-Chloroaniline, 3-Nitroaniline, and 4-Nitroaniline were
excluded from the study because they could not be effectively extracted and analyzed. Detailed
information about the holding time study can be found in Attachment F.
Charts 1 through 6 are of rolling averages broken into groupings of analytes by internal standard.
The previous two days' analyses and the current day's analysis are grouped for the rolling
average. For example, "to day 14" is an average of days 5, 8, and 14; "to day 28" is an average
of days 14, 21, and 28. The problem compounds Benzoic Acid, 4-Chloroaniline, 3-Nitroaniline,
and 4-Nitroaniline have been excluded from the following rolling averages. Visual inspection of
the charts confirms that the analytes are stable for at least 14 days. We observed a drop in
concentration of the late eluting compounds, especially PAHs, as seen by the decrease of
concentration after Day 14 in Charts 5 and 6. Because system optimization continued during
these 28 days, the holding time study may need to be repeated to see if analytes within the
PDMS phase on the Twisters™ are stable for more than 14 days.
Chart 1. Semi-volatile Compounds in Group 1 (Dichlorobenzene-D4 - IS)
80.0 n
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
—~— Phenol
—bis(2-Chloroethyl)et
1,3-Dichlorobenzene
2-Chlorophenol
—*— 1,4-Dichlorobenzene
—•— 1,2-Dichlorobenzene
—i— Benzyl
—2-Methy Iphenol
—bis(2-Chloroisopropy
Hexachloroethane
4-Methylphenol
N-nitroso-di-n-propy
D-25
-------
Chart 2. Semi-volatile Compounds in Group 2 (Naphthalene-D8 - IS)
dayO to day 1 today2 today5 today8 to day 14 today21 today28
- Nitrobenzene
- Isophorone
2-Nitrophenol
2,4-Dimethylphenol
- bis(2-Chloroethoxy)m
-2,4-Dichlorophenol
-1,2,4-Trichlorobenze
Naphthalene
Hexachlorobutadiene
4-Chloro-3-methylphe
2-Methylnaphthalene
Chart 3. Semi-volatile Compounds in Group 3 (Acenaphthene-DIO - IS)
70.0
day 0 to day 1 to day 2 to day 5 to day 8 to day 14 to day 21 to day 28
- Hexachlorocyclopenta
-2,4,6-Trichloropheno
2,4,5-Trichloropheno
2-Chloronaphthalene
-2-Nitroaniline
-Acenaphthylene
- Dimethylphthalate
-2,6-Dinitrotoluene
Acenaphthene
2,4-Dinitrophenol
Dibenzofuran
4-Nitrophenol
2,4-Dinitrotoluene
Fluorene
Diethylphthalate
- 4-Chloropheny l-pheny
Chart 4. Semi-volatile Compounds in Group 4 (Phenanthrene-DIO - IS)
60.0
50.0
40.0
30.0
20.0
10.0
- 4,6-Dinitro-2-methyl
- N-nitrosodiphenylami
4-Bromophenyl-phenyl
Hexachlorobenzene
- Pentachlorophenol
- Phenanthrene
-Anthracene
- Carbazole
Di-n-butylphthalate
Fluoranthene
day 0 to day 1 to day 2 to day 5 to day 8 to day 14 to day 21 to day 28
D-26
-------
Chart 5. Semi-volatile Compounds in Group 5 (Chrysene-D12 - IS)
Chart 6. Semi-volatile Compounds in Group 6 (Perylene-D12 - IS)
G. Comparability Study
The comparability of our study had two parts. First, since this study utilized a mobile laboratory
to extract the majority of the field samples, we wanted to be able to show that extractions that
took place in that mobile laboratory were essentially the same as those that may take place in a
traditional laboratory setting. Second, we wanted to show that the Twister™ extraction and
thermal desorption analysis is essentially the same as or an improvement of the traditional
liquid/liquid extraction and liquid injection analysis.
For the comparison of field data to laboratory data, we evaluated field spiked samples with
laboratory spiked samples. Also, we looked at method blanks that were prepared in the field and
in the lab. Because of the variation we were naturally getting during analysis in the months of
July 2009 and August 2009, it was difficult to show comparability. The internal standards were
fluctuating such that replicate analyses of the same sample (duplicate continuing calibration
checks) did not necessarily agree with each other. We continued to optimize instrument
D-27
-------
conditions, and did not have perfect conditions on the days that field and laboratory spikes were
analyzed. In the end, we determined that we were spiking too much material into the samples for
the Twister™ technique. We should be able to get better results with lower concentrations.
For the comparison of the Twister™ technique to the traditional laboratory method, we were
more successful because in September 2009 we showed that the PAHs and several other
compounds gave data of a quality that is equivalent to EPA Method 625. This data is sufficient to
apply for an Alternate Test Procedure (ATP) in Region 7. For a detailed discussion of the data,
see Attachment G.
II. Matrix Interferences (Brenner Heights)
During the research project, one sample site exhibited matrix interferences. The sample, labeled
1001 and BH1, was from a small urban stream in Kansas City, Kansas named Brenner Heights
Creek. This sampling location has repeatedly posed problems with bacteria sampling results,
giving zero E. coli colonies each of the last three sampling seasons, while upstream portions of
the creek yield high levels of E. coli. As a result, residual chloride measurements were taken this
season, and the downstream results were significantly higher than the upstream portions of the
creek. There is a large influx of chlorine into the creek somewhere between BH1 and the next
upstream sample, labeled 1005 or BH2. That is the only significant difference between the three
sampling sites on this creek.
During the first analysis on July 31, 2009, there was a dramatic drop in the last two internal
standards for the Brenner Heights samples run with dual stir bars. However, these samples were
run during the time of high internal standard concentrations which resulted in variability in the
last two internal standard areas for all samples. On September 17, 2009, Brenner Heights was re-
sampled and residual chloride levels were also measured. These samples were run several times
over the course of the next week with various modifiers. The results show that while methanol
modifier does not seem to be affected, the samples that used salt and/or acid modifier gave poor
precision for the last two internal standards from Brenner Heights. It appears that changing the
ionic concentration of the sample with either acid or salt caused significant matrix affects,
especially toward the latter part of the chromatograph. It appears that residual chloride in surface
water samples interferes with the analyses, especially when salt or acid are used as a modifier.
This phenomenon should be further investigated.
IV. Cost/Benefit Analysis
During this project, EPA Region 7 tested and validated the extraction method without purchasing
the needed equipment. The basic Gerstel equipment lists for $55,688 which was loaned to us for
90 days. The Gerstel equipment with autosampler lists for approximately $75,000. We purchased
some liners and stir bars for approximately $1000.
EPA evaluated the realistic savings in time, materials, and manpower during this project. With
an average sample load of 140 water samples per year for semi-volatile organic analysis using
the Gerstel Twister™ with an autosampler, we would be able to
• Reduce dichloromethane usage by 6-13% or by $700-$1400,
• Reduce staff time per sample, and effectively increase sample capacity by 25%,
• Reduce glassware costs for sample containers from $890 to $128, and
D-28
-------
• Reduce sample shipping costs by 75% or more.
As other analytical methods are tested, additional cost savings may be realized. The sample data
would have to be evaluated in a new way. For example, a water sample could be analyzed for
most semivolatile compounds, pesticides, and PAHs within a single analysis instead of 3-4
separate extractions and analyses. Those few water soluble compounds that do not perform well
using the Twister™ would have to be analyzed by a different technique such as Liquid
Chromatography/Multi-sector Mass Spectrometer (LC/MS/MS).
Specific examples with immediate impact include:
1. The UAA stream samples are collected during the summer, typically 40-60 samples
spread out over several months. Without additional funding, all sample preparation and analysis
must be performed by EPA staff only. The maximum samples that could be collected in any
given two week period are 15 samples—not because of the collection process but because of the
extraction process in the laboratory. Samples are analyzed for SVOC, PAH, and Pesticides in
both water and soil. With 7 and 14 day extraction holding times respectively, the first two weeks
after sample collection is spent extracting, cleaning, and concentrating extracts prior to analysis.
With this type of schedule, the samples for this project are collected over a minimum of two
months with about 60 days to analyze and report the data after the receipt of the last sample. This
project takes three people full time and one part time technician for a period of 4 months to
complete the work.
By performing the analysis using Gerstel on the water samples for this one project, one chemist
can extract and analyze water samples, potentially for all analytes on a weekly basis. This will
leave the remaining two chemists to perform extractions on the solid samples only, prior to
analysis. Samples could be collected on a weekly basis (15/week), thus shortening the analysis
and reporting time to 2.5-3 months. This is a savings of staff time of three FTEs in one month, or
more.
2. A second stream study occurs every summer that looks at water quality issues. In 2009, it
was the Kansas tributaries project. This project collects 3-4 samples per week until 35-50
samples are collected. Because these samples trickle in, sample sets are very small resulting in
50% or more analyses being QC data instead of field sample data. Using the traditional method,
approximately 200 staff hours were spent extracting and analyzing samples for this project. Our
projections of staff time using the Gerstel system would be only 80 hours—mainly the time to
analyze samples.
D-29
-------
Table 13. Summary o
' Cost Benefits of Twister
Process
Per Sample
Savings
Annual Savings (a)
Comments
Solvent Reduction
300 mL —~ lmL
(S) 10 gals
(A) 32 gals
Hazardous Waste
Disposal
Sample Transport
Costs
1-3 L —~ 40 mL
(A) 109 gals -> 4
gals
Staff exposure to
solvents
Almost no exposure
Staff Time
3 days —~ minutes
Per batch of samples
(A) 65 days —~ 4
days
Sample Turn-around
and scheduling
Potential to increase
load with same
number of staff by
40% or more
Glassware
(S) 140 gallon jars
($890) —>140 VOA
vials ($128)
Twisters
Replace as needed.
Estimate 50 or more
uses within life of
stir bar.
(a) Annual savings is based on either (S) 140 samples per year for SVOCs or (A) 375
samples for pesticides, PAHs, and SVOCs. These numbers include the required quality control
samples assuming maximum batch size of 20 samples.
V. Conclusions
We were able to complete all the tasks we had planned and answer the questions listed in Section
II. The results from the tests indicate that some compounds can be extracted from water samples
no matter how the sample was modified, while other compounds are very particular and succeed
with only one type of modifier.
Essentially, all 18 PAHs met EPA Method 625 and CARB Method 3230.2 performance criteria,
some of the remaining semi-volatile compounds met or nearly met method criteria. There were
12 semi-volatile compounds that were not amenable to the Twister™ extraction, resulting in a
potential of 54 semi-volatile compounds meeting method performance criteria. These
compounds generally were water soluble and either an acid or a base. We had time to also test
pesticides used for UAA evaluations, pharmaceutical and personal care products and brominated
flame retardants. Preliminary data for the UAA pesticides indicate that at least 14 of the 18
pesticides can meet the Method 608 performance criteria. The remaining four pesticides need to
be evaluated. Preliminary data for the pharmaceutical and personal care products gave very
mixed results, many compounds not meeting default method criteria (calibration curve,
D-30
-------
extraction recovery). Review of these compounds indicates that they have a low logKo/w. These
compounds may benefit from another analytical technique such as LC/MS. Preliminary data for
the brominated flame retardants show that these compounds are amenable to the Twister™
technology. In fact, they tend to perform better at lower concentrations (ppt).
Some of the overarching accomplishments and findings include:
• Neutral compounds that are insoluble in water can be extracted with Twisters™.
• Method detection limits were lower in all compound classes by a factor of 10
compared to the liquid/liquid extraction technique.
• Many classes of compounds can be analyzed in the same sample simultaneously.
• All but two pesticides from the UAA list met method performance criteria.
• Process was fast and efficient, resulting in a fast turn-around capability for screening
unknown, aqueous samples.
• Process eliminated hazardous solvents and tedious bench work, reducing staff time
and strain.
• Process may allow us to meet the risk level concentrations for PAHs for Superfund
projects.
• Draft standard operating procedure documented the optimum conditions for the
Twister™ technique.
Alternate Test Procedure for UAA list for stream studies
Most of the UAA list pesticides that were attempted were successful in either a methanol or salt-
modified extraction. The MDLs rivaled those published in CARB Methods 3240.2 and 3250.4,
using GC/ECD methods. The Twister™ technique gave data that met method performance
criteria in these two methods. A formal demonstration of proficiency has not been completed.
One change for future work may include finding pesticide standards in an alternate solvent that
causes less interference than iso-octane. Based on the preliminary success of the pesticide
analysis, the full pesticide compound list may be evaluated, and the successful compounds will
be submitted for an Alternate Test Procedure.
Alternate Test Procedure for PAHs
Enough data was produced to show equivalency to EPA Method 625 for the PAH target
compounds. The method detection limits are between 5 and 25 times lower than those listed in
CARB Method 3230.2. The Twister™ technique was shown to be accurate enough and
reproducible enough to meet the method performance requirements of both EPA Method 625
and CARB Method 3230.2. We will be submitting this report with raw data to management and
Region 7 Regional Administrator for review and approval as an Alternate Test Procedure for
PAHs. Table 14 summarizes the compounds that meet or nearly meet the requirements for the
ATP.
D-31
-------
Table 14. Summary of Proposed Compounds for ATP
Compound
Group
(a)
Modifier
(b)
MDL
(ug/L) (c)
Comment (d)
bis(2-Chloroethyl)ether
svoc
Salt
0.23
1,3 -Di chl orob enzene
svoc
Methanol
0.14
2-Chlorophenol
svoc
Acid
Need MDL
1,4-Di chl orob enzene
svoc
Methanol
0.15
1,2-Di chl orob enzene
svoc
Methanol
0.18
bi s(2-Chl oroi sopropyl)ether
svoc
Methanol
Need MDL
Hexachl oroethane
svoc
Methanol
0.10
Nitrobenzene
svoc
Methanol
0.50
Isophorone
svoc
Salt
0.10
2,4-Dimethylphenol
svoc
Salt
0.24
bis(2-Chloroethoxy)m ethane
svoc
Methanol
0.53
2,4-Dichlorophenol
svoc
Acid
Need MDL
1,2,4-Trichlorobenzene
svoc
Methanol
0.14
Naphthalene
PAH
Methanol
0.07
Hexachl orobutadi ene
svoc
Salt
0.19
Need IDP
4-Chloro-3-methylphenol
svoc
Methanol
Need MDL
2-Methylnaphthalene
PAH
Methanol
0.07
Hexachl orocy cl opentadi ene
svoc
Methanol
0.06
2,4,6-Trichlorophenol
svoc
Acid
0.36
2,4,5-Trichlorophenol
svoc
Acid
2-Chl oronaphthal ene
PAH
Methanol
0.06
2-Nitroaniline
svoc
Salt
Acenaphthylene
PAH
Methanol
0.06
Dimethylphthalate
SVOC
Methanol
0.72
Need IDP
2,6-Dinitrotoluene
svoc
Methanol
0.19
Acenaphthene
PAH
Methanol
0.04
Dibenzofuran
PAH
Methanol
0.08
2,4-Dinitrotoluene
SVOC
Methanol
0.51
Fluorene
PAH
Methanol
0.04
Diethylphthalate
SVOC
Salt
0.22
Need IDP
4-Chlorophenyl-phenyl ether
SVOC
Methanol
0.07
4,6-Dinitro-2-methylphenol
SVOC
Salt
0.21
N-nitrosodiphenylamine
SVOC
Methanol
0.15
4-Bromophenyl-phenyl ether
SVOC
Methanol
0.08
Hexachl orob enzene
svoc
Methanol
0.13
Pentachlorophenol
svoc
Acid
0.25
Phenanthrene
PAH
Methanol
0.04
Anthracene
PAH
Methanol
0.04
Carbazole
PAH
Methanol
0.36
Di-n-butylphthalate
SVOC
Methanol
0.09
Fluoranthene
PAH
Methanol
0.05
D-32
-------
Table 14 (continued). Summary of Proposed Compounds for ATP
Compound
Group
(a)
Modifier
(b)
MDL
(ug/L) (c)
Comment (d)
Pyrene
PAH
Methanol
0.05
Butylbenzylphthalate
SVOC
Methanol
0.06
Benzo(a)anthracene
PAH
Methanol
0.06
Chrysene
PAH
Methanol
0.05
3,3'-Dichlorobenzidine
SVOC
Salt
0.52
Need MS
bi s(2-Ethylhexyl)phthalate
SVOC
Salt
0.79
Di-n-octylphthalate
SVOC
Methanol
0.29
Benzo(b)fluoranthene
PAH
Methanol
0.05
Benzo(k)fluoranthene
PAH
Methanol
0.05
Benzo(a)pyrene
PAH
Methanol
0.07
Indeno( 1,2,3 -cd)pyrene
PAH
Methanol
0.28
Dib enz(a, h)anthracene
PAH
Methanol
0.25
B enzo(g,h,i)peryl ene
PAH
Methanol
0.20
(a) SVOC = Semi-volatile compound, PAH = Polyaromatic Hydrocarbon
(b) Sample modifier that provides the best data: Methanol, Salt, or Acid
(c) Method Detection Limit (MDL) obtained, using the lowest observed during the study
(d) If a comment is provided, this compound could be added to the final ATP. Generally, these compounds
need to have the optimum calibration range and spiking amount in order to get acceptable data. IDP =
Initial Demonstration of Proficiency, MS = Matrix spike.
For a detailed discussion of the data for the ATP and the Standard Operating Procedure, see
Attachments H and I.
VI Future Work/Development
The Twister™ extraction and thermal desorption analysis can likely be used for many more
compounds than those that we tried during the trial period. Future work could fall into three
categories: refine what we have accomplished, develop a screening method, and pushing the
detection limits.
The three areas that need more work are the 12 semi-volatile compounds that should be
successful, pesticide analysis, and the brominated flame retardants. The remaining semi-volatiles
that met three of four criteria need some additional work where different concentrations are used,
generally lower concentrations. Pesticides that are neutral and relatively water insoluble should
perform well. We could investigate the full 608 pesticide list and develop an ATP for pesticides.
We should keep in mind that the typical solvent, iso-octane, would need to be substituted with
another that is amenable to Twister™. Brominated flame retardants consist of many more
isomers that could be tested and evaluated. The full list would need to be identified prior to
starting any work in this area.
Occasionally, we have emergencies where answers to sample data are needed as fast as possible.
The traditional method, working 24 hours a day can provide analytical data results in 3 days.
Using the Twister™ technique and an autosampler, answers for up to a dozen samples could be
provided in 24 hours working 10-12 hour day. If screening data is all that may be required, the
D-33
-------
rugged QC procedures we adhered to need not be followed, and a quick estimate can be provided
to customers within hours. Also, this technique could be used to screen samples prior to full
extraction using the traditional techniques. The data from the screened samples could provide an
immediate determination while the data from the traditional method would provide the
compliance data for court litigation.
If low detection limits are needed, sample volume and extraction time can be increased to gain
orders of magnitude on the MDLs that we obtained from a mere 5 mL sample. Region 7 has been
asked by our Superfund customers in the past to do method development to be able to get much
lower detection limits for the PAH compounds. Their request is based on Region 9 preliminary
remediation goals (PRGs) which are based on human risk assessments. The goals for the PAH
compounds are summarized Table 15. The lowest goal is Benzo(a)Pyrene and
Dibenz(a,h)anthracene at 0.0092 ug/L in water. During our study, we were able to show an MDL
for these compounds of 0.07 and 0.25 ug/L, respectively, using a 5 mL sample volume. It is
reasonable to think that the Region 9 PRG can be nearly met by increasing the sample volume
and extraction time.
Table 15. Summary of PE
.Gs and MDLs for PAHs
Compound
Region 9
PRG, ug/L
Traditional
technique MDL,
ug/L (a)
Twister™ MDL, ug/L
(5 mL sample)
Acenaphthene
370
1.0
0.04
Anthracene
1800
1.0
0.04
Benzo(a)anthracene
0.092
0.93
0.06
Benzo(b)fluoranthene
0.092
0.82
0.05
Benzo(k)fluoranthene
0.92
0.93
0.05
Benzo(a)pyrene
0.0092
0.84
0.07
Chrysene
0.56
0.99
0.05
Dib enz(a, h)anthracene
0.0092
0.76
0.25
Fluoranthene
1500
0.98
0.05
Fluorene
240
1.0
0.04
Indeno( 1,2,3 -cd)pyrene
0.092
1.0
0.28
Naphthalene
0.093
1.0
0.07
Pyrene
180
0.96
0.05
(a) CARB Method 3230.2
VII Acknowledgement and Disclaimer
We would like to thank Gerstel and its employees for the opportunity to evaluate the equipment
for 3 months, specifically, Ed Pfannkoch, Tim Pence, and Terry Stevens. Ed Pfannkoch was very
helpful in trouble shooting and method development, spending countless hours on the phone
assisting us with instrument conditions. Tim Pence was a life-saver when we had a TDU
malfunction, because he quickly provided us a spare TDU and 506C controller the next day.
Terry Stevens was crucial in making the loan agreement a reality and a success.
D-34
-------
We would like to acknowledge the support of EPA management in Environmental Services
Division who allowed us to pursue this project, and support of the Organic Chemistry Section
who helped by sharing more of the sample load in order for the project team to be free to carry
on this project. Special note of appreciation goes to Karen Johnson for helping with the
traditional semi-volatile analysis, assisting with the Gerstel analysis, and reviewing the final
report. Special note of appreciation goes to Jenn Boggess who helped with glassware washing so
that the chemists could perform an inordinately large number of extractions in the normal time
frame
The views of the scientists who worked on this project are not necessarily the views of the
Environmental Protection Agency.
Mention of specific products does not constitute an endorsement of the product by the United
States Environmental Protection Agency. It should be noted that the Gerstel Twister™ is
patented technology, and is sold by a single company. At this time there is no other product of
this type available.
VIII. References
[1] Prieto, A; Zuloaga, O.; Usobiaga, A.; Etxebarria, N.; Fernandez, L.A.: "Development of a
stir bar sorptive extraction and thermal desorption-gas chromatography-mass spectrometry
method for the simultaneous determination of several persistent organic pollutants in water
samples.", Journal of Chromatography A, 1174 (2007) 40 - 49.
[2] Heiden, A.; Hoffmann, A.; Kolahgar, B.: "Application of stir bar sorptive extraction to
the determination of polycyclic aromatic hydrocarbons in aqueous samples.", Journal of
Chromatography A, 963 (2002) 225-230.
[3] Perez-Carrera, E.; Leon Leon, V.M.; Gomez Parra, A.; Gonzalez-Mazo, E.:
"Simultaneous determination of pesticides, polycyclic aromatic hydrocarbons and
polychlorinated biphenyls in seawater and interstitial marine water samples, using stir bar
sorptive extraction-thermal desorption-gas chromatography-mass spectrometry.", Journal of
Chromatography A, 1170 (2007) 82 - 90.
[4] Kansas Urban Streams, http//:kcwaters.org.
D-35
-------
Appendix E
EPA Offices Method Validation References
E-l
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
Office of Air and Radiation (OAR)
Office of Radiation
and Indoor Air,
National Air and
Radiation
Environmental
Laboratory
Method Validation Guide
for Qualifying Methods
Used by Radiological
Laboratories Participating
in Incident Response
Activities
June 2009
httpsi//www, epa.gov/site
s/production/files/2015-
05/documents/method v
alidation final with web
cover 6~24~03.pdf
Radiochemical
Method validation and
peer review policies and
guidelines
Methods for processing
samples during a response to a
radiochemical incident,
including radiochemical
incidents of national
significance.
Office of Radiation
and Indoor Air,
National Air and
Radiation
Environmental
Laboratory
Radiological Laboratory
Sample Analysis Guide for
Incident Response -
Radionuclides in Soil
EPA 402-R-12-006
September 2012
httpsi//www, epa.gov/site
s/production/files/2015-
05/documents/402-r-12-
006 soil guide sept 201
2.pdf
Radiochemical
General method
development guidelines
Methods for processing
samples during a response to a
radiochemical incident,
including radiochemical
incidents of national
significance.
Office of Radiation
and Indoor Air,
National Air and
Radiation
Environmental
Laboratory
Radiological Laboratory
Sample Analysis Guide for
Incident Response -
Radionuclides in Water
EPA 402-R-07-007
January 2008
https://nepis.epa.gov/Exe
/ZyPDF.cgi/60000LAW.PD
F?Dockev=60000LAW.PD
F
Radiochemical
General method
development guidance
Methods for processing
samples during a response to a
radiochemical incident,
including radiochemical
incidents of national
significance.
Office of Radiation
and Indoor Air,
National Air and
Radiation
Environmental
Laboratory
Radiological Laboratory
Sample Analysis Guide for
In ciden ts of Nation al
Significance -
Radionuclides in Air
EPA 402-R-09-007
June 2009
https://cfpub.epa.gov/si/
si public record report.c
fm?Lab=ORIA&dirEntryld
=2.31510
Radiochemical
General method
development guidance
Methods for processing
samples during a response to a
radiochemical incident,
including radiochemical
incidents of national
significance.
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
OAR
Method Detection and
Quantitation - Program
Use and Needs
October 2010
https://www.epa.gov/site
s/production/fiies/2016-
11/documents/mth det
auant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
E-2
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
ORD, OSA, FEM - OAR
Calibration Curves -
Program Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/fiIes/2014-
05/documents/calibratio
n-guide-ref-final-
oct2010.pdf
Chemical
Method quantitation
Common practices for
instrument calibration
EPA (General) - OAR
Detection Limit and
Quantitation Limit
Summary Table
November 2016
httpsi//www, epa.gov/site
s/production/fiIes/2016-
11/documents/mdlmql-
toolbox-
final nov2016 0.pdf
Chemical
Method quantitation
Common practices for
determining detection limits
OAR; Office of Air
Quality, Planning and
Standards (OAQPS);
Ambient Air
Regulatory Program
Ambient Air Monitoring
Reference and Equivalent
Methods, 40 CFR Part 53
July 2002
httpsi //www .ecfr.gov/cgi
-bin/text-
idx?SID=6711885e63e97d
cf6c9c2c94b0237ab9&mc
=true&node=pt40,6,S3&r
gn=div5#sp40.6.53.c
Chemical
Method validation and
peer review policies and
guidelines
Procedural requirements for
demonstrating reference and
equivalent methods for
ambient air
OAR, OAQPS, Air
Quality Assessment
Division,
Measurement
Technology Group
Method 301 - Field
Validation of Pollutant
Measurement Methods
from Various Waste Media
March 2018
https://www.epa.gov/site
s/production/files/2018-
03/documents/method 3
01 3-26-2018 l.pdf
Chemical
Method validation and
peer review policies and
guidelines
Procedures for validating an
alternative candidate test
method for an affected source
Office of Chemical Safety and Pollution Prevention (OCSPP)
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
OCSPP
Method Detection and
Quantitation - Program
Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
ORD, OSA, FEM -
OCSPP
Calibration Curves -
Program Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2014-
05/documents/calibratio
n-guide-ref-final-
oct2010.pdf
Chemical
Method quantitation
Common practices for
instrument calibration
E-3
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
EPA (General) - OCSPP
Detection Limit and
Quantitation Limit
Summary Table
November 2016
httpsi//www, epa.gov/site
Chemical
Method quantitation
Common practices for
determining detection limits
s/production/files/2016-
11/documents/mdlmql-
toolbox-
Office of Enforcement and Compliance Assurance (OECA)
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
Office of Enforcement
and Compliance
Assurance (OECA)
Method Detection and
Quantitation - Program
Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/files/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
ORD, OSA, FEM -
OECA
Calibration Curves -
Program Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2014-
05/documents/calibratio
n-guide-ref-final-
oct2010.pdf
Chemical
Method quantitation
Common practices for
instrument calibration
EPA (General), OCSPP
Detection Limit and
Quantitation Limit
Summary Table
November 2016
https://www.epa.gov/site
s/production/fiies/2016-
11/documents/mdlmql-
toolbox-
Dffice of Land and Emergenc
Chemical
y Management ((
Method quantitation
DLEM)
Common practices for
determining detection limits
Office of Resource
Conservation and
Recovery (ORCR)
Guidance for Methods
Development and Methods
Validation for the Resource
Conservation and Recovery
Act (RCRA) Program
https://www.epa.gov/site
s/production/fiies/2015-
10/documents/methdev,
pdf
Chemical
General method
development guidance
SW- 846 Methods
E-4
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
OLEM
Method Detection and
Quantitation - Program
Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/fiIes/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
ORD, OSA, FEM -
OLEM
Calibration Curves -
Program Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/files/2014-
05/documents/calibratio
n-guide-ref-final-
oct2010.pdf
Chemical
Method quantitation
Common practices for
instrument calibration
EPA (General)-OLEM
Detection Limit and
Quantitation Limit
Summary Table
November 2016
https://www.epa.gov/site
s/production/files/2016-
11/documents/mdlmql-
toolbox-
final nov2016 O.pdf
Chemical
Method quantitation
Common practices for
determining detection limits
Office of Research and Development (ORD)
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurements (FEM)
Flexible Approaches to
Environmental
Measurement
2008, 2015
https://www.epa.gov/site
s/production/files/2014-
OS/documents/faem-
webinar-presentation-
notes.pdf;
https://www.epa.gov/me
asurements-
modeling/flexible-
approaches-
environmental-
measurement
Chemical,
Biological,
Radiochemical
General method
development guidance
Flexible approaches to
measurements and methods
E-5
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
National Homeland
Security Research
Center (NHSRC)
Magnuson, M.,
Campisano, R., Griggs, J.,
Fitz-James, S., Hall, K.,
Mapp, L., Mullins, M.,
Nichols, T., Shah, S.,
Silvestri, E. and Smith, T.,
2014. Analysis of
environmental
contamination resulting
from catastrophic
incidents: Part 2. Building
laboratory capability by
selecting and developing
analytical methodologies.
Environment international,
72, pp.90-97.
httpsi//www,sciencedirec
t.com/science/article/pii/
S0160412014000263
Chemical and
biological
General method
development guidance
Method Development
development process for
chemical, biologicals, and
biotoxins
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM)
Method Detection and
Quantitation - Program
Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
OSA, FEM
Calibration Curves -
Program Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2014-
05/documents/calibratio
Chemical
Method quantitation
Common practices for
instrument calibration
n-guide-ref-final-
oct2010.pdf
EPA (General) - ORD
Detection Limit and
Quantitation Limit
Summary Table
November 2016
https://www.epa.gov/site
s/production/files/2016-
11/documents/mdlmql-
toolbox-
final nov2016 O.pdf
Chemical
Method quantitation
Common practices for
determining detection limits
OSA, FEM
Validation and Peer
Review of U.S.
Environmental Protection
Agency Chemical Methods
https://www.epa.gov/site
s/production/files/2015-
01/documents/chemmet
hod validity guide.pdf
Chemical
Method validation and
peer review policies and
guidelines
Chemical methods of analysis
E-6
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
of Analysis
October 2005
OSA, FEM
Validation and Peer
Review of U.S.
Environmental Protection
Agency Radiochemical
Methods of Analysis
September 2006
httpsi//www, epa.gov/site
Radiochemical
Method validation and
peer review policies and
guidelines
Radiochemical methods of
analysis
s/production/files/2014-
05/documents/radioche
mmethod validity guide,
pdf
OSA, FEM
Emergency Response
Methods Validation Policy
July 2010, 2016
httpsi//www, epa.gov/site
s/production/files/2016-
11/documents/emergenc
y response validity polic
y reaffirmed nov2016,pd
f
Chemical,
Biological,
Radiochemical
Method validation and
peer review policies and
guidelines
All environmental methods of
analysis (chemical,
radiochemical,
microbiological) developed for
emergency response situation
(e.g., natural disaster,
homeland security)
National Exposure
Research Laboratory,
Human Exposure and
Atmospheric Sciences
Division
Guidelines for FRM and
FEM Applicants
September 2011
https://www3.epa,gov/tt
n/amtic/files/ambient/cri
teria/frmfemguidelines.p
df
Chemical
General method
development guidance
NAAQS (Clean Air Act)-
Manual Reference Methods
National Homeland
Security Research
Center (NHSRC)
Risk-Based Criteria to
Support Validation of
Detection Methods for
Drinking Water and Air
October 2008
httpsi//cfpub, epa.gov/si/
si public file download.c
fm?p download id =4982
47&La b=NHSRC
Chemical,
Biological,
Radiochemical
Method validation and
peer review policies and
guidelines
Validation of analytical
methods for threat
contaminants under the U.S.
EPA NHSRC program.
E-7
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
Air Pollution
Prevention and
Control Division,
National Risk
Management
Research Laboratory
Interlaboratory Validation
of the Leaching
Environmental Assessment
Framework (LEAF) Method
1314 and Method 1315
September 2012
httpsi//nepis, epa.gov/Ad
obe/PDF/PlOOFAFC.pdf
Chemical,
Radiochemical
Method validation and
peer review policies and
guidelines
Validation report to
summarize a validation study
that was performed on
Methods 1314 and 1315 (for
EPA to review and approve for
the purpose of inclusion into
EPA's Test Methods for
Evaluating Solid Waste,
Physical/Chemical Methods
[SW-846 Methods])
Office of Water (OW)
Office of Ground
Water and Drinking
Water
Methods Development for
Unregulated Contaminants
in Drinking Water: Public
Meeting and Webinar
June 2018
httpsi//www, epa.gov/site
s/production/files/2018-
07/documents/method-
development-
unregulated-
contaminants-drinking-
water-meeting-materials-
june2018.pdf
Chemical
General methods
development guidance
Safe Drinking Water Act 500
series
Office of Science and
Technology (OST),
Engineering and
Analysis Division
(EAD),
Engineering and
Analytical Support
Branch
Quality Assurance and
Quality Control
Requirements in Methods
Not Published by EPA
May 2009
https://www.epa.gov/site
s/production/files/2015-
10/documents/qa-qc-in-
methods memo 05-07-
2009.pdf
Chemical
Method quantitation
40 CFR Part 136 Methods
OST,
Engineering and
Analytical Support
Branch/ EAD (4303T),
Clean Water Act
(CWA) Methods Team
Definition and Procedure
for the Determination of
the Method Detection
Limit, Revision 2
December 2016
https://www.epa.gov/site
s/production/files/2016-
12/documents/mdl-
procedure rev2 12-13-
2016.pdf
Chemical
Method quantitation
Revision 2 of MDL procedure
from 40 CFR Part 136
Appendix B
E-8
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
EAD
Protocol for Review and
Validation of New
Methods for Regulated
Organic and Inorganic
Analytes in Wastewater
under EPA's Alternate Test
Procedure Program
February 2018
httpsi//www, epa.gov/site
s/production/files/2018-
03/documents/chemical-
Chemical
General methods
development guidance
New methods for organic and
inorganic analytes used in
CWA programs (specifically,
while operating within the
Alternate Test Procedure
Program)
atp-protocol feb-
2018.pdf
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
OW
Method Detection and
Quantitation - Program
Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/files/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
Chemical
Method quantitation
Common practices for
determining method detection
and quantitation
ORD, OSA, FEM - OW
Calibration Curves -
Program Use and Needs
October 2010
https://www.epa.gov/site
s/production/files/2014-
05/documents/calibratio
n-guide-ref-final-
oct2010.pdf
Chemical
Method quantitation
Common practices for
instrument calibration
EPA (General)-OW
Detection Limit and
Quantitation Limit
Summary Table
November 2016
https://www.epa.gov/site
s/production/files/2016-
11/documents/mdlmql-
toolbox-
final nov2016 O.pdf
Chemical
Method quantitation
Common practices for
determining detection limits
EPA (General)-OW
Protocol for the Evaluation
of Alternate Test
Procedures for Analyzing
Radioactive Contaminants
in Drinking Water
February 2015
http://nepis.epa.gov/Exe/
ZvPDF.cgi?Dockev=P100
MESN.txt
Radiochemical
General method
development guidance;
Method validation and
peer review policies and
guidelines
New drinking water methods
for Safe Drinking Water Act
compliance monitoring
submitted to the Drinking
Water Alternate Test
Procedure Program
E-9
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
EPA (General)-OW
Protocol for the Evaluation
of Alternate Test
Procedures for Organic
and Inorganic Analytes in
Drinking Water
February 2015
https://nepis.epa.gov/Exe
/ZyPDF,cgi?Dockey=P100
MERX.txt
Chemical
General method
development guidance;
Method validation and
peer review policies and
guidelines
New drinking water methods
for Safe Drinking Water Act
compliance monitoring
submitted to the Drinking
Water Alternate Test
Procedure Program
EPA General
EPA (General)
Performance Based
Measurement System, 62
Fed. Reg. 193 (October 6,
1997)
October 1997
httpsi//archive,epa,gov/e
pawaste/hazard/web/pdf
Z97-26443.pdf
Chemical,
Biological,
Radiochemical
General method
development guidance
Physical, Chemical, and
Biological measurements.
(Performance Based
Measurement System
Approach does not apply when
a specific method is prescribed
in a regulation itself)
Office of Research and
Development (ORD),
Office of the Science
Advisor (OSA), Forum
on Environmental
Measurement (FEM) -
Office of Cooperative
Environmental
Management (OCEM)
Method Detection and
Quantitation - Program
Use and Needs
October 2010
httpsi//www, epa.gov/site
s/production/files/2016-
11/documents/mth det
quant-guide-ref-
revision nov2016.pdf
General
Method quantitation
Common practices for
determining method detection
and quantitation
Federal Advisory
Committee on
Detection and
Quantitation
Approaches and Uses
in Clean Water Act
Programs (FACDQ)
Report of the Federal
Advisory Committee on
Detection and
Quantitation Approaches
and Uses in Clean Water
Act Programs
December 2007
httpsi//www, epa.gov/site
s/production/files/2015-
10/documents/detection-
quant-faca final-
report 2012.pdf
Chemical
Method quantitation
CWA Methods
E-10
-------
Office/ Organization
Document or Reference
URL
Category
Guidance Type
Statute/ Application
EPA (This document
was developed in
conjunction with
multiple federal
agencies)
Multi-Agency Radiological
Laboratory Analytical
Protocols Manual
(MARLAP)
July 2004
httpsi//www, epa.gov/rad
Radiochemical
General method
development guidance;
method quantitation;
method validation and
peer review policies and
guidelines
Performance-based approach
to radioanalytical methods
iation/marlap-manual-
and-supporting-
documents
Forum of
Environmental
Management (FEM)
Validation and Peer
Review of U.S. EPA
Sampling Methods for
Chemical and
Radiochemical Parameters
February 2016
https://www.epa, gow/site
s/production/files/2016-
02/documents/radioche
m method guide revise
Chemical,
Radiochemical
Method validation and
peer review policies and
guidelines
Validation of new sampling
methods for chemical and
radiochemical parameters
before publication for general
use
d 020316 00000002.t>df
Forum of
Environmental
Management (FEM)
Validation and Peer
Review of U.S.
Environmental Protection
Agency Chemical Methods
of Analysis
February 2016
https://www.epa.gov/site
s/production/files/2016-
02/documents/chemical
method guide revised 0
Chemical
Method validation and
peer review policies and
guidelines
Validation of new chemical
methods of analysis before
publication for general use
20316.pdf
E-ll
-------
Appendix F
Non-EPA Method Validation References
F-l
-------
Office/Organization
Information Source or
Reference
URL
Category
Guidance Type
Statute/Application
Non-EPA Federal Agencies
US Department of Health and Human Services (HHS), Centers for Disease Control and Prevention (CDC)
Centers for Disease
Control and Prevention
(CDC),
National Institute for
Occupational Safety and
Health (NIOSH)
NIOSH Manual of Analytical
Methods - Chapter E -
Development and
Evaluation of Methods
April 2016
https://www.cdc.g
ov/niosh/nmam/p
Chemical
General method
development guidelines
and method validation and
peer review policies and
guidelines
Generalized set of
evaluation criteria
prepared by NIOSH
researchers for the
evaluation of sampling
and analytical
methodology
dfs/NMAM 5thEd
EBook.pdf
CDC,
NIOSH
Guidelines for Air Sampling
and Analytical Method
Development and
Evaluation
May 1995
https://www.cdc.g
ov/niosh/docs/95-
liy/defaulthtml
Chemical
General method
development guidelines
and method validation and
peer review policies and
guidelines
Sampling and analytical
methods for workplace
compliance
determinations
CDC,
NIOSH
Development and
Validation of Methods for
Sampling and Analysis of
Workplace Toxic Substances
September 1980
https://www.cdc.g
ov/niosh/docs/80-
133/pdfs/80~
133.pdf
Chemical
General method
development guidelines
and method validation and
peer review policies and
guidelines
Sampling and analytical
methods of workplace
toxic substances
National Institutes of Health (NIH)
National Institutes of
Health (NIH), The
National Institute of
Environmental Health
Sciences (NIEHS),
National Toxicology
Program (NTP)
Validation and Regulatory
Acceptance ofToxicological
Test Methods - A Report of
the ad hoc Interagency
Coordinating
March 1997
httpsi//ntp,niehs,n
ili.gov/iccvam/docs
/about docs/valid
ate.pdf
Chemical
Method validation and
peer review policies and
guidelines
Criteria and processes for
validation and regulatory
acceptance of
toxicological testing
methods
United States Department of Commerce, National Institute of Standards and Technology (NIST)
F-2
-------
Office/Organization
National Institute of
Standards and
Technology (NIST)
Information Source or
Reference
Sander, L.C., 2017. Liquid
Chromatography:
Introduction to Method
Development. Journal of
Research of the National
Institute of Standards and
Technology, 122.
March 2017
URL
https://nvlpubs.nis
t.gov/nistpubs/ires
/122/ires, 122,018,
pdf
Category
Guidance Type
Statute/Application
Liquid chromatography
and analytical
instrumentation methods
Chemical
General method
development guidelines
NIST
Procedure for Method
Validation
2018
https://www.nist, g
ov/system/files/do
cuments/2018/01/
12,/procedure-for-
method-validation-
20180101.pdf
Chemical
Method validation and
peer review policies and
guidelines
Laboratory methods
United States Department of Labor, Occupational Safety and Health Administration (OSHA)
Occupational Safety and
Health Administration
(OSHA)
Validation Guidelines for Air
Sampling Methods Utilizing
Chromatographic Analysis
May 2010
https://www.osha.
gov/dts/sltc/meth
ods/chromguide/c
hromguide.pdf
Chemical
Method validation and
peer review policies and
guidelines
Air sampling methods
utilizing chromatographic
analysis
OSHA, Methods
Development Team,
Industrial Hygiene
Chemistry Division
Evaluation Guidelines for Air
Sampling Methods Utilizing
Spectroscopic Analysis
October 2010
https://www.osha.
gov/dts/sltc/meth
ods/spectroguide/s
pectroguide.pdf
Chemical
Method validation and
peer review policies and
guidelines
Air sampling methods
utilizing spectroscopic
analysis
U.S. Department of
Health and Human
Services, FDA, Center for
Drug Evaluation and
Research (CDER), Center
for Veterinary Medicine
(CVM)
Bioanalytical Method
Validation - Guidance for
Industry
May 2018
https://www.fda. g
ov/files/drugs/publ
ished/Bioanalytical
-Method-
Validation-
Guidance-for-
lndustry.pdf
Biological and
Chemical
Method validation and
peer review policies and
guidelines
Bioanalytical procedures
F-3
-------
Office/Organization
US FDA, Office of Foods
and Veterinary Medicine
Information Source or
Reference
Guidelines for the
Validation of Chemical
Methods for the FDA FVM
Program, 2nd Edition
April 2015
URL
https://www.fda,g
ov/media/81810/d
ownload
Category
Guidance Type
Statute/Application
Chemical methods for
foods and feeds
Chemical
Method validation and
peer review policies and
guidelines
US FDA, ORA Laboratory
Procedure
Methods, Method
Verification and Validation -
ORA-LAB.5.4.5
Revised 08-29-14
https://www.fda,g
ov/media/73920/d
ownload
Chemical and
Biological
Method validation and
peer review policies and
guidelines
United States Geological Survey (USGS)
United States Geological
Survey (USGS)
https://minerals.us
gs.gov/science/ana
lytical-methods-
development/inde
N/A
General method
development guidelines
Macro analytical methods
development
Non-EPA State Agencies
Florida Department of
Environmental Protection
(FDEP)
New and Alternative
Laboratory Methods DEP-
QA-001/01
February 2004
— , .. MV
s.org/gateway/rea
dRefFile,asp?refld=
4359&filename=Ne
w%20and%20Alter
native%20Laborato
ry%20Methods.pdf
Chemical and
Biological
Method validation and
peer review policies and
guidelines
Analytical laboratory
methods
FDEP
Procedure And Policy For
Demostration Of Capability
For Methods, Instruments,
And Laboratory Staff
November 2018
https://fldeploc.de
p.state.fl.us/sop/so
pl.asp?sect=BURE
AU
Chemical and
Biological
Method validation and
peer review policies and
guidelines
Analytical laboratory
methods
FDEP
Quality Manual for State of
Florida Environmental
Protection Laboratory
January 2019
http://publicfiles,d
ep.state.fl.us/dear/
labs/lab qualityma
nual 19.pdf
Chemical and
Biological
Method validation and
peer review policies and
guidelines
Analytical laboratory
methods
F-4
-------
Office/Organization
Information Source or
Reference
URL
Category
Guidance Type
Statute/Application
California Department of
Pesticide Regulation,
Environmental Hazards
Assessment Branch
California Department of
Pesticide Regulation - SOP -
Chemistry Laboratory
Quality Control
July 1995
https://www.wate
rboards, ca.gov/wat
er issues/program
s/tmdl/records/reg
i 08/ref244
2.pdf
Chemical
General method
development guidelines
and Method validation and
peer review policies and
guidelines
Analytical laboratory
methods
California Department of
Pesticide Regulation,
Environmental
Monitoring Branch
California Department of
Pesticide Regulation - SOP -
Guide For Analytical
Method Development
2017
https://www.cdpr.
ca.gov/docs/emon
Chemical
General method
development guidelines
and Method validation and
peer review policies and
guidelines
Quantification of
pesticides in aqueous
and/or sediment samples
/pubs/sops/qaqcO
1200.pdf
New Jersey Department
of Health, Public Health
and Environmental
Laboratories
Quality Manual,
Environmental and
Chemical Laboratory
Services
July 2014
https://www.ni.go
v/health/phel/doc
uments/ecls qm 7
-2014.pdf
Chemical,
Radiochemical
Method validation and
peer review policies and
guidelines
Analytical laboratory
methods
Private Sector
ASTM International
ASTM International
ASTM E2857-11, Standard
Guide for Validating
Analytical Methods
Reapproved 2016
N/A
Chemical
Method validation and
peer review policies and
guidelines
Chemical and
spectrochemical
analytical methods for
metals, ores, and related
materials
ASTM International
ASTM E691 - 19el,
Standard Practice for
Conducting an
Interlaboratory Study to
Determine the Precision of a
Test Method
April 2020
N/A
Chemical
Method validation and
peer review policies and
guidelines
Interlaboratory study
F-5
-------
Office/Organization
Information Source or
Reference
URL
Category
Guidance Type
Statute/Application
ASTM International
ASTM E1169 - 18, Standard
Practice for Conducting
Ruggedness Tests
April 2018
N/A
Chemical,
Biological, and
Radiochemical
Method validation and
peer review policies and
guidelines
Ruggedness tests
ASTM International
ASTM E1601 - 19, Standard
Practice for Conducting an
Interlaboratory Study to
Evaluate the Performance
of an Analytical Method
November 2019
N/A
Chemical
Method validation and
peer review policies and
guidelines
Interlaboratory study for
analytical method
International Union of Pure and Applied Chemistry (IUPAC)
International Union of
Pure and Applied
Chemistry (IUPAC)
Thompson, M., Ellison, S.L
and Wood, R., 2002.
Harmonized guidelines for
single-laboratory validation
of methods of analysis
(IUPACTechnical Report).
Pure and Applied Chemistry,
74(5), pp.835-855.
httpi//publications
.iupac.org/pac/200
2/pdf/740Sx083S, p
df
Chemical
Method validation and
peer review policies and
guidelines
Methods of analysis
Eurachem
Eurachem
The Fitness for Purpose of
Analytical Methods - A
Laboratory Guide to
Method Validation and
Related Topics
2014
httpsi//www,eurac
hem.org/imaKes/st
ories/Guides/pdf/
MV guide 2nd ed
EN.pdf
Chemical
Method validation and
peer review policies and
guidelines
Analytical chemistry
AOAC International
AOAC International
How to Meet ISO 17025
Requirements for Method
Verification
2007
httpi//www,aoac,Q
)ac prod imis
/aoac docs/Iptp/al
acc guide 2008.pd
|
Chemical and
Biological
Method Verification
guidance (Not to be
confused with method
validation guidance)
Chemical and
microbiological methods
F-6
-------
Office/Organization
Information Source or
Reference
URL
Category
Guidance Type
Statute/Application
AOAC International
Appendix F Guidelines for
Standard Method
Performance Requirements
2016
httpi//www,eoma,
aoac.org/app f.pdf
Chemical and
Biological
Method validation and
peer review policies and
guidelines
Chemical and biological
methods
AOAC International
Appendix D: Guidelines for
Collaborative Study
Procedures To Validate
Characteristics of a Method
of Analysis
2005
httpi//www,aoac,o
rg/aoac prod imis
Chemical
Method validation and
peer review policies and
guidelines
Methods of analysis
/AOAC Docs/Stand
ardsDevelopment/
Collaborative Stud
V Validation Guid
elines.pdf
National Association of Testing Authorities (NATA)
National Association of
Testing Authorities
(NATA)
Technical Note 17 -
Guidelines for the validation
and verification of
quantitative and qualitative
test methods
2004, 2012
httpi//www,demar
cheisol7025.com/
document/Guideli
nes%20for%20the
%20validation%20
and%20verifi cation
%20of%20quantita
tive%20and%20qu
alitative%20test%2
0methods.pdf
Chemical,
Biological
Method validation and
peer review policies and
guidelines
Quantitative and
qualitative test methods
F-7
------- |