v>EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/8-91/005
February 1991
Preparation Aids for the
Development of
Category III Quality
Assurance Project Plans
-------
-------
EPA/600/8-91/005
February 1991
PREPARATION AIDS
FOR THE DEVELOPMENT OF
CATEGORY III
QUALITY ASSURANCE PROJECT PLANS
by
Guy F. Simes
Risk Reduction Engineering Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
Printed on Recycled Paper
-------
This material has been funded by the United States Environmental
Protection Agency. Although it has been subject to the Agency's review,
and has been approved for publication as an EPA document, it does not
necessarily reflect EPA policy
-------
Foreword
Today's rapidly developing and changing technologies and industrial
products and practices frequently carry with them the increased generation
of materials that, if improperly dealt with, can threaten both public health
and the environment. The U.S. Environmental Protection Agency is
charged by Congress with protecting the Nation's land, air, and water
systems. Under a mandate of national environmental laws, the Agency
strives to formulate and implement actions leading to a compatible balance
between human activities and the ability of natural systems to support and
nurture life. These laws direct the EPA to perform research to define our
environmental problems, measure the impacts, and search for solutions.
The Risk Reduction Engineering Laboratory is responsible for planning
implementing, and managing the research, development and
demonstration programs to provide an authoritative, defensible
engineering basis in support of the policies, programs, and regulations of
the EPA with respect to drinking water, wastewater, pesticides, toxic
substances, solid and hazardous wastes, and Superfund-related activities
This publication is one of the products of a quality assurance outreach
program that provides a vital communication link between the researcher
and the user community.
Perhaps the greatest benefit to the users of this document comes from its
emphasis on up-front planning - do the right thing, right, the first time.
While this sounds straightforward, managers know that determining the
right course in a complex and uncertain situation is anything but simple
Determining customer requirements up-front, and then having the
processes and procedures in place to accomplish them, averts costly
mistakes. Resources are conserved in two ways: by avoiding rework to
correct efforts that do not initially meet management or cl^mer
specifications; and by performing to the specifications required and not
beyond them. In these ways, this "Preparation Aids" document can help
management achieve its mission with more effective utilization of
diminishing resources.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
Abstract
Data collection activities performed for the Risk Reduction Engineering
Laboratory (RREL) of the U.S. Environmental Protection Agency are
divided into four categories, depending on the intended use of the data.
Quality Assurance (QA) Project Plans are written to ensure that project
needs will be met and that quality control procedures are sufficient for
obtaining data of known quality. The level of QA required, however,
depends on the project category selected for a given project. Projects that
produce results for the purpose of evaluating and selecting basic options,
or performing feasibility studies or preliminary assessments of unexplored
areas which might lead to further work are identified as Category III
projects.
To assist professional scientists and engineers in preparing QA Project
Plans, separate guidance manuals in an easy-to-read format have been
developed for each category. The Category III manual contains detailed
descriptions of each of the 11 required elements of a Category III QA
Project Plan. Also included are definitions and explanations of frequently
used terms, examples of QA forms and charts, sample equations, and
numerous types of tables suggested for summarizing information.
IV
-------
Contents
Page
Foreword ^
Abstract iv
Figures vi
Tables JZZI vii
Acknowledgments
0.0 Introduction j
1.0 Project Description 7
2.0 Quality Assurance Objectives 17
3.0 Site Selection and Sampling Procedures 25
4.0 Analytical Procedures and Calibration 33
5.0 Data Reduction, Validation, and Reporting 41
6.0 Internal Quality Control Checks 43
7.0 Performance and Systems Audits 47
8.0 Calculation of Data Quality Indicators 49
9.0 Corrective Action 53
10.0 Quality Control Reports to Management 55
11.0 References 57
-------
Figures
Number Page
0-1 QA Project Plan Approval Form 4
1-1 Project organization and names of responsible individuals 15
VI
-------
Tables
Number
Page
1-1 Summary of Critical and Noncritical Measurements for the
Evaluation of an Hypothetical Electric Arc Incinerator 9
1-2 Summary of Physical Measurements for the Evaluation of an
Hypothetical Electric Arc Incinerator 10
1-3 Summary of Planned Analyses (Including QC) for Chemical
Treatment of Water , 2
1-4 Total Number of Analyses (Not Including QC) for each Treatment
Condition: Hypothetical Solidification Process 13
2-1 QA Objectives for Precision, Accuracy, and Method Detection Limits..... 19
2-2 QA Objectives for Precision, Accuracy, and Detection Limits -
Alternate Form 2Q
2-3 Required Detection Limits for Volatile Chlorinated Organic
Compounds 2i
3-1 Summary of Number of Samples Required for Hypothetical
Incineration Test 27
3-2 Summary of Laboratory Analyses and Sample Quantity Requirements .... 28
3-3 Required Containers, Preservation Techniques, and Holding Times.. 30
4-1 Typical Summary Table of Standard Methods and Procedures 34
4-2 Typical Summary Table of Calibration Requirements for Physical
Process Measurements 3o
6-1 Scheduled QC and Calibration 45
Vll
-------
Acknowledgments
The extensive technical contributions of Dr. John R. Wallace of Maxwell
Laboratories and the assistance of Robert Banner, Ann Kern, and William
Mueller of the USEPA in reviewing the document are gratefully
acknowledged. Grateful appreciation is, also, extended to the excellent
editorial and technical comments provided by Dr. Monica Nees.
-------
SECTION 0.0
INTRODUCTION
0.1 PURPOSE OF THE QUALITY ASSURANCE PROJECT PLAN (QAPJP)
The purpose of a QA Project Plan is to relate project objectives to specific measurements
required to achieve those objectives. The QA Project Plan must provide sufficient detail to
demonstrate the following:
Intended measurements are appropriate for achieving project objectives;
Quality control procedures are sufficient for obtaining data of known and
adequate quality; and
Such data will be defensible if challenged technically or legally.
Environmental projects require the coordinated efforts of numerous individuals, including
regulators, managers, engineers, scientists, statisticians, and economists. The QA Project Plan
must integrate the requirements of everyone involved, in a form that permits an easy and painless
review. It must also provide unambiguous instructions to the sampling team, the analytical
laboratory, and any other parties responsible for data generation. Finally, the QA Project Plan
must provide sufficient detail to allow a thorough, detailed review by an independent party not
involved in the project implementation.
0.2 QUALITY ASSURANCE CATEGORIES
Because the end use of the data determines the degree of quality assurance that is required,
RREL uses four categories in its QA Program:
CATEGORY I PROJECTS require the most rigorous and detailed QA, since
the resulting data must be both legally and scientifically defensible. Category I
projects include enforcement actions and projects of significant national or
congressional visibility. Such projects are typically monitored by the Adminis-
trator. Category I projects must produce results that are autonomous; that is,
results that can prove or disprove a hypothesis without reference to comple-
mentary projects.
-------
CATEGORY n PROJECTS are those producing results that complement other
inputs. These projects are of sufficient scope and substance that their results
could be combined with those from other projects of similar scope to produce
information for making rules, regulations, or policies. In addition, projects that do
not fit this pattern, but have high visibility, would also be included in this
category.
CATEGORY m PROJECTS are those producing results used to evaluate and
select basic options, or to perform feasibility studies or preliminary assessments
of unexplored areas which might lead to further work.
CATEGORY IV PROJECTS are those producing results for the purpose of
assessing suppositions.
The RREL Technical Project Manager is responsible for assigning the category that accurately
reflects the intended use of the data and the type of work being done.
This document addresses only Category HI requirements; separate documents are available for
each of the remaining categories. Be sure to use the Preparation Aids document for the correct
category.
0.3 UNDERLYING LOGIC AND ORGANIZATION OF A QA PROJECT PLAN
A QA Project Plan, regardless of category, must cover the following topics:
Process being tested and the objectives of the test (i.e., the hypothesis to be
tested);
Measurements that will be taken to achieve those objectives;
Quality of data that will be required and how that quality will be obtained; and
How data will be recorded, calculated, reviewed, and reported in a defensible
manner.
The QA Project Plan is organized according to the topics discussed in the body of this report.
-------
Be sure to include:
* Any project-specific information necessary for the sampling team, analyt-
ical laboratory, data reduction team, and ail other project participants.
However, assume these parties are familiar with standard methods,
ป Any deviations from standard methods or procedures, or clarification of
such documents whenever there are ambiguities,
* Equations for all project-specific calculations and data reductions*
But don't be repetitious:
Bo not discuss the same subject matter more than once in a QA Project
Plan, Although the required elements of the QA Project Plan may appear
to overlap in some cases, repetition leads to confusion, and to documents
too long for effective communication, If you are unsure where a subject
should be treated, discuss it once and cross-reference that discussion in the
other sections.
Do not repeat material from a Sampling Plan, If a separate Sampling
Plan has been generated, reference the section and page number in the QA
Project Plan, as needed*
* Do not repeat material from standard methods that are available from the
EPA or the Code of Federal Regulations. Provide detailed citations and
..assume that the analyst and reviewer have these methods on-hand,
0.4 CATEGORY IE FORMAT NOTES
All Category HI QA Project Plans must incorporate the following format requirements:
ซ Title page.
QA Project Plan Approval Form (see Figure 0-1).
-------
QUALITY ASSURANCE PROJECT PLAN APPROVAL FORM
for
RREL Contracts/IAGs/Cooperatlve Agreements/In-house Projects
RREL QA ID No:.
Contractor:
RREL Project Category: _
RREL Lab Workplan No:.
QA Project Plan Title:.
Revision Date:
COMMITMENT TO IMPLEMENT THE ABOVE QA PROJECT PLAN:
Contractor's Project/Task Manager (print)
Contractor's QA Manager (print)
Other as Appropriate/Affiliation* (print)
Other as Appropriate/Affiliation* (print)
Signature
Signature
Signature
Signature
Date
Date
Date
Date
Other as Appropriate/Affiliation* (print)
' Commitment signature Is required tor any ancillary sampling, analytical, or data
gathering support provided by a subcontractor or RREL principal Investigator.
APPROVAL TO PROCEED IN ACCORDANCE TO THE ABOVE QA PROJECT PLAN:
RREL Technical Project Manager (print)
Signature
Date
CONCURRENCES:
RREL Section or Branch Chief (print)
Guv Slmes/Davld Smith
RREL QA Manager
Signature
Signature
Date
Date
RREL (QAPJP AF)
(March 1989)
Figure 0-1. Quality Assurance Project Plan Approval Form
-------
Distribution list to ensure that key personnel will receive current copies and
updates.
Table of Contents as shown at the beginning of this document. If a given
element is not applicable, the text of the QA Project Plan should explain its
omission.
Document control format. Section number, revision, date, and page should be
recorded in an upper corner of each page. This format requirement is illus-
trated below.
Section No.
Revision:
Date:
Page:
1.0
0
October 20, 1989
3 of 5
0.5
QA PROJECT PLAN APPROVAL FORM (SIGNATURE PAGE)
It is important that the project principals understand and agree on the experimental approach. For this
reason, the QA Project Plan Approval Form must be signed by key project personnel as well as key
personnel from any subcontractors. These signatures -- which must be obtained before the final QA
Project Plan is submitted - indicate that the key personnel have read the appropriate sections of the QA
Project Plan and are committed to implementing the plan.
-------
-------
SECTION 1.0
PROJECT DESCRIPTION
This section describes the process or environmental system that is to be tested, the project
objectives, a summary of the experimental design, and the proposed project schedule. This
section typically contains the subsections listed below, plus others at the principal investigator's
discretion.
1.1 GENERAL OVERVIEW
This section provides a brief synopsis of the overall project. In one or two paragraphs, describe
the programmatic and regulatory setting in which the project is going to be carried out. The
purpose(s) of the study should be described, along with the decisions that are to be made and the
hypothesis to be tested. Other anticipated uses of the data should also be noted. Explain the
consequences of drawing incorrect conclusions or making invalid decisions. The type of process
or environmental system that is to be tested should also be described briefly.
1.1.1 The Process, Site, Facility, or System
This section describes the process, site, facility, or environmental system that will be tested.
Include flow diagrams, maps, charts, etc., as needed. Approximate mass or volumetric flow rates
should be indicated to permit proper evaluation of the sampling and process monitoring
procedures. Indicate sampling points with alphanumeric designations, either in this section or in
Section 3.0 (Site Selection and Sampling Procedures). Additional diagrams are often included to
unambiguously describe sampling points. Such detailed diagrams can be included either in this
section or in Section 3.0.
This section should contain enough material to permit a technical reviewer who is unfamiliar
with the specific project to assess the proposed sampling strategy.
1.1.2 Statement of Project Objectives
Category III projects typically have multiple objectives. These project objectives should be
summarized and stated clearly in this section. Avoid scattering statements of project objectives
throughout the sampling and analytical sections of the QA Project Plan, or among several
documents. Statements of objectives, when scattered among various sections, often result in an
unfocused effort.
-------
Project objectives should be stated in numerical terms whenever possible:
Poor Statement: The bio-oxidation procedure will be characterized with respect
to its ability to remove semivolatile organic compounds.
Better Statement: The major objective is to demonstrate a bio-oxidation
efficiency of 95 percent for the compounds listed in Table...
Best Statement: The objective is to demonstrate a bio-oxidation efficiency of 90
percent or higher at a confidence level of 95 per cent for the compounds listed in
Table...
Some objectives cannot be stated in entirely quantitative terms. For example, one objective for a
bio-treatment project might be to determine the dominant microbiological species in a reactor.
Or, for a test of a new incinerator design, one might wish to determine the range of combustion
conditions that result in a molten ("slagging") ash.
It is common to rank project objectives according to importance (e.g., into primary and
secondary objectives). Although this ranking is not essential, it does help focus effort on the
primary goals of the project.
1.2 EXPERIMENTAL DESIGN
List all measurements that will be made during the project, then classify them as critical or
noncritical measurements. Critical measurements are those that are necessary to achieve
project objectives; they may include either on-site process measurements or chemical
measurements. Thus, for a test of an electric arc furnace as a means of incinerating organics, the
rates of consumption of shield gas (e.g., argon) and electric power are as important as the feed
rate and concentration of pollutant. Noncritical measurements are those used for process
control or background information. Table 1-1 and Table 1-2 illustrate two different ways of
designating critical and noncritical measurements.
It is common practice to test pollution control equipment or processes in different operating
modes or stages. For example, an incinerator might be operated under warm-up conditions,
followed by incineration tests at three different feed rates. The various test conditions should be
summarized in this section. Indicate the length of time anticipated for each test condition.
-------
Table 1-1. Summary of Critical and Noncritical Measurements for
the Evaluation of an Hypothetical Electric Arc Incinerator
A. CRITICAL MEASUREMENTS
Chemical
Critical SVOCs3 in ail feed and discharge
streams
PCBs in all feed and discharge streams
Particulates, NOX, and SO2 in stock gas
TCLP/critical metalsb in slag and feed
TCLP/critical SVOCs3 in slag
Alkalinity of scrubber water
HC1 in stack gas
Organic carbon in feed
Process
Electrical power consumption of arc furnace
Argon and oxygen consumption (rate and total)
Mass feed rate of soil
Total mass of slag discharged
Cooling requirements (temperatures and flow rates of
cooling water)
B. NONCRITICAL MEASUREMENTS
Chemical
Noncritical SVOCs in all feed and discharge streams
Permanent gases in stack: N2ป ฉ2, CO2, CO
Percent moisture in stack
Ash fusion temperature of slag
Process
All temperatures and flows not listed above
Opacity of stack gas
Operating pressures
Critical SVOCs (semivolatile organic compounds) are pentachlorophenol, the three tetrachlorophenol isomers,
chlorophenol, 1- and 2-methylnaphthalene, anthracene, acenaphthalene, and acetophene. These compounds have
been detected during preliminary studies on the untreated wastes or are suspected decomposition products.
Noncritical SVOCs are all additional compounds detected by Method 8270, as described in Section of this
QA Project Plan.
Critical metals are As, Ba, Cd, Cr, Pb, Hg, Ni, Se, and Ag.
Tins table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type.
-------
Table 1-2. Summary of Physical Measurements for the Evaluation
of an Hypothetical Electric Arc Incinerator
Measurement
Measurement
Classification
Measurement
Site (Designation)3
Measurement
Frequency
Electric Power to Arc Furnace
Current Critical
Voltage Critical
kWh Critical
Ar and 02 Flow
Mass of Soil/Slag
Critical
Critical
Power leads to furnace (PI)
Feeds to furnace (Ql)
Drum balances @ feeder and
discharge (Ml, M2)
15 minutes
15 minutes
Before/after each run
15 minutes
Each batch
Opacity of Stack Gas
Temperatures
Noncritical Stack (S3)
Noncritical (Test points T3-T12)
Continuous
5 minutes
a Alphanumeric designations of test points are shown in Figure.
This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type,
10
-------
When appropriate, indicate how equilibrium conditions will be established before sample
collection begins.
mamQle: Total hydrocarbons in the stack gas will be monitored continuously
Sampling will not begin until THC concentrations are stable within ฑ 5 percent
over a three-hour period.
Summarize in tabular form all measurements that are planned for each sample. Ideally, this table
should indicate the total number of samples for each sample point, including quality control and
reserve samples, as illustrated in Table 1-3 for a hypothetical chemical treatment project Relate
the individual samples to the sampling points shown in the process diagram, if applicable For
projects involving a large number of samples or analyses, it may not be possible to include all
QC and reserve samples in a single table. For such cases, summarize the expected number of
non-QC analyses as in Table 1-4, and describe elsewhere the frequency of the various QC and
reserve samples. Estimates of the number of samples is needed for performing a comprehensive
review, as well as for estimating analytical costs.
For some Category III projects it may not be possible to estimate beforehand the total number of
samples of each type. For such projects, describe how the number of samples will be determined
during project execution.
1.3 SCHEDULE
Indicate start-up and ending dates, including those for preliminary studies and field and
laboratory activities.
1.4 PROJECT ORGANIZATION AND RESPONSIBILITIES
This section demonstrates that the project organization is adequate to accomplish the project
goals, and that all responsibilities have been assigned.
Provide a table or chart illustrating project organization (including subcontractors) and lines of
authority. For each organizational entity or subcontractor, identify the key personnel The
organizational chart should identify quality assurance managers and illustrate their relationship to
other project personnel. The QA Managers should be organizationally independent of project
management to minimize the risk of conflict of interest.
11
-------
Table 1-3. Summary of Planned Analyses (Including QC) for Chemical Treatment of Water
Preliminary
Samples
Number of Tests Performed3
Influent
Water
Cl
Effluent
Water
Sample Points:
E2
Sludge
S3
Total
Semivolatile Organic Compounds (SVOC)b
Non-QC (Primary)
Field Sampling Blank0
Field Duplicates
Laboratory Duplicates
Matrix Spikes (MSs)
Matrix Spike Duplicates (MSDs)
Spare Samples*1
Independent Check Standard
Metals6
Non-QC (Primary)
Field Sampling Blank0
Field Duplicates
Laboratory Duplicates
Matrix Spikes (MSs)
Matrix Spike Duplicates (MSDs)
Independent Check Standard
Spare Samples**
3
1
0
0
1
1
2
0
3
1
0
1
1
0
0
2
60
1
0
0
3
3
10
1
60
1
0
3
3
0
1
10
60
0
0
0
10
10
10
0
60
0
0
10
10
0
0
10
10
0
5
0
2
2
0
0
Grand Total
10
0
5
2
2
0
0
0
Grand Total
133
2
5
0
16
16
22
1
195
133
2
5
16
16
0
1
22
195
a Samples will be divided evenly among the ten anticipated runs. More QC samples are planned for the effluent
than for the influent, since the former samples are expected to exhibit more variability. A matrix spike/matrix
spike duplicate or a laboratory duplicate/matrix spike of the effluent stream will be determined for each treatment
condition, because the effluent matrix may vary significantly with each treatment
b Refers to the twelve compounds listed in Table .
c Trip blanks will be collected but not analyzed unless field blanks indicate a contamination problem.
d Not analyzed unless required.
e Refers to the eight metals in Table .
This table i$ meant to illustrate typical format. - _
The tests listed are NOT necessarily those that would be employed
for a process of this type.
12
-------
Table 1-4. Total Number of Analyses (Not Including QC)
for Each Treatment Condition: Hypothetical Solidification Process
Treatment Condition
Measurement3
Treated Long-Termb
Measurement Raw Soil Reagent Treated
Classification Soil (28 Days) Mix Soils
On-Site Tests
Slump of Portland Cement Concrete
Homogeneity of Mixing
Leaching Tests/Analyses^
TCLP/Metals
TCLP-ZHE/VOCs
WILT/TOC + VOC + Metalsd
ANS-16.1/TOC + VOC + Metalsc
Chemical Tests
PH
Metals - Total
VOCs- Total
Engineering Tests
Particle Size Analysis
Bulk Density
Freeze/Thaw Durability
Unconfined Compressive Strength
Noncritical
Noncritical
Critical
Critical
Critical
Critical
Critical
Noncritical
Critical
Critical
Noncritical
Critical
Critical
Critical
9
9
0
0
9
9
9
9
3
3
0
0
AN0
9
9
9
3
3
9
9
9
9
0
3
6
6
1
1
0
0
0
0
0
0
a VOC = Volatile Organic Compounds
TCLP = Toxicity Characteristic Leaching Procedure
TCLP-ZHE = Zero headspace modification of TCLP
WILT = Waste Interface Leaching Test
ANS-16.1 = American Nuclear Society 16.1.
TOC = Total Organic Carbon
b Test performed at 6, 24, and 60 months. Five samples will be analyzed at the end of each period.
AN = As needed until adequate consistency is achieved.
15
15
15
15
15
15
15
Ihmnf ^S P6^6? each, yiel? several ^P^te leachate solutions. However, as described in the text
samples will be composited, resulting in two sets of analyses per leaching procedure.
This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type*
13
-------
Describe the responsibilities of all project participants, including QA Managers. Be sure to
indicate responsibility for each type of analysis, physical measurement, and process measure-
ment.
Describe the frequency and mechanisms of communications among the contractor, the
contractor's QA Manager, and the EPA Project Manager, as well as among contractor and
subcontractors. Give the regular schedule for progress reports, site visits, and teleconferences,
and mention what special occurrences would trigger additional communication.
Figure 1-1 shows a project organizational chart with an independent QA Manager.
Be sure to:
Include enough information in this section to permit a technical person
unfamiliar with your project to evaluate the sampling and'ana'iytieaV
nnnrnfK'h , " ; - s * "".-,',s A ,$. r , ..jx-oss^v ,.,, .,,*,
iippi uS v^* -'-
Cite applicable regulations, if any.
14
-------
EPA
PROJECT OFFICER
CONTRACTOR
PROJECT MANAGER
CONTRACTOR
ENGINEERING
ASSESSMENTS
CONTRACTOR
QA MANAGER
SUBCONTRACTOR:
(COMPANY NAME)
LABORATORY
MANAGER
CONTRACTOR
LABORATORY
SUPERVISOR
SAMPLE
PREPARATION
TRACE METALS
AND CHLORIDE
ANALYSES
DATA
REDUCTION
SAMPLING
GC
ANALYSIS
pH, CEC, RsCI2
ACIDITY, ALKALINITYr
. ANDTDS
REPORT
GENERATION
Figure 1-1. Project organization and names of-responsible individuals
15
-------
-------
SECTION 2.0
QUALITY ASSURANCE OBJECTIVES
Quality assurance objectives are specifications that measurements must meet in order to achieve
project objectives. For example, in ordering a pump for a chemical plant, factors such as capac-
ity, pressure, and materials of construction must be specified. Similarly, precision, accuracy,
detection limits, and completeness must be specified for physical/chemical measurements.
Additional analytical requirements are described qualitatively in terms of representativeness and
comparability. Quality assurance objectives are needed for all critical measurements (see Section
1.2) and for each type of sample matrix (soil, water, biota, etc.). Be sure to include quality
assurance objectives for physical as well as chemical measurements.
Once QA objectives are set, the measurement systems are designed to meet them. The project
manager, analytical chemists, and other principals must agree on the feasibility and appro-
priateness of these objectives.
2.1 DETERMINING QA OBJECTIVES
QA objectives must be defined in terms of project requirements, and not in terms of the
capabilities of the intended test methods. Of course, the QA objectives must be achievable by
available methods, and for this reason it is important that the laboratory review the QA
objectives. When QA objectives exceed the capabilities of available methods, either the methods
must be modified or the test plan must compensate for these deficiencies. Modifications may
often be as simple as collecting a larger sample. However, if nonstandard or significantly
modified test methods are required, the QA Project Plan must include laboratory validation data
to prove that the method is capable of achieving the desired performance.
The following are examples of how QA objectives can be determined:
Example: Low temperature stripping of carbon disulfidefrom soil.
The first objective of this project is to identify procedures capable of removing 99 percent
of the CS2from the contaminated soil. The untreated soils contain 150 mg/kg ofCS2,
and consequently the treated soils should contain no more than 1.0% of this amount, or
$ 1.5 mg/kg ofCS2. To allow a reasonable margin of safety, the detection limit of 0.1
mg/kg is required for total CS2 in soils.
17
-------
2.2
A second objective of this project is to demonstrate that leachate from the treated soil
contains less than 4.8 mg/L ofCS2 as required by land disposal regulations (40 CFR
268.41). For this purpose a detection limit of 0.5 mg/L is required for 82 in leachate.
Example: Biological treatment of pentachlorophenol (PCP).
Previous experience with similar processes has indicated that the concentration of
pollutant in the effluent typically varies by a factor of two under stable operating
conditions with uniform feed. To assure that the analytical error is significantly less than
the natural process variability, the precision of the analytical methods should be
significantly less than this value, or approximately ฑ 30 percent or less.
QUANTITATIVE QA OBJECTIVES: PRECISION, ACCURACY, METHOD
DETECTION LIMIT, AND COMPLETENESS
QA objectives for precision, accuracy, method detection limit, and completeness should be
presented in a QA objectives table similar to those shown in Tables 2-1 through 2-3. Be sure to
include QA objectives for all matrix types and to indicate the units in which these QA objectives
are given. Summary tables are very helpful to the laboratory which must meet these objectives.
Because precision and accuracy can be measured in various ways, explain the method to be used.
If precision, for instance, is to be determined by duplicates, explain whether sample splitting
will occur in the laboratory, during sampling, or at some other stage. Then summarize all such
information in either text or tabular format.
The following statements are examples of descriptions for precision, accuracy, method detection
limits, and completeness:
Precision objectives for all the listed methods except pH are presented as relative
percent difference (RPD) of field duplicates. Precision objectives for pH are listed in
pH units and expressed as limits for field duplicates.
Precision objectives for unconfined compressive strength are given as relative
standard deviation for triplicate sets.
Accuracy objectives for organic compounds and metals are given as percent recovery
range of laboratory matrix spikes. Accuracy objectives for temperature measurements
are absolute deviations in ฐC.
Detection limits are defined as the method detection limit (MDL) multiplied by the
dilution factor required to analyze the sample. MDLs will be determined by replicate
extraction and analysis of seven identical spiked samples ofXAD resin.
18
-------
c?
^
CO
4 ป
'S
3
ง
s
B
i
1
o
o
<
c~
.2
OT
'o
w
ฃ
LJ
ซง
jectives
&
O
S
r <
1
CN
JQ
J3
cU
H
i
s
V
"E,
/|
|l
g.
9-x
1
S
o
<ฃ
>J
2
a
iง
.S1
"fi
1
OH
"8
I
1
1
c
S
i
|
1
B
w^ ID o
o o o
O 0 0
IO 10 >O
o o o
VI VI VI
si s
ซj S tj
^ ^ ^
cs ซs o
OO OO "O
S3 S3
i> co ^
C/l
3
2 ^
G* S
O W)
S oj
TO ฃ2
|u | I
"ง^ 1 i*
CO ^.^ ^^ o
ปn v> ซo
O\ ON ON
ปo in
ON ON
o o o o o
cs cs cs cs cs
o o o o o
OO OO OO 00 OO
n 10 n ~ o >o^-i
M W M M M
O O ซ-i O i-i
^O OO CO O\ CM
O O 1-1 t-( rt
r~- r~ r- t r
ISIII
o o o o o
t/5 &0 CO CO 00
s
*C ffl U
I
ill
.S
ง I
O T3
19
-------
Table 2-2. QA Objectives for Precision, Accuracy, and Detection Limits - Alternate Form
Data
Quality Parameter
Method of
Determination
Frequency
Required '.'
Objective3^1)
Precision
-> Water
- TCLP leachates
Accuracy
- Water & TCLP leachates
- Water & TCLP leachates
- Water
- MM5 train
Detection limits
- Water
- MM5 train
Semivolatile Organic Compounds
Field duplicate
Duplicate leaching
of laboratory - split
sample
Laboratory matrix
spike
Surrogate
NIST standard
reference materials
Spike of XAD resin
7 analyses of
spiked clean water
3 analyses of
spiked XAD
I/test condition
I/test condition
1 water/test
condition
1 leachate/test
condition
All samples
I/project
3/project
1 before project
1 before project
RPD<50
RPD<60
Recovery = 50-150%
Recovery = 50-150%
(b)
Recovery = 50-150%
Recovery = 50-150%
MDL i 10 Aig/L, neutrals
MDL ฃ 20 tfg/L, phenols
MDL i 100 ng/L, bases
MDL * 20 vg
a RPD = relative percent difference.
MDL = method detection limit.
b As specified in Method 8270.
Note (1) Objectives must be met for all critical measurements. (See Section 1.2 of this document.)
Note (2) MDLs must be calculated according to Equation (8), Section 8.0 of this document, which takes into
account the number of low-level samples analyzed.
% This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type- ' ' "
20
-------
Table 2-3. Required Detection Limits for Volatile Chlorinated Organic Compounds*
Compound
Regulatory
Threshold Ug/L)
Required
MDL
1,1,1 -Trichloroethane
1,1 -Dichloroe thane
1,1 -Dichloroethene
Vinyl chloride
1,2-Dichloroethane
Perchloroethylene
5
5
5
2
1
5
0.5
0.5
0.5
0.2
0.1
0.5
Method detection limits for these compounds are critical and will be determined experi-
mentally by the laboratory before sample collection is started. MDLs must be well below the
Regulatory Threshold in order to demonstrate that treated waters meet discharge limits.
This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type.
For Category III projects, completeness is defined as the number of measure-
ments judged valid compared to the total number of measurements. For this
project 10 samples will be collected and analyzed for copper for each
operating condition. Because 8 copper determinations must be judged valid
for the test run to be successful, the completeness objective is 80 percent. If
this completeness objective is not realized, the test run will be repeated until
the 80% completeness objective is achieved.
When analyzing for a large number of organic compounds by GC or GC/MS, it is usually
unnecessary to list QA objectives for each compound separately. Instead, list QA objectives
according to compound type, as was done for the semivolatile detection limits in Table 2-2. In
other cases, for example, where detection limits are derived from applicable regulations, list
detection limits for individual compounds, as in Table 2-3.
21
-------
Finally, the author of the QA Project Plan must explain how the QA objectives are to be
interpreted in a statistical sense. QA objectives are often interpreted in a sense that all data must
fall within these goals; for such projects any data that fail to satisfy the QA objectives are
rejected and corrective action is undertaken. However, other interpretations are possible. For
example, the project requirements may be satisfied if the average recovery is within the
objectives; that is, excursions beyond the objectives might be permitted, but the average recovery
would have to satisfy the goals described in this table. Whatever the case, it is important to
describe in this section how tabulated QA objectives will be interpreted.
2.3 QUALITATIVE QA OBJECTIVES: COMPARABILITY AND
REPRESENTATIVENESS
Comparability is the degree to which one data set can be compared to another. For instance, to
evaluate an environmental cleanup process, analyses of the feed and discharge streams must be
comparable. Similarly, to perform a nationwide environmental survey, methods used at different
locations must be comparable. Comparability is achieved by the use of consistent methods and
by traceability of standards to a reliable source.
Representativeness is the degree to which a sample or group of samples is indicative of the
population being studied. An environmental sample is representative of a parameter of interest
when the average value obtained from a group of such samples tends towards the true value of
that parameter in the actual environment, as the number of representative samples is increased.
Representativeness is normally achieved by collecting a sufficiently large number of unbiased
samples.
The QA Project Plan should demonstrate that adequate comparability and representativeness will
be achieved.
22
-------
2.4 OTHER QA OBJECTIVES
Some projects may require additional QA objectives, such as mass balances. Requirements for
all additional QA measurements should be stated in this section.
Example: Soil Washing for Removal of PCP.
A mass balance will be calculated according to the following expression:
Mass Balance = Mass of PCP in output streams/mass of PCP in input streams
For this test to be successful, mass balance should be between 50 and 150
percent. The specific equations for calculating the influent and effluent masses in
terms of the measured quantities are given in Section _._ of this QA Project Plan.
2.5 WHAT IF QA OBJECTIVES ARE NOT MET?
This section should include a discussion of the impact of not meeting one or more QA objectives.
Will the project be a complete loss? Will some, but not all, of the project goals still be realized?
Will the statistical confidence level be reduced? Are there legal or regulatory ramifications?
Answers to such questions help provide a critical perspective on the QA Program.
23
-------
-------
SECTION 3.0
SITE SELECTION AND SAMPLING PROCEDURES
This section must describe a plan for site selection and sampling which is responsive to those
project objectives stated in Section 1.1.2. A detailed discussion of the sampling plan devel-
opment process is presented in Volume 2 of SW-846, Test Methods for Evaluating Solid Waste
(3rd Ed.). Since SW-846 will likely undergo revisions, be sure to refer to the latest revision or
edition.
This section explains the overall sampling strategy, the specific sampling procedures that will be
employed, sample custody, record keeping, and shipping requirements.
3.1 SAMPLING SITE SELECTION
For most Category III projects, the sampling sites will have been identified already, and pre-
liminary data will be available. However, for certain projects such as those involving surveys at
a number of geographically separate sites, it may be impossible to explicitly delineate sampling
locations in the QA Project Plan. For these projects, explain how sampling sites will be selected
during the project. The following information should be provided:
A qualitative statement of the sample population to be represented by the
samples;
A description of the statistical method or scientific rationale to be used in
selecting sample sites;
Descriptions of the type of sampling strategy (e.g., simple, stratified, sys-
tematic random sampling);
A description of the sample types (air, water, soil, biota);
A qualitative statement regarding potential sources of sample contamination;
Guidelines for determining sampling frequency and number for each sample
type; and
A discussion of the extent to which site selection will affect the validity of the
resulting data and project objectives.
25
-------
3.2 SAMPLING SITE DESCRIPTION
This section includes charts, maps, sampling grids, or tables specifying the exact sampling sites.
Note any site modifications or additions that will be needed prior to sampling, such as the
extension of duct work or the addition of special valves. Also describe all locations and access
points for critical process measurements such as flow rates, pressures, or temperatures. Discuss
any site-specific factors that may affect sampling procedures.
For each analyte and each sampling point, list the frequency of sample collection and the total
number of samples collected. This summary can best be prepared in tabular form, as shown in
Table 3-1. Note that the numbers given in this table may differ from the number of analyses,
since some samples may be analyzed in replicate, while others may be analyzed only on a
contingency basis. Explain the statistical basis for the sampling scheme, as necessary.
3.3 SAMPLING PROCEDURES
This section describes the specific procedures that will be used for collecting and preserving
samples.
Discuss each sampling procedure that will be employed. For EPA-approved
procedures, a reference is sufficient. Other sampling procedures should be
summarized in the text, and additional details should be provided in an
Appendix. (Copies of ASTM sampling procedures are frequently appended.)
Prepare a list of analytes, sample volumes to be collected, and the amount of
sample that is required for each analysis (see Table 3-2). Note whether the
required amount is intended for matrix spike/matrix spike duplicate determin-
ations or for a single determination only. Be sure to have your laboratory
manager review this table to ensure that the sample volume or mass is suf-
ficient for all intended analyses.
Describe any compositing or sample splitting procedures that will be employed
in the field or laboratory.
Describe any sampling equipment that will be used, and how this equipment
will be calibrated.
26
-------
Table 3-1. Summary of Number of Samples Required for Hypothetical Incineration Test
Slurry
Feed
Description/Use (Fl)
SVOCs3
Non-QC Samples + MS + MSDb
Sample Duplicates
Sample Blanks
Spare Samples0
Modified Method 5
Non-QC Samples
Sample Blanks (see text)
VOCsa
Non-QC Samples
Sample Blanks
Trip Blanks'1
Spare Samples0
10
3
1
5
0
0
20
1
1
20
Scrubber
Sump
(SI)
10
0
1
5
0
0
20
1
1
20
Scrubber
Blowdown
(S2)
10
0
0
5
,0
0
20
1
1
20
Ash
(S3)
5
0
0
5
0
0
10
1
1
10
Stack
(S4)
0
0
0
0
4
1
0
0
0
0
Total
35
3
2
20
4
1
70
4
4
70
As designated in Section.
Each sample will provide enough material for an original sample plus a matrix spike and a
matrix spike duplicate.
Not analyzed unless first analysis fails.
Not analyzed unless sample blank is contaminated.
This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type.
27
-------
Table 3-2. Summary of Laboratory Analyses and Sample Quantity Requirements
Sample Quantity
Stream Sampling Method Analysis Parameter Container Size Required for Analysis^1)
Test soil
Treated soil
Scrubber makeup
scrubber liquor
Scrubber solids
Stack
Shelby tubes
ASTM-C172/
composite' '
Grab
Thief
Method 5
Midget M-S^2'
Metals trains
Modified Method 5
NIOSH-1003^
Semivolatiles
Volatiles
Metals scan
Dioxins/furans
EP toxicity
TCLP
Higher heating value
Chlorine
Moisture
Semivolatiles
Volatiles
Metals scan
Dioxins/furans
TCLP
Semivolatiles
Volatiles
Metals scan
Dioxins/furans
Semivolatiles
Volatiles
Metals scan
Dioxins/furans
Paniculate
HC1
Metals
Semivolatiles
Dioxins/furans
Volatiles
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
250 mL
1 L
40 mL VGA vial
1 L
1 L
250 mL
250 mL
250 mL
250 mL
50 g
50 g
100 g
50 g
250 g
500 g
250 g
250 g
100 g
50 g
50 g
100 g
50 g
500 g
1 L
40 mL VGA vial
1 L
1 L
50 g
50 g
100 g
50 g
900 L
900 L
900 L
3000 L
3000 L
20 L
Note (1) Indicate whether the listed quantity suffices for replicates, spikes, and other QC purposes, or if it is
sufficient for only a single analysis. In the latter case, indicate how extra material will be provided for
QC purposes.
Note (2) Copies of these methods would have to be appended because they are not readily available.
This table is meant to illustrate typical format
The tests listed are NOT necessarily those that would be employed
for a process of this type.
28
-------
Explain how sample containers will be cleaned to prevent sample contami-
nation, and how new sample containers will be checked for,contaminants.
Describe the containers used for sample collection, transport, and storage for
each sample type. Include sample preservation methods, noting specific
reagents, equipment, supplies, etc., required for sample preservation, and the
specific time requirements for shipping samples to the laboratory. Note
refrigeration conditions and holding times that will be employed. (See, for
example, Table 3-3.)
Describe the procedures used to record sample history, sampling conditions,
and any other pertinent information; include examples of forms that will be
employed. Describe the numbering sequence to ensure that each sample will
be assigned a unique number.
Include an example of the sample label to be used.
How are you doing so far? At the completion of this section,
you should be able to calculate the required number
of each type of sample container.
3.4 SAMPLE CUSTODY
Occasionally samples are spilled, contaminated, accidentally evaporated to dryness, or otherwise
compromised before or during sampling and analysis. Sample custody allows the detection of
any such problems should they occur and minimizes such occurrences by assigning responsibility
for all stages of sample handling. . .
This section should answer the following questions: :
Who will be responsible for samples during field sampling and at each of the
laboratories?
, How will samples be stored and shipped to avoid contamination and tam-
- .'. pering? ; ' ,.,. " -- -
29
-------
Table 3-3. Required Containers, Preservation Techniques, and Holding Times
Measurement
Type3
Preservation1*
Maximum Holding Times0
Extendable organics
Pesticides, PCBs
Metals (except mercury
and chromium VI)
G Coolto4ฐC,
Teflon-lined septum protect from light
G Coolto4ฐC,
Teflon-lined septum pH 5-9
P,G
HNO3topH<2
7 days until extraction,
40 days after extraction
7 days until extraction,
40 days after extraction
6 months
Mercury
Chromium VI
PH
Residue
Organic carbon, total
Sulfide
P,G
P,G
P,G
P,G
P,G
P,G
HNO3topH<2
Cool,48C
None required
Cool, 4 "C
Cool,4ฐC,HClor
H2SO4topH<2
Coolto4'C;
add zinc acetate plus
NaOHtopH>9
28 days
24 hours
Analyze water immediately
(on site); none specified for soil
7 days
28 days
7 days
a Polyethylene (P) or glass (G).
b Sample will be preserved immediately upon sample collection.
c Samples will be analyzed as soon as possible after collection. The times listed are the maximum times that
samples will be held before analysis and still be considered valid. All data obtained beyond the maximum
holding times will be flagged.
This table is meant to illustrate typical format, -/ - - ,
Preservation requirements, although typical, may not be appropriate for every project
30
-------
How will sample history be recorded? How will problems such as sample loss,
possible contamination, or other problems affecting sample integrity be
reported?
Do you want to save time and money?
Use Standard Operating Procedures (SOPs)! Certain information required
in this section - such as sample custody procedures and sampling methods -
is not project-specific, but applies to a wide range of sampling efforts.
Typically, a sampling team or a laboratory will use the same chain-of-
custody procedures from one project to the next. Such procedures may be
written as SOPs, which can then be appended to subsequent QA Project
Plans. Be sure to use document control format in the SOPs to alert the
reviewer to any recent changes in the procedure.
31
-------
-------
SECTION 4.0
ANALYTICAL PROCEDURES AND CALIBRATION
This section of the QA Project Plan must describe all of the field and laboratory procedures used
for both chemical and physical measurements. Sample preparation methods and cleanup
procedures (such as extraction, digestion, column cleaning) should also be included.
All methods must be appropriate for their intended use and be described in sufficient detail. This
section, when coupled with QC procedures described in Section 6.0, should provide enough
detail to permit analytical chemists or other measurement specialists to carry out their procedures
unambiguously.
Requirements of this section can often be satisfied by referencing appropriate standard methods.
Most RREL projects rely heavily on EPA-approved methods that have been validated for
environmental samples. Standardized procedures from other organizations, such as the ASTM
and the American Public Health Association, are also commonly employed when approved,
validated EPA methods are unavailable. When standard methods are unavailable, nonstandard
methods must be used and described in detail. It is essential to consult the analytical
laboratory and other measurement specialists for guidance during preparation of this
section because they know the intricacies and limitations of the methods.
Begin this section with a summary table of all measurements to be made. For projects involving
many kinds of measurements, consider providing separate tables for each type of test. Specify
the parameter to be measured, the sample type, the method number (when available), the title, a
method type, and a reference. (See Table 4-1 as an example.) Note the revision number of the
method or the edition number of a publication, since nominally identical methods may be
modified by these updates. Provide additional project-specific information in the text of this
section.
33
-------
s
s
8-8
o
oo
3
34
-------
i
J
4
I
a
O
I
ฃ
"8
S
?
g
iS
I
lg
1 o
"+3 ฃ3
t
>ป '
"o -
cu PQ
SO
o o
OT U
-I
0,
I
III
W $ W
1
o
oo
1!
a
11
M
S
eo O >
'.a
f
3
42
"3
"2
2
ซ
a S
05
I
8
o
.S
fe
ง
a
a
K
o
*..
II o
*rt S **
1 I-8>
S| S
r-| P 0<
II II II
&o P S
S
-I
I
IS
c1
2
i-J i-5
n ii
a
S
J
35
-------
4.1 EPA-APPROVED OR OTHER VALIDATED STANDARD METHODS
EPA-approved or similar validated methods can be incorporated by reference, thereby greatly
minimizing the effort required to complete this section. Once a method is cited, do not repeat
information that is already found in the method. Some additional information is almost always
required, however, to assure that project-specific requirements will be met. The following
considerations apply:
Some EPA-promulgated methods contain only general procedure descriptions
and lack specific QC requirements or applicable validation data. For example,
EPA Method 18 provides general requirements for GC analysis of stack gases,
but contains no QC requirements. For such methods, validation data pertinent
to the specific project must be appended. Alternately, a preliminary method
validation can be specified as a subtask of the project; in that case, however,
specific procedures and acceptance criteria for method validation must also be
included as part of the QA Project Plan.
Other EPA-approved standard methods, such as those found in Standard
Methods for the Examination of Water and Wastewater, give operating
procedures but omit most QC and calibration requirements. This missing
information must be provided either in this section or in Section 6.0 of the QA
Project Plan. As a minimum, be sure to specify the frequency, acceptance
criteria, and corrective action plans for all QC procedures and calibrations.
It is not necessary to include copies of methods or sections of methods from
SW-846, the Code of Federal Regulations, Standard Methods for the
Examination of Water and Wastewater, or Methods for Chemical Analysis of
Water and Waste because these sources are readily available. However, do
append ASTM and NIOSH procedures because they may not be so readily
available to other project principals or reviewers.
EPA-approved or similarly validated methods that are significantly modified
are considered to be unvalidated methods and are treated as described in
Section 4.2.
Certain EPA methods such as those found in SW-846 specify most operating
details, including quality control and calibration requirements. Such
procedures, however, frequently allow the user to specify certain other options
36
-------
to satisfy project objectives. For example, for multianalyte methods such as
GC/MS, the user will typically specify the target compound list that is required
for the project. Matrix spike compounds are then chosen from the compounds
of interest to the particular project, and are not necessarily those recommended
in the method. Also list, the project-specific target compounds to be used as
calibration check compounds and matrix spike compounds. Specify the
acceptance criteria for matrix spike compounds. In certain cases, isotopically
labeled forms of the project analytes may be included as surrogates.
The following is an example of how one might specify various options allowed by Method 8080,
a validated method from SW-846 which contains QC and calibration requirements:
Example: Method 8080.
This method will be employed to determine the 19 pesticides listed in Table of
this QA Project Plan. Although PCBs can also be determined by this method, the
GC will not be calibrated for these analytes unless they are observed. The matrix
spike compounds will be the six critical pesticides listed previously in this QA
Project Plan. The surrogate will be tetrachlorometaxylene (not dibutyl-
chlorendate). Detection will be by ECD. Quantitation will be by external
calibration. Acceptance criteria for surrogate recovery will not be determined by
control charts, but must be within a 50-150 percent range. Acceptance criteria for
matrix spike/matrix spike duplicates are those stated in Table of this QA
Project Plan. Extraction and cleanup procedures are described below...
4.2 NONSTANDARD OR MODIFIED METHODS
Any nonstandard procedure must be described in detail in the form of a Standard Operating
Procedure (SOP) that is appended to the QA Project Plan. Validation data applicable to the
expected samples must also be included. Validation data must demonstrate that the analytes of
interest can be determined without interferences in the expected matrices, and that precision,
accuracy, and detection limits will be adequate for the intended use of the data. The method is
validated only for samples expected from the specific project, and not for general environmental
usage. Once the SOP is written and the validation data are accepted, the method is
validated only for use on that specific project and on the specific sample matrix associated
with that project. This validation process for a project-specific analysis does not constitute
EPA approval for other projects or matrix types.
37
-------
If the method is not yet validated, describe the intended procedure, state how it will be validated,
and specify what criteria must be met before it is accepted.
4.3 CALIBRATION PROCEDURES AND FREQUENCY
This section covers calibration procedures for each analytical or measurement system used to
obtain data for critical measurements. Each description should include the specific calibration
procedure to be used and the frequency of calibration verification.
In the case of standard EPA-approved methods that include calibration
procedures, a reference to those methods suffices. Simply list the required
frequency and acceptance criteria in a summary table, such as that shown in
Table 6-1. For nonstandard methods that are appended to the QA Project Plan,
reference can be made to the SOP. Describe all other calibration procedures in
detail.
A list of calibration standards, including source, traceability, and verification
of purity, must be included.
For process measurements (e.g., flow, mass, etc.) and for chemical or physical
analyses, specify the planned frequency of initial and continuing calibration
checks and the acceptance criteria for all calibration measurements. Table 4-2
illustrates a typical summary for process measurements. For routine,
scheduled calibrations, include the calibration frequency and acceptance
criteria in Section 6.0, along with a summary of QC requirements.
For physical measurements such as temperature and pressure, calibration statements can be quite
brief.
Example: All thermocouples intended for use in the range 50"Cto 350" C will be
calibrated versus an NIST-traceable thermometer at a minimum of two
temperatures by placing both in an oven simultaneously and noting any difference
between readings. The thermocouple readout must be within 2ฐ C of the
corrected mercury thermometer or the thermocouple device will be replaced or
corrected.
38
-------
Table 4-2. Typical Summary Table of Calibration Requirements
for Physical Process Measurements
Parameter
Flow rate of dopant
feed
Secondary volume
standard
Secondary time
standard
Temperature
(50-300 "C)
Temperature
(300-800 ฐC)
Flow rate in stack
Measurement
Classification
Critical
Critical
Critical
Noncritical
Critical
Critical
Device
Mass flow
meter
Dry test
meter
Stopwatch
Thermometer
Thermocouple
Pilot tube
and manometer
Calibration
Procedure
Compare to cali-
brated dry test
meter
NIST-traceable
spirometer
MIST time
base
Comparison to
certified ther-
mometerฎ
2 temperatures
Comparison to
NIST-calibrated
thermocouple @
2 temperatures
Measure pilot
orifice with NIST-
Frequency
Before/after
field test;
weekly
Before/after
field test
Before field
test
Before project
initiation
Before field
test
Before field
test
Acceptance3
Criteria
5%
2%
.05 sec/min
2'C
5ฐC
1%
traceable micro-
meter; compare
manometer mark-
ings to NIST-
calibrated meter
stick
a Maximum allowable deviation from standard.
This table is meant to illustrate typical format.
The tests listed are NOT necessarily those that would be employed
for a process of this type,
39
-------
-------
SECTION 5.0
DATA REDUCTION, VALIDATION, AND REPORTING
This section describes how data will be reduced, validated and reported. Deliverables that will
be required from the analytical laboratory must also be specified.
5.1 DATA REDUCTION
5.2
Name the individual responsible for data reduction.
Summarize the data reduction procedures that are specific to this project. Data
reduction procedures that are part of standard methods or SOPs cited in
Section 4.0 should not be repeated here other than to note any deviations.
Summarize the planned statistical approach, including formulas, units, and
definition of terms. Do not simply reference a "standard text."
Explain how results from blanks will be treated in calculations.
DATA VALIDATION
Name the individuals responsible for data validation at each stage of data
reduction.
Describe the procedures that will be used for determining outliers. Describe
the guidelines that will be employed for flagging or validating data. (This
requirement is normally satisfied by Section 6.0 with respect to routine quality
control.) In Section 5.0, discuss criteria that are not covered in Section 6.0.
5.3 DATA REPORTING
Name individuals responsible for the various types of reports.
Indicate the units for each measurement and each matrix, and whether data will
be reported on a wet, dry, or some other reduced basis. If this requirement has
been satisfied by information found in other sections (e.g., Table 2-1), do not
repeat it here.
Indicate data storage requirements that will be expected of the laboratory once
the project is complete. Will the laboratory need to maintain a complete set of
raw data for six months? One year? Five years? Also indicate how long the
actual samples will be stored, in case reanalysis is required.
41
-------
List the deliverables that will be expected from the laboratory and from the
field operations. Will the data package include all raw data sufficient to
recalculate any result, if need be? What type of QC data will be reported?
Reporting requirements may vary, depending on the intended use of the data.
* Summarize the data that will be included in the final report. What QC data
will be included? Will analytical and other measurement data be partially
reduced before they are reported or will all individual measurements be
reported?
Example: The final report will contain the following analytical data:
All analytical results from all primary samples. Data judged to be outliers will be
included, along with a justification for excluding these outliers from further
interpretation.
All individual results from standard reference materials, independent check
samples, replicates, and matrix spikes, including initial and final concentrations.
Data from the continuous emission monitors will be reported as 15-minute
averages and, when appropriate, will be summarized in graphical form.
For Category III projects, this section should contain a statement that the final report will include
a QA section that documents QA/QC activities and results. These results must be readily
correlated to the primary data and must clearly indicate the limitations of the data and the range
of validity of the conclusions. The final report should also include a summary of the original QA
objectives, and a statement regarding whether these objectives were met. If QA objectives were
not met, include an explanation of the impact of not meeting the project's QA objectives.
42
-------
SECTION 6.0
INTERNAL QUALITY CONTROL CHECKS
This section describes all internal quality control (QC) checks that will be used throughout the
project, including field and laboratory activities of all organizations involved. The QC
procedures that are specified should follow from the QA objectives stated in Section 2.0. Thus,
Section 2.0 specifies the analytical requirements, while Section 6.0 describes how these specifi-
cations will be met.
6.1 TYPES OF QC CHECKS
Examples of QC checks that should be considered include the following:
Samples
- Collocated, split, replicate
Spikes
- Matrix spikes and matrix spike duplicates
- Spiked blanks
- Surrogates and internal standards
Blanks
- Sampling, field, trip, method, reagent, instrument
- Zero and span gases
Others
- Standard reference materials (complex natural materials, pure solutions)
- Mass tuning for mass analysis
- Confirmation on second column for gas chromatographic analyses
- Control charts
- Independent check standards
- Determinations of detection limits
- Calibration standards
- Proficiency testing of analysts
- Any additional checks required by the special needs of your project
Include QC checks for process measurements as well.
43
-------
Be sure to:
. Identify the stage at which replication and spiking occitf. Avoid using
terms such as "sample replicate" without explaining how sueu replication^
will be performed,
-K^Ml " v^JWXrtv
Explain exactly how blanks will be prepared.
6.2 ITEMS TO INCLUDE
Most information for this section can be summarized in a table, as shown in Table 6-1. This
table should designate the types of QC procedures, required frequencies, associated acceptance
criteria and corrective action that will occur if the acceptance criteria are not.met. When QC
procedures are referenced to a standard method that describes an exact procedure, additional
discussion is not normally needed. However, standard methods lacking exact QC procedures or
nonstandard methods require a detailed explanation, either in this section of the QA Project Plan
or in an appended Standard Operating Procedure.
The tabular format shown in Table 6-1 is also convenient for summarizing routine and ongoing
calibration requirements. If routine calibration is summarized in this section, a reference to that
effect should be included in Section 4.0.
The accompanying text must assure that there are no ambiguities. Particularly troublesome terms
are "duplicate" or "replicate." The text should explain precisely how and when replicates are
taken. Do replicate samples, for instance, refer to samples collected simultaneously or
sequentially in the field; to samples collected at the same sample point but at different times; to
samples that are split upon receipt in the laboratory? The term "QC check sample" must also be
carefully defined. Indicate at which point matrix spiking occurs.
Exact procedures for preparing the numerous kinds of blanks must be described fully in the text.
Never assume that a term such as "field blank" will mean the same to the sampling team or to a
reviewer that it does to you.
44
-------
ฃ 1
.s
co
co
!Q Ti
! b ?.
IซS 10
CO rj;
CO cซS
r- H
II
S 2
If
i
'ง
I
ฃ
3
o
I C3
I VI
tq
vq
od
VO
00
M
If!
m g 0
ll
o o
S S;
VO
od
cf
od
45
-------
Either in this section or in Section 4.0, be sure to specify what compounds or elements will be
employed as matrix spikes and surrogates. Some standard methods recommend surrogate and
matrix spike compounds, which may be incorporated by reference, when appropriate. More
typically for RREL projects, some of the matrix spike compounds must be selected on a project-
specific basis.
In some cases, it may also be necessary to provide additional discussion of potential problems
that might be expected with certain QC procedures, along with the proposed solutions. For
example, spiking samples in the field is frequently less reliable and more difficult than spiking m
the laboratory, due to contamination and less controlled conditions. If field spiking is required,
then a discussion of procedures that minimize such problems is also required.
In as much as standard methods often include extensive QC requirements, it is natural to ask why
such QC procedures must be summarized in this section. Why not simply state that QC will be
performed as required in the method? In some cases, EPA standard methods are sufficiently
complete for this approach, but frequently they are not. First, many EPA-approved methods do
not include specific QC procedures. Second, even the more complete EPA methods allow
options, such as the choice of matrix spike compounds, or the use of either control charts or fixed
acceptance limits. Third, the analytical and measurement requirements of RREL are often
project-specific and require project-specific QC guidelines.
Thus this section of the QA Project Plan must provide an unambiguous description of QC
procedures. There is no need, however, to repeat in the text any material that is already descnbed
in a standard method.
46
-------
SECTION 7.0
PERFORMANCE AND SYSTEMS AUDITS
Each Category III QA Project Plan should describe the QA audits planned for monitoring the
system to be used for obtaining critical measurements. A schedule of all contractor-planned
audits should be included, along with identification of the responsible personnel. This section
should also indicate what audit reports will be generated and who will receive these reports. If
no audits are planned, include an explanation.
QA audits for RREL projects may include one or more Technical Systems Audits (TSAs),
Performance Evaluation Audits (PEAs), and Audits of Data Quality (ADQs).
A TSA is a qualitative evaluation of all components of the total measurement system, including
technical personnel and QA management. This type of audit includes a careful evaluation of
both field and laboratory QC procedures. TSAs are normally performed before, or shortly after,
measurement systems are operational; they should also be performed on a regularly scheduled
basis throughout the lifetime of the project.
After measurement systems are operational and begin generating data, PEAs are conducted
periodically to determine the bias of the total measurement system or component parts. As part
of a PEA, the laboratory analyzes a performance evaluation sample. QA Project Plans should
also indicate any scheduled participation in other interlaboratory performance evaluation studies.
Long-term projects should provide for regularly scheduled PEAs.
ADQs are retrospective evaluations of data. Typically, a representative portion of the results in
an analytical report is reviewed in detail, starting with raw data and chromatograms, and
proceeding through the calculation of final results. ADQs are often used to resolve specific
questions regarding the quality of a data set.
47
-------
-------
SECTION 8.0
CALCULATION OF DATA QUALITY INDICATORS
This section describes how data quality indicators will be calculated and reported. As a mini-
mum, equations must be provided for precision, accuracy, completeness, and method detection
limits. In addition, equations must be given for other project-specific calculations, such as mass
balance, emission rates, confidence ranges, etc.
Make sure that this section complements Section 2.0 (QA Objectives) and Section 6.0 (Internal
QC Checks) of the QA Project Plan. Section 2.0 .specifies which particular data quality indi-
cators will be employed. Section 6.0 then uses these specific indicators to generate acceptance
criteria. Because groups can use different equations to calculate data quality indicators,
Section 8.0 must provide the exact equations to avoid future misunderstandings among future
data users.
Listed below are general guidelines for calculating the more common data quality indicators.
8.1 COMMON DATA QUALITY INDICATORS
8.1.1 Precision
If calculated from duplicate measurements, relative percent difference is the normal measure of
precision:
RPD =
(C-L - C2) x 100%
C2)/2
(1)
where:
RPD = relative percent difference
C-]_ = larger of the two observed values
C2 = smaller of the two observed values.
If calculated from three or more replicates, use relative standard deviation rather than RPD:
RSD = (s/y) x .100%
where: RSD = relative standard deviation
s = standard deviation
y = mean of replicate analyses.
(2)
49
-------
Standard deviation is defined as follows:
o ss
n
Z
1=1
(yi
n
-5>2
- 1
(3)
where:
s = standard deviation
r j = measured value of the ith replicate
y = mean of replicate measurements
n = number of replicates.
For measurements, such as pH, where the absolute variation is more appropriate, precision is
usually reported as the absolute range, D, of duplicate measurements:
D =
- m I
(4)
where:
D = absolute range
-L = first measurement
i = second measurement
The standard deviation, s, given above, can also be used.
8.1.2 Accuracy
For measurements where matrix spikes are used, calculate the percent recovery as follows:
%R = 100% x
S - U
sa
(5)
where:
%R =
O
U =
csa =
percent recovery
measured concentration in spiked aliquot
measured concentration in unspiked aliquot
actual concentration of spike added.
50
-------
When a standard reference material (SRM) is used:
%R = 100% x
m
srm
where: %R =
Cm =
csrra =
8.1.3 Completeness
percent recovery
measured concentration of SRM
actual concentration of SRM.
(6)
Completeness is defined as follows for all measurements:
%C = 100% x
V
T
(7)
where:
%c
V
T
percent completeness
number of measurements judged valid
total number of measurements.
8.1.4 Method Detection Limit (MPLI
MDL is defined as follows for all measurements:
MDL -
-a=0. 99)
(8)
where: MDL
s
t(n-l,l-a=0.99)
method detection limit
standard deviation of the replicate analyses
Students' t-value for a one-sided 99% confidence level and a
standard deviation estimate with n-1 degrees of freedom.
51
-------
8.2 PROJECT-SPECIFIC INDICATORS
RREL projects frequently incorporate data quality indicators in addition to those discussed in the
previous sections. The following is an example of a project-specific data quality indicator:
Example: Mass balance calculation for a soil washing process being tested with
PC?-contaminated soils. Mass balance (MB) will be calculated according to
(9)
where MQUt and Mir, denote the total mass ofPCP in the output and input
MB=Mout/Min
note ;
streams for each test condition.
52
-------
SECTION 9.0
CORRECTIVE ACTION
Each QA Project Plan must incorporate a corrective action plan. This corrective action plan must
include the predetermined acceptance limits, the corrective action to be initiated whenever such
acceptance criteria are not met, and the names of the individuals responsible for implementing
the plan.
Routine QC procedures already included in Section 6.0 need not be repeated here. This section
is reserved primarily for nonroutine corrective action not described elsewhere. Nonroutine
corrective action may result from common monitoring activities, such as:
Performance evaluation audits,
Technical systems audits, and
Intel-laboratory comparison studies.
It may also arise from conditions unique to specific projects, as seen below.
Example: Bio-oxidation study of hazardous waste in municipal sludge carried
out over a one-year period.
The prime contractor's Quality Assurance Manager will submit to the laboratory
a blind performance evaluation (PE) sample containing some or all of the target
analytes before any analytical work begins and on a monthly basis thereafter.
The average percent recovery of all target analytes must be between 80 and 120
percent, with no outliers less than 50 or greater than 150 percent. If these limits
are exceeded, analytical work will stop until the problems are identified and
solved. Before work is restarted, another blind PE sample must be analyzed and
results must meet the acceptance criteria. Results of these PE samples will be
included in the final report.
53
-------
-------
SECTION 10.0
QUALITY CONTROL REPORTS TO MANAGEMENT
This section of a Category III QA Project Plan identifies the individuals responsible for QA
reports, and describes the type and frequency of reports (weekly oral presentations and discus-
sions, monthly written reports, etc.) that will be used to keep project management informed. As
a minimum, such reports include:
Changes in the QA Project Plan
Summary of QA/QC programs, training, and accomplishments
Results of technical systems and performance evaluation audits
Significant QA/QC programs, recommended solutions, and results of correc-
tive actions
ซ Data quality assessment in terms of precision, accuracy, representativeness
completeness, comparability, and method detection limit
on
Discussion of whether the QA objectives were met, and the resulting impact
decision making
Limitations on use of the measurement data.
Managers receiving these detailed reports will then be able to monitor data quality easily
and effectively. y
55
-------
-------
SECTION 11.0
REFERENCES
References, if any, can be included in the body of the text, as footnotes, or collected in this
section. If a reference is not readily available, attach a copy to the QA Project Plan. References
must uniquely identify the cited material. In particular, when citing various compendia of
standard methods published by the EPA, ASTM, American Public Health Association, etc., be
sure to include the edition number since such methods can change substantially from one edition
to the next.
For example, the following are references applying to Table 4-1 of this document.
1. American Society for Testing and Materials. Annual Book of Standards. ASTM
Philadelphia, PA. Parts 14 and 19 (updated yearly).
2. U.S. Environmental Protection Agency. "A Procedure for Estimating MonofUled Solid
Waste Leachate Composition." Technical Resource Document SW-924, 2nd ed
Hazardous Waste Engineering Research Laboratory, Office of Research and
Development. Cincinnati, OH, and Office of Solid Waste and Emergency Response
Washington, D.C., 1986.
3.
4.
5.
6.
American Nuclear Society. ANSI/ANS-16.1-1988. American National Standard
Measurement of the Leachability of Solidified Low-Level Radioactive Wastes by a Short-
Term Test Procedure. American Nuclear Society. LaGrange Park, IL, 1986.
U.S. Environmental Protection Agency. Office of Solid Waste. Test Methods for
Evaluating Solid Waste, 3rd ed. Available from U.S. Government Printine Office
Washington, D.C., 1986.
Alford-Stevens, A., T. A. Bellar, J. W. Eichelberger, and W. L. Budde. Method 680
Determination of Pesticides and PCB in Water and Soil/Sediment by Gas
ChromatographylMass Spectrometry. Available from the U.S. Environmental Protection
Agency. Cincinnati, OH, 1985.
Klute, A. (Ed.). Methods of Soil Analysis, Part I. American Society of Agronomv
Madison, WI, 1986.
57
&U.S. GOVERNMENT PRINTING OFFICE: 1991 - 548-187/20545
-------
-------
-------
-------
-------
3- rt>
<
(0
CO
o
o
> m c
CO 3 3
o> <. ~
3 -i (D
O O Q.
(D 0>
E ป
5) en
>
o
!
o
I
O
Ul
in o
o to
ง0
ft)
: o
i
3.0 3
D 5 Q)
o - =
r* QJ _.
3 ป
O- 0
-, o
28
C Q]
3-3
S|
r* 03
3- D
5'3 SปL
3 to _J
.
(Dl
0)1
-o
m
T3
O
no
=1 _ m r-
9
CO
01
S3m
13
------- |