REGION II
CERCLA QUALITY ASSURANCE MANUAL
Final Copy
Revision 1
October 1989
-------
FORWARD
This CERCLA Quality Assurance Manual has been prepared by the Monitoring
Management Branch of the Environmental Services Division for use by the Emergency
and Remedial Response Division Project Managers of the Region II Superfund
(CERCLA/SARA) program in their daily working with contractors. Its use is intended to
ensure that quality assurance and quality control practices (QA/QC) are fully built into all
monitoring project designs.
Part I of this Manual, Administrative Procedures, describes the roles and responsibilities
of the various Region II organizational units involved in the Superfund program. Part I is
designed to be used as a guide so that Project Managers understand what quality
assurance activities are required, who the key individuals responsible for carrying out
these activities are, and where and how to obtain information and assistance in meeting
these requirements.
Part II of this Manual provides Region ll's quality assurance/quality control requirements
for CERCLA sampling and analysis. Quality assurance procedures are used to verify
that field and laboratory measurement systems operate within acceptable, defined
limits. The effectiveness of the overall Quality Assurance Program demands that all
personnel are aware of the QA/QC requirements for any investigation and that the
quality assurance objectives are understood. This Part outlines all aspects of quality
control in a monitoring program: minimum requirements necessary, and the rationale
behind the requirements. The quality control procedures outlined in Part II of this
Manual should be incorporated into all field/project/site operations plans and/or quality
assurance project plans prepared for CERCLA work. Recommendations and
requirements presented herein should be incorporated into project designs to the fullest
extent possible. Where deviations from these recommendations and requirements is
necessary, full justification must be presented in writing.
This Manual is meant to be a dynamic document. It will periodically be reviewed and
updated, however it is not meant to provide definitive answers to all site-specific
concerns. This is, rather, an attempt to provide the rationale behind the most common
site-specific concerns which could be extrapolated for use in new situations.
-------
Disclaimer
This Manual has been prepared for use by the Environmental Services and the
Emergency and Remedial Response Divisions of the USEPA, Region II. Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use by the Agency.
-------
REGION II CERCLA QUALITY ASSURANCE MANUAL
TABLE OF CONTENT
Forward
Disclaimer
Table of Content
PART I: ADMINISTRATIVE PROCEDURE
I. CERCLA QA Program 6
A. Definition of Quality Assurance 6
B. Basis for Quality Assurance in the CERCLA Program 6
II. CERCLA QA Organization 7
A. Emergency and Remedial Response Division 7
B. Environmental Services Division 8
C. Superfund Contractor Services 9
III. Data Quality Objectives 11
A. Definition 11
B. Roles and Responsibilities of Key Personnel 11
C. DQO Guidance 11 ,
IV. Preparation of QA Project Plans 12
A. Definition 12
1. Site-Specific QAPjP 12
2. Generic QAPjP . 12
B. Procedures for Preparation, Submittal, and Approval of QAPjPs 12
1.EPA-Lead Sites 12
2. State-Lead Sites 12
3. Federal Facility Sites 13
4. Responsible Party Sites 13
C. Guidance Documents for Preparing QAPjP's 14
D. QAPjP Pre-Development Meetings • 14
V. Audit Program 15
A. Management Systems Audit ' 15
B. Technical Systems Audit 15
C. Data Quality Audit 15
D. Performance Evaluations 16
-------
VI. Superfund Contract Laboratory Program 17
VII. Data Validation / Data Useability 19
A. Data Validation 19
B. Data Useability 19
Acronym List 20
PART II: QUALITY CONTROL HANDBOOK FOR CERCLA SAMPLING AND
ANALYSIS 22
I. Sampling Design and Strategy 23
A. Sampling Plan Components 23
B. Purpose and Objective of Sampling 24
C. Types of Samples 24
1. Environmental 25
2. Hazardous 25
D. Types of Measurement 25
1. Laboratory Measurement , 25
a. Grab 25
b. Composite 25
2. In-Situ Measurement 26
II. Analytical Methods, Preservation and Holding Times 27
A. Methodology Available for use in the CERCLA Program 27
B. Analytical References 28
1. Aqueous/Solid Matrices 28
2. Air 29
3. NAPL 29
C. Preservation, Methodology and Holding Times 30
D. QC Criteria for 40 CFR Part 136 and SW-846 Third Edition Methods 32
III. Documentation Procedures 33
A. Chain-of-Custody 33
1. Definition and Reference 33
2. Recordkeeping and Procedures 33
a. General 33
b. CLP 34
B. Field Records 35
IV. Glassware Requirements 37
A. Bottle Suppliers 37
B. Volume and Type of Container 37
C. Quality Control and Storage 37
-------
V. Field/Laboratory Decontamination of Sampling Apparatus 38
A. General Considerations 38
B. Decontamination Procedures 38
VI. Decontamination of Peripheral Equipment 39
A. Well Evacuation Equipment 39
B. Well Casings 39
C. Field Instrumentation 39
D. Drilling Equipment and Other Large Pieces of Equipment 39
VII. Monitoring Well Design and Construction 40
A. Well Drilling and Development Methods 40
1. General Discussion and Preferred Methods 40
B. Well Filterpack and Annular Sealant 41
C. Well Casing Selection 42
1. General Discussion 42
2. Selection Criteria SOP 44
D. Evaluation of Existing Wells 44
VIII. Sample Collection'Devices, Materials and Quality Control Practices 45
A. References for Selection of a Sampling Device 45
B. Ground water 45
1. Sampling Design 45
2. Well Evacuation 45
3. Sampling Organics and Inorganics 48
4. Microbiological Sampling 49
C. Surface water 51
1. Sampling Design 51
2. Sampling Devices 51
D. Sediment 53
1. Sampling Design 53
2. Sampling Devices 53
E. Soil . 54
1 .Sampling Design 54
2. Sampling Devices 54
F. Potable Water 55
G. Dust/Wipes 55
H. Dioxin 56
I. Drums 56
IX. Methods of Sample Preparation 57
A. Homogenization 57
B. Compositing 57
C. Splitting 58
-------
X. Field Quality Control Samples 59
A. All Analyses Except Dioxin 59
1. Duplicates 59
2. Blanks 59
a. Trip 60
b. Field 60
3. Matrix Spike/Matrix Spike Duplicate Analyses 61
B. Dioxin 61
XI. Filtered and Non-Filtered Fractions of Ground Water Samples 63
A. General Discussion 63
B. Procedures for Filtration of Aqueous Metals Samples 64
1. Decontamination of Apparatus . 64
2. Filtration Procedure and Preservation 65
XII. Laboratory Qualifications 66
A. Use of CLP vs. non-CLP Laboratories 66
XIII. Use of Mobile Laboratories 67
A. Qualifications and Methods 67
XIV. Validation of Data 68
A. CLP 68
B. non-CLP 68
XV. Field Auditing and Oversight 70
A. Audits Initiated by EPA and Primary Contractor 70
B. Contractors in an Oversight Capacity 70
C. Audits Performed by States 70
Bibliography 71
List of Appendices 72
-------
PART I
Administrative Procedures
-------
1. CERCLA QUALITY ASSURANCE PROGRAM
A. Definition of Quality Assurance
Quality Assurance (QA) is the review and oversight at the planning, implementation,
and completion stages of an environmental data collection activity to assure that when
data is provided to data users it is of the quality needed. A QA program is a system of
documented checks that ensures monitoring data is valid.
B. Basis for Quality Assurance in the CERCLA Program
All activities associated with the collection of physical and chemical data are required to
be incorporated into the Agency's formal quality assurance program (EPA
Administrator's Policy Statement, EPA Order 6350.1, May 30, 1979). These activities
include all phases of sampling, analysis, and data handling that can affect the validity of
the data. Any monitoring activity that generates data for EPA use must be collected
under a formal QA program that adheres to Agency requirements (40 CFR Part 300.68
(k) of the National Contingency Plan).
The QA requirements for Region II and CERCLA are spelled out in the Region II Quality
Assurance Program Plan (QAPP). The QAPP is approved annually by the Regional
Administrator, concurred by the Region II Division Directors and distributed to regional
managers and staff personnel. A QA program is needed in CERCLA to ensure that the
data collected are of the type and quality required for the specific decision to be made,
and to ensure defensible quality.
-------
II. CERCLA Quality Assurance Organization
A. Emergency and Remedial Response Division (ERRD) - ERRD is responsible for the
development, implementation and coordination of Regional activities under
CERCLA/SARA. ERRD manages a comprehensive program for site evaluation,
planned and immediate removals and long-term remedial actions including cost
recovery actions. ERRD serves as the focal point for all emergency response and
emergency contingency planning activity. ERRD is responsible for spill control and
monitoring programs under Section 311 of the Clean Water Act. Within ERRD there are
seven branches:
- Removal Action Branch (RAB)
- Response and Prevention Branch (RPB)
- N.Y./Caribbean Remedial Action Branch (NY/CRAB)
- New Jersey Remedial Action Branch (NJRAB)
- N.Y./Caribbean Compliance Branch (NY/CCB)
- New Jersey Compliance Branch (NJCB)
- Program Support Branch (PSB)
The Removal Action Branch and the Response and Prevention Branch together
comprise the Removal and Emergency Preparedness Program which responds to
spills, manages the Spill Prevention, Countermeasure and Control Program (SPCC),
manages the Superfund removal program, manages the Technical Assistance Team
(TAT) and the Emergency Response Cleanup Service (ERCS) contracts, develops
contingency plans and ensures response capabilities at remedial action sites.
The New Jersey and New York/Caribbean Remedial Action Branches carry out
remedial investigations (RIs), conduct feasibility studies (FSs), manage EPA remedial
contractor activities, overview remedial design (RD) activities, serve as liaisons with the
Corps of Engineers and manage EPA responsibilities for operation and maintenance of
facilities constructed at Superfund sites.
The New Jersey Compliance Branch and the NY/Caribbean Compliance Branch
conduct hazard assessments, manage CERCLA technical enforcement activities, draft
administrative orders, carry out remedial investigation/feasibility study (RI/FS)
responsibilities for the sites classified as enforcement lead, develop cost recovery
cases, negotiate settlements, and follow up on compliance by responsible parties with
orders and negotiated settlements.
The Program Support Branch is responsible for several Superfund program tracking
and data management systems as well as policy dissemination and implementation.
The Branch also manages the Superfund site assessment and investigation programs,
federal facility Superfund sites, dioxin sites, and site deletion activities. Many site
investigations (Sis) are conducted by the FIT contractor in which case the projects are
managed by the Environmental Services Division (ESD). The Branch also manages the
REM, ARCS, and TES contracts.
-------
B. Environmental Services Division (ESD) - ESD is responsible for setting
priorities and assuring that resources are available to collect environmental samples,
analyze collected samples, and evaluate the resulting data in support of monitoring
programs. ESD directs and coordinates the field and laboratory support, directs the
implementation of the QAPP, and directs special studies, investigations and surveys to
support the Regional enforcement actions or define environmental quality programs.
Within the ESD there are four branches:
.4
- Technical Support Branch (TSB)
- Pesticides and Toxic Substances Branch (PTSB)
- Monitoring Management Branch (MMB)
- Surveillance and Monitoring Branch (SMB)
The Technical Support Branch consists of an Organic Chemistry Section, an Inorganic
Chemistry Section, and a Sanitary Chemistry and Microbiology Section. This Branch is
responsible for chemical and microbiological testing of pollutants in support of
CERCLA, RCRA, CWA, TSCA, etc. activities.
The Pesticides and Toxic Substances Branch is responsible for the implementation of
the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) and the Toxic
Substances Control Act (TSCA). The FIFRA program has been delegated to all Region
II States and is basically an oversight program providing both technical and financial
assistance. The TSCA program is an enforcement effort with major emphasis on
polychlorinated biphenyl (PCB) chemicals and friable asbestos in primary and
secondary schools.
The Monitoring Management Branch is responsible for conducting the Region's quality
assurance and data management programs. MMB plans, coordinates, provides
technical assistance and evaluates activities with EPA, State, local and other Federal
and private laboratory and field operations. MMB develops quality assurance and data
management plans and agreements with State and local agencies; plans and develops
quality assurance and data management programs for activities of all Agency programs
carried out in the Region.
MMB also carries out reviews of data quality objectives, QAPjPs and standard operating
procedures; provides management and technical systems audits and evaluations of
data quality; and operates a proficiency testing program and acts as a focal point for
EPA methodology requirements and quality assurance services. In addition, the
monitoring management function maintains an inventory of all monitoring projects in the
Region, provides the data necessary for the Environmental Status Reports and reviews
all State grantee outputs for monitoring activity.
t
The Surveillance and Monitoring Branch is responsible for the collection and evaluation
of environmental data in all Agency monitoring and enforcement programs. SMB
conducts investigations and studies of surface and ground water and air quality, RCRA
-------
regulated facilities and industrial and municipal waste site field investigations and
studies in support of Superfund remedial, removal, enforcement, and emergency
response activities. 8MB provides technical assistance to municipal treatment plants to
improve operating procedures.
.C. Superfund Contractor Services - Federal Contractors are major participants in
Superfund environmental monitoring and data collection. Contractors provide support
for removal, pre-remedial, and remedial activities. All contractors must implement a
QAPP which will ensure data of known and adequate quality to meet the requirements
of their Statements of Work (SOW). The development and implementation of a QAPP
and individual QAPj'Ps also include auditing and corrective actions within which
contractor auditors report to contractor corporate management. The following are brief
descriptions of Superfund contractor activities.
Technical Enforcement Support (TES) - provides technical support such as health
endangerment assessments, hydrological/geological studies and other special studies
in support of litigation and negotiations. TES collects very little environmental data.
Future TES contracts are expected to provide more collection of environmental data,
including oversight of responsible party actions. Contract oversight is provided by
ERRD.
TES V - Camp Dresser and McKee Incorporated
TES VI - Alliance Technologies Corporation
Environmental Services Assistance Team (ESAT) - provides analytical support, data
review, logistical and administrative support, QA/QC support, and management and
reporting services to ESD. R.F. Weston is the ESAT contractor/Oversight of ESAT
contractors is provided by TSB/ESD.
Emergency Response Cleanup Services (ERCS) - provides implementation support to
the removal program. ERCS contractors supply all personnel, material, and equipment
for conducting removal operations. Mini-ERCS provide the same services for projects
with smaller dollar amounts. OH Materials is the principal ERCS contractor. Oversight
of ERCS contractors is done by the RPB/ERRD.
Technical Assistance Team (TAT) - provides technical and management support to
EPA during removal actions. TAT activities include site assessments, sampling and
monitoring, documenting project costs, QA, data management and reporting,
enforcement support, community relations, and contingency planning. Roy F. Weston
is the principal TAT contractor at this time, and is overseen by RPB/ERRD.
Field Investigation Team (FIT) - provides technical services in support of preliminary
site investigations and a variety of special studies at remedial sites. NUS Corporation is
the principal FIT contractor. FIT contractors are overseen by the SMB/ESD, especially
through the RSCC.
-------
Remedial Planning Team (REM) - provides support to enforcement and remedial
response activities. REM services encompass remedial investigations, data
management, engineering feasibility studies, technical support, and oversight of
responsible parties, remedial design and planning and implementation activities.
Contract oversight is provided by the PSB/ERRD. REM contractors are divided by
region as below:
REM I - NUS Corporation (expired 9/86)
REM II - Camp, Dresser and McKee, Inc. (expired 6/89)
REM III - Ebasco Services, Inc.
REM IV - CH2M Hill, Inc.
REM V - Williams, Russell, and Johnson, Inc.
REM VI - Peer Consultants
Alternate Remedial Contracting Strategy (ARCS). Provides support to enforcement and
remedial response activities from the RI/FS stage to the remedial designs and
implementation activities. Contract oversight is provided by the PSB/ERRD. ARCS
contractors are divided by the amount of work designated in each contract:
Major
ARCS Ebasco Services, Incorporated
ARCS Camp Dresser and McKee Incorporated
ARCS Malcom Pirnie
Minor
ARCS TAMS Consulting
ARCS ICF
ARCS R.F. Weston
-------
III. Data Quality Objectives
A. Definition - "Data Quality Objectives (DQOs) are qualitative and quantitative
statements which specify the quality of data required to support Agency decisions
during remedial response activities. DQOs are determined based on the end uses of
the data to be collected." (Guidance on Data Quality Objectives, Quality Assurance
Management Staff, September 25,1984) Therefore, a unique set of DQOs must be
developed for each site and integrated into the project planning process for each data
collection activity. DQOs apply to fund lead, federal or state enforcement lead, and
potentially responsible party lead projects especially during the remedial investigation
(Rl), but should also be applied to some degree during feasibility studies (FS), remedial
designs (RD), and remedial actions (RA).
B. Roles and Responsibilities of Key Personnel - Decision Maker (ERRD Program
Director) - The decision maker is the individual responsible for deciding on the remedial
design and remedial action to be taken for NPL sites, and deciding on the "level of risk
DQO" needed.
Data User (ERRD Regional Project Manager) - The data user is the individual
responsible for deciding on the amount and quality of data needed to determine degree
and extent of contamination, the remedial design, and the remedial action to be taken.
The data user is responsible for developing DQOs for each decision to be made.
Quality Assurance Officer (Regional within MMB) - The Quality Assurance Officer
(QAO) advises and assists on the data collection design. The QAO prescribes the
appropriate QA/QC measurements employed to determine the quality of data needed
for the decision to be made.
C. DQO Guidance - Guidance on the development of DQO statements can be found in
two documents:
1. Data Quality Objectives for Remedial Response Activities - Development
Process, EPA/540/G-87/003, March 1987.
2. Data Quality Objectives for Remedial Response Activities - Example
Scenario:RI/FS Activities at a Site with Contaminated Soils and Ground Water,
EPA/540/G-87/004, March 1987.
-------
IV. Preparation of Quality Assurance Project Plans (QAPjP)
A. Definition - A QAPjP presents in specific terms the policies, organization objectives,
functional activities, and specific QA/QC activities designed to achieve the data quality
goals or objectives of a specific project or continuing operation. All removal,
pre-remedial, and remedial contractors involved in environmental monitoring or
sampling activities must develop QAPjPs. QAPjPs should be prepared according to
"Interim Guidelines and Specifications for Preparing QA Project Plans", QAMS-005/80,
December 29, 1980, and "Guidance For Preparation of Combined Work/Quality
Assurance Project Plans for Water Monitoring", OWRS-1, May 1984, provided in
Appendix I. In order to save duplication of effort, it is allowable for QAPjPs to be
incorporated into Project/Site Operations Plans (Plan).
1. Site-specific QA Project Plans - A site-specific QAPjP is required for each site
which requires a focused approach that is unique to the site
(e.g. - number of monitoring wells, monitoring of waste lagoons, waste piles,
or metal drums containing waste).
2. Generic QA Project Plans - When a group of monitoring projects are carried
out for the same purpose and with the same personnel and/or procedures,
a "generic" QAPjP may be used to address this group of projects. Generic QAPjPs for
Superfund activities may be developed only for site investigations and emergency
removals.
B. Procedures for Preparation, Submittal, and Approval of QAPjPs
i
1. Federal-Lead Sites - RPMs are responsible for ensuring that EPA contractors
prepare QAPjPs for site investigations, remedial investigations/feasibility studies
(RI/FS), remedial design (RD), remedial .actions (RA), and emergency removals.
QAPjPs are first submitted to the RPM for review according to guidance criteria. Once
the RPM is satisfied that all QA elements have been addressed, the plan is submitted to
the Chief, THWS, MMB/ESD for review and approval at least 30 days prior to
commencement of sampling. THWS will approve or request changes in writing within
15 workdays of receipt. It is important to keep in mind that QAPjPs must be approved
by MMB before sampling may begin. Once MMB approves a plan, the QAO assigned
to the project notifies the RSCC and RPM of approval and submits the approved
analytical methods and parameter list to the RSCC to facilitate the CLP sample booking
procedures. The RSCC will not schedule CLP analyses until they receive this approval
memo; and the RSCC must be given appropriate lead time in order to obtain laboratory
space for sample bookings. Lead time is discussed in Section VI of Part I.
2. State-Lead Sites (Remedial and Enforcement) - State project managers are
responsible for ensuring that their contractors prepare QAPjPs. These QAPjPs are
submitted to the state Superfund QAO for review and approval. Copies of the final,
approved QAPjPs and .the review comments are kept on file with the state and reviewed
periodically by MMB/ESD.
-------
3. Federal Facility Sites - These include the Department of Defense, U.S. Army
Toxic and Hazardous Materials Agency, the Department of Energy, the Department of
the Interior, and the United States Army Corps of Engineers. RPMs are responsible for
ensuring that federal facilities prepare QAPjPs which follow the guidance described in
this Manual for Responsible Party Sites. The QAPjP produced is submitted through the
PSB to MMB for review and approval. Federal agencies are required under Section
120 of Superfund Amendments and Reauthorization Act (SARA) to comply with
CERCLA both procedurally and substantively. This includes adherence to guidance
and provisions set forth in the National Contingency Plan (NCP). Section 300.68 k(2) of
the NCP states: "In fund-financed actions or actions under CERCLA Section 106, the
quality assurance/site sampling plan must be reviewed and approved by the Remedial
Project Manager with a coordination signature from the Quality Assurance Officer."
4. Responsible Party Sites - With the enactment of the SARA, Congress gave EPA
specific authority to settle CERCLA actions, and it gave specific instructions respecting
the type of settlement, administrative or judicial, to be used for each kind of response
activity, as follows:
a) cost recovery actions may be settled either with a judicial consent decree or an
administrative order.
b) abatement actions (abatement of a "danger" or "threat") must be settled
through judicial consent decrees.
c) actions (removal actions and investigative activities, including Rl's and FS's)
may be settled with the use of either judicial consent decrees or administrative orders.
- Administrative Orders - Administrative orders (AO) are issued either by consent of the
responsible parties ("administrative orders on consent") after negotiation of the terms,
or on a mandatory basis by the Agency ("unilateral orders"). Negotiations are
conducted between EPA and the potentially responsible parties (PRPs), within the
jurisdiction of the Agency. Under administrative orders, the RPMs are responsible for
ensuring that the PRP or their contractors prepare QAPjPs. The RPM reviews this
document, and when satisfied that all QA elements have been addressed, submits it to
the Chief of THWS for review and approval at least 30 days prior to commencement of
sampling. THWS will approve the QAPjP or request changes in writing within 15
workdays. ,
- Judicial Consent Decrees - Consent decrees differ from administrative orders, in that
the monitoring requirements for the QAPjPs are "negotiated" by EPA, the Department
of Justice, and the PRP under the authority of a federal district court, usually in the
jurisdiction in which the Superfund site is located. For this reason it is important for the
RPM to request assistance from the ESD at negotiation sessions on technical
requirements for monitoring and quality assurance. The RPM is responsible for
ensuring that the PRP submits a QAPjP that addresses the available guidance and the
recommendations of the ESD technical negotiators.
-------
- EPA Oversight of Responsible Parties - If EPA oversight contractors take split
samples from a PRP's contractor, procedures for split sampling and analysis must be
prepared and submitted to MMB in a Short Form for approval. The Short Form is
discussed in Part II of this Manual.
C. Guidance Documents for the Preparation of QAPjPs - QAPj'Ps should be prepared
according to "Interim Guidelines and Specification For Preparing Quality Assurance
Project Plans", QAMS-005/80, December 29, 1980, or "Guidance For Preparation of
Combined Work/Quality Assurance Project Plans for Environmental Monitoring",
OWRS, May, 1984, provided in Appendix I, and Part II of this Manual. A QAPjP can be
combined into another document, such as a Project Operations Plan, Site Operations
Plan, or Field Sampling Plan (generally referred to as the Plan) to avoid duplication.
This is the preferred practice.
D. QAPjP Pre-Development Meetings - In order to reduce the time needed to resolve
deficiencies in QAPjPs, up-front meetings are held between personnel responsible for
preparing QAPjPs (RPMs, contractors, PRPs, consultants) and EPA technical
personnel responsible for monitoring and quality assurance. Pre-development meetings
should be convened prior to QAPjP submittal and at least 60 days prior to sampling.
-------
V. Audit Programs
In general, QA audits will be performed by MMB staff. The various types of audits
conducted by the Region are defined as follows.
A. Management Systems Audit (MSA) - An MSA is generally a QAPP review, checking
managerial implementation of the approved QAPP. It evaluates the unit's QA
organization and its activities, as well as the degree of management support afforded
the program. Specifically, the audit covers the following:
- QAPjP development and approval,
- DQO development (where implemented),
- Standard Operating Procedure (SOP) development and approval,
- audit schedules and procedures,
- tracking systems for QA activities and corrective actions,
- managerial support, including financial and resource support.and
- QA responsibilities of personnel including Project Managers,
field and laboratory staff, and QA staff.
B. Technical Systems Audit (TSA) - A TSA is generally an on-site audit of
project-specific monitoring activities, either field, laboratory or both. It focuses on actual
quality control activities of environmental data collection systems, and uses the
approved QAPjP as a reference. TSAs are conducted on intramural projects and
extramural monitoring which the Region supports or requires.
Specific activities vary with the scope of the audit, but can include review of:
- sample collection and analytical activities,
- equipment calibration techniques and records, .
- decontamination and equipment cleaning,
- equipment suitability and maintenance/repair,
- background and training of personnel,
- laboratory control charts and support systems,
- QC samples such as duplicates, trip and field blanks, method
blanks, unknown performance evaluation samples,
-sample containers, preservation techniques, and chain-of-custody,
- data logs, data transfer, data reduction and data validation,and
- monitor siting.
EPA REM, ARCS, and FIT contractors are required by contract to conduct TSAs of their
field activities.
C. Data Quality Audit (DQA) - A DQA focuses on collected data by evaluating whether
sufficient information exists to support the assessment of the data's quality. It will
determine if the data set undergoes any validation procedure by the collector or data
user to establish whether the data can be used to support the decision making process.
-------
If DQOs exist, the'data set(s) will be evaluated against them. If deficiencies are found,
DQAs will be able to determine the causes, whether technical, managerial or both.
DQAs are to be distinguished from data validation activities which are considered to be
part of the data generation process. In Region II, DQAs are not normally conducted
alone, but rather are conducted as part of TSAs.
D. Performance Evaluations (PE) - A PE is a means of evaluating laboratory
performance with a sample of specific contaminants in an appropriate matrix. The true
values of performance evaluation components are unknown to the laboratory analyst.
In some cases, PEs are performed on field operations that also perform analytical
functions. This includes, for example, audits of ambient air monitoring operations in
which cylinders of gas mixtures or audit devices are sent to field operators to check the
calibration/operation of the monitors.
On an as-needed basis, MMB arranges for PEs to be submitted to all laboratories
providing CERCLA analytical services. PE analyses are required for laboratories which
participate in CLP every three months. PEs are also required for all non-CLP
laboratories which have not analyzed PEs within the last six months. MMB sends the
results of their PE audits to ERRD project managers along with their recommendation
of capability. See Part II Section XII for further information.
-------
VI. Superfund Contract Laboratory Program
The majority of Superfund sample analyses are performed within the Superfund
Contract Lab Program (CLP). The program was created to handle the excess analytical
demand that could not be handled by the regional laboratories.
Each Region has a Regional Sample Control Center (RSCC). In Region II, the RSCC
resides in the SMB/ESD and can be reached at FTS 340-6705. The RSCC schedules
sample analyses with CLP laboratories through the Sample Management Office (SMO)
in Alexandria, VA. Each contractor is assigned a contact who may book samples
through the RSCC. The contractor is responsible for ensuring that their subcontractors
understand how to properly use the CLP. Primary contractors may not refer their
subcontractors to the RSCC for guidance or to schedule samples. The December 1988
"User's Guide to The Contract Laboratory Program" (available through the RSCC)
contains a thorough discussion of the CLP program.
After a QAPjP or Plan and all SAS requests have been approved by MMB, samples
may be booked with the RSCC. The required lead time for all requests is :
1. Routine Analytical Services (RAS)- Verbal requests for RAS must be submitted to
the RSCC on the Tuesday two weeks prior to the scheduled date of sample collection.
2. Special Analytical Services (SAS)- SAS requests must be submitted to the RSCC,
in writing, by the Tuesday four weeks prior to the scheduled date of sample collection.
(In instances of restricted funding longer lead time may be required.)
3. Exception SAS Requests- Exception SAS requests require a four to five week
lead time. SMO must obtain program approval through Headquarters before soliciting
laboratories.
Examples of exception SAS requests are the following:
a. All non-Superfund requests.
b. Non-invitation for bid (IFB) parameters or compounds requested for the first time.
c. All requests requiring use of non-approved analytical procedures (methods which
are not routinely requested through CLP).
d. All requests specifying specific remedial parameters for waste samples (for
example, compatibility, grain size distribution, permeability, viscosity, Atterburg limits,
etc.)
e. Requests which will incur a cost greater than $25,000.
-------
In addition, the CLP Deputy Project Officer (DPO), who resides in MMB, is available to
resolve technical problems on sample analysis or contract compliance. The DPO may
also be contacted with questions on methodology.
-------
VII. Data Validation / Data Useability
A. Data Validation - It is the policy of Region II that all data generated be validated.
Data validation is a systematic process for reviewing a body of data against a set of
criteria to provide assurance that the data are adequate for their intended use. Data
validation consists of data editing, screening, checking, auditing, verifying, certifying,
and reviewing. Data are validated according to Region II standard operating
procedures (SOPs). The Region II SOPs are based on National CLP Functional
Guidelines for CLP Data Review. At this time data validation is carried out by both EPA
and contractor personnel.
In the case of data generated by the Contract Laboratory Program, data packages
(identified by a specific case number) are sent to the RSCC from the CLP laboratories
that performed the analyses. If the package is complete, the RSCC forwards it to the
data validators. The package is sent back to the RSCC after validation, whereupon it is
sent to the data requestor (RPM or EPA contractor).
All non-CLP data, such as PRP generated data, must be validated. PRP contracted
laboratories or outside parties must validate their data according to Region II SOPs.
Project Managers may request MMB to validate PRP data if it is deemed necessary. To
do this, the data should be sent to the Toxic and Hazardous Waste Section of MMB. In
addition, all the CLP required deliverable information must accompany the data, so that
MMB can adequately evaluate it. See Section XIV of Part II for more information.
B. Data Useability - In addition to reviewer/RPM interactions data useability meetings
can be arranged with the MMB. These meetings can assist RPMs in the use of data for
other than enforcement or risk assessment purposes. Data reviewers are required to
reject data with, questionable quantitative value. However, even though the true value is
unknown, the data may still be of such value as to allow the user to make decisions
based on maximum or minimum possible concentration.
-------
ACRONYM LIST
AWS - Air and Water Section
AO - Administrative Order
ARCS - Alternate Remedial Contracting Strategy
CERCLA- Comprehensive Environmental Response Compensation and Liability
Act
CFR - Code of Federal Regulations
CLP - Contract Laboratory Program
DPO - Deputy Project Officer
DQA - Data Quality Audit
DQO - Data Quality Objective
ERGS - Emergency Response Clean-up Services
ERRD - Emergency and Remedial Response Division
ERT - Emergency Response Team
ESAT - Environmental Services Assistance Team
ESD - Environmental Services Division
FIFRA - Federal Insecticide, Fungicide and Rodenticide Act
FIT - Field Investigation Team
FOP - Field Operations Plan
FR - Federal Register
FS - Feasibility Study
MMB - Monitoring Management Branch
MSA - Management Systems Audit
NCP - National Contingency Plan
NPL - National Priority List
NJRAB - New Jersey Remedial Action Branch
NY/CRAB- New York/Caribbean Remedial Action Branch
OSC - On-Scene Coordinator
PCB - Polychlorinated Biphenyl
PE - Performance Evaluation
POP - Project Operations Plan
PRP - Potentially Responsible Party
PSB - Public Support Branch
PTSB - Pesticides and Toxic Substances Branch
QA - Quality Assurance
QAO - Quality Assurance Officer
QAPjP - Quality Assurance Project Plan
QAPP - Quality Assurance Program Plan
QA/QC - Quality Assurance/Quality Control
RA - Remedial Action
RD - Remedial Design
REM - Remedial Contractor Program
Rl - Remedial Investigation
RPB - Response and Prevention Branch
RPM - Regional Project Managers
-------
RSCC - Regional Sample Control Center
SAP - Sampling and Analysis Plan
SARA - Superfund Amendments and Reauthorization Act
SCB - Site Compliance Branch
SI - Site Investigation
SMB - Surveillance and Monitoring Branch
SMO - Sample Management Office
SOP - Standard Operating Procedure
SOW - Statement of Work
SPCC - Spill Prevention, Countermeasure and Control Program
TAT - Technical Assistance Team
TES - Technical Enforcement Support
THWS - Toxic and Hazardous Waste Section
TSA - Technical Systems Audit
TSB - Technical Services Branch
TSCA - Toxic Substances Control Act
-------
PART II
Quality Control Handbook for CERCLA Sampling and Analysis
-------
I. SAMPLING DESIGN AND STRATEGY
A. Sampling Plan Components
Detail of sampling and analysis is a necessary part of each field/site/project operations
plan, sampling and analysis plan, or quality assurance project plan (from hereon
referred to as the Plan) in order to ensure uniform and acceptable sampling and
analytical protocol for each project. The Plan describes the objectives and details how
the individual tasks of a sampling and analytical effort will be performed. The Plan must
discuss the following topics at a minimum, and it must include a Parameter Table, an
example of which is included in Appendix IV. For EPA oversight contractors, a
completed "Combined Work/QA Short Form" is the only required documentation that
must be submitted for review and approval. An example of the "Short Form" is attached
in Appendix I. Also, when booking samples through CLP SAS, a copy of the SAS
Request Form must be submitted to MMB for review and approval, along with the Plan.
All analytical methodology and special analytical requests should be incorporated by
reference and/or attached to the plan.
* Objectives of sampling design and selection of representative sampling sites.
Discussion of site history and sampling design rationale must be provided, so that
reviewers of the Plan have the necessary information. The discussion should include
topics such as the history of the contamination, the matrices involved, the dimensions of
the site, etc.
* Sampling Design
* Selection of Parameters to be measured.
Parameters to be measured are usually dictated by the purpose of an investigation and
should be based on knowledge of the problem being investigated. An in-depth
discussion of parameter selection is out of the scope of this document as it is a process
requiring much background and expertise in dealing with hydrogeologic systems,
chemistry, and engineering. No criteria for parameter selection can be put forth in the
format of a standard operating procedure.
* Selection and preparation of sampling equipment.
* Sampling equipment construction materials.
* Required sample volumes.
N.
* Selection and preparation of sample containers.
* Sample collection and handling.
-------
* Sample preservation.
* Sample chain-of-custody and identification.
* Use of field instrumentation.
* Field quality control samples.
* Choice of laboratories and validation of data.
These topics are further discussed within this Manual.
B. Purpose and Objective of Sampling
The basic objective of any sampling campaign is to collect a sample which is
representative of the media under investigation. More specifically, the purpose of
sampling at hazardous waste sites is to acquire information that will aid investigators in
determining the presence and identity of on-site contaminants and the extent to which
they have become integrated into the surrounding environment. This information can
then be used as support for future litigations or as input to remedial investigations and
risk assessments.
The validity of environmental data is dependent in part on the integrity of the field
procedures employed in obtaining a sample. Proper sampling techniques must be
employed to obtain a sample which is representative of the area or container of interest.
A sample is representative if it possesses the same qualities or properties as the
material under consideration. Due to the complexity of most hazardous substances and
site conditions, no universal sampling methods can be recommended. Procedures must
be adapted for use in various matrices and site-specific restrictions.
C. Types of Samples
Before defining the general sample types, the nature of the media or materials under
investigation must be discussed. Materials can be described as homogeneous or
heterogeneous. Homogeneous materials are generally defined as having uniform
composition throughout. In this case, any sample increment can be considered
representative of the material. On the other hand, heterogeneous samples present
problems to the sampler because of changes in composition of the material over
distance and time.
When discussing types of samples, it is important to distinguish between the type of
media to be sampled and the sampling technique that yields a specific type of sample.
In relation to the media to be sampled, two basic types of samples can be considered:
the environmental sample and the hazardous sample.
-------
1. Environmental Samples
Environmental samples (ambient air, soils, surface water, groundwater, sediment or
biota) are generally dilute (in terms of pollutant concentration) and usually do not
require the special handling procedures used for concentrated wastes. However, in
certain instances, environmental samples can contain elevated concentrations of
pollutants and in such cases would have to be handled as hazardous samples.
2. Hazardous Samples
Hazardous or concentrated samples are those collected from drums, tanks, lagoons,
pits, waste piles, fresh spills, etc., and require special handling procedures because of
their potential toxicity or hazard. These samples can be further subdivided based on
their degree of hazard; however, care should be taken when handling any wastes
believed to be concentrated, regardless of the degree.
D. Types of Measurement
In general, two basic types of sample measurements are recognized, both of which can
be used for either environmental or concentrated samples. They are: 1) samples which
are collected and subsequently analyzed in the laboratory and, 2) samples which are
analyzed in-situ.
1. Laboratory Measurement
There are two types of samples which are collected and analyzed in a laboratory. These
are grab samples and composite samples.
/ '
a. Grab Samples
A grab sample is defined as a single sample representative of a specific location at a
given point in time. The sample is collected all at once and at one particular point in the
sample medium. The representativeness of such samples is defined by the nature of
the materials being sampled. In general, as sources vary over time and distance, the
representativeness of grab samples will decrease.
b. Composite Samples
Composites are combinations of more than one sample collected at various sampling
locations and/or different points in time. Analysis of composite yields an average value
and can, in certain instances, be used as an alternative to analyzing a number of
individual grab samples and calculating an average value. It should be noted, however,
that compositing can mask problems by diluting isolated concentrations of some
hazardous compounds below detection limits.
For sampling situations involving hazardous wastes, grab sampling techniques are
-------
generally preferred because grab sampling minimizes the amount of time sampling
personnel must be in contact with the wastes, reduces risks associated with
compositing unknowns, and eliminates chemical changes that might occur due to
compositing. Compositing is often used for environmental samples, including dioxin
samples, to determine vertical or horizontal spatial variability of parameters. This
procedure provides data that can be useful by providing an average concentration over
a number of locations and can serve to keep analytical costs down; however, it is
important to understand that sensitivity is sacrificed when samples are composited due
to dilution of individual grab samples. If contamination occurs in "hot" spots on site and
"hot" grabs are composited with clean samples, a true vertical or horizontal distribution
of contamination will appear to be a uniform distribution at a level lower than the true
value of any one individual component^). This is especially a concern when doing
dioxin sampling with an action level of 1 ppb. Compositing is further discussed in
Section IX.
2. In-Situ Measurement
In-situ measurements are made on samples in the environment. Measurements for pH,
conductivity, and temperature must be taken in the field. Use of instrumentation such as
an OVA, HNu and other gas analyzers is also considered in-situ measurement.
-------
II. ANALYTICAL METHODS, PRESERVATION AND HOLDING TIMES
A. Methodology Available for use in the CERCLA Program
The CERCLA program has no legally mandated analytical methods. Methods from
other programs or methods which are proven to be scientifically valid can be used for
CERCLA work. The bulk of the analytical methods used presently are, however, from
the Contract Laboratory Program (CLP).
The Contract Laboratory Program supports the Agency's Superfund effort by providing
a range of chemical analytical services on a high volume, cost effective basis. Its
purpose is to provide legally defensible analytical data. The program is managed by the
National Program Office in Headquarters, and the Contractor-operated Sample
Management Office receives the analytical requests from the Regions and coordinates
and schedules sample analyses. Analytical Statements of Work exist for organics,
inorganics and dioxin in water and soil\sediment matrices.
In addition to standardized analyses provided under the Routine Analytical Services
(RAS) program, the CLP's Special Analytical Services (SAS) program provides clients
with limited customized or specialized analyses, different from or beyond the scope of
the RAS contract protocols. Services provided by SAS include: quick turn around
analyses, verification analyses, analyses requiring lower detection limits than RAS
methods provide, identification and quantification of non-Target Compound List (TCL)
constituents, general waste characterizations, analysis of non-standard matrices, and
other specific analyses. Consult the "User's Guide to the Contract Laboratory Program",
December 1988 for further information.
As stated above, the CLP parameters of interest for RAS were titled under the 10/86
Statement of Work (SOW) the 'Target Compound List". Under previous SOWs the TCL
was titled the "Hazardous Substance List". Neither of these lists has been published in
the Federal Register (FR) or Code of Federal Regulations (CFR) and thus are strictly
lists defined by contract.
The "Priority Pollutant List" was established in the consent decree of the Natural
Resources Defense Council (NRDC) vs. Train, in 1976. Although not published in the
Federal Register under that title, it was published in a more generalized form in 44 FR
44502, July 30,1979 as the "Toxic Pollutants" list under the Federal Water Pollution
Control Act, and was amended in 46 FR 2266, January 8,1981, and 46 FR 10724,
February 4,1981. Thus, the "Priority Pollutant" list as it now stands is comprised of 126,
compounds or elements.
The Priority Pollutant and Target Compound Lists are presented in Appendix II which
shows the differences between the two. They are not interchangeable; neither list is a
subset of the other, both contain compounds not found on the other. The only Priority
Pollutants which are not on TCL are: asbestos, benzidine, 1,2-diphenylhydrazine, N-
nitrosodimethylamine, endrin aldehyde, 2-chloroethyl vinyl ether, acrolein, and
-------
acrylonitrile. Refer to the Appendix for those compounds which are on TCL but not
considered Priority Pollutants.
A discussion of available references, methods, holding times and preservatives follows.
B. Analytical References
The following is a listing of selected analytical references containing methods available
for use in the CERCLA program.
1. Aqueous/solid matrices
a. 40 CFR Part 136, as updated yearly.
b. 7/88 CLP Statement of Work for Inorganics, the 9/87 Statement of Work for
Dioxin, the 2/88 revision of the Organics Statement of Work, and as updated.
The CLP standardized organic analytical methods are based on the CFR
methods 608, 613, 624, and 625 modified for use in the analysis of both water
and soil matrices. The standardized inorganic analytical methods are based on FR
methods, EPA Methods for Chemical Analysis of Water and Wastes (MCAWW),
and Test Methods for Evaluating Solid Waste (SW-846), Third Edition for the
analysis of water and soil matrices. Appendix II provides a listing of the CLP organic
and inorganic Target Compound/Analyte Lists as taken from the most recent
SOWs and includes the RAS detection limits.
The dioxin Routine Analytical Services (RAS) contract method determines the
presence of the 2,3,7,8-tetrachloro-dibenzo-p-dioxin isomer in water and
soil/sediment matrices.
c. Standard Methods, 15th and 16th eds., or as revised.
d. Methods for Chemical Analysis of Water and Wastes(MCAWW), Revised 1983,
EPA 600/4-79-020.
e. American Society for Testing and Materials.
f. Test Methods for Evaluating Solid Waste-SW-846, Third Edition, November
1986.
g. Procedures for Handling and Chemical Analysis of Sediment and Water
Samples, May 1981, Technical Report CE/81-1, NTIS#AD-A103788.
h. Methods for the Determination of Organic Compounds in Drinking Water
(500 Series Methods), December 1988, EPA 600/4-88/039.
2. Air
-------
Detailed analytical procedures for air analyses must be provided with any QAPj'P.
These procedures must accompany the SAS request and the air samples to the
laboratory.
The following references contain the most commonly used lab methods for measuring
various air contaminants. Whenever practicable EPA methods should be used for air
analyses. However if EPA methods do not exist for the compounds of interest, other
common methods, eg. NIOSH methods, may be specified provided that they have been
validated at the detection limits of interest to the project.
Air Monitoring References
a. Riqgin. R.M. Technical Assistance Document for Sampling and Analysis
of Toxic Organic Compounds in Ambient Air. EPA-600/4-83-027. U.S. Environmental
Protection Agency, Research Triangle Park, NC, 1983.
b. NIOSH Manual of Analytical Methods. Volumes 1-2, 3rd edition.
P. Eller, ed. National Institute for Occupational Safety and Health, Cincinnati, OH, 1984.
c. Riqqin. R.M. Compendium of Methods for the Determination of Toxic Organic
Compounds in Ambient Air. EPA-60074-89-017. U.S. EPA, Research Triangle Park,
NC, 1988.
i
d. 40 CFR Part 50 Appendices'A-J.
e. SOP for the GC/MS Determination of Volatile Organic Compounds Collected on
Tenax, June 1984. EMSL/RTP-SOP-EMD-020. .
3. Non-Aqueous Phase Liquids
a. Test Methods for Evaluating Solid Waste, SW-846, Third Edition, November
1986.
b. Interim Methods for the Measurement of Organic Priority Pollutants in Sludges,
Revised Draft June 1980.
c. Determination of Polychlorinated Biphenyls in Transformer Fluid and Waste
Oils, Sept. 1982. EPA 600/4-81-045.
C. Preservation, Methodology and Holding Times
-------
Complete and unequivocal preservation of samples is a practical impossibility.
Regardless of the nature of the sample, complete stability for every constituent can
never be achieved. At best, preservation techniques can only retard the chemical and
biological changes that inevitably continue after the sample is removed from the parent
source. The changes that take place in a sample are either chemical or biological. In
the former case, certain changes occur in the chemical structure of the constituents that
are a function of physical conditions. Metal cations may precipitate as hydroxides or
form complexes with other constituents; cations or anions may change valence states
under certain reducing or oxidizing conditions; other constituents may dissolve or
volatilize with the passage of time. Metal cations such as iron and lead may also adsorb
onto surfaces (glass, plastic, quartz, etc.). Biological changes taking place in a sample
may change the valence of an element or a radical. Soluble constituents may be
converted to organically bound materials in cell structures, or cell lysis.may result in
release of cellular material into solution. The well known nitrogen and phosphorus
cycles are examples of biological influence on sample composition.
Methods of preservation are relatively limited and are intended generally to (1) retard
biological action, (2) retard hydrolysis of chemical compounds and complexes, (3)
reduce volatility of constituents, and (4) reduce absorption effects. Preservation
methods are generally limited to pH control, chemical addition, refrigeration and
freezing(8).
Appendix IV contains a copy of Table 2 from 40 CFR Part 136, July 1,1987 and a
tabular presentation of the CLP holding time and preservation requirements. The Table
is comprised of approved "conventional" parameter (meaning those analyses not
considered part of the most commonly used "organic" and "inorganic" sets of analyses)
methodology for an aqueous matrix. The holding times and preservation requirements
of these conventional parameters are to be followed.
Appendix IV also includes two tables of holding time and preservation requirements.
One table applies to the Contract Laboratory Program's contractual stipulations and the
requirements listed must be adhered to when using the CLP. Holding times begin at
the Verified Time of Sample Receipt (VTSR) when the Contract Laboratory Program
has been engaged. The second table contains requirements for all analyses which do
not entail utilizing the CLP. Holding times on this table begin on the date of sample
collection.
Various studies including a project funded by EPA and the Department of Defense and
performed by the Oak Ridge Laboratory determined that aqueous volatile organic
-------
samples (VOAs) can be held for extended periods of time with preservation using
hydrochloric acid (HCI) to a pH less than 2 without significant loss of constituents. It is
now a requirement when using CLP that all samples taken for volatile organics analysis
be preserved with hydrochloric acid to a pH less than 2. When not using CLP, if
non-preservation is chosen, then samples must be analyzed within 7 days of the
sampling date. If preservation methods are chosen, then samples must be preserved
as for CLP, ie. acidify with HCI to pH<2 as per the procedure below.
The following procedure, adapted from the drinking water methods should be used for
acidification of volatile organic samples with HCI to a pH less than 2.
Adjust the pH of the sample to <2 by carefully adding 1:1 HCI drop
by drop to the required 2 (40 ml) VOA sample vials. The number of
drops of 1:1 HCI required should be determined on a third portion
of sample water of equal volume.
It should be noted that if acidification of the sample causes effervescence, the sample
should be submitted without preservation except for cooling to 4 degrees C. This
sample property should be appropriately noted when present. When adding sodium
thiosulfate to samples containing residual chlorine, the thiosulfate should be added to
the vial prior to addition of the sample followed by addition of HCI. The 1:1 HCI solution
should be made up with concentrated HCI (12N) and demonstrated analyte-free
deionized water.
The pH testing and the determination of the appropriate volume of acid required to
achieve pH<2 must be performed at each groundwater monitoring well prior to sampling
unless monitoring is performed on the same wells on a continuous basis.
When samples are to be iced to 4 degrees C., it is intended that the sample bottle be
surrounded by bags of ice or by ample packets of "Blue Ice" to ensure that the proper
temperature is achieved and maintained during transport. It is not acceptable to put
bags of ice around only the necks of the bottle or to use only a few packets of "Blue Ice"
since these techniques do not ensure the attainment of the proper temperature. All
samples must be shipped to the lab within 24 hours from the time of collection. Further
information can be found in the User's Guide to the CLP of December 1988.
When booking samples through CLP SAS, a copy of the SAS Client Request must be
attached to the Plan.
D. QC Criteria for 40CFR Part 136 and SW-846 Third Edition Methods
If the Contract Laboratory Program is used for the analysis of target compound list
-------
(TCL) analytes, the laboratory analyzing the samples will perform specific quality control
procedures to assure that the data is valid. However, Superfund PRP and RCRA
analyses will usually be performed outside of the CLP program. The methods listed in
40 CFR Part 136 and SW-846 Third Edition do not specify detailed quality control
procedures, so data analyzed under those methods may not be amenable to validation.
In order to use 40 CFR Part 136 or SW-846 Third Edition methods for the analysis of
TCL compounds, the Plan must specify proposed quality control criteria for spikes,
blanks, surrogates, detection limits, and internal standards. We strongly recommend
that CLP procedures be used to analyze all TCL compounds.
When SW-846 Third Edition methods are used, the generic QC procedures listed in
Chapter One, and the specific QC procedures listed in Section Eight of each method
must be followed. For the analysis of TCL analytes, the lab may use the CLP criteria
for surrogates, internal standards, pesticide linearity and retention time shift shown on
the Chapter One QC forms, or the lab may propose their own QC criteria based on
SW-846 protocol. When non-TCL analytes are analyzed by SW-846 methods, the lab
must propose all QC criteria before samples are analyzed.
These proposals must be submitted to MMB in the Quality Assurance Project Plan for
review and approval, prior to the commencement of the sampling event.
-------
I. DOCUMENTATION PROCEDURES
A. Chain-of-Custody
1. Definition and Reference
According to the USEPA Office of Enforcement and Compliance Monitoring National
Enforcement Investigations Center (NEIC) Policies and Procedures, May 1978 revised
May 1986, a sample is under custody if:
1. it is in your possession, or
2. it is in your view, after being in your possession, or
3. it was in your possession and you locked it up, or
4. it is in a designated secure area.
Possession must be traceable from the time the samples are collected.
2. Recordkeeping and Procedures
a. General
The method of sample identification utilized depends on the type of sample collected.
In-situ field analyses are those conducted for specific field analyses or measurements
where the data are recorded directly in bound field logbooks or recorded directly on the
chain-of-custody record, with identifying information, while in the custody'of the
sampling team. Examples of such in-situ field measurements and analyses include pH,
temperature, and conductivity. Also included in this category are those field
measurements or analyses such as flow measurements, geophysical measurements,
surveying measurements, etc. that are made with field instruments or analyzers, where
no sample is actually collected.
Samples, other than those collected for in-situ field measurements or analyses, are
identified by using a standard sample tag which is attached to the sample container. In
some cases, particularly with biological samples, the sample tag may have to be
included with or wrapped around the sample and waterproofed. The sample tags are
sequentially numbered and are accountable documents after they are completed and
attached to a sample or other physical evidence. The following information shall be
included on the sample tag: N
a. site name
b. field identification or sample station number
c. date and time of sample collection
d. designation of the sample as a grab or composite
e. type of sample (matrix), and a brief description of the sampling location
f. the signature of the sampler
-------
g. whether the sample is preserved or unpreserved
h. the general types of analyses to be conducted
If a sample is split with another party, sample tags with identical information shall be
attached to each of the sample containers.
The chain of custody record is used to record the custody of samples. It must
accompany samples at all times. The following information must be supplied to
complete the chain of custody record:
a. project name
b. signature of samplers
c. sampling station number, date and time of collection, grab or
composite sample designation, and a brief description of the type of sample and
sampling location,
d. tag numbers
e. signatures of individuals involved in sample transfer, i.e.,
relinquishing and accepting samples. Individuals receiving the samples shall sign,
date and note the time that they received the samples on the form.
Sample analysis request sheets serve as official communication to the laboratory of the
particular analyses required for each sample and provide further evidence that the chain
of custody is complete.
Shipping containers should be secured to ensure samples have not been disturbed
during transport by using nylon strapping tape and EPA custody seals. The custody
seals should be placed on the container so that it cannot be opened without breaking
the seal.
b. CLP
The CLP documentation system provides the means not only to track and identify each
sample, but to support the use of sample data in potential enforcement actions.
Appendix V provides copies of CLP documentation described below.
RAS organic and inorganic samples are documented with corresponding CLP sample
Traffic Reports (TRs), a four part carbonless form. Each TR may document up to
twenty samples shipped to one CLP laboratory under one Case Number and one RAS
analytical program. Samplers must complete the appropriate TRs for every shipment of
RAS samples to a CLP laboratory. An adhesive sample label printed with the sample
number is affixed to each container, and, in order to protect the label from water and
solvent attack, each label is covered with clear waterproof tape. The sample labels
permanently identify each sample collected and link each sample component
throughout the analytical process. A custody seal is then placed over the lid of the
container to ensure the samples are not opened prior to arrival at the laboratory.
Sample documentation for the RAS dioxin program utilizes the CLP Dioxin Shipment
-------
Record (DSR) and samples are individually numbered using pre- printed labels.
For SAS samples, a SAS Packing List (PL) is used along with adhesive sample labels.
Sample tags, containing the necessary information as required by NEIC, are attached to
each sample container at the time of collection. Following sample analysis, sample tags
are retained by the laboratory as physical evidence of sample receipt and analysis.
The Chain-of-Custody Record is employed as physical evidence of sample custody.
One Record accompanies each cooler shipped from the field to the laboratory. In
Region II, the Environmental Services Division Chain-of- Custody Record is used.
Shipping coolers are secured and custody seals placed across cooler openings. As long
as custody forms are sealed inside the sample cooler and the custody seals remain
intact, commercial carriers are not required to sign off on the custody form.
Whenever samples are split with a source or government agency, a separate
Chain-of-Custody Record must be prepared for those samples, indicating with whom
the samples are being split and sample tag serial numbers from splits (13).
If errors are made when completing any of these forms, the error must be crossed out
with a single line and initialed and dated by the sampler.
Information regarding the information contained within, completion of, or obtaining these
forms can be found in the CLP User's Guide available from the Region II RSCC.
B. Field Records
Field records must be kept by contractor personnel for each site. All aspects of sample
collection and handling as well as visual observations must be documented in the
logbooks. The following information must be recorded:
1. sample collection equipment;
2. field analytical equipment;
3. any other equipment used to make field measurements;
4. calculations;
5. results, and;
6. calibration data for equipment.
All entries must be dated, initialed, and legible (9).
All maintenance and calibration records for equipment must be traceable through field
records to the person using the instrument and to the specific piece of instrumentation
itself. Equipment should be labeled with the calibration date and when it is due for the
next calibration. The calibration of the pH and conductivity meters must be checked
-------
daily. Appendix VI describes the required quality assurance procedures for field
analysis and equipment.
Standard operating procedures (SOPs) for use of any field instrumentation must be
provided in the form of a manual or individually in the Plan itself. The SOPs should
address calibration, maintenance and use of the instrumentation and should reflect
what is currently being done in the field.
-------
IV. GLASSWARE REQUIREMENTS
A. Bottle Suppliers
The CLP Sample Bottle Repository (SBR) provides cleaned, contaminant-free sample
containers for use by groups performing hazardous waste sample collection activities
under the Superfund program. Within this contract, sample containers are cleaned by
defined procedures and representative containers undergo strict quality testing prior to
shipment. This contributes to the integrity of sample data and supports its viability for
use in enforcement case actions. The contractor uses approved techniques and
instrumentation to procure., prepare, clean, label, store, package and ship sample
containers and component materials. Appendix VII contains the Statement of Work for
Maintenance of a Quality-Controlled Prepared Sample Container Repository, dated
4/87, revised 7/87 and 8/87, in which the specific requirements for quality control are
delineated.
It is the policy of this office at this time that bottles supplied by any party performing
Superfund work who does not obtain those bottles from the SBR must be cleaned and
quality controlled in the same manner as is defined in the SBR SOW. Therefore, if
containers are not being procured from the Repository, the container construction,
cleaning and quality controlling must be the same as that described in the SOW
presented in Appendix VII. A statement that the bottle supplier will follow the SBR SOW
must be included in the Plan.
B. Volume and Type of Container
The volume of sample obtained should be sufficient to perform all required analyses
with an additional amount collected to provide for quality control needs, split samples, or
repeat analyses.
The sample container requirements may be found in the SBR SOW, the CLP User's
Guide, or in 40 CFR Parti 36.
NOTE: It is the policy of this office at this time that the 40 ml glass vial with Teflon
septum must be used to collect volatile organics in a soil/sediment matrix.
C. Quality Control and Storage .
As stated above, the SBR Statement of Work must be followed when it comes to
procuring, preparing, cleaning, labeling, storing and quality controlling containers. This
involves analysis/testing of one or more representative containers from each lot or
batch after they have been cleaned and designation of a storage QC container for
future analysis if contamination should be suspected at a later time. All storage .QC
containers should be kept in a separate contaminant-free area, See Appendix VII for
detail. Contractors who store containers for any period of time must also comply with
-------
storage requirements.
-------
V. FIELD/LABORATORY DECONTAMINATION OF SAMPLING APPARATUS
A. General Considerations
All sampling apparatus must be properly decontaminated prior to its use in the field to
prevent cross-contamination. The equipment should be pre- cleaned in a laboratory
situation, or if the duration of the sampling event prohibits pre-cleaning in a lab, then
equipment should be decontaminated once a day in an area outside of the
contaminated zone. Enough equipment must be available to be dedicated to sampling
points each day.
Also to avoid cross-contamination, disposable gloves must be worn by the sampling
team and changed between sampling points. While performing the decontamination
procedure, phthalate-free gloves must be used in order to prevent phthalate
contamination of the sampling equipment by interaction between the gloves and the
organic solvent(s).
B. Decontamination Procedures
The required decontamination procedure for all sampling equipment is:
a. wash and scrub with low phosphate detergent
b. tap water rinse
c. rinse with 10% HN03, ultrapure
d. tap water rinse
e. an acetone only rinse or a methanol followed by hexane rinse (solvents
must be pesticide grade or better)
f. thorough rinse with deionized demonstrated analyte free water**
g. air dry, and
h. wrap in aluminum foil for transport
*See page 59 for water criteria.
+The volume of water used during this rinse must be at least five times the volume
of solvent used in Step e.
Tap water may be used from any municipal water treatment system. The use of an
untreated potable water supply is not an acceptable substitute. If metals samples are
not being collected, the 10% nitric acid (HN03) rinse may be omitted, and, conversely,
if organics samples are not being taken, the solvent rinse may be omitted.
When it is necessary to use split spoon sampling devices which are composed of
carbon steel instead of stainless steel, the nitric acid rinse may be lowered to a
concentration of 1% instead of 10% so as to reduce the possibility of leaching metals
from the spoon itself. ,
-------
VI. DECONTAMINATION OF PERIPHERAL EQUIPMENT
A. Well Evacuation Equipment
All tubing and evacuation equipment such as submersible pumps which are put into the
borehole must be rinsed with soapy water and deionized water before use. All tubing
must be dedicated to individual wells, i.e., tubing cannot be reused. If bailers are used
to evacuate wells they must be decontaminated with the same procedure listed in
Section V.
B. Well Casings
Well casings must be steam cleaned prior to installation to ensure that all oils, greases,
and waxes have been removed. Because of the softness of casings and screens made
of fluorocarbon resins, these materials should be detergent washed, not steam cleaned
prior to installation. They should be rested on clean polyethylene sheeting to keep the
possibility of contamination to a minimum.
C. Field Instrumentation
Instrumentation should be cleaned as per manufacturer's instructions. Probes such as
those used in pH and conductivity meters must be rinsed after each use with deionized
water.
/-
D. Drilling Equipment and Other Large Pieces of Equipment
All drilling equipment that comes in contact with the soil must be steam cleaned before
use and between boreholes. This includes drill rods, bits and augers, dredges, or any
other large piece of equipment. Sampling devices such as split spoons and shelby
tubes must be decontaminated as per Section V between boreholes.
-------
VII. MONITORING WELL DESIGN AND CONSTRUCTION
A. Well Drilling and Development Methods
1. General Discussion and Preferred Methods
There are various well drilling methods for application in geologic conditions ranging
from hard rock to unconsolidated sediments. Particular drilling methods have become
dominant in certain areas because they are most effective in penetrating the local
formation.
The recommended types of ground water well drilling techniques are presented in
Appendix VIII. This list has been adopted from "The Practical Guide to Ground Water
Sampling" (4). The selected methods for drilling ground water monitoring wells is a site
specific decision. The Hollow Stem Augering technique is the preferred method for
drilling into unconsolidated sediments of up to one hundred feet in depth, because this
process potentially disturbs the formation the least. If a fluid rotary method is used,
clean water should be used (see water criteria, page 59), and the fluids carefully
controlled to minimize impact on the ground water system.
The mud rotary method has been the alternative chosen most often in the region, but
may add foreign materials into the formation. Effects of this can be minimized by
utilizing high-grade pure bentonite drilling fluids, coupled with rigorous well
development, and purging to remove the fluid residues from the formation. This method
should only be used if site subsurface conditions warrant it's use.
Procedures designed to maximize well yield are included in the term "well
development". Development has two broad objectives: 1) repair damage done to the
formation by the drilling operation so that the natural hydraulic properties are restored,
and 2) alter the basic physical characteristics of the aquifer near the borehole so that
water will flow more freely to a well. These objectives are accomplished by applying
some form of energy to the screen and formation (10). More importantly wells must be
developed to provide water free of suspended solids for sampling. Improperly
developed monitoring wells will produce samples containing suspended sediments that
will bias the chemical analysis of the collected samples (4).
The first step in well development involves the movement of water at alternately high
and low velocity into and out of the wellscreened gravel pack to break down the mud
pack on the well bore and loosen fines in the materials being monitored. This step is
followed by pumping to remove these materials from the well and the immediate area
outside the well screen. This procedure should be continued until the water pumped
from the well is visually free of suspended materials or sediments (4). Methods of
development include overpumping, backwashing, mechanical surging, high velocity
jetting, and air development procedures (10). Of these methods, high velocity jetting
and air development procedures are unacceptable without modification.
-------
High velocity jetting involves the use of a horizontal water or air stream forced through
the well screen to agitate and rearrange the particles surrounding the screen. Although
this is an effective method of development, its major disadvantage is the introduction of
either air or water into the formation. Water jetting is acceptable only if the water used
has a controlled source so cross contamination does not occur. If potable water is used
for water jetting development, analysis of a water blank is required to ensure that the
water is not introducing contaminants into the borehole. Air jetting is acceptable only if
the air injected into the well has a controlled source. Due to frequent contamination of
formations with petroleum hydrocarbons from the air jetting process, the use of an oil
filter between the compressor pump and the borehole to control the purity of the air
introduced downhole is a requirement.
As each monitoring well represents a unique circumstance involving formation
characteristics, well parameters and pumping requirements, current USEPA policy does
not require a minimum waiting period between development and sampling for most
development procedures, but relies on the technical expertise of the drilling contractors
to define the time required for the aquifer to return to stability. For the processes of high
velocity jetting and air development, however, a ten to fourteen day waiting period has
been defined as necessary by the Robert S. Kerr Environmental Research Laboratory
and the New Jersey Department of Environmental Protection Geologic Survey, and is
therefore required by Region II, for the stabilization of aquifer flow and to allow recovery
of the aquifer from the stresses of development.
B. Well Filter Pack and Annular Sealant
The materials used to construct the filter pack should be chemically inert (e.g., clean
quartz sand, silica, or glass beads), well rounded, and dimensionally stable. Natural
gravel packs are acceptable, provided that a sieve analysis is performed to establish
the appropriate well screen slot size and determine chemical inertness of the filter pack
materials in anticipated environments.
The materials used to seal the annular space must prevent the migration of
contaminants to the sampling zone from the surface or intermediate zones and prevent
cross contamination between strata. The materials should be chemically compatible
with the anticipated waste to ensure seal integrity during the life of the monitoring well
and chemically inert so they do not affect the quality of the ground water samples. An
example of an appropriate use of annular sealant material is using a minimum of two
feet of certified sodium bentonite pellets immediately over the filter pack when in a
saturated zone. The pellets are most appropriate in a saturated zone because they will
penetrate the column of water to create an effective seal. Coarse grit sodium bentonite
is likely to hydrate and bridge before reaching the filter pack. A cement and bentonite
mixture, bentonite chips, or anti-shrink cement mixtures should be used as the annular
sealant in the unsaturated zone above the certified-bentonite pellefseal and below the
frost line. Again, the appropriate clay must be selected on the basis of the environment
-------
in which it is to be used. In most cases, sodium bentonite is appropriate^).
The selected seal must not interfere with the water chemistry. Bentonite clay has
appreciable ion exchange capacity which may interfere with the chemistry on collected
samples when grout seal is in close proximity to the screen or well intake. Similarly,
expanding cement which does not harden properly may affect the pH of water from
monitoring wells when in close proximity to the well screen or intake.
To minimize these potential interferences, a 1-foot layer of silica sand should be placed
above the selected gravel pack. Then, if possible, 1-2 feet of pure bentonite pellets
should be placed in the hole to prohibit the downward migration of bentonite slurry or
neat cement(4).
The untreated sodium bentonite seal should be placed around the casing either by
using a tremie pipe or, if a hollow-stem auger is used, putting the bentonite between the
casing and the inside of the auger stem. Both of these methods present a potential for
bridging. In shallow monitoring wells, a tamping device should be used to reduce this
potential. In deeper wells, it may be necessary to pour a small amount of clean water
down the casing to wash the bentonite down the hole.
The cement-bentonite mixture should be prepared using clean water and placed in the
borehole using a tremie pipe. The tremie method ensures good sealing of the borehole
from the bottom.
The remaining annular space should be sealed with expanding cement to provide for
security and an adequate surface seals. Locating the interface between the cement and
bentonite-cement mixture below the frost line serves to protect the well from damage
due to frost heaving. The cement should be placed in the borehole using the tremie
method(5).
C. Well Casing Selection
1. General Discussion
Well construction materials must be durable enough to resist degradation thereby
retaining their long-term stability and structural integrity and be relatively inert to
minimize alteration of ground water and collected samples.
In general, the more inert (i.e., less reactive) the casing material, the more assured one
is that the ground water sample withdrawn from the well is representative of the actual
ground water. The major potential alterations of the sample resulting from interactions
with the well casing/screen materials are: (1) adsorption/absorption reactions, both of
organics and inorganics; and (2) desorption reactions, meaning leaching of chemical
constituents from the well casing material into the ground water or desorption of newly
adsorbed material. Casing materials can also be affected by chemical attack, i.e.,
corrosion/deterioration, and microbial colonization and attack(4).
-------
These processes may lead to the observation of false trends in analyte concentrations,
highly variable water chemistry and the identification of artifacts resultant from surface
release or sorptive interactions. As with the errors which sampling mechanisms can
introduce into the chemical data, casing materials' related errors can be quite significant
and difficult to predict. Appropriate choice of materials for each application must be
made on the basis of long-term durability, cleanability, and the minimization of the
secondary effects of sorption or leaching. Structural integrity over time is, therefore, the
primary criterion for making reliable material choices. The materials must neither be
attacked nor degraded during the course of the monitoring program(4).
A variety of construction materials have been used for casing and well screens,
including virgin fluorocarbon resins (Teflon), stainless steel (304 or 316), cast iron,
galvanized steel, polyvinylchloride (PVC), polyethylene and polypropylene. Many of
these materials, however, may affect the quality of ground water samples and may not
have the long-term structural characteristics necessary for site specific needs. For
example, steel casing deteriorates in corrosive environments; PVC deteriorates when in
contact with ketones, esters and aromatic hydrocarbons; polyethylene deteriorates in
contact with aromatic and halogenated hydrocarbons; and polypropylene deteriorates in
contact with oxidizing acids, aliphatic hydrocarbons, and aromatic hydrocarbons. In
addition, steel, PVC, polyethylene and polypropylene may adsorb and leach
constituents that may affect the quality of ground water samples(5).
The selection of well casing and screen material should be made with due consideration
to geochemistry, anticipated lifetime of the monitoring program, well depth, chemical
parameters to be monitored and other site specific factors. Fluorocarbon resins or
stainless steel should be specified for use in the saturated zone when volatile organics
are to be determined during long term monitoring. Where high corrosion potential exists
or is anticipated, fluorocarbon resins are preferable to stainless steel. National
Sanititation Foundation (NSF) or ASTM-approved polyvinylchloride (PVC) well casing
and screens may be appropriate if only trace metals or non-volatile organics are the
contaminants anticipated(S).
Any well casing material may be used in the vadose zone, however, one combination
that should be avoided is the use of dissimilar metals, such as stainless steel and
galvanized steel, without an electrically isolating (dielectric) bushing. If such dissimilar
metals are in direct contact in the soil, a potential difference is created and leads to
accelerated corrosion of the galvanized steel (in this example). More generically, in the
Galvanic series the less noble metal becomes the anode to the more noble metal and is
corroded at an accelerated rate. In well construction, this acceleration in corrosion at
the point of connection will lead to failure of the construction materials. Thus, a
dielectric coupling should be used for connecting dissimilar metals above the saturated
zone.
Plastic pipe sections must be flush threaded or have the ability to be connected by
another mechanical method that does not introduce contaminants such as glue or
-------
solvents into the well(5).
2. Selection Criteria SOP
Appendix X presents the "Standard Operating Procedure for Selecting Ground Water
Well Construction Material at CERCLA Sites", dated December 15, 1986. The appendix
to the SOP provides a "Summary Table for Comparing Features of Various Ground
Water Well Construction Materials", which was used to develop the criteria presented
for selecting the appropriate casing material. The considerations involved in the process
include duration of intended well use, use of data, desired detection limits, and known
site conditions and contaminants. The numerical cut-off values presented except for the
chloride and pH conditions on page 4 of the SOP were designed to be ball-park figures
intended to guide decision making but they were not intended to be absolute limitations.
They were, as all of the criteria were, devised after digestion of the current literature and
using best professional judgement.
The Summary Form presented on page 5 of the SOP should be filled out by the EPA
Project Manager and presented with the Plan for review for each site.
D. Well Screens
Monitoring well screens should be 5 to 10 feet in length to avoid dilution of the
contaminated groundwater with water from less contaminated zones in the aquifer (5).
If site specific circumstances dictate the use of longer screens, this issue should be
clearly discussed in the Plan. The well depth should ensure that the screened section
is always submerged, considering seasonal water level fluctuations. In cases where
light, non-aqueous phase liquids are expected to pose a significant problem, the
screened interval should intersect the water table throughout the year. If dense,
non-aqueous phase liquids or denser than water dissolved contaminants are suspected
to pose a problem, then the screened interval for the deeper wells should be positioned
immediately above any significant confining layer.
E. Evaluation of Existing Wells
If it is desired that a well already existing on-site be sampled in conjunction with newly
installed wells, the Project Manager should consider the ramifications of utilizing data
from those wells if, according to the "SOP for Selecting Ground Water Well
Construction Material", the existing well casing is not compatible with the type of ground
water contamination or the sensitivity of analysis needed.
-------
VIII. SAMPLE COLLECTION DEVICES, MATERIALS AND QUALITY CONTROL
PRACTICES
A. References for Selection of Sampling Devices
Sampling at hazardous waste sites requires many different types of sampling
devices. Selection of a device should be based on practicality, economics,
representativeness, compatibility with analytical considerations, and safety. There are
many documents which compile sampling methods and materials suitable to address
most needs that arise during investigations. The following is a list of the most commonly
used references compiling sampling equipment and methodology, however it is not
meant to be an exhaustive listing of all the references available.
1. Characterization of Hazardous Waste Sites-A Methods Manual: Volume II. Available
Sampling Methods, Second Edition. EPA-600/4-84-076. December 1984. Available
from ORD Publications in Cincinnati at (513)569-7562.
2. Handbook for Sampling and Sample Preservation of Water and Wastewater.
EPA-600/4-82-029. September 1982. Available from ORD Publications.
3. Samplers and Sampling Procedures for Hazardous Waste Streams. EPA-600/2-
80-018. January 1980. Available from ORD Publications.
4. Practical Guide for Ground-Water Sampling. EPA 600/2-85/104. September 1984.
Available from ORD Publications.
5. RCRA Ground-Water Monitoring Technical Enforcement Guidance Document,
September, 1986. Office of Waste Programs Enforcement and Office of Solid Waste
and Emergency Response. Available from RCRA Hotline at 800-424-9346.
6. A Guide to the Selection of Materials for Monitoring Well Construction and Ground
Water Sampling. Barcelona, Gibb, Miller. Illinois State Water Survey, Champaign,
Illinois. January 1984. NTIS publication #PB84-126929.
7. Test Methods for Evaluating Solid Waste, Physical and Chemical Methods.
SW-846, Third Edition. Office of Solid Waste and Emergency Response. GPO
publication #955-001-00000-1, at (202) 783-3238.
8. Compendium of Methods for the Determination of Toxic Organic Compounds in
Ambient Air, April 1984. EPA 600/4-84-041. Available from ORD Publications.
9. SOP for the GC/MS Determination of Volatile Organic Compounds Collected on
Tenax. June 27, 1984. EMSL/RTP-SOP-EMD-020. Available through USEPA at
Research Triangle Park, NC at (919)541-2777.
10. USEPA Dioxin Strategy. November 28,1983. Office of Water Regulations and
-------
Standards and the Office of Solid Waste and Emergency Response in conjunction with
the Dioxin Strategy Task Force, Washington, D.C., 20460.
11. Sampling Guidance Manual for the National Dioxin Study, May 16,1984. Office of
Water Regulations and Standards, Washington, D.C.
12. Soil Sampling Quality Assurance User's Guide, Draft. May 1984. EMSL-LV. EPA
600/4-84-043. Available from ORD Publications.
13. Data Quality Objectives for Remedial Response Activities. Development Process.
EPA/540/G-87/003. March 1987.
14. Data Quality Objectives for Remedial Response Activities. Example Scenario:RI/FS
Activities at a Site With Contaminated Soils and Ground Water. EPA/540/G-87/004.
March 1987.
B. Groundwater
1. Sampling Design
Samples from a monitoring well represent a small part of the horizontal and vertical
extent of the aquifer. Unlike its surface counterpart, where a sample can be arbitrarily
taken at any point in the system, moving a ground water sampling point implies the
installation of additional monitoring wells. There is a need to be concerned not with the
point data as an end in itself, but as a component of a network approach wherein
information on the ground water system is developed as a basis for extrapolating
information to areas where samples were not collected and/or for predicting the effects
of natural and man-made stresses on the subsurface systems(2). Discussion of the
areas of consideration for location of ground water sampling points can be found in
references listed above.
2. Well Evacuation
In order to obtain a representative sample of ground water, the water that has stagnated
and stratified in the well casing must be purged or evacuated. Prior to evacuating the
well, however, the presence or absence of immiscible phases (i.e., "floaters" and
"sinkers") must be determined. "Floaters" are those relatively insoluble organic liquids
that are less dense than water and which spread across the potentiometric surface.
"Sinkers" are those relatively insoluble organic liquids that are more dense than water
and tend to migrate vertically through the sand and gravel aquifers to the underlying
confining layer. The detection of these immiscible layers requires specialized equipment
that must be used before the well is evacuated for conventional sampling. The Plan
should specify the device to be used to detect light phases and dense phases, as well
as the procedures to be used for detecting and sampling these contaminants.
Procedures for identifying and sampling "floaters" and "sinkers" can be found in section
4.2.2 of the RCRA Ground-Water Monitoring Technical Enforcement Guidance
-------
Document dated September 1986.
Water that has remained in the well casing for extended periods of time (i.e., more than
about two hours) has the opportunity to exchange gases with the atmosphere and to
interact with the well casing material. The chemistry of the water stored in the well
casing is dissimilar to that of the aquifer and, thus, should not be samp led (4).
Evacuating the well allows for fresh formation ground water to enter the well. When
indicator parameters such as pH, temperature and specific conductance are observed
to vary less than 10% over the removal of two successive well volumes, the well is
presumed to be adequately flushed for representative sampling. Evacuation of at least
3-5 well volumes is required for high yielding wells, however, in wells with very low
recoveries this may not be practical. In this case the well may be evacuated to near
dry ness once and allowed to recover sufficiently for samples to be taken. A well must
be sampled within three hours of evacuation. If a well is allowed to sit longer than three
hours after evacuation, it should be re-evacuated since the water contained in the
casing may no longer be representative of the aquifer conditions(4).
Any device used to evacuate the well must be cleaned as per Section V to ensure that
cross contamination between wells does not occur. When a pump is used to evacuate,
the tubing which comes in contact with water should be made of polyethylene or Teflon,
and must be dedicated to individual wells. The intake should be placed just below the
water level and lowered as the water level lowers while pumping to ensure that all the
water within the well bore is exchanged with fresh water.
A bailer may be used to evacuate the well and to sample it. Bailers must be constructed
of Teflon or stainless steel. Bailer cords are to be stainless steel single stranded wire,
or polypropylene monofilament such as fishing line. Braided or twisted cords of any
type are not allowed as complete decon would be difficult resulting in possible cross
contamination between wells. Ten foot leaders may be used of these acceptable
materials, with nylon cord above. ANY down- hole equipment having neoprene fittings,
PVC, tygon tubing, silicon rubber bladders, neoprene impellers, or viton are not
acceptable. A bailer which is used to evacuate the well may also be used to sample it
without any additional cleaning.
Bailer cord must be cleaned with soap and deionized water before use. Cord can be
reused; it is not necessary to dedicate it to individual wells. If a ten foot or greater length
leader (any cord of unacceptable material may not contact the water) is being used,
only the leader need be cleaned. See Section VIII for acceptable bailer cord materials.
Any water that is removed from the well during evacuation can no longer be considered
a representative portion of the aquifer and should not be reintroduced into the well after
sampling.
3. Sampling Considerations
After evacuation of the required volume of water from the well, sampling may occur.
-------
The unstable nature of many chemical, physical and microbial constituents in ground
water limit the sample collection options. Certain factors should be considered when
collecting representative samples:
* Temperatures are relatively constant in the subsurface, therefore the sample
temperature may change significantly when brought to the surface. This change can
alter chemical reaction rates, reverse cationic and anionic ion exchanges on solids, and
change microbial growth rates.
* A change in pH can occur due to carbon dioxide adsorption and subsequent changes
in alkalinity. Oxidation of some compounds may occur.
* Dissolved gases may be lost at the surface.
* The integrity of organic samples may be affected by problems associated with either
adsorption or contamination from sampling materials and volatility(2).
The only acceptable sampling devices for pH sensitive and volatile parameters are:
1. Teflon or stainless steel bladder pumps having adjustable flow control;,
2. Teflon or stainless steel bottom-filling bailers; and,
3. Teflon or stainless steel syringe bailers.
Appropriate operating precautions for these sampling devices include:
* Bladder pumps must be operated in a continuous manner so that they do not produce
pulsating samples that are aerated in the return tube or upon discharge. Pumping rates
should not exceed. 100ml/min when samples are being taken for dissolved gases,
volatile organic constituents, TOX and TOG;
* Check valves must be designed and inspected to ensure that fouling problems do not
reduce delivery capabilities or result in aeration of the sample;
* Sampling equipment (especially bailers) must never be dropped into the well because
this will cause degassing of the water upon impact;
* The bailer's contents must be transferred to a sample container in a way that will
minimize agitation and aeration without transferring the sample to an intermediate
container, or utilizing a mechanical device; the bailer should not be "acclimated" by
discharging the first bailer- full of water onto the ground since this unnecessarily
agitates the water column prior to the volatile sensitive parameters being taken;
* Clean sampling equipment must not be placed directly on the ground or other
contaminated surface. When not in use, these devices should be placed on
polyethylene sheeting or aluminum foil.
-------
When sampling, heavy gauge aluminum foil or polyethylene sheeting should be placed
on the ground around each well to prevent contamination of sampling equipment in the
event that equipment is dropped or otherwise comes in contact with the ground.
When a number of rounds or phases of sampling will take place, the same type of
sampling equipment should be used consistently throughout the entire project in order
to increase the reproducibility in analytical results by eliminating the variability in sample
collection technique.
Other sampling devices, including positive displacement pumps, gas lift devices,
centrifugal pumps, and venturi pumps, may be used for collection of non-volatile or
non-pH sensitive parameters (most parameters ARE pH- sensitive), provided that they
are constructed of Teflon or stainless steel(5).
The preferred order of sample collection is as follows:
1. In-situ measurements:temperature, pH, specific conductance
2. Volatile organics (VOA)
3. Purgeable organic carbon (POC)
4. Purgeable organic halogens (POX)
5. Total organic halogens (TOX)
6. Total organic carbon (TOG)
7. Extractable organics
8. Total metals
9. Dissolved metals
10. Phenols
11. Cyanide
12. Sulfate and Chloride
13. Turbidity
14. Nitrate and Ammonia
15. Radionuclides
Detailed discussions of sample evacuation and collection procedures can be found in
references #1,2,4,5 found in the Bibliography of this document.
For a discussion on the collection of filtered and non-filtered sample fractions for metals
analysis, see Section XI.
4. Microbiological Sampling
There are several different methods for obtaining a ground water sample. Each of these
methods differ in their advantages and disadvantages for obtaining samples for
microbiological analyses.
The majority of ground water samples obtained for microbiological analysis are obtained
-------
using preexisting wells which have existing in-place pumps. This limits the precautions
the sampler must take to ensure a non- contaminated sample. Samples should be
obtained from outlets as close as possible to the pump and should not be collected
from leaky or faulty spigots or spigots that contain screens or aeration devices. The
pump should be flushed for 5 to 10 minutes before the sample is collected. A steady
flowing water stream at moderate pressure is desirable in order to prevent splashing
and dislodging particles in the faucet or water line.
To collect the sample, remove the cap or stopper carefully from the sample bottle. Do
not lay the bottle closure down or touch the inside of the closure. Avoid touching the
inside of the bottle with your hands or the spigot. The sample bottle should not be
rinsed out and it is not necessary to flame the spigot. The bottle should be filled directly
to within 2.5 cm (1 inch) from the top. The bottle closure and closure covering should be
replaced carefully and the bottle should be placed in a cooler (4 to 10 degrees C)
unless the sample is going to be processed immediately in the field.
If a well does not have an existing in-place pump, samples can be obtained by either
using a portable surface or submersible pump or by using a bailer. Each method
presents special problems in obtaining an uncontaminated sample.
The main problem in using a sterilized bailer is obtaining a representative sample of the
aquifer water without pumping or bailing the well beforehand to exchange the water in
the bore for fresh formation water. This is difficult since such pre-sampling activities
must be carried out in such a way as to not contaminate the well. Care must also be
taken with bailers to not contaminate the sample with any scum on the surface of the
water in the well. This is usually done by using a weighted, sterilized sample bottle
suspended by a cord of acceptable material and lowering the bottle rapidly to the
bottom of the well.
The use of portable pumps provides a way of pumping out a well before sampling and
thus providing a more representative sample, but presents a potential source of
contamination if the pumping apparatus cannot be sterilized beforehand. The method of
sterilization will depend on what other samples are taken from the well since the use of
many disinfectants may not be feasible if the well is also sampled for chemical
analyses. If disinfection is not ruled out by other considerations, a method of sterilizing
a submersible pump system is to submerge the pump, and any portion of the pump
tubing which will be in contact with the well water, into a disinfectant solution and
circulating the disinfectant through the pump and tubing for a recommended period of
time.
The most widely used method of disinfection is chlorination due to its simplicity.
Chlorine solutions may be easily prepared by dissolving either calcium or sodium
hypochlorite in water. Calcium hypochlorite, Ca(OCI)2, is available in a granular or
tablet form usually containing about 70% of available chlorine by weight and should be
stored under dry and cool conditions. Sodium Hypochlorite, NaOCI, is available only in
liquid form and can be bought in strengths up to 20% available chlorine. Its most
-------
available form is household laundry bleach, which has a strength of about 5% available
chlorine, but should not be considered to be full strength if it is more than 60 days old.
The original percentage of available chlorine will be on the label. Fresh chlorine
solutions should frequently be prepared because the strength will diminish with time.
After disinfection the pump should be carefully placed in the well and then pumped to
waste until the chlorine is thoroughly rinsed from the pump system.
If the pump cannot be disinfected, then the pump and tubing should be carefully
handled to avoid gross surface contamination and the well should be pumped for 3 to
10 bore volumes before taking a sample. It may be desirable after pumping to pull the
pump and take the sample with a sterile bailer.
In those cases where the water level in the well is less than 20 to 30 feet below the
surface, a surface vacuum pumping system can be used for flushing out the well and
withdrawing a sample.
Lastly, interpretation of analytic results may be difficult in some cases since surface
contamination of wells due to^poor drilling and completion practices is common. In
cases where drinking water supplies are involved, a thorough inspection of the well is
required to eliminate surface contamination down the well bore as a source of
contaminants. Disinfection of the well by approved methods and resampling may be
advisable, if disinfection will not affect the well for other sampling purposes(2).
C. Surface Water
1. Sampling Design
There are at present no quality assurance requirements for site location for sampling
surface waters. References #1 and #2 in Part A of this Section can be consulted for
design guidance.
2. Sampling Devices
Appendix XI provides a tabular comparison of surface water sample collection devices.
Samples from shallow depths can be readily collected by merely submerging the
sample container. However, preservatives cannot be present in the container when it .is
lowered into the water. The method is advantageous when the sample might be
significantly altered during transfer from a collection vessel into another container. This
is the case with samples collected for oil and grease analysis since considerable
material may adhere to the sample transfer container and as a result produce
inaccurately low analytical results. Similarly the transfer of a liquid into a small sample
container for volatile organic analysis, if not done carefully, could result in significant
aeration and resultant loss of volatile species. Though simple, representative, and
generally free from substantial material disturbances, it has significant shortcomings
-------
when applied to a hazardous waste, since the external surface of each container would
then need to be decontaminated.
In general the use of a sampling device constructed of a nonreactive material such as
glass, stainless steel, or Teflon, is the most prudent method. The device should have a
capacity of at least 500 ml, if possible, to minimize the number of times the liquid must,
be disturbed, thus reducing agitation of any sediment layers.
A 1-liter stainless steel beaker with pour spout and handle or large stainless steel ice
scoops and ladles available from commercial kitchen and laboratory supply houses can
be used.
It is often necessary to collect liquid samples at some distance from shore or the edge
of the containment. In this instance an adaptation which extends the reach of the
sampler is advantageous. Such a device is the pond sampler. It incorporates a
telescoping heavy-duty aluminum pole with an adjustable beaker clamp attached to the
end. The beaker previously described, a disposable glass container, or the actual
sample container itself, can be fitted into the clamp.
It may on occasion be necessary to sample large bodies of water where a near surface
sample will not sufficiently characterize the body as a whole. In this instance a
peristaltic pump may be used in which the sample is drawn in through heavy walled
Teflon tubing and pumped directly into the sample container. This method, however, is
not suitable for pH sensitive or volatile samples in which stripping would occur.
Situations may still arise where a sample must be collected from depths beyond the
capabilities of a peristaltic pump. In this instance an at-depth sampler may be required,
such as a Kemerer, ASTM Bomb (Bacon Bomb) or Van Dorn sampler. These devices
work well; however, care must be utilized in selecting devices that are made of
materials that will not contaminate the sample. Van Dorn samplers are not generally
recommended for organics as they rely on an elastic closing mechanism that can effect
samples. They are readily available in a totally nonmetallic design which is very useful
for sample collection for trace metal analysis.
Kemerer samplers are available on special order or adaptable for sample collection for
organic analysis by substituting Teflon for the rubber or plastic stoppers. If the device is
further ordered with stainless steel metallic parts in addition to Teflon stoppers it
becomes a very versatile sampler(1). .
Consult "Characterization of Hazardous Waste Sites-A Methods ManuahVolume II.
Available Sampling Methods, Second Edition" for specific collection techniques using
these devices. See Part B.4 of this Section for collection of microbial samples.
D. Sediment
-------
Sediment is defined as the deposited material underlying a body of water.
1. Sampling Design
Streams, lakes, and impoundments will likely demonstrate significant variations in
sediment composition with respect to distance from inflows, discharges or other
disturbances. It is important, therefore, to document exact sampling location by means
of triangulation with stable reference points on the banks of the stream or lake.
At the present time, however, there are no quality control requirements for choosing site
location when sampling sediments. Reference #2 in Part A of this Section can be
consulted and sediment transport and deposition modeling along with statistical
considerations can be found in "Sediment Sampling Quality Assurance User's Guide,"
EPA 600/4-85-048,July 1985, NTIS #PB85- 233542.
2. Sampling Devices
Appendix XII provides a tabular comparison of some bottom grab and coring devices.
Samples can be taken with stainless steel spoons or trowels or the sample container
itself if there is little or no water on top of the sediment. If the water above the sediment
is a few feet deep a stainless steel corer or corer having a removable glass or Teflon
liner may be used. This will help ensure the integrity of the surface layer of sediment
and will minimize the loss of fine grained material. In deeper water the bottom grab
samplers presented in Appendix XII may be used(1).
If at any time surface water samples are being taken in conjunction with sediment
samples, the water samples should be taken first. The sampler should approach the
location from the downstream direction with the container pointed upstream to ensure
collection of an undisturbed sample.
Sediment samples collected for all analyses except VOA and TOX should be thoroughly
mixed before being placed in appropriate sample containers. Rocks, twigs, and other
debris should be removed from the sample prior to homogenization, if they are not
considered part of the environmental sample. In this way stratification of constituents
will not affect analytical results. VOA and TOX samples should be taken as individual
grab samples, never homogenized. See Section IX for the procedure used for
homogenization of samples.
E. Soil
1. Sampling Design
Statistical techniques for obtaining accurate and precise samples are relatively simple
and easy to implement. In the sampling of a soil matrix, sampling accuracy is usually
achieved by some form of random sampling. In random sampling, every unit in the
-------
population has a theoretically equal chance of being sampled and measured.
Consequently, statistics generated by the sample are unbiased (accurate) estimators of
true population parameters. In other words, the sample is representative of the
population. One of the most common methods of selecting a random sample is to
divide the population by an imaginary grid, assign a series of consecutive numbers to
the units of the grid, and select the numbers to be sampled through the use of a
random-numbers table (such a table can be found in any text on basic statistics). It is
important to emphasize that a haphazardly selected sample is not a suitable substitute
for a randomly selected sample. That is because there is no assurance that a person
performing undisciplined sampling will not consciously or subconsciously favor the
selection of certain units of the population, thus causing the sample to be
unrepresentative of the population. A detailed discussion of random sampling can be
found in SW-846, Third Edition, November 1986, Volume 2, Chapter 9.
In a biased sampling approach, site history, available information and professional
judgement are used to determine the sampling locations in which contamination is
expected to be found. In this case the most information is gotten from the least number
of samples thereby allowing a cost savings. However, this approach to sampling design
is more appropriate to a site investigation than a remedial investigation (Rl) because it
is not a comprehensive view of site conditions(T). Use of a biased approach during an
Rl must be justified in the Plan.
2. Sampling Devices
When sampling soil, stainless steel, Teflon or glass utensils should be used. The only
exception is split spoons which are most commonly available in carbon steel. These are
acceptable for use provided they are not excessively rusty.
A stainless steel spatula should be used to remove sample from the opened spoons,
not the sampler's fingers, as the gloves may introduce organic interferences into the
sample. As per water sampling, volatile organics and TOX samples should be taken
immediately upon opening the spoon. Any rocks, twigs, leaves or other debris should be
removed from the sample before homogenization, if they are not considered part of the
environmental sample. All samples except those for volatile organics and TOX must be
homogenized before being put into sample containers. Section IX provides a discussion
on homogenization. Samples for VOA and TOX should never be homogenized as they
both contain volatile parameters.
F. Potable Water ,' '
When sampling potable water the same quality control requirements apply as for
sampling groundwater, where applicable.
The "Manual for the Certification of Laboratories Analyzing Drinking Water," EPA
570/9-82-002, October 1982, states that the sample needs to be representative of the
potable water system, and that the tap water must be sampled after maintaining a
-------
steady flow of 2 or 3 minutes to clear the service line. This flushing period is a minimum
requirement and more time can be allowed if a large holding tank is in place and it is
desireable to purge an entire tank volume. A tap must be free of any aerator, strainer,
hose attachment, or water purification devices. When sampling chlorinated waters for
organics and bacterial analyses, sodium thiosulfate should be added to the sample
bottles. Consult the reference for details.
G. Dust/Wipe
The following Standard Operating Procedure has been developed by Region II for use
in taking wipe samples.
1. Materials needed:
a. cotton swabs, solvent rinsed and completely air dried;
(use of synthetic materials requires checking for compatibility with
solvents)
b. acetone, pesticide grade;
c. hexane, pesticide grade;
d. deionized water;
e. HCI or HNO3, redistilled
f. stainless steel clamps or plastic clamps (only for taking metals
samples);
g. appropriate sample bottles.
2. A square area, of a size sufficient to give the required amount of sample for each
fraction as prbvided in the analytical methodology to be used, should be marked off.
This may require taking cotton swabs and a balance into the field, wiping a certain area
and weighing the swab before and after to determine how much area should be wiped
to give the required weight of sample.
3. Swabs for semi-volatile, pesticide and PCB samples should be moistened in a 1:4
acetone/hexane mixture. Swabs for volatile organic samples should be moistened with
hexane alone, and those for metals with deionized water. While holding the swab in a
clean, metal clamp, moisten the cotton swab with the appropriate solution.
4. While still holding the cotton swab in the clamp, wipe the sampling area back and
forth repeatedly, applying moderate pressure. Wipe the entire area so that all the
sample material is picked up.
5. Place the used swab in the appropriate sample container and seal.
6. Clean the clamps between each sample with both solvent and 10% HCL or
HN03.
7. As a blank, moisten a clean swab with the solvent or water (for each collection
medium), place it in a separate jar, and submit it with the other samples,
-------
8. When samples are submitted for analysis, the laboratory should be told to rinse
the sample jars with solvent or 10% HCL or HN03, depending on the analysis to be
performed, when transferring sample to the extraction glassware.
The samples should be analyzed with the appropriate methodology for a soil/sediment
matrix, and a sufficient quantity of material must be collected as called for in the
analytical methodology in order for method detection limits to be achieved.
H. Dioxin
For guidance on conducting dioxin sampling events refer to "Sampling Guidance
Manual for the National Dioxin Study", May 16,1984. The sampling material
requirements are the same as those present in this document for the matrix being
sampled, with the following additions.
The equipment decontamination procedure is the same as that presented in Section V
however samplers must use trichloroethylene as the solvent rinse in step e of the
procedure.
Homogenization must be performed on dioxin samples of a solid matrix.
Homogenization may be performed in a laboratory or in the field and it should be done
using either the coning and quartering method (see Section IX) or by using stainless
steel blenders. Homogenization of wet sediment samples is more easily accomplished
using the coning and quartering method. Specific quality control samples for dioxin
sampling events are presented in Section X.
»
I. Drums
For sampling of non-homogenous or multi-phased materials in drums, tanks, waste
piles, lagoons, etc., refer to SW-846, Third Edition, Volume 2 for guidance. The Plan
must include a detailed approach in terms of basic strategy and sampling equipment if
such containers are to be sampled.
-------
IX. METHODS OF SAMPLE PREPARATION
A. Homogenization
The homogenization of a sample is the process of mixing individual grab samples in
order to minimize any bias of sample representativeness introduced by the natural
stratification of constituents within the sample.
To homogenize a sample of a soil/sediment matrix, first rocks, twigs, leaves and other
debris should be removed if they are not considered part of the sample. The
soil/sediment should be removed from the sampling device and placed in a stainless
steel pan, then thoroughly mixed using a stainless steel spoon. The sediment in the
pan should be scraped from the sides, corners and bottom of the pan, rolled to the
middle of the pan, and initially mixed. The sample should then be quartered and moved
to the four corners of the pan. Each quarter of the sample should be mixed individually,
and then rolled to the center of the container and the entire sample mixed again.
Homogenization of an aqueous sample is only necessary if stratification of constituents
is of concern, for example when sampling a lagoon or containerized liquid. Then
homogenization would be performed by mixing in a stainless steel bowl.
B. Compositing
Compositing of samples is performed when samplers desire to obtain an average
concentration of contaminants over a certain number of sampling points. Anytime
compositing is performed, the concentration of contaminant in individual grab samples
is diluted proportionately to the number of samples taken. Not only is the contaminant
diluted, the detection limits for each individual sample are raised proportionally to the
number of samples added to the composite. For instance, if a sampler wishes to
composite two discreet samples into one, and the method detection limit for a target
compound is 330 ppb, the detection limit for the target compound does not change for
the composite, however, the detection limit for the compound in the individual samples
which make up the composite is two times the normal detection limit or 2 * 330 = 660
ppb. This is important to keep in mind because it is possible that if a contaminant were
present in only one of the two composited samples, and if it were at a level between
330 and 660 ppb, that contaminant would not be quantified or possibly even identified
due to the effective dilution of the contaminant concentration in the composite. This
concept should be taken into account when determining the data quality objectives of a
composite sampling event, to ensure that useful data is collected. It is advisable that if a
positive identification is made in the course of analyzing a composite sample, that the
discreet samples then be analyzed individually to determine the true distribution of
contaminant throughout each component of the composite.
Compositing of a solid matrix is accomplished by mixing equal volumes of grab samples
in stainless steel pans with stainless steel spoons. Compositing is never performed on
sample8 for volatile organics analysis (or any analysis involving a volatile portion such
-------
as TOX), and, for a solid matrix, should never be done by placing equal portions of grab
samples directly into sample jars, as the occurrence of error introduced by the sampler
is highly probable.
C. Splitting
Splitting of samples is performed when two or more parties wish to have a portion of the
same sample. They are most often collected in enforcement actions to ensure that
sample results generated by Potentially Responsible Parties (PRPs) are accurate.
Splitting of samples, however, is not as useful as providing blind performance
evaluations samples to a laboratory since analytical performance and accuracy differs
from laboratory to laboratory, and therefore one laboratory cannot be considered a
"referee" whose performance can be considered the standard against which another's
can be measured. Performance evaluation samples provide information on a
laboratory's performance based upon analysis of that sample which contains
parameters of a known and defined concentration.
Soil/sediment samples taken for volatile analysis cannot be split. In this case samples
must be taken as co-located grabs. Then,' a large quantity of material can be collected,
homogenized, split and used to fill the remaining containers. Note that enough sample
must be collected at one time in order to fill all the necessary sample containers. It may
be necessary to co- locate or depth integrate collection so enough sample volume is
available. A description of this process should be provided in the Plan.
When splitting aqueous samples, homogenization of the sample is only necessary if
heterogeneity is suspected, for example when sampling a small lagoon or containerized
liquid; however VOA and TOX samples are never homogenized. It is not generally
necessary to homogenize ground water or surface water samples when splitting, and it
is likewise generally unnecessary to divide a bailer's contents among several bottles.
Please note, duplicate samples, trip blanks and field equipment rinse blanks must be
included as part of those samples which are split between the two or more parties
involved.
-------
X. FIELD QUALITY CONTROL SAMPLES
A. All Analyses Except Dioxin
*
1. Duplicates
Environmental duplicate samples are collected to demonstrate the reproducibility of
sampling technique. Environmental duplicate samples must be taken at a frequency of
at least 5% (1 in 20). This is a separate duplicate from the duplicates a laboratory must
run, and cannot be replaced by a laboratory generated duplicate. This applies to every
matrix sampled. Environmental duplicates are representative of field sampling precision,
whereas laboratory duplicates are a measure of analytical precision. Both pieces of
information are essential to determining the quality of data generated for a project.
2. Blanks
Blank water generated for use in the Region II CERCLA program must be
"demonstrated analyte-free". By this term we mean water of a known quality which is
'defined by the Quality Assurance office.
The criteria for analyte-free water is as follows. The assigned values for the Contract
Required Detection Limits (CRDLs) and Contract Required Quantitation Limits (CRQLs)
can be found in the most recent CLP SOWs or in Appendix III enclosed. These criteria
apply to §Ji blank water used for fund-lead as well as PRP projects whether or not EPA
CLP analytical methods are employed. If the levels of detection needed on a specific
site are lower than the CLP CRDLs/CRQLs, then those levels are used to define the
criteria for analyte free water.
purgeable organics <10 ppb
semi-volatile organics
-------
personnel during a field audit. In addition, the contractors shall maintain quality control
records for each source of water which is used. These records shall demonstrate over
time the presence/absence and level of contaminants found in each water supply. EPA
personnel will randomly audit throughout the year the records kept by the
generators/contractors.
If potable water is used during the drilling process and is introduced into the borehole, a
sample of the water should be collected and analyzed to ensure contaminants are not
being introduced via the water supply.
a. Trip Blanks
When sampling for volatile organics, a trip blank, consisting of demonstrated analyte
free water sealed in 40 ml Teflon lined septum vials, must be taken into the field where
sampling is going on. It should be taken at a minimum frequency of one per day when
volatile organics in an aqueous matrix are being collected. Note that it is not necessary
to take an aqueous trip blank when a non-aqueous medium is being sampled. Trip
blanks are used to determine if any on-site atmospheric contaminants are seeping into
the sample vials, or if any cross contamination of samples is occurring during shipment
or storage of sample containers. Trip blanks are only analyzed for volatile organics.
b. Field Rinse Blanks
Rinse blanks consist of pouring demonstrated analyte free water over decontaminated
sampling equipment as a check that the decontamination procedure has been
adequately carried out and that there is no cross-contamination of samples occurring
due to the equipment itself. Analysis of rinse blanks is performed for all analytes of
interest. One blank should be collected for each type of equipment used each day a
decontamination event is carried out. It is required that rinse blanks be performed on
bowls and pans used to homogenize samples as well as on any filtration device used
on aqueous samples being analyzed for "dissolved" metals. It is permissible to use the
same aliquot of water on all equipment associated to a particular sample matrix for
analysis of semi-volatile organics, pesticides, RGBs and inorganics. This rinse must be
performed sequentially on all sampling equipment. However, a separate field rinse
blank must be collected for each piece of equipment associated to a particular sample
matrix which will be analyzed for volatile organics.
The blank should be collected at the beginning of the day prior to the sampling event
and that blank must accompany those samples which were taken that day. This is a
necessary procedure so that the blank will be associated with the proper samples
during data validation. If all samples collected that day are not validated with the field
rinse blank sample, it is the contractor's responsibility to ensure the application of the
blank's results to the group of samples. It is also the contractor's responsibility to
monitor the field rinse blank results over time in order to assess the performance of the
sampling team with respect to the adequacy of the decontamination procedure. This
will help reduce the number of samples needing reanalysis as well as the number of
-------
results being qualified and/or rejected due to contamination of the field rinse blank
sample.
3. Matrix Spike/Matrix Spike Duplicate Analyses
When performing CLP organic extractable analysis, the laboratory must be supplied
with triple sample volume for each Sample Delivery Group (SDG) in order to perform
matrix spike and matrix spike duplicate analyses. This does not include field or trip
blanks. Blanks do not require separate matrix spike or duplicate analyses regardless of
their matrix.
As stated in the SOW, the limits on an SDG are:
*each Case of field samples, or
*each 20 field samples within a Case, or
*each fourteen calendar day period during which field samples in a Case are
received (said period beginning with the receipt of the first sample in the SDG),
whichever comes first.
B. Dioxin
When dioxin sampling is performed a batch of quality control samples must accompany
environmental samples whether the samples are sent through the Contract Laboratory
Program or to a privately contracted laboratory.
The quality control sample requirements for every group of 24 dioxin samples collected
or every 24 samples collected over a period of two days, whichever comes first, are:
a. two performance evaluation samples from EMSL-LV fortified with a known amount
of native 2,3,7,8-TCDD (unknown to the laboratory);
b. one environmental duplicate - designated as a duplicate by the sampling team;
c. one blind blank (blind blank to the laboratory)-this sample does not go through the
hornogenization on-site;
d. one known blank - this is an uncontaminated soil/sediment and/or water sample
that is to be fortified by the lab with 2,3,7,8-TCDD at a concentration of 1 ug/kg for soils
and 10 ng/L for aqueous samples. This sample must be designated by the contractor
sampling team to the laboratory as "for spiking".
e. one blender blank-this is blank soil homogenized in the field to check for
cross-contamination during the blending process/and is only necessary if blenders are
used to homogenize the samples.
f. rinsate blank - rinsate blanks consist of pouring spectrophotometric grade
-------
trichloroethene (minimum 200 mis) over decontaminated sampling equipment as a
check that the decontamination procedure has been adequately carried out and that no
cross contamination of samples has occurred due to the equipment itself. Analysis of
rinsate blanks is performed for all isomers of interest.
Samples a, c, and d above will be ordered by the Regional Quality Assurance Officer
from EMSL-LV at the request of the EPA Project Manager only.
-------
XI. FILTERED AND NON-FILTERED FRACTIONS OF GROUND WATER
SAMPLES
A. General Discussion
In certain situations it may by desireable to consider filtration of ground water samples
in order to obtain information about the "dissolved" specie of the contaminant in solution
as opposed to that portion of contaminant which may adhere to silt or clay particles.
The real concern in this issue is whether or not the contaminant is a component of the
ground water, which implies that if the contaminant is truly a component of the ground
water, then it must flow with the ground water. If the contaminant flows with the ground
water on or under a site then there is the potential for the contaminant to be moved to
and from impact areas outside of the site.
There has been a general assumption that water and soil are the only distinct
constituents of an aquifer system; there is also a false assumption that water and
completely solvated solutes are the only constituents of the system that are mobile. In
fact, components of the solid phase in the colloidal size range may be mobile in
subsurface environments(11). The colloidal state refers to a two phase system in which
one phase in a very finely divided state is dispersed through a second. In ground water,
colloidal particles are generally smaller than one micrometer (1 um) in diameter. Since
the clay fraction is defined by the USDA as being 2 um and smaller, not all clay is
strictly colloidal, but even the larger clay particles have colloid-like properties (12).
There is ample evidence, as can be seen in the literature, that colloid particles can
move in aquifers (11).
There are two distinct types of colloidal matter, inorganic and organic, which exist in
intimate intermixture. The inorganic is present almost exclusively as clay minerals of
various kinds; the organic is represented by humus (12). These colloidal particles can
sorb organic and inorganic contaminants and stabilize them in the mobile phase of the
aquifer. Association of contaminants with mobile colloidal particles may enhance the
transport of highly adsorbed pollutants, or deposition of colloidal particles in porous
media may decrease permeability and reduce contaminant transport (11).
The separation of "dissolved" and "particulate" sample fractions is most commonly
accomplished by filtration through a 0.45 um membrane filter. The convention of using a
0.45 um pore size filter was borrowed from the microbiological science where it was
used as the separation point for filtering bacteria out of aqueous media. This convention
was borrowed by other organizations for use in the analysis of aqueous metals
samples, and now is specified as the pore size through which will pass those
constituents that are "dissolved". Those constituents which are retained on the 0.45 um
filter are labeled "suspended", and "total" metals is the sum of the "dissolved" and
"suspended" fractions.
-------
It would be more accurate, however, to term the fractions "filterable" and "non-filterable"
instead of "dissolved" and "total" or "particulate" given the operational nature of the
separation. The 0.45 um distinction is not useful when one is concerned with true
soil/water chemistry but has its sole value at present in the fact that it is convention and
is used as such in ground water characterization throughout the country.
The policy in Region II on filtering ground water has been that only samples for metals
analyses may be filtered, and when taking metals samples, "total" metals should be
taken with the option to take a "dissolved" sample if so desired. The filtration of
aqueous samples for organics analyses has not been allowed in the Region since,
1)volatile contaminants would be released during filtration, and 2)the membrane filters
used for the filtration of metals samples will adsorb the organics, thereby giving falsely
negative results.
The rationale behind the policy is this: rather than rely totally on "dissolved" metals data,
which will generally give results that are lower than the true amount of contaminant
which moves with ground water, the Region has chosen to be more conservative in its
use of metals data by preferring to consider "total" metals data, thereby erring on the
side of finding more metals in a sample than actually may be mobile in the ground water
phase. Obviously, in some cases when samples are silty, the "total" metals values will
be high due to the addition of metals which were bound to particles of greater size than
the colloidal range. Unfortunately the use of either "total" or "dissolved" metals data
alone is inappropriate when one wishes to consider the true portion and constituents
which move with ground water.
The Regional policy as presented here will continue to be enforced in spite of the
limitations until such time as a technically well-founded alternative is developed. If
exceptions or modifications to this policy are desired based on site specific needs, the
Project Manager should consult the Quality Assurance Officer assigned to the project.
B. Procedures for Filtration of Aqueous Metals Samples
1. Decontamination of Apparatus
When filtering aqueous metals samples, a device made of polyethylene, polypropylene
or borosilicate glass should be used. The apparatus should be pre-cleaned by rinsing
with a 10% HN03 solution, followed by a demonstrated analyte-free deionized water
rinse, and should be cleaned in the same manner between samples. Also, a field rinse
blank must be collected for this apparatus. See Section X for further details.
2. Filtration Procedures and Preservation
The filter used should be a cellulose-based membrane filter of 0.45 um nominal pore
-------
size. Samples must be filtered immediately after their collection to minimize changes in
the concentration of the substances of interest. Samples are only passed through the
filtration apparatus once, they are not to be passed through repeatedly until they are
free of turbidity. Samples are then preserved immediately with undiluted ultrapure
HNO3 and the pH checked to ensure proper pH has been attained. No samples for
cyanide, conventional parameters, or organics may be filtered in this manner. All
paperwork accompanying the samples to the laboratory should clearly state that the
samples have been field filtered, in order to avoid a second filtration at the lab.
-------
XII. LABORATORY QUALIFICATIONS
A. Use of CLP vs. non-CLP Laboratories
Most analytical work performed for Federal fund-lead CERLCA sites within the Region
utilizes the USEPA Contract Laboratory Program. However, it is not mandatory that all
analyses supporting the CERCLA program be performed by a Contract Laboratory,
whether the project is a fund-lead or enforcement-lead site. Laboratories which do not
participate in CLP may be used at any time, provided they adhere to Region II QA/QC
requirements which are described here.
If a non-CLP laboratory is used, that laboratory must supply to the Regional Quality
Assurance Officer (QAO), a copy of their in-house QA/QC manual which is applicable to
the analyses to be performed. The QA/QC manual should cover the following topics:
resumes
personnel training and experience
organizational structure
equipment available
reference materials/reagents
control charts
standard operating procedures
data reduction/reporting
chain-of-custody
glassware preparation
Also, that laboratory must perform acceptably on performance evaluation samples
supplied by the EPA or a State certified program for those parameters of interest to the
project. A formal request for EPA performance evaluation samples should be sent from
the EPA Project Manager for the site to the EPA QAO in MMB.
A non-CLP laboratory must also undergo a technical systems audit performed by the
party independent to the analysis in order to evaluate the laboratory's capability to.
perform the work. A copy of the resultant report should be sent directly to the EPA
QAO. A State audit report.'outlining the laboratory's performance within the last year,
will be acceptable. The format of the audit checklist can be taken from the CLP
Invitation for Bid (IFB). Agreement from the laboratory to perform these tasks must be
made before the plan is approved. Only after this information has been provided and
found to be acceptable, can sampling and analysis begin. The CLP IFBs are available
from the Sample Management Office at (703) 557-2490.
-------
XIII. USE OF MOBILE LABORATORIES
A. Qualifications and Methods
There has been a growing demand throughout the Region for use of field analytical
laboratories in order to screen samples and generate real-time results which can be
used to make decisions in the field. These laboratories commonly use "quick and dirty"
techniques since strict precision and accuracy requirements are not necessary for the
intended use of the data, and since, in most cases, critical sampling locations, where
quality data is important, are split for analysis by a CLP laboratory also.
However, there are certain quality control requirements for use of mobile laboratories.
First, the data quality objectives (DQOs) for the screening event must be determined
and documented. The DQOs should take into account the fact that "quick and dirty"
screening methods do not generally give high quality precision and accuracy and some
confirmational analysis with a CLP or other commercial laboratory is necessary.
Confirmational analysis should be run on those sample locations which are most critical
to the project, for instance, the boundary area of a removal action. Secondly, if a
methodology is developed for use, that method must be documented and validated.
Proof of the validation must be provided to the EPA QAO and the QAO must consider it
satisfactory before the method can be used. The method validation should address the
following points, where applicable.
a. analytical objectives
b. method detection limits
c. analytical procedure
d. precision and accuracy
e. calibration
f. quantitation
g. data reduction/validation
h. holding times
Finally, as for any data generated within the Region and which is not validated by
Regional personnel, the contractor's quality assurance officer must sign a summary
statement which describes the quality control measures followed, the quality control
sample results, what data was rejected due to exceedances, etc. This statement should
be supplied to the EPA Project Manager. .
-------
XIV. VALIDATION OF DATA
A. Contract Laboratory Program
All data generated for use by Region II which is produced by CLP is validated by the
Region with in-house protocol. These data validation standard operating procedures for
organics, inorganics and dioxin data are updated yearly for the current set of CLP
contracts. Application of a protocol which standardizes data useability criteria ensures
that all data which is used in the Region is of comparable and acceptable quality and
utility. The Region II data validation standard operating procedures are presented in
Appendix XIII.
B. Non-CLP
Data which is generated outside of CLP for use in Region II must be validated in the
same manner as all other data is validated so that a standard useability criteria is
applied to all data used in the Region. All data should be of comparable quality.
The Plan must identify the laboratory to be used if the laboratory is not to be engaged
through the CLP program or if the laboratory does not participate in CLP. All data
produced by laboratories which do not participate in CLP (or if they do participate but
are not directly contracted by EPA) must be validated by the laboratory or primary
engineering contractor according to Region II validation SOPs. Note that these SOPs
apply only to the CLP methodologies and that if different analytical method references
are used (such as SW-846 or Methods for Chemical Analysis of Water and Wastes) the
validation criteria will have to be modified according to the quality control criteria called
for in that methodology used. The QAO of the laboratory must be identified and a
detailed discussion of the validation criteria to be used when data is generated must be
provided in the Plan.
After data is generated a validation report must be provided to the EPA Quality
Assurance Officer in MMB assigned to the site. This will ensure that the data has been
validated in accordance with the Region II protocol, or, if quality control criteria had to
be established according to the dictates of the method, the party performing the
validation is responsible for establishing precision and accuracy protocol and for
validating the data and meeting criteria based on that protocol. Quality assurance
summary sheets, and, if applicable, the Region II SOP validation checklist should be
provided to the Project Manager. The summary sheets should be taken from the CLP
SOW, the third edition of SW-846 or be based thereon. Data analysis sheets must be
provided for each environmental sample listing quantities found or detection limits.
If the Project Manager wishes to have the validation of data checked by the MMB staff,
or if upon review of the validation report provided a revalidation of a certain percentage
of data is deemed necessary, CLP deliverables must be made available. This may
mean notifying the PRP in advance that CLP deliverables may be requested from their
-------
laboratory at a future date.
-------
XV. FIELD AUDITING AND OVERSIGHT
A. Audits Initiated by EPA and Primary Contractor
On-site audits of EPA contracted and PRP contracted field sampling teams takes place
on a random basis within the Region. EPA personnel performing the audits look for
good sampling technique and ensure that approved Plans are being implemented in the
field. Auditors do make suggestions to contractors in the field but they do not stop work
unless a discrepancy severe enough to invalidate data results is observed. If a severe
discrepancy is observed the Project Manager is notified by phone when the auditors
return to the office that day.
EPA personnel send written audit reports to the EPA Project Manager following the
audit and request written response from the contractor when inadequacies are found.
B. Contractors in an Oversight Capacity
Contractors retained in an oversight capacity should be looking for good sampling
technique and adherance to an approved Plan when overseeing other contractors in the
field. Logbooks should be kept by the oversight contractor and any poor practices or
discrepancies with the Plan should be noted and the EPA Project Manager notified of
the findings by phone.
C. Audits Performed by States
State Quality Assurance personnel routinely perform audits of Federal-lead sites.
Although State auditors can make suggestions to improve sampling technique and to
ensure that approved Plans are being implemented, they do not have the authority to
stop work in the field or to change any part of the Plan. If a disagreement with State
auditors arises in the field, the contractor should contact the Project Manager for
guidance.
-------
BIBLIOGRAPHY
1 . Characterization of Hazardous Waste Sites-A Methods Manual: Volume II. Available
Sampling Methods, Second Edition; EPA 600/4-84-076; December 1984.
2. Handbook for Sampling and Sample Preservation of Water and Wastewater; EPA
600/4-82-029; September 1982.
3. Samplers and Sampling Procedures for Hazardous Waste Streams; EPA 600/2-
80-018; January 1980.
4. Practical Guide to Ground Water Sampling: EPA 600/2-84/104; September 1984.
5. RCRA Ground Water Monitoring Technical Enforcement Guidance Document,
September 1986. Office of Waste Programs Enforcement and Office of Solid Waste
and Emergency Response.
6. A Guide to the Selection of Materials for Monitoring Well Construction and Ground
VVater Sampling. Barcelona, Gibb, Miller; Illinois State Water Survey, Champaign,
Illinois. January 1984.
7. Test Methods for Evaluating Solid Waste SW-846, Third Edition. Office of Solid
Waste and Emergency Response. November 1986.
8. Methods for Chemical Analysis of Water and Wastes. EPA 600/4-79-020. Revised
March 1983.
9. NEIC Policies and Procedures. United States Environmental Protection Agency
Office of Enforcement and Compliance Monitoring. EPA 330/9-78-00 1-R. May 1978,
revised May 1986.
10. Ground Water and Wells. Driscoll, Fletcher G. Second Ed. 1986. Johnson Division,
5t. Paul, Minnesota, 55112.
11. Summary RepoitTransport of Contaminants in the Subsurface: The Role of
Organic and Inorganic Colloidal Particles, ISIS Seminar, USDOE, October 6-9, 1986,
Ivlanteo, NC.
12. The Nature and Properties of Soils, Nyle C. Brady, MacMillan Publishing Co., NY,
Edition, 1984.
1 3. User's Guide to the Contract Laboratory Program, USEPA Office of Emergency and
Remedial Response, December 1988.
-------
LIST OF APPENDICES
I. "Interim Guidelines and Specifications for Preparing QA Project Plans",
QAMS-005/80, December 29,1980; "Guidance for Preparation of Combined
Work/Quality Assurance Project Plans for Water Monitoring", OWRS-1, May 1984.
II. Priority Pollutant vs. Target Compound List
III. CLP Statement of Work Target Compound/Element Lists
IV. 40 CFR Part 136, Table 2
CLP and Non-CLP Holding Time and Preservation Requirements
V. CLP Documentation
VI. Quality Assurance Procedures for Field Analysis and Equipment
VII. Sample Bottle Repository Statement of Work
VIII. Recommended Well Drilling Techniques
IX. "The Effects of Grouts, Sealants, and Drilling Fluids on the Quality of Ground
Water Samples", K. Jennings
X. Region II Standard Operating Procedure for Selecting Ground Water Well
Construction Material at CERCLA Sites
XI. Comparison of Surface Water Collection Devices
XII. Comparison of Bottom Grab and Coring Devices
XIII. Region II Data Validation Standard Operating Procedures
-------
APPENDIX I
-------
APPENDIX II
-------
APPENDIX III
-------
APPENDIX IV
-------
APPENDIX V
-------
APPENDIX VI
-------
APPENDIX VII
-------
APPENDIX VIII
-------
APPENDIX IX
-------
APPENDIX X
-------
APPENDIX XI
-------
APPENDIX XII
-------
APPENDIX XIII
-------
REGION II
CERCLA QUALITY ASSURANCE MANUAL
PART II: QUALITY CONTROL HANDBOOK FOR CERCLA SAMPLING AND
ANALYSIS
I. Forward
A. Intent
B. Disclaimer
II. Sampling Design and Strategy
A. Sampling Plan Components
B. Purpose and Objective of Sampling
C. Types of Samples
1. Environmental
2. Hazardous
D. Types of Measurement
1. Laboratory Measurement
a. Grab
b. Composite
2. In-Situ Measurement
III. Analytical Methods, Preservation and Holding Times
A. Methodology Available for use in CERCLA Program
1. Aqueous/Solid Matrices
2. Air
3. NAPL
B. Analytical References
C. Preservation, Methodology and Holding Times
IV. Documentation Procedures
A. Chain-of-Custody
1. Definition and Reference
2. Recordkeeping and Procedures
a. General
b.CLP
B. Field Records
V. Glassware Requirements
A. Bottle Suppliers
B. Volume and Type of Container
C. Quality Control and Storage
VI. Field/Laboratory Decontamination of Sampling Apparatus ...
-------
A. General Considerations
B. Decontamination Procedures
VII. Decontamination of Peripheral Equipment
A. Well Evacuation Equipment
B. Well Casings
C. Field Instrumentation
D. Drilling Equipment
VIII. Monitoring Well Design and Construction
A. Well Drilling and Development Methods
1. General Discussion and Preferred Methods
B. Well Filterpack and Annular Sealant
C. Well Casing Selection
1. General Discussion
2. Selection Criteria SOP
D. Evaluation of Existing Wells
IX. Sample Collection Devices, Materials and Quality Control Practices
A. References for Selection of a Sampling Device
B. Ground water
1. Sampling Design
2. Well Evacuation
3. Sampling Organics and Inorganics
4. Sampling Microbiological
C. Surface water
1. Sampling Design
2. Sampling Devices
D. Sediment
1. Sampling Design
2. Sampling Devices
E. Soil
1. Sampling Design
2. Sampling Devices
F. Potable Water
G. Dust/Wipes
H. Dioxin
I. Drums
X. Methods of Sample Preparation
A. Homogenization
B. Compositing
C. Splitting
XI. Field Quality Control Samples
A. All Analyses Except Dioxin
-------
1. Duplicates
2. Blanks
a. Trip
b. Field
3. Matrix Spike/Matrix Spike Duplicate Analyses
B. Dioxin
XII. Filtered and Non-Filtered Fractions of Ground Water Samples
A. General Discussion
B. Procedures for Filtration of Aqueous Metals Samples
1. Decontamination of Apparatus
2. Filtration Procedure and Preservation
XIII. Laboratory Qualifications
A. Use of CLP vs. non-CLP Laboratories
XIV. Use of Mobile Laboratories
A. Qualifications and Methods
XV. Validation of Data
A. CLP
B. non-CLP
XVI. Field Auditing and Oversight
A. Audits Initiated by EPA and Primary Contractor
B. Contractors in an Oversight Capacity
C. Audits Performed by States
-------
I. FORWARD
A. Intent
This CERCLA Quality Assurance Manual has been prepared by the Monitoring
Management Branch of the Environmental Services Division for use by the Emergency
and Remedial Response Division Project Managers of the Region II Superfund
(CERCLA) program in their daily working with contractors. It's use is intended to ensure
that quality assurance and quality control practices (QA/QC) are fully built into all
monitoring project designs.
This Manual provides Region M's quality assurance/quality control requirements for
CERCLA sampling and analysis. Quality assurance procedures are used to verify that
field and laboratory measurement systems operate within acceptable, defined limits.
The effectiveness of the overall Quality Assurance Program demands that all personnel
are aware of the QA/QC requirements for any investigation and that the quality
assurance objectives are understood. This Volume outlines of all aspects of QC in a
monitoring program, minimum requirements necessary, and the rationale behind the
requirements. This Manual is meant to be a dynamic document. It will periodically be
reviewed and updated, however it is not meant to provide definitive answers to all
site-specific concerns. This is, rather, an attempt to provide a rationale behind the most
common site-specific concerns which could be extrapolated for use in new situations.
The quality control procedures outlined in this Manual should be incorporated into all
field/project/site operations plans and/or quality assurance project plans prepared for
CERCLA work. Recommendations and requirements presented herein should be
incorporated into project designs to the fullest extent possible. Where deviations from
these recommendations and requirements is necessary, full justification must be
presented in writing.
-------
B. Disclaimer
This Manual has been prepared for use by the Environmental Services and the
Emergency and Remedial Response Divisions of the USEPA, Region II. Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use by the Agency.
-------
II. SAMPLING DESIGN AND STRATEGY
A. Sampling Plan Components
Detail of sampling and analysis is a necessary part of each field/site/project operations
or quality assurance project plan (from hereon referred to as the Plan) in order to
ensure uniform and acceptable sampling and analytical protocol for each project. The
plan describes the objectives and details how the individual tasks of a sampling and
analytical effort will be performed. The Plan must include the following topics as a
minimum.
* Objectives of sampling design and selection of representative sampling sites.
Discussion of site history and sampling design rationale must be provided, so that
reviewers of the Plan have the necessary information. The discussion should include
topics such as the history of the contamination, the matrices involved, the dimensions of
the site, etc.
* Sampling Design
* Selection of Parameters to be measured.
parameters to be measured are usually dictated by the purpose of an investigation
and should be based on knowledge of the problem being investigated. An in-depth
discussion of parameter selection is out of the scope of this document as it is a process
requiring much background and expertise in dealing with hydrogeologic systems,
chemistry, and engineering, and no criteria for parameters can be put forth in the format
of an SOP.
* Selection and preparation of sampling equipment.
* Sampling equipment construction materials.
* Required sample volumes.
* Selection and preparation of sample containers.
* Sample collection and handling.
* Sample preservation.
* Sample chain-of-custody and identification.
* Use of field instrumentation.
* Field quality control samples.
-------
* Choice of laboratories and validation of data.
These topics are further discussed within this Manual.
B. Purpose and Objective of Sampling
The basic objective of any sampling campaign is to collect a sample which is
representative of the media under investigation. More specifically, the purpose of
sampling at hazardous waste sites is to acquire information that will aid investigators in
determining the presence and identity of on-site contaminants and the extent to which
they have become integrated into the surrounding environment. This information can
then be used as support for future litigations or as input to remedial investigations and
risk assessments.
The validity of environmental data is dependent in part on the integrity of the field
procedures employed in obtaining a sample. Proper sampling techniques must be
employed to obtain a sample which is representative of the area or container of interest.
A sample is representative if it possesses the same qualities or properties as the
material under consideration. Due to the complexity of most hazardous substances and
site conditions, no universal sampling methods can be recommended. Procedures must
be adapted for use in various matrices and site-specific restrictions.
C. Types of Samples
Before defining the general sample types, the nature of the media or materials under
investigation must be discussed. Materials can be described as homogeneous or
heterogeneous. Homogenous materials are generally defined as having uniform
composition throughout. In this case, any sample increment can be considered
representative of the material. On the other hand, heterogeneous samples present
problems to the sampler because of changes in composition of the material over
distance and time.
When discussing types of samples, it is important to distinguish between the type of
media to be sampled and the sampling technique that yields a specific type of sample.
In relation to the media to be sampled, two basic types of samples can be considered:
the environmental sample and the hazardous sample.
1. Environmental Samples
Environmental Samples (ambient air, soils, surface water, groundwater, sediment or
biota) are generally dilute (in terms of pollutant concentration) and usually do not
require the special handling procedures used for concentrated wastes. However, in
certain instances, environmental samples can contain elevated concentrations of
pollutants and in such cases would have to be handled as hazardous samples.
-------
2. Hazardous Samples
Hazardous or concentrated samples are those collected from drums, tanks, lagoons,
pits, waste piles, fresh spills, etc., and require special handling procedures because of
their potential toxicity or hazard. These samples can be further subdivided based on
their degree of hazard; however, care should be taken when handling any wastes
believed to be concentrated, regardless of the degree.
D. Types of Measurement
In general, two basic types of sample measurements are recognized, both of which can
be used for either environmental or concentrated samples. They are: 1) samples which
are collected and subsequently analyzed in the laboratory and, 2) samples which are
analyzed in-situ.
1. Laboratory Measurement
There are two types of samples which are collected and analyzed in a laboratory. These
are grab samples and composite samples.
a. Grab Samples
A grab sample is defined as a single sample representative of a specific location at a
given point in time. The sample is collected all at once and at one particular point in the
sample medium. The representativeness of such samples is defined by the nature of
the materials being sampled. In general, as sources vary over time and distance, the
representativeness of grab samples will decrease.
b. Composite Samples
Composites are combinations of more than one sample collected at various sampling
locations and/or different points in time. Analysis of composite yields an average value
and can, in certain instances, be used as an alternative to analyzing a number of
individual grab samples and calculating an average value. It should be noted, however,
that compositing can mask problems by diluting isolated concentrations of some
hazardous compounds below detection limits.
For sampling situations involving hazardous wastes, grab sampling techniques are
generally preferred because grab sampling minimizes the amount of time sampling
personnel must be in contact with the wastes, reduces risks associated with
compositing unknowns, and eliminates chemical changes that might occur due to
compositing. Compositing is often used for environ-mental samples, including dioxin
samples, to determine vertical or horizontal spatial variability of parameters. This
procedure provides data that can be useful by providing an average concentration over
a number of locations and can serve to keep analytical costs down; however, it is
-------
important to understand that sensitivity is sacrificed when samples are composited due
to dilution of individual grab samples. If contamination occurs in "hot" spots on site and
"hot" grabs are composited with clean samples, a true vertical or horizontal distribution
of contamination will appear to be a uniform distribution at a level lower than the true
value of any one individual component(7). This is especially a concern when doing
dioxin sampling with an action level of 1 ppb. Information on methods of compositing
are presented in Section X.
2. In-Situ Measurement
In-situ measurements are made on samples in the environment. Measurements for pH,
conductivity, and temperature must be taken in the field. Use of instrumentation such as
an OVA, HNu and other gas analyzers is a|so considered in-situ measurement.
-------
III. ANALYTICAL METHODS, PRESERVATION AND HOLDING TIMES
A. Methodology Available for use in the CERCLA Program
The CERCLA program has no legally mandated analytical methods. Methods from
other programs or methods which are proven to be scientifically valid can be used for
CERCLA work. The bulk of the analytical methods used presently are, however, from
the Contract Laboratory Program (CLP).
The Contract Laboratory Program supports the Agency's Superfund effort by providing
a range of chemical analytical services on a high volume, cost effective basis. Its
purpose is to provide legally defensible analytical data. The program is managed by the
National Program OfficeMn Headquarters, and the Contractor-operated Sample
Management Office receives the analytical requests from the Regions and coordinates
and schedules sample analyses. Analytical Statements of Work exist for organics,
inorganics and dioxin in water and soil\sediment matrices.
In addition to standardized analyses provided under the Routine Analytical Services
(RAS) program, the CLP's Special Analytical Services (SAS) program provides clients
with limited customized or specialized analyses, different from or beyond the scope of
the RAS contract protocols. Services provided by SAS include: quick turn around
analyses, verification analyses, analyses requiring lower detection limits than RAS
methods provide, identification and quantification of non-Target Compound List (TCL)
constituents, general waste characterizations, analysis of non-standard matrices, and
other specific analyses. Consult the "User's Guide to the Contract Laboratory Program",
Oecember 1986 for further information.
As stated above, the CLP parameters of interest for RAS were titled under the 10/86
Statement of Work (SOW) the "Target Compound List". Under previous SOWs the TCL
was titled the "Hazardous Substance List". Neither of these lists has been published in
the Federal Register (FR) or Code of Federal Regulations (CFR) and thus are strictly
lists defined by contract.
The "Priority Pollutant List" was established in the consent decree of the Natural
Resources Defense Council (NRDC) vs. Train, in 1976. Although not published in the
Federal Register under that title, it was published in a more generalized form in 44 FR
44502, July 30,1979 as the "Toxic Pollutants" list under the Federal Water Pollution
Control Act, and was amended in 46 FR 2266, January 8,1981, and 46 FR 10724,
February 4,1981. Thus, the "Priority Pollutant" list as it now stands is comprised of 126
compounds or elements.
The Priority Pollutant and Target Compound Lists are presented in Appendix I to
which shows the differences between the two. They are not interchangeable terms;
neither list is a subset of the other, both contain compounds not found on the other. The
only Priority Pollutants which are not on TCL are: asbestos, benzidine,
1,2-diphenylhydrazine, N-nitrosodimethylarhine, endrin aldehyde, 2-chloroethyl vinyl
-------
ether, acrolein, and acrylonitrile. Refer to the Appendix for those compounds which are
on TCL but not considered Priority Pollutants.
A discussion of available references, methods, holding times and preservatives follows.
B. Analytical References
The following is a listing of seleceted analytical references containing methods available
for use in the CERCLA program.
1. Aqueous/solid matrices
1. 40 CFR Part 136, as updated yearly.
2. 7/87 CLP Statement of Work for Inorganics, the 10/86 Statement of Work for Dioxin,
the 1/87 revision of the Organics Statement of Work, and as updated.
<_
The CLP standardized organic analytical methods are based on the CFR methods 608,
613,624, and 625 modified for use in the analysis of both water and soil matrices. The
standardized inorganic analytical methods are based on FR methods, EPA Methods for
Chemical Analysis of Water and Wastes (MCAWW), and Test Methods for Evaluating
Solid Waste (SW-846), Third Edition, or as revised, for the analysis of water and soil
matrices. Appendix II provides a listing of the CLP organic and inorganic Targeted
Compound/Element Lists as taken from the most recent SOWs and includes the RAS
detection limits.
The dioxin Routine Analytical Services (RAS) contract method determines the presence
of the 2,3,7,8-tetrachloro-dibenzo-p-dioxin isomer in water and soil/sediment matrices.
3. Standard Methods, 15th and 16th eds., or as revised.
4. Methods for Chemical Analysis of Water and Wastes(MCAWW), Revised 1983, EPA
600/4-79-020.
5. American Society for Testing and Materials.
6. Test Methods for Evaluating Solid Waste-SW-846, Third Edition, November 1986.
7. Procedures for Handling and Chemical Analysis of Sediment and Water Samples,
May 1981, Technical Report CE/81-1,NTIS#AD-A103788.
2. Air :
1. Compendium of Methods for the Determination of Toxic Organic Compounds in
Ambient Air, April 1984. EPA 600/4-84-041.
-------
2. SOP for the GC/MS Determination of Volatile Organic Compounds Collected
on Tenax, June 1984. EMSL/RTP-SOP-EMD-020.
3. Non-Aqueous Phase Liquids
1. Test Methods for Evaluating Solid Waste, SW-846, Third Edition, November
1986.
2. Interim Methods for the Measurement of Organic Priority Pollutants in Sludges,
Revised Draft June 1980.
3. Determination of Polychlorinated Biphenyls in Transformer Fluid and Waste
Oils, Sept. 1982. EPA 600/4-81-045.
C. Preservation, Methodology and Holding Times
Complete and unequivocal preservation of samples is a practical impossibility.
Regardless of the nature of the sample, complete stability for every constituent can
never be achieved. At best, preservation techniques can only retard the chemical and
biological changes that inevitably continue after the sample is removed from the parent
source. The changes that take place in a sample are either chemical or biological. In
the former case, certain changes occur in the chemical structure of the constituents that
are a function of physical conditions. Metal cations may precipitate as hydroxides or
form complexes with other constituents; cations or anions may change valence states
under certain reducing or oxidizing conditions; other constituents may dissolve or
volatilize with the passage of time. Metal cations such as iron and lead may also adsorb
onto surfaces (glass, plastic, quartz, etc.). Biological changes taking place in a sample
may change the valence of an element or a radical. Soluble constituents may be
converted to organically bound materials in cell structures, or cell lysis may result in
release of cellular material into solution. The well known nitrogen and phosphorus
cycles are examples of biological influence on sample composition.
Methods of preservation are relatively limited and are intended generally to (1) retard
biological action, (2) retard hydrolysis of chemical compounds and complexes, (3)
reduce volatility of constituents, and (4) reduce absorption effects. Preservation
methods are generally limited to pH control, chemical addition, refrigeration and
freezing(8).
Appendix III contains a copy of Table 2 from 40 CFR Part 136, July 1,1987 and a
tabular presentation of the CLP holding time and preservation requirements. The Table
is comprised of approved "conventional" parameter (meaning those analyses not
considered part of the most commonly used "organic" and "inorganic" sets of analyses)
methodology for an aqueous matrix. The holding times and preservation requirements
of these conventional parameters are to be followed.
Appendix III also includes a table of CLP holding time and preservation requirements.
-------
These requirements are to be followed not only for CLP methodology, but also for 40
CFR Part 136 and SW-846 methodology used for organic and inorganic analyses of
aqueous or solid matrices. Holding times begin from the time of sample collection
unless the Contract Laboratory Program has been engaged, in which case holding
times begin at the Verified Time of Sample Receipt (VTSR) due to contractual
requirements.
Recent studies including a project funded by EPA and the Department of Defense and
performed by the Oak Ridge National Labortory determined that aqueous volatile
organic samples (VOAs) could be held for extended periods of time with preservation
with hydrochloric acid (HCI) to pH less than 2 without significant loss of constituents.
Thus, it is now a requirement that all samples taken for volatile organics analysis be
preserved with hydrochloric acid to pH less than 2. This applies to samples analyzed by
CLP or any other laboratory using any accepted methodology. Note, however, that this
does not change the CLP holding time requirements.
The following procedure, adapted from the drinking water methods should be used for
acidification of volatile organic samples with HCI to a pH less than 2.
Adjust the pH of the sample to <2 by carefully adding 1:1 HCI drop
by drop to the required 2 (40 ml) VOA sample vials. The number of
drops of 1:1 HCI required should be determined on a third portion
of sample water of equal volume.
It should be noted that if acidification of the sample causes effervescence, the sample
should be submitted without preservation except for cooling to 4 degrees C. This
sample property should be appropriately noted when present. Also, the 1:1 HCI solution
should be made up with demonstrated analyte-free deionized water.
When samples are to be iced to 4 degrees C., it is intended that the sample bottle be
surrounded by bags of ice to ensure that the proper temperature is achieved and
maintained during transport. It is not acceptable to put bags of ice around only the
necks of the bottle or to use "Blue Ice" since the use of these devices does not ensure
the attainment of the proper temperature.
When booking samples through CLP SAS, a copy of the SAS request must be attached
to the parameter table in the Plan.
-------
IV. DOCUMENTATION PROCEDURES
A. Chain-of-Custody
1. Definition and Reference
According to the USEPA Office of Enforcement and Compliance Monitoring National
Enforcement Investigations Center (NEIC) Policies and Procedures, May 1978 revised
May 1986, a sample is under custody if:
1. it is in your possession, or
2. it is in your view, after being in your possession, or
3. it was in your possession and you locked it up, or
4. it is in a designated secure area.
Possession must be traceable from the time the samples are collected.
2. Recordkeeping and Procedures
a. General
The method of sample identification utilized depends on the type of sample collected.
In-situ field analyses are those conducted for specific field analyses or measurements
where the data are recorded directly in bound field logbooks or recorded directly on the
chain-of-custody record, with identifying information, while in the custody of the
sampling team. Examples of such in-situ field measurements and analyses include pH,
temperature, and conductivity. Also included in this category are those field
measurements or analyses such as flow measurements, geophysical measurements,
surveying measurements, etc. that are made with field instruments or analyzers, where
no sample is actually collected.
Samples, other than those collected for in-situ field measurements or analyses, are
identified by using a standard sample tag which is attached to the sample container. In
some cases, particularly with biological samples, the sample tag may have to be
included with or wrapped around the sample and waterproofed. The sample tags are
sequentially numbered and are accountable documents after they are completed and
attached to a sample or other physical evidence. The following information shall be
included on the sample tag:
a. site name
b. fie Id identification or sample station number
c. date and timeof sample collection
d. designation of the sample as a grab or composite
e. type of sample (matrix), and a brief description of the sampling location
f. the signature of the sampler
-------
g. whether the sample is preserved or unpreserved
h. the general types of analyses to be conducted
If a sample is split with another party, sample tags with identical information shall be
attached to each of the sample containers.
The chain of custody record is used to record the custody of samples. It must
accompany samples at all times. The following information must be supplied to
complete the chain of custody record:
a. project name
b. signature of samplers
c. sampling station number, date and time of collection, grab or composite
sample designation, and a brief description of the type of sample and sampling location,
d. tag numbers
e. signatures of individuals involved in sample transfer, i.e., relinquishing and
accepting samples. Individuals receiving the samples shall sign, date and note the
time that they received the samples on the form.
Sample analysis request sheets serve as official communication to the laboratory of the
particular analyses required for each sample and provide further evidence that the chain
of custody is complete.
Shipping containers should be secured to ensure samples have not been disturbed
during transport by using nylon strapping tape and EPA custody seals. The custody
seals should be placed on the container so that it cannot be opened without breaking
the seal.
b. CLP
The CLP documentation system provides the means not only to track and identify each
sample, but to support the use of sample data in potential enforcement actions.
Appendix IV provides copies of CLP documentation described below.
A sample Traffic Report (TR), which has a unique sample identification number, is
assigned to each sample collected. An adhesive sample label printed with the TR
sample number is affixed to each container, and, in order to protect the label from water
and solvent attack, each label is covered with clear waterproof tape. The sample labels,
which bear the TR identification number, permanently identify each sample collected
and link each sample component throughout the analytical process. A custody seal is
then placed over the lid of the container to ensure the samples are not opened prior to
arrival at the laboratory.
Sample documentation for the RAS dioxin program utilizes the CLP Dioxin Shipment
Record (DSR) and samples are individually numbered using pre- printed labels.
-------
For SAS samples, a SAS Packing List (PL) is used along with adhesive sample labels.
Sample tags, containing the necessary information as required by NEIC, are attached to
each sample container at the time of collection. Following sample analysis, sample tags
are retained by the laboratory as physical evidence of sample receipt and analysis.
The Chain-of-Custody Record is employed as physical evidence of sample custody.
One Record accompanies each cooler shipped from the field to the laboratory. In
Region II, the Environmental Services Division Chain-of- Custody Record is used.
Shipping coolers are secured and custody seals placed across cooler openings. As long
as custody forms are sealed inside the sample cooler and the custody seals remain
intact, commercial carriers are not required to sign off on the custody form.
Whenever samples are split with a source or government agency, a separate
Chain-of-Custody Record should be prepared for those samples, indicating with whom
the samples are being split and sample tag serial numbers from splits (13).
Information regarding the information contained within, completion of, or obtaining these
forms can be found in the CLP User's Guide available from the Region II RSCC.
B. Field Records
Field records should be kept by contractor personnel for each site. All aspects of
sample collection and handling as well as visual observations should be documented in
the logbooks. The following information should be recorded:
1. sample collection equipment;
2. field analytical equipment;
3. any other equipment used to make field measurements;
4. calculations;
5. results, and;
6. calibration data for equipment.
AH entries should be dated and initialed and must be legible (9).
All maintenance and calibration records for equipment should be traceable through field
records to the person using the instrument and to the specific piece of instrumentation
itself. Equipment should be labeled with the calibration date and when it is due for the
next calibration. The calibration of the pH and conductivity meters must be checked
daily. Appendix V describes the required quality assurance procedures for field analysis
and equipment.
Standard operating procedures (SOPs) for use of any field instrumentation must be
provided in the form of a manual or individually in the Plan itself. The SOPs should
address calibration, maintenance and use of the instrumentation and should reflect
what is currently being done in the field.
-------
V. GLASSWARE REQUIREMENTS
A. Bottle Suppliers
The CLP Sample Bottle Repository (SBR) provides cleaned, contaminant-free sample
containers for use by groups performing hazardous waste sample collection activities
under the Superfund program. Within this contract, sample containers are cleaned by
defined procedures and representative containers undergo strict quality testing prior to
shipment. This contributes to the integrity of sample data and supports its viability for
use in enforcement case actions. The contractor uses approved techniques and
instrumentation to procure, prepare, clean, label, store, package and ship sample
containers and component materials. Appendix VI is the Statement of Work for
Maintenance of a Quality-Controlled Prepared Sample Container Repository, dated
4/87, revised 7/87 and 8/87, in which the specific requirements for quality control are
delineated.
It is the policy of this office at this time that bottles supplied by any party performing
Superfund work who does not obtain those bottles from the SBR must be cleaned and
quality controlled in the same manner as is defined in the SBR SOW. Therefore, if
containers are not being procured from the Repository, the container construction,
cleaning and quality controlling must be the same as that described in the SOW
presented in Appendix VI. A statement that the bottle supplier will follow the SBR SOW
must be included in the Plan.
B. Volume and Type of Container
The volume of sample obtained should be sufficient to perform all required analyses
with an additional amount collected to provide for quality control needs, split samples, or
repeat analyses.
The sample container requirements may be found in the SBR SOW, the CLP User's
Guide, or in 40 CFR Part 136.
C. Quality Control and Storage
As stated above, the SBR Statement of Work must be followed when it comes to
procuring, preparing, cleaning, labeling, storing and quality controlling containers. This
involves analysis/testing of one or more representative containers from each lot or
batch after they have been cleaned and designation of a storage QC container for
future analysis if contamination should be suspected at a later time. All storage QC
containers should be kept in a separate contaminant-free area. See Appendix V for
detail. Contractors who store containers for any period of time must also comply with
storage requirements.
-------
VI. FIELD/LABORATORY DECONTAMINATION OF SAMPLING APPARATUS
A. General Considerations
All sampling apparatus must be properly decontaminated prior to its use in the field to
prevent cross-contamination. The equipment should be pre- cleaned in a laboratory
situation, or if the duration of the sampling event prohibits pre-cleaning in a lab, then
equipment should be decontaminated once a day in an area outside of the
contaminated zone. Enough equipment must be available to be dedicated to sampling
points each day.
B. Decontamination Procedures
The required decontamination procedure for all sampling equipment is:
a. wash and scrub with low phosphate detergent
b. tap water rinse
c. rinse with 10% HNO3, ultrapure
d. tap water rinse i
e.an acetone rinse or a methanol followed by hexane rinse (solvents must be
pesticide grade or better)
f. deionized demonstrated analyte free water rinse
g. air dry, and
h. wrap in aluminum foil, shiny side out, for transport
Tap water may be used from any municipal water treatment system. The use of an
untreated potable water supply is not an acceptable substitute. If metals samples are
not being collected, the 10% nitric acid (HN03) rinse may be omitted, and, conversely,
if organics samples are not being taken, the solvent rinse may be omitted.
When it is necessary to use split spoon sampling devices which are composed of
carbon steel instead of stainless steel, the nitric acid rinse may be lowered to a
concentration of 1% instead of 10% so as to reduce the possibility of leaching metals
from the spoon itself.
Bailer cord must be cleaned with soap and deionized water before use. Cord can be
reused; it is not necessary to dedicate it to individual wells. If a ten foot or greater length
leader (any cord of unacceptable material may not contact the water) is being used,
only the leader need be cleaned. See Section IX for acceptable bailer cord materials.
-------
VII. DECONTAMINATION OF PERIPHERAL EQUIPMENT
A. Well Evacuation Equipment
AH tubing and evacuation equipment such as submersible pumps which are put into the
borehole must be rinsed with soapy water and deionized water before use. All tubing
must be dedicated to individual wells, i.e., tubing cannot be reused. If bailers are used
to evacuate wells they must be decontaminated with the same procedure listed in
Section VI.
B. Well Casings
Well casings must be steam cleaned prior to installation to ensure that all oils, greases,
and waxes have been removed. Because of the softness of casings and screens made
of fluorocarbon resins, these materials should be detergent washed, not steam cleaned
prior to installation. They should be rested on clean polyethylene sheeting to keep the
possibility of contamination to a minimum.
C. Field Instrumentation
Instrumentation should be cleaned as per manufacturer's instructions. Probes such as
those used in pH and conductivity meters must be rinsed after each use with deionized
water.
D. Drilling Equipment and Other Large Pieces of Equipment
All drilling equipment that comes in contact with the soil must be steam cleaned before
use and between boreholes. This includes drill rods, bits and augers, dredges, or any
other large piece of equipment. Sampling devices such as split spoons and shelby
tubes must be decontaminated as per Section VI between boreholes.
-------
VIII. MONITORING WELL DESIGN AND CONSTRUCTION
A. Well Drilling and Development Methods
1. General Discussion and Preferred Methods
There are various well drilling methods for application in geologic conditions ranging
from hard rock to unconsolidated sediments. Particular drilling methods have become
dominant in certain areas because they are most effective in penetrating the local
formation.
The recommended types of ground water well drilling techniques are presented in
Appendix VII. This list has been adopted from "The Practical Guide to Ground Water
Sampling" (4). The most widely used method of drilling in the Region is the mud rotary
technique, however, as presented in Appendix VII, "The Effects of Grouts, Sealants,
and Drilling Fluids on the Quality of Ground-Water Samples," drilling fluids and additives
may introduce contamination into the well which may persist even after development
and can affect the chemical and biological quality of the samples. The use of mud
rotary is not recommended, particularly for investigation of organic contaminant
situations. Where possible, hollow stem auger, cable tool, or air rotary should be used
to install wells. Where fluid rotary methods are employed, clean water should be used,
and fluids carefully controlled to minimize impact on the ground water system. Vigorous
development and purging should be employed to remove the filtrable solid residue from
the wall cake and invasive filtrate (10).
Procedures designed to maximize well yield are included in the term "well
development". Development has two broad objectives: 1) repair damage done to the
formation by the drilling operation so that the natural hydraulic properties are restored,
and 2) alter the basic physical characteristics of the aquifer near the borehole so that
water will flow more freely to a well. These objectives are accomplished by applying
some form of energy to the screen and formation (10). More importantly wells must be
developed to provide water free of suspended solids for sampling. Improperly
developed monitoring wells will produce samples containing suspended sediments that
will bias the chemical analysis of the collected samples (4).
The first step in well development involves the movement of water at alternately high
and low velocity into and out of the wellscreened gravel pack to break down the mud
pack on the well bore and loosen fines in the materials being monitored. This step is
followed by pumping to remove these materials from the well and the immediate area
outside the weir screen. This procedure should be continued until the water pumped
from the well is visually free of suspended materials or sediments (4). Methods of
development include overpumping, backwashing, mechanical surging, high velocity
jetting, and air development procedures (10). Of these methods, high velocity jetting
and air development procedures are unacceptable without modification.
-------
High velocity jetting involves the use of a horizontal water or air stream forced through
the well screen to agitate and rearrange the particles surrounding the screen. Although
this is an effective method of development, its major disadvantage is the introduction of
either air or water into the formation. Water jetting is acceptable only if the water used
has a controlled source so cross contamination does not occur. If potable water is used
for water jetting development, analysis of a water blank is required to ensure that the
water is not introducing contaminants into the borehole. Air jetting is acceptable only if
the air injected into the well has a controlled source. Due to frequent contamination of
formations with petroleum hydrocarbons from the air jetting process, the use of an oil
filter between the compressor pump and the borehole to control the purity of the air
introduced downhole is a requirement.
As each monitoring well represents a unique circumstance involving formation
characteristics, well parameters and pumping requirements, current USEPA policy does
not require a minimum waiting period between development and sampling for most
development procedures, but relies on the technical expertise of the drilling contractors
to define the time required for the aquifer to return to stability. For the processes of high
velocity jetting and air development, however, a ten to fourteen day waiting period has
been defined as necessary by the Robert S. Kerr Environmental Research Laboratory
and the New Jersey Department of Environmental Protection Geologic Survey, and is
therefore required by Region II, for the stabilization of aquifer flow and to allow recovery
of the aquifer from the stresses of development.
B. Well Filter Pack and Annular Sealant
The materials used to construct the filter pack should be chemically inert (e.g., clean
quartz sand, silica, or glass beads), well rounded, and dimensionally stable. Natural
gravel packs are acceptable, provided that a sieve analysis is performed to establish
the appropriate well screen slot size and determine chemical inertness of the filter pack
materials in anticipated environments.
The materials used to seal the annular space must prevent the migration of
contaminants to the sampling zone from the surface or intermediate zones and prevent
cross contamination between strata. The materials should be chemically compatible
with the anticipated waste to ensure seal integrity during the life of the monitoring well
and chemically inert so they do not affect the quality of the ground water samples. An
example of an appropriate use of annular sealant material is using a minimum of two
feet of certified sodium bentonite pellets immediately over the filter pack when in a
saturated zone. The pellets are most appropriate in a saturated zone because they will
penetrate the column of water to create an effective seal. Coarse grit sodium bentonite
is likely to hydrate and bridge before reaching the filter pack. A cement and bentonite
mixture, bentonite chips, or anti-shrink cement mixtures should be used as the annular
sealant in the unsaturated zone above the certified-bentonite pellet seal and below the
frost line. Again, the appropriate clay must be selected on the basis of the environment
in which it is to be used. In most cases, sodium bentonite is appropriate^).
-------
The selected seal must not interfere with the water chemistry. Bentonite clay has
appreciable ion exchange capacity which may interfere with the chemistry on collected
samples when grout seal is in close proximity to the screen or well intake. Similarly,
expanding cement which does not harden properly may affect the pH of water from
monitoring wells when in close proximity to the well screen or intake.
To minimize these potential interferences, a 1-foot layer of silica sand should be placed
above the selected gravel pack. Then, if possible, 1-2 feet of bentonite pellets should be
placed in the hole to prohibit the downward migration of bentonite slurry or neat
cement(4).
The untreated sodium bentonite seal should be placed around the casing either by
using a tremie pipe or, if a hollow-stem auger is used, putting the bentonite between the
casing and the inside of the auger stem. Both of these methods present a potential for
bridging. In shallow monitoring wells, a tamping device should be used to reduce this
potential. In deeper wells, it may be necessary to pour a small amount of formation
water down the casing to wash the bentonite down the hole.
The cement-bentonite mixture should be prepared using clean water and placed in the
borehole using a tremie pipe. The tremie method ensures good sealing of the borehole
from the bottom.
The remaining annular space should be sealed with expanding cement to provide for
security and an adequate surface seals. Locating the interface between the cement and
bentonite-cement mixture below the frost line serves to protect the well from damage
due to frost heaving. The cement should be placed in the borehole using the tremie
method(5).
C. Well Casing Selection
1. General Discussion
Well construction materials must be durable enough to resist degradation thereby
retaining their long-term stability and structural integrity and be relatively inert to
minimize alteration of ground water and collected samples.
In general, the more inert (i.e., less reactive) the casing material, the more assured one
is that the ground water sample withdrawn from the well is representative of the actual
ground water. The major potential alterations of the sample resulting from interactions
with the well casing/screen materials are: (1) adsorption/absorption reactions, both of
organics and inorganics; and (2) desorption reactions, meaning leaching of chemical
constituents from the well casing material into the ground water or desorption of newly
adsorbed material. Casing materials can also be affected by chemical attack, i.e.,
corrosion/deterioration, and microbial colonization and attack(4).
-------
These processes may lead to the observation of false trends in analyte concentrations,
highly variable water chemistry and the identification of artifacts resultant from surface
release or sorptive interactions. As with the errors which sampling mechanisms can
introduce into the chemical data, casing materials' related errors can be quite significant
and difficult to predict. Appropriate choice of materials for each application must be
made on the basis of long-term durability, cleanability, and the minimization of the
secondary effects of sorption or leaching. Structural integrity over time is, therefore, the
primary criterion for making reliable material choices. The materials must neither be
attacked nor degraded during the course of the monitoring program(4).
A variety of construction materials have been used for casing and well screens,
including virgin fluorocarbon resins (Teflon), stainless steel (304 or 316), cast iron,
galvanized steel, polyvinylchloride (PVC), polyethylene and polypropylene. Many of
these materials, however, may affect the quality of ground water samples and may not
have the long-term structural characteristics necessary for site specific needs. For
example, steel casing deteriorates in corrosive environments; PVC deteriorates when in
contact with ketones, esters and aromatic hydrocarbons; polyethylene deteriorates in
contact with aromatic and halogenated hydrocarbons; and polypropylene deteriorates in
contact with oxidizing acids, aliphatic hydrocarbons, and aromatic hydrocarbons. In
addition, steel, PVC, polyethylene and polypropylene may adsorb and leach
constituents that may affect the quality of ground water samples(5).
The selection of well casing and screen material should be made with due consideration
to geochemistry, anticipated lifetime of the monitoring program, well depth, chemical
parameters to be monitored and other site specific factors. Fluorocarbon resins or
stainless steel should be specified for use in the saturated zone when volatile organics
are to be determined during long term monitoring. Where high corrosion potential exists
or is anticipated, fluorocarbon resins are preferable to stainless steel. National
Sanititation Foundation (NSF) or ASTM-approved polyvinylchloride (PVC) well casing
and screens may be appropriate if only trace metals or non-volatile organics are the
contaminants anticipated(5).
Any well casing material may be used in the vadose zone, however, one combination
that should be avoided is the use of dissimilar metals, such as stainless steel and
galvanized steel, without an electrically isolating (dielectric) bushing. |f such dissimilar
metals are in direct contact in the soil, a potential difference is created and leads to
accelerated corrosion of the galvanized steel (in this example). More generically, in the
Galvanic series the less noble metal becomes the anode to the more noble metal and is
corroded at an accelerated rate. In well construction, this acceleration in corrosion at
the point of connection will lead to failure of the construction materials. Thus, a
dielectric coupling should be used for connecting dissimilar metals above the saturated
zone.
plastic pipe sections must be flush threaded or have the ability to be connected by
another mechanical method that does not introduce contaminants such as glue or
solvents into the well(5).
-------
2. Selection Criteria SOP
Appendix IX presents the "Standard Operating Procedure for Selecting Ground Water
Well Construction Material at CERCLA Sites", dated December 15,1986. The Appendix
to the SOP provides a "Summary Table for Comparing Features of Various Ground
Water Well Construction Materials", which was used to develop the criteria presented
for selecting the appropriate casing material. The considerations involved in the process
include duration of intended well use, use of data, desired detection limits, and known
site conditions and contaminants. The numerical cut-off values presented except for the
chloride and pH conditions on page 4 of the SOP were designed to be ball-park figures
intended to guide decision making but they were not intended to be absolute limitations.
They were, as all of the criteria were, devised after digestion of the current literature and
using best professional judgement.
The Summary Form presented on page 5 of the SOP should be filled out by the EPA
Project Manager and presented with the Plan for review for each site.
D. Evaluation of Existing Wells
If it is desired that a well already existing on-site be sampled in conjunction with newly
installed wells, the Project Manager should consider the ramifications of utilizing data
from those wells if, according to the "SOP for Selecting Ground Water Well
Construction Material", the existing well casing is not compatible with the type of ground
water contamination or the sensitivity of analysis needed.
-------
IX. SAMPLE COLLECTION DEVICES, MATERIALS AND QUALITY CONTROL
PRACTICES
A. References for Selection of Sampling Devices
Sampling at hazardous waste sites requires many different types of
sampling devices. Selection of a device should be based on practicality, economics,
representativeness, compatibility with analytical considerations, and safety. There are
many documents which compile sampling methods and materials suitable to address
most needs that arise during investigations. The following is a list of the most commonly
used references compiling sampling equipment and methodology, however it is not
meant to be an exhaustive listing of all the references available.
1. Characterization of Hazardous Waste Sites-A Methods Manual: Volume II. Available
Sampling Methods, Second Edition. EPA-600/4-84-076. December 1984. Available
from ORD Publications in Cincinnati at (513)569-7562.
2. Handbook for Sampling and Sample Preservation of Water and Wastewater.
EPA-600/4-82-029. September 1982. Available from ORD Publications.
3. Samplers and Sampling Procedures for Hazardous Waste Streams. EPA-600/2-
80-018. January 1980. Available from ORD Publications.
4. Practical Guide for Ground-Water Sampling. EPA 600/2-85/104. September 1984.
Available from ORD Publications.
5. RCRA Ground-Water Monitoring Technical Enforcement Guidance Document,
September, 1986. Office of Waste Programs Enforcement and Office of Solid Waste
and Emergency Response. Available from RCRA Hotline at 800-424-9346.
6. A Guide to the Selection of Materials for Monitoring Well Construction and Ground
Water Sampling. Barcelona, Gibb, Miller. Illinois State Water Survey, Champaign,
Illinois. January 1984. NTIS publication #PB84-126929.
7. Test Methods for Evaluating Solid Waste, Physical and Chemical Methods.
SW-846, Third Edition. Office of Solid Waste and Emergency Response. GPO
publication #955-001-00000-1, at (202) 783-3238.
8. Compendium of Methods for the Determination of Toxic Organic Compounds in
Ambient Air, April 1984. EPA 600/4-84-041. Available from ORD Publications.
9. SOP for the GC/MS Determination of Volatile Organic Compounds Collected on
Tenax. June 27,1984. EMSL/RTP-SOP-EMD-020. Available through USEPA at
Research Triangle Park, NC at (919)541-2777.
10. USEPA Dioxin Strategy. November 28,1983. Office of Water Regulations and
Standards and the Office of Solid Waste and Emergency Response in conjunction with
-------
the Dioxin Strategy Task Force, Washington, D.C., 20460.
11. Sampling Guidance Manual for the National Dioxin Study,
May 16, 1984. Office of Water Regulations and Standards, Washington, D.C.
12. Soil Sampling Quality Assurance User's Guide, Draft. May 1984. EMSL-LV. EPA
600/4-84-043. Available from ORD Publications.
13. Data Quality Objectives for Remedial Response Activities. Development Process.
EPA/540/G-87/003. March 1987.
14. Data Quality Objectives for Remedial Response Activities. Example Scenario:RI/FS
Activities at a Site With Contaminated Soils and Ground Water. EPA/540/G-87/004.
March 1987.
B. Ground water
1. Sampling Design
Samples from a monitoring well represent a small part of the horizontal and vertical
extent of the aquifer. Unlike its surface counterpart, where a sample can be arbitrarily
taken at any point in the system, moving a ground water sampling point implies the
installation of additional monitoring wells. There is a need to be concerned not with the
point data as an end in itself, but as a component of a network approach wherein
information on the ground water system is developed as a basis for extrapolating
information to areas where samples were not collected and/or for predicting the effects
of natural and man-made stresses on the subsurface systems(2). Discussion of the
areas of consideration for location of ground water sampling points can be found in
references listed above.
2. Well Evacuation
In order to obtain a representative sample of ground water, the water that has stagnated
and stratified in the well casing must be purged or evacuated. Prior to evacuating the
well, however, the presence or absence of immiscible phases (i.e., "floaters" and
"sinkers") must be determined., "Floaters" are those relatively inslouble organic liquids
that are less dense than water and which spread across the potentiometric surface.
"Sinkers" are those relatively insoluble organic liquids that are more dense than water
and tend to migrate vertically through the sand and gravel aquifers to the underlying
confining layer. The detection of these immiscible layers requires specialized equipment
that must be used before the well is evacuated for conventional sampling. The Plan
should specify the device to be used to detect light phases and dense phases, as well
as the procedures to be used for detecting and sampling these contaminants.
Procedures for identifying and sampling "floaters" and "sinkers" can be found in section
4.2.2 of the RCRA Ground-Water Monitoring Technical Enforcement Guidance
Document dated September 1986.
-------
Water that has remained in the well casing for extended periods of time (i.e., more than
about two hours) has the opportunity to exchange gases with the atmosphere and to
interact with the well casing material. The chemistry of the water stored in the well
casing is dissimilar to that of the aquifer and, thus, should not be sampled(4).
Evacuating the well allows for fresh formation ground water to enter the well. When
indicator parameters such as pH, temperature and specific conductance are observed
to vary less than 10% over the removal of two successive well volumes, the well is
presumed to be adequately flushed for representative sampling. Evacuation of at least
3-5 well volumes is required for high yielding wells, however, in wells with very low
recoveries this may not be practical. In this case the well may be evacuated to near
dryness once and allowed to recover sufficiently for samples to be taken. A well must
be sampled within three hours of evacuation. If a well is allowed to sit longer than three
hours after evacuation, it should be re-evacuated since the water contained in the
casing may no longer be representative of the aquifer conditions(4).
Any device used to evacuate the well must be cleaned as per Section VI to ensure that
cross contamination between wells does not occur. When a pump is used to evacuate,
the tubing which comes in contact with water should be made of polyethylene or Teflon,
and must be dedicated to individual wells. The intake should be placed just below the
water level and lowered as the water level lowers while pumping to ensure that all the
water within the well bore is exchanged with fresh water.
A bailer may be used to evacuate the well and to sample it. Bailers must be constructed
of Teflon or stainless steel with cords made of Teflon coated wire, stainless steel wire or
polypropylene monofilament. Ten foot leaders may be used of these acceptable
materials, with nylon cord above. ANY down- hole equipment having neoprene fittings,
PVC, tygon tubing, silicon rubber bladders, neoprene impellers, orviton are not
acceptable. A bailer which is used to evacuate the well may also be used to sample it
without any additional cleaning.
Any water that is removed from the well during evacuation can no longer be considered
a representative portion of the aquifer and should not be reintroduced into the well after
sampling.
3. Sampling Organics and Inorganics
After evacuation of the required volume of water from the well, sampling may occur.
The unstable nature of many chemical, physical and microbial constituents in ground
water limit the sample collection options. Certain factors should be considered when
collecting representative samples:
* Temperatures are relatively constant in the subsurface, therefore the sample
temperature may change significantly when brought to the surface. This change can
alter chemical reaction rates, reverse cationic and anionic ion exchanges on solids, and
change microbial growth rates.
-------
* A change in pH can occur due to carbon dioxide adsorption and subsequent changes
in alkalinity. Oxidation of some compounds may occur.
* Dissolved gases may be lost at the surface.
* The integrity of organic samples may be.affected by problems associated with either
adsorption or contamination from sampling materials and volatility(2).
The only acceptable sampling devices for pH sensitive and volatile parameters are:
1. Teflon or stainless steel bladder pumps having adjustible flow control;
2. Teflon or stainless steel bottom-filling bailers; and,
3. Teflon or stainless steel syringe bailers.
Appropriate operating precautions for these sampling devices include:
* Bladder pumps must be operated in a continuous manner so that they do not produce
pulsating samples that are aerated in the return tube or upon discharge. Pumping rates
should not exceed 100ml/min when samples are being taken for dissolved gases,
volatile organic constituents, TOX and TOC; ,
* Check valves must be designed and inspected to ensure that fouling problems do not
reduce delivery capabilities or result in aeration of the sample;
* Sampling equipment (especially bailers) must never be dropped into the well because
this will cause degassing of the water upon impact;
* The bailer's contents must be transferred to a sample container in a way that will
minimize agitation and aeration without transferring the sample to an intermediate
container, or utilizing a mechanical device; the bailer should not be "acclimated" by
discharging the first bailer- full of water onto the ground since this unnecessarily
agitates the water column prior to the volatile sensitive parameters being taken;
* Clean sampling equipment must not be placed directly on the ground or other
contaminated surface. When not in use, these devices should be placed on '
polyethylene sheeting or aluminum foil.
When sampling, heavy gauge aluminum foil or polyethylene sheeting should be placed
on the ground around each well to prevent contamination of sampling equipment in the
event that equipment is dropped or otherwise comes in contact with the ground.
When a number of rounds or phases of sampling will take place, the same type of
sampling equipment should be used consistently throughout the entire project in order
to increase the reproduceability in analytical results by eliminating the variability in
sample collection technique.
-------
Other sampling devices, including positive displacement pumps, gas lift devices,
centrifugal pumps, and venturi pumps, may be used for collection of non-volatile or
non-pH sensitive parameters (most parameters ARE pH- sensitive), provided that they
are constructed of Teflon or stainless steel(5).
The preferred order of sample collection is as follows:
1. In-situ measurements:temperature, pH, specific conductance
2. Volatile organics (VGA)
3. Purgeable organic carbon (POC)
4. Purgeable organic halogens (POX)
5. Total organic halogens (TOX)
6. Total organic carbon (TOC)
7. Extractable organics
8. Total metals
9. Dissolved metals
10. Phenols
11. Cyanide
12. Sulfate and Chloride
13. Turbidity
14. Nitrate and Ammonia
15. Radionuclides
Detailed discussions of sample evacuation and collection procedures can be found in
references #1,2,4,5 found in the Bibliography of this document.
For a discussion on the collection of dissolved and particulate sample fractions for
metals analysis, see Section XII.
4. Sampling Microbiological
There are several different methods for obtaining a ground water sample. Each of these
methods differ in their advantages and disadvantages for obtaining samples for
microbiological analyses.
\
The majority of ground water samples obtained for microbiological analysis are obtained
using preexisting wells which have existing in-place pumps. This limits the precautions
the sampler must take to ensure a non- contaminated sample. Samples should be
obtained from outlets as close as possible to the pump and should not be collected
from leaky or faulty spigots or spigots that contain screens or aeration devices. The
pump should be flushed for 5 to 10 minutes before the sample is collected. A steady
flowing water stream at moderate pressure is desirable in order to prevent splashing
and dislodging particles in the faucet or water line.
To collect the sample, remove the cap or stopper carefully from the sample bottle. Do
-------
not lay the bottle closure down or touch the inside of the closure/Avoid touching the
inside of the bottle with your hands or the spigot. The sample bottle should not be
rinsed out and it is not necessary to flame the spigot. The bottle should be filled directly
to within 2.5 cm (1 inch) from the top. The bottle closure and closure covering should be
replaced carefully and the bottle should be placed in a cooler (4 to 10 degrees C)
unless the sample is going to be processed immediately in the field.
If a well does not have an existing in-place pump, samples can be obtained by either
using a portable surface or submersible pump or by using a bailer. Each method
presents special problems in obtaining an uncontaminated sample.
The main problem in using a sterilized bailer is obtaining a representative sample of the
aquifer water without pumping or bailing the well beforehand to exchange the water in
the bore for fresh formation water. This is difficult since such pre-sampling activities
must be carried out in such a way as to not contaminate the well. Care must also be
taken with bailers to not contaminate the sample with any scum on the surface of the
water in the well. This is usually done by using a weighted, sterilized sample bottle
suspended by a cord of acceptable material and lowering the bottle rapidly to the
bottom of the well.
The use of portable pumps provides a way of pumping out a well before sampling and
thus providing a more representative sample, but presents a potential source of
contamination if the pumping apparatus cannot be sterilized beforehand. The method of
sterilization will depend on what other samples are taken from the well since the use of
many disinfectants may not be feasible if the well is also sampled for chemical
analyses. If disinfection is not ruled out by other considerations, a method of sterilizing
a submersible pump system is to submerge the pump, and any portion of the pump
tubing which will be in contact with the well water, into a disinfectant solution and
circulating the disinfectant through the pump and tubing for a recommended period of
time.
The most widely used method of disinfection is chlorination due to its simplicity.
Chlorine solutions may be easily prepared by dissolving either calcium or sodium
hypochlorite in water. Calcium hypochlorite, Ca(OCI)2, is available in a granular or
tablet form usually containing about 70% of available chlorine by weight and should be
stored under dry and cool conditions. Sodium hypochlorite, NaOCI, is available only in
liquid form and can be bought in strengths up to 20% available chlorine. Its most
available form is household laundry bleach, which has a strength of about 5% available
chlorine, but should not be considered to be full strength if it is more than 60 days old.
The original percentage of available chlorine will be on the label. Fresh chlorine
solutions should frequently be prepared because the strength will diminish with time.
After disinfection the pump should be carefully placed in the well and then pumped to
waste until the chlorine is thoroughly rinsed from the pump system.
If the pump cannot be disinfected, then the pump and tubing should be carefully
handled to avoid gross surface contamination and the well should be pumped for 3 to
10 bore volumes before taking a sample. It may be desirable after pumping to pull the
-------
pump and take the sample with a sterile bailer.
In those cases where the water level in the well is less than 20 to 30 feet below the
surface, a surface vacuum pumping system can be used for flushing out the well and
withdrawing a sample.
Springs are unlikely to yield representative samples of an aquifer due to surface
contamination close to a spring's discharge unless the spring has an extremely fast flow
and the outlet is protected from surface contamination.
Lastly, interpretation of analytic results may be difficult in some cases since surface
contamination of wells due to poor drilling and completion practices is common. In
cases where drinking water supplies are involved, a thorough inspection of the well is
required to eliminate surface contamination down the well bore as a source of
contaminants. Disinfection of the well by approved methods and resampling may be
advisable, if disinfection will not affect the well for other sampling purposes(2).
C. Surface Water
1. Sampling Design
There are at present no quality assurance requirements for site location for sampling
surface waters. References #1 and #2 in Section IX can be consulted for design
guidance.
2. Sampling Devices
Appendix X provides a tabular comparison of surface water sample collection devices.
Samples from shallow depths can be readily collected by merely submerging the
sample container. However, preservatives cannot be present in the container when it is
lowered into the water. The method is advantageous when the sample might be
significantly altered during transfer from a collection vessel into another container. This
is the case with samples collected for oil and grease analysis since considerable
material may adhere to the sample transfer container and as a result produce
inaccurately low analytical results. Similarly the transfer of a liquid into a small sample
container for volatile organic analysis, if not done carefully, could result in significant
aeration and resultant loss of volatile species. Though simple, representative, and
generally free from substantial material disturbances, it has significant shortcomings
when applied to a hazardous waste, since the external surface of each container would
then need to be decontaminated.
In general the use of a sampling device constructed of a nonreactive material such as
glass, stainless steel, or Teflon, is the most prudent method. The device should have a
capacity of at least 500 ml, if possible, to minimize the number of times the liquid must
be disturbed, thus reducing agitation of any sediment layers.
-------
A 1-liter stainless steel beaker with pour spout and handle or large stainless steel ice
scoops and ladles available from commercial kitchen and laboratory supply houses can
be used.
It is often necessary to collect liquid samples at some distance from shore or the edge
of the containment. In this instance an adaptation which extends the reach of the
sampler is advantageous. Such a device is the pond sampler. It incorporates a
telescoping heavy-duty aluminum pole with an adjustable beaker clamp attached to the
end. The beaker previously described, a disposable glass container, or the actual
. sample container itself, can be fitted into the clamp.
It may on occasion be necessary to sample large bodies of water where a near surface
sample will not sufficiently characterize the body as a whole. In this instance a
peristaltic pump may be used in which the sample is drawn in through heavy walled
Teflon tubing and pumped directly into the sample container. This method, however, is
not suitable for pH sensitive or volatile samples in which stripping would occur.
Situations may still arise where a sample must be collected from depths beyond the
capabilities of a peristaltic pump. In this instance an at-depth sampler may be required,
such as a Kemerer, ASTM Bomb (Bacon Bomb) or Van Dorn sampler. These devices
work well; however, care must be utilized in selecting devices that are made of
materials that will not contaminate the sample. Van Dorn samplers are not generally
recommended for organics as they rely on an elastic closing mechanism that can effect
samples. They are readily available in a totally nonmetallic design which is very useful
for sample collection for trace metal analysis.
Kemerer samplers are available on special order or adaptable for sample collection for
organic analysis by substituting Teflon for the rubber or plastic stoppers. If the device is
further ordered with stainless steel metallic parts in addition to Teflon stoppers it
becomes a very versatile sampler(1).
Consult "Characterization of Hazardous Waste Sites-A Methods ManuahVolume II.
Available Sampling Methods, Second Edition for specific collection techniques using
these devices. See Section IX.B.4 for collection of microbial samples.
D. Sediment
Sediment is defined as the deposited material underlying a body of water.
1. Sampling Design
Streams, lakes, and impoundments will likely demonstrate significant variatoins in
sediment composition with respect to distance from inflows, discharges or other
disturbances. It is important, therefore, to document exact sampling location by means
of triangulation with stable reference points on the banks of the stream or lake.
-------
At the present time, however, there are no quality control requirements for choosing site
location when sampling sediments. Reference #2 in Section IX can be consulted and
sediment transport and deposition modeling along with statistical considerations can be
found in "Sediment Sampling Quality Assurance User's Guide," EPA 600/4-85-048,July
1985,NTIS#PB85-233542.
2. Sampling Devices
Appendix XI provides a tabular comparison of some bottom grab and coring devices.
Samples can be taken with stainless steel spoons or trowels or the sample container
itself if there is little or no water on top of the sediment. If the water above the sediment
is a few feet deep a stainless steel corer or corer having a removable glass or Teflon
liner may be used. This will help ensure the integrity of the surface layer of sediment
and will minimize the loss of fine grained material. In deeper water the bottom grab
samplers presented in Appendix XI may be used(1).
If at any time surface water samples are being taken in conjunction with sediment
samples, the water samples should be taken first. The sampler should approach the
location from the downstream direction with the container pointed upstream to ensure
collection of an undisturbed sample.
Sediment samples collected for all analyses except VGA and TOX should be thoroughly
mixed before being placed in appropriate sample containers. Rocks, twigs, and other
debris should be removed from the sample prior to homogenization. In this way
stratification of constituents will not affect
analytical results. VOA and TOX samples should be taken as individual grab samples,
never homogenized. See Section X for the procedure used for homogenization of
samples.
E. Soil
1. Sampling Design
Statistical techniques for obtaining accurate and precise samples are relatively simple
and easy to implement. In the sampling of a soil matrix, sampling accuracy is usually
achieved by some form of random sampling. In random sampling, every unit in the
population has a theoretically equal chance of being sampled and measured.
Consequently, statistics generated by the sample are unbiased (accurate) estimators of
true population parameters.
In other words, the sample is representative of the population.
One of the most common methods of selecting a random sample is to divide the
population by an imaginary grid, assign a series of consecutive numbers to the units of
the grid, and select the numbers to be sampled through the use of a random-numbers
-------
table (such a table can be found in any text on basic statistics). It is important to
emphasize that a haphazardly selected sample is not a suitable substitute for a
randomly selected sample. That is because there is no assurance that a person
performing undisciplined sampling will not consciously or subconsciously favor the
selection of certain units of the population, thus causing the sample to be
unrepresentative of the population. A detailed discussion of random sampling can be
found in SW-846, Third Edition, November 1986, Volume 2, Chapter 9.
In a biased sampling approach, site history, available information and professional
judgement are used to determine the sampling locations in which contamination is
expected to be found. In this case the most information is gotten from the least number
of samples thereby allowing a cost savings. However, this approach to sampling design
is more appropriate to a site investigation than a remedial investigation (Rl) because it
is not a comprehensive view of site conditions(7). Use of a biased approach during an
Rl must be justified.
2. Sampling Devices
When sampling soil, stainless steel, Teflon or glass utensils should be used. The only
exception is split spoons which are most commonly available in carbon steel. These are
acceptable for use provided they are not excessively rusty.
A stainless steel spatula should be used to remove sample from the opened spoons,
not the sampler's fingers, as the gloves may introduce organic interferences into the
sample. As per water sampling, volatile organics and TOX samples should be taken
immediately upon opening the spoon. Any rocks, twigs, leaves or other debris should be
removed from the sample before homogenization. All samples except those for volatile
organics and TOX must be homogenized before being put into sample containers.
Section X provides a discussion on homogenization. Samples for VOA and TOX should
never be homogenized as they both contain volatile parameters.
F. Potable Water
When sampling potable water the same quality control requirements apply as for
sampling ground water, where applicable.
The "Manual for the Certification of Laboratories Analyzing Drinking Water," EPA
570/9-82-002, October 1982, states that the sample needs to be representative of the
potable water system, and that the tap water must be sampled after maintaining a
steady flow of 2 or 3 minutes to clear the service line. This flushing period is a minimum
requirement and more time can be allowed if a large holding tank is in place and it is
desireable to purge an entire tank volume. A tap must be free of any aerator, strainer,
hose attachment, or water purification devices. When sampling chlorinated waters for
organics, cyanide and bacterial analyses, sodium thiosulfate should be added to the
sample bottles.
-------
F. Dust/Wipe
The following Standard Operating Procedure has been developed by Region II for use
in taking wipe samples.
1. Materials needed:
a. cotton swabs, solvent rinsed and completely air dried;
(use of synthetic materials requires checking for compatibility with
solvents)
b. acetone, pesticide grade;
c. hexane, pesticide grade;
d. deionized water;
e. stainless steel clamps or plastic clamps (only for taking metals samples);
f. appropriate sample bottles.
2. A square area, of a size sufficient to give the required amount of sample, for each
fraction as provided in the analytical methodology to be used, should be marked off.
This may require taking cotton swabs and a balance into the field, wiping a certain area
and weighing the swab before and after to determine how much area should be wiped
to give the required weight of sample.
3. Swabs for semi-volatile, pesticide and PCB samples should be moistened in a 1:4
acetone/hexane mixture. Swabs for volatile organic samples should be moistened with
hexane alone, and those for metals with deionized water. While holding the swab in a
clean, metal clamp, moisten the cotton swab with the appropriate solution.
4. While still holding the cotton swab in the clamp, wipe the sampling area back and
forth repeatedly, applying moderate pressure. Wipe the entire area so that all the
sample material is picked up.
5. Place the used swab in the appropriate sample container and seal.
6. Clean the clamps between each sample with both solvent and 10% HCL.
7. As a blank, moisten a clean swab with the solvent or water (for each collection
medium), place it in a separate jar, and submit it with the other samples.
8. When samples are submitted for analysis, the laboratory should be told to rinse
the sample jars with solvent or 10% HCL, depending on the analysis to be performed,
when transferring sample to the extraction glassware.
The samples should be analyzed with the appropriate methodology for a soil/sediment
matrix, and a sufficient quantity of material must be collected as called for in the
analytical methodology, in order for method detection limits to be achieved.
-------
H. Dioxin
For guidance on conducting dioxin sampling events refer to "Sampling Guidance
Manual for the National Dioxin Study", May 16, 1984. The sampling material
requirements are the same as those present in this document for the matrix being
sampled, with the following additions.
The equipment decontamination procedure is the same as that presented in Section VI
and samplers have the option of using methanol/hexane, acetone, or
1,1,1-trichloroethane as solvents.
Homogenization must be performed on dioxin samples of a solid matrix.
Homogenization may be performed in a laboratory or in the field and it should be done
using either the coning and quartering method (see Section X) or by using stainless
steel blenders. Homogenization of wet sediment samples is more easily accomplished
using the coning and quartering method. Specific quality control samples for dioxin
sampling events are presented in Section XI.
I. Drums
For sampling of non-homogenous or multi-phased materials in drums, tanks, waste
piles, lagoons, etc., refer to SW-846, Third Edition, Volume 2 for guidance. The Plan
must include a detailed approach in terms of basic strategy and sampling equipment if
such containers are to be sampled.
-------
X. METHODS OF SAMPLE PREPARATION
A. Homogenization
The homogenization of a sample is the process of mixing individual grab samples in
order to minimize any bias of sample representativeness introduced by the natural
stratification of constituents within the sample.
To homogenize a sample of a soil/sediment matrix, first rocks, twigs, leaves and other.
debris should be removed if they are not considered part of the sample. The
soil/sediment should be removed from the sampling device and placed in a stainless
steel pan, then thoroughly mixed using a stainless steel spoon. The sediment in the
pan should be scraped from the sides, corners and bottom of the pan, rolled to the
middle of the pan, and initially mixed. The sample should then be quartered and moved
to the four corners of the pan. Each quarter of the sample should be mixed individually,
and then rolled to the center of the container and the entire sample mixed again.
Homogenization of an aqueous sample is only necessary if stratification of constituents
is of concern, for example when sampling a lagoon or containerized liquid. Then
homogenization would be performed by mixing in a stainless steel bowl.
B. Compositing
Compositing of samples is performed when samplers desire to obtain an average
concentration of contaminants over a certain number of sampling points. Anytime
compositing is performed, the concentration of contaminant in individual grab samples
is diluted proportionately to the number of samples taken. Not only is the contaminant
diluted, the detection limits for each individual sample are raised proportionally to the
number of samples added to the composite. For instance, if a sampler wishes to
composite two discreet samples into one, and the method detection limit for a target
compound is 330 ppb, the detection limit for the target compound does not change for
the composite, however, the detection limit for the compound in the individual samples
which make up the composite is two times the normal detection limit or 2 * 330 = 660
ppb. This is important to keep in mind because it is possible that if a contaminant were
present in only one of the two composited samples, and if it were at a level between
330 and 660 ppb, that contaminant would not be quantified or possibly even identified
due to the effective dilution of the contaminant concentration. This concept should be
taken into account when determining the data quality objectives of a composite
sampling event, to ensure that useful data is collected. It is advisable that if a positive
identification is made in the course of analyzing a composite sample, that the discreet
samples then be analyzed individually to determine the true distribution of contaminant
throughout each component of the composite.
Compositing of a solid matrix is accomplished by mixing equal volumes of grab samples
in stainless steel pans with stainless steel spoons. Compositing is never performed on
samples for volatile organics analysis (or any analysis involving a volatile portion such
-------
as TOX), and, for a solid matrix, should never be done by placing equal portions of grab
samples directly into sample jars, as the occurrence of error introduced by the sampler
is highly probable.
C. Splitting
Splitting of samples is performed when two or more parties wish to have a portion of the
same sample. They are most often collected in enforcement actions to ensure that
sample results generated by Potentially Responsible Parties (PRPs) are accurate.
Splitting of samples, however, is not as useful as providing blind performance
evaluations samples to a laboratory since analtyical performance and accuracy differs
from laboratory to laboratory, and therefore one laboratory cannot be considered a
"referee" who's performance can be considered the standard against which another's
can be measured. Performance evaluation samples provide information on a
laboratory's performance based upon analysis of that sample which is of a known and
defined concentration.
Soil/sediment samples taken for volatile analysis cannot be split. In this case samples
must be taken as co-located grabs. Then a large quantity of material can be collected,
homogenized, split and used to fill the remaining containers. Note that enough sample
must be collected at one time in order to fill all the necessary sample containers. It may
be necessary to co locate or depth integrate collection so enough sample volume is
available. A description of this process should be provided in the Plan.
When splitting aqueous samples, homogenization of the sample is only necessary if
heterogeneity is suspected, for example when sampling a small lagoon or containerized
liquid; however VOA and TOX samples are never homogenized. It is not generally
necessary to homogenize ground water or surface water samples when splitting, and it
is likewise generally unnecessary to divide a bailer's contents among several bottles.
-------
XI. FIELD QUALITY CONTROL SAMPLES
A. All Analyses Except Dioxin
1. Duplicates
*
Environmental duplicate samples are collected to demonstrate the reproducibility of
sampling technique. Environmental duplicate samples must be taken at a frequency of
at least 5% (1 in 20). This is a separate duplicate from the duplicates a laboratory must
run, and cannot be replaced by a laboratory generated duplicate. This applies to every
matrix sampled. Environmental duplicates are representative of field sampling precision,
whereas laboratory duplicates are a measure of analytical precision. Both pieces of
information are essential to determining the quality of data generated for a project.
2. Blanks
Blank water generated for use in the Region II CERCLA program must be
"demonstrated analyte-free". By this term we mean water of a known quality which is
defined by the Quality Assurance office.
The criteria for analyte-free water is as follows. The assigned values for the Contract
Required Detection Limits (CRDLs) can be found in the most recent CLP SOWs.
purgeable organics <10 ppb
semi-volatile organics
-------
a. Trip Blank
When sampling for volatile brganics, a trip blank, consisting of demonstrated analyte
free water sealed in 40 ml Teflon lined septum vials, must be taken into the field where
sampling is going on. It should be taken at a frequency of one per day per matrix •
sampled when volatile organics in an aqueous matrix are being collected. Note that it is
not necessary to take an aqueous trip blank when a non-aqueous medium is being
sampled. Trip blanks are used to determine if any on-site atmospheric contaminants are
seeping into the sample vials, or if any cross contamination of samples is occurring
during shipment or storage of sample containers. Trip blanks are only analyzed for
volatile organics.
b. Field Rinse Blanks
Rinse blanks consist of pouring demonstrated analyte free water over decontaminated
sampling equipment as a check that the decontamination procedure has been
adequately carried out and that there is no cross- contamination of samples occurring
due to the equipment itself. Analysis of rinse blanks is performed for all analytes of
interest. One blank should be taken for each type of equipment used each time a
decontamination event is carried out, whether that be daily or weekly. It is required also
that rinse blanks be performed on bowls and pans used to homogenize samples.
The blank should be done at the beginning of the day prior to the sampling event and
that blank must accompany those samples which were taken that day. This is a
necessary procedure so that the blank will be associated with the proper samples for
the purpose of data validation.
3. Matrix Spike/Matrix Spike Duplicate Analyses
When performing CLP organic extractable analysis, the laboratory must have triple
sample volume for each Sample Delivery Group (SDG), which does not include field or
trip blanks. Blanks do not require separate matrix spike or duplicate analyses
regardless of their matrix.
As stated in the SOW, the limits on an SDG are:
*each Case of field samples, or
*each 20 field samples within a Case, or
*each fourteen calendar day period during which field samples in a Case are
received (said period beginning with the receipt of the first sample in the SGD),
whichever comes first. .
B. Dioxin
When dioxin sampling is performed a batch of quality control samples must accompany
-------
environmental samples whether the samples are sent through the Contract Laboratory
Program or to a privately contracted laboratory.
The quality control sample requirements for every 20 dioxin samples taken are:
a. two performance evaluation samples from EMSL-LV containing 2,3,7,8-TCDD;
b. one environmental duplicate;
c. one blind blank (blind blank to the laboratory)-this sample does not go through the
homogenization on-site;
d. one known blank (the lab will spike with 1 ppb TCDD)-this does not go through
homogenization on-site;
e. one blender blank-this is blank soil homogenized in the field to check for
cross-contamination during the blending process, and is only necessary if blenders
are used to homogenize the samples.
These samples will be ordered by the Regional Quality Assurance Officer from
EMSL-LV at the request of the EPA Project Manager only.
-------
XII. FILTERED AND NON-FILTERED FRACTIONS OF GROUND WATER
SAMPLES
A. General Discussion
In certain situations it may by desireable to consider filtration of ground water samples
in order to obtain information about the "dissolved" specie of the contaminant in solution
as opposed to that portion of contaminant which may adhere to silt or clay particles.
The real concern in this issue is whether or not the contaminant is a component of the
ground water, which implies that if the contaminant is truly a component of the ground
water, then it must flow with the ground water. If the contaminant flows with the ground
water on or under a site then there is the potential for the contaminant to be moved to
and to impact areas outside of the site.
There has been a general assumption that water and soil are the only distinct
constituents of an aquifer system; there is also a false assumption that water and
completely solvated solutes are the only constituents of the system that are mobile. In
fact, components of the solid phase in the colloidal size range may be mobile in
subsurface environments^1). The colloidal state refers to a two phase system in which
one phase in a very finely divided state is dispersed through a second. In ground water,
colloidal particles are generally smaller than one micrometer (1 um) in diameter. Since
the clay fraction is defined by the USDA as being 2 um and smaller, not all clay is
strictly colloidal, but even the larger clay particles have colloid-like properties (12).
There is ample evidence, as can be seen in the literature, that colloid particles can
move in aquifers (11).
There are two distinct types of colloidal matter, inorganic and organic, which exist in
intimate intermixture. The inorganic is present almost exclusively as clay minerals of
various kinds; the organic is represented by humus (12). These colloidal particles can
sorb organic and inorganic contaminants and stabilize them in the mobile phase of the
aquifer. Association of contaminants with mobile colloidal particles may enhance the
transport of highly adsorbed pollutants, or deposition of colloidal particles in porous
media may decrease permeability and reduce contaminant transport (11).
The separation of "dissolved" and "particulate" sample fractions is most commonly
accomplished by filtration through a 0.45 um membrane filter. The convention of using a
0.45 um pore size filter was borrowed from the microbiological science where it was
used as the separation point.for filtering bacteria out of aqueous media. This convention
was borrowed by other organizations for use in the analysis of aqueous metals
samples, and now is specified as the pore size through which will pass those
constituents that are "dissolved". Those constituents which are retained on the 0.45 um
filter are labeled "suspended", and "total" metals is the sum of the "dissolved" and
"suspended" fractions.
It would be more accurate, however, to term the fractions "filterable" and "non-filterable"
-------
instead of "dissolved" and "total" or "particulate" given the operational nature of the
separation. The 0.45 um distinction is not useful when one is concerned with true
soil/water chemistry but has its sole value at present in the fact that it is convention and
is used as such in ground water characterization throughout the country.
The policy in Region II on filtering ground water has been that only samples for metals
analyses may be filtered, and when taking metals samples, "total" metals should be
taken with the option to take a "dissolved" sample if so desired. The filtration of
aqueous samples for organics analyses hasd not been allowed in the Region since,
1)volatile contaminants would be released during filtration, and 2)the membrane filters
used for the filtration of metals samples will adsorb the organics, thereby giving falsely
negative results.
The rationale behind the policy is this: rather than rely totally on "dissolved" metals data,
which will generally give results that are lower than the true amount of contaminant
which moves with ground water, the Region has chosen to be more conservative in its
use of metals data by preferring to consider "total" metals data, thereby erring on the
side of finding more metals in a sample than actually may be mobile in the ground water
phase. Obviously, in some cases when samples are silty, the "total" metals values will
be high due to the addition of metals which were bound to particles of greater size than
the colloidal range. Unfortunately the use of either "total" or "dissolved" metals data
alone is inappropriate when one wishes to consider the true portion and constituents
which move with ground water.
The Regional policy as presented here will continue to be enforced in spite of the
limitations until such time as a technically well-founded alternative is developed. If
exceptions or modifications to this policy are desired based on site specific needs, the
Project Manager should consult the Quality Assurance Officer assigned to the project.
B. Procedures for Filtration of Aqueous Metals Samples
1. Decontamination of Apparatus
When filtering aqueous metals samples, a device made of polyethylene or borosilicate
glass should be used. The apparatus should be pre-cleaned by rinsing with a 10%
HNO3 solution, followed by a demonstrated analyte-free deionized water rinse, and
should be cleaned in the same manner between samples.
2. Filtration Procedures and Preservation
The filter used should be a cellulose-based membrane filter of 0.45 um nominal pore
size. Samples must be filtered immediately after their collection to minimize changes in
the concentration of the substances of interest. Samples are only passed through the
filtration apparatus once, they are not to be passed through repeatedly until they are
free of turbidity. Samples are then preserved immediately with undiluted ultrapure
HN03 and the pH checked to ensure proper pH has been attained. No samples for
-------
cyanide, conventional parameters, or organics may be filtered in this manner.
-------
XIII. LABORATORY QUALIFICATIONS
A. Use of CLP vs. non-CLP Laboratories
Most analytical work performed for Federal fund-lead CERLCA sites within the Region
utilizes the USEPA Contract Laboratory Program. However, it is not mandatory that all
analyses supporting the CERCLA program be performed by a Contract Laboratory,
whether the project is a fund-lead or enforcement-lead site. Laboratories which do not
participate in CLP may be used at any time, provided they adhere to Region II QA/QC
requirements which are described here.
If a non-CLP laboratory is used, that laboratory must supply to the Regional Quality
Assurance Officer (QAO), a copy of their in-house QA/QC manual which is applicable to
the analyses to be performed. The QA/QC manual should cover the following topics:
resumes
personnel training and experience
organizational structure
equipment available
reference materials/reagents
control charts
standard operating procedures
data reduction/reporting
chain-of-custody
glassware preparation
Also, that laboratory must perform acceptably on performance evaluation samples
supplied by EPA for those parameters of interest to the project. A formal request for
performance evaluation samples should be sent from the EPA Project Manager for the
site to the EPA QAO.
A non-CLP laboratory must also undergo a technical systems audit performed by the
primary engineering contractor in order to evaluate the laboratory's capability to perform
the work. A copy of the resultant report should be sent to the EPA Project Manager and
subsequently to the EPA QAO. The format of the audit checklist can be taken from the
CLP Invitations for Bid (IFBs). Only after this information has been provided and is
found to be acceptable can sampling and analysis begin. The CLP IFBs are available
from the Sample Management Office at (703)557-2490.
-------
XIV. USE OF MOBILE LABORATORIES
A. Qualifications and Methods
There has been a growing demand throughout the Region for use of field analytical
laboratories in order to screen samples and generate real-time results which can be
used to make decisions in the field. These laboratories commonly use "quick and dirty"
techniques since strict precision and accuracy requirements are not necessary for the
intended use of the data, and since, in most cases, critical sampling locations, where
quality data is important, are split for analysis by a CLP laboratory also.
However, there are certain quality control requirements for use of mobile laboratories.
First, the data quality objectives (DQOs) for the screening event must be determined
and documented. The DQOs should take into account the fact that "quick and dirty"
screening methods do not generally give high quality precision and accuracy and some
confirmational analysis with a CLP or other commercial laboratory is necessary.
Confirmational analysis should be run on those sample locations which are most critical
. to the project, for instance, the boundary area of a removal action. Secondly, if a
methodology is developed for use, that method must be documented and validated.
Proof of the validation must be provided to the EPA QAO and the QAO must consider it
satisfactory before the method can be used. The method validation should address the
following points, where applicable.
a. analytical objectives
b. method detection limits
c. analytical procedure
d. precision and accuracy
e. calibration
f. quantitation
g. data reduction/validation
h. holding times
Finally, as for any data generated within the Region and which is not validated by
Regional personnel, the contractor's quality assurance officer must sign a summary
statement which describes the quality control measures followed, the quality control
sample results, what data was rejected due to exceedances, etc. This statement should
be supplied to the EPA Project Manager.
-------
XV. VALIDATION OF DATA
A. Contract Laboratory Program
All data generated for use by Region II which is produced by CLP is validated by the
Region with in-house protocol. These data validation standard operating procedures for
organics, inorganics and dioxin data are updated yearly for the current set of CLP
contracts. Application of a protocol which standardizes data useability criteria ensures
that all data shich is used in the Region is of comparable and acceptable quality and
utility. The Region II data validation standard operating procedures are presented in
Appendix XI I.
B. Non-CLP
Data which is generated outside of CLP for use in Region II must be validated in the
same manner as all other data is validated so that a standard useability criteria is
applied to all data used in the Region. All data should be of comparable quality.
The Plan must identify the laboratory to be used if the laboratory is not to be engaged
through the CLP program or if the laboratory does not participate in CLP. All data
produced by laboratories which do not participate in CLP (or if they do participate but
are not directly contracted by EPA) must be validated by the laboratory or primary
engineering contractor according to Region II validation SOPs. Note that these SOPs
apply only to the CLP methodologies and that if different analytical method references
are used (such as SW-846 or Methods for Chemical Analysis of Water and Wastes) the
validation criteria will have to be modified according to the quality control criteria called
for in that methodology used. The QAO of the laboratory must be identified and must
provide a signed document to the EPA Project Manager stating that he/she has
validated the data in accordance with the Region II protocol, or, if quality control criteria
had to be established according to the dictates of the method, the laboratory is
responsible for establishing precision and accuracy protocol and for validating the data
and meeting criteria based on that protocol. A document delineating the criteria used
must be provided along with quality assurance summary sheets, and, if applicable, the
Region II SOP validation checklist. The summary sheets should be taken from the CLP
SOW, the third edition of SW-846 or be based thereon. Data analysis sheets must be
provided for each environmental sample listing quantities found or detection limits.
-------
XVI. FIELD AUDITING AND OVERSIGHT
A. Audits Initiated by EPA and Primary Contractor
On-site audits of EPA contracted and PRP contracted field sampling teams takes place
on a random basis within the Region. EPA personnel performing the audits look for
good sampling technique and ensure that approved Plans are being implemented in the
field. Auditors do make suggestions to contractors in the field but they do not stop work
unless a discrepancy severe enoung to invalidate data results is observed. If a severe
discrepancy is observed the Project Manager is notified by phone when the auditors
return to the office that day.
EPA personnel send written audit reports to the EPA Project Manager following the
audit and request written response from the contractor when inadequacies are found.
B. Contractors in an Oversight Capacity
Contractors retained in an oversight capacity should be looking for good sampling
technique and adherance to an approved Plan when overseeing other contractors in the
field. Logbooks should be kept by the oversight contractor and any poor practices or
discrepancies with the Plan should be noted and the EPA Project Manager notified of
the findings by phone.
C. Audits Performed by States
State Quality Assurance personnel routinely perform audits of Federal-lead sites.
Although State auditors can make suggestions to improve sampling technique and to
ensure that approved Plans are being implemented, they do not have the authority to
stop work in the field or to change any part of the Plan. If a disagreement with State
auditors arises in the field, the contractor should contact the Project Manager for
guidance.
-------
LIST OF APPENDICES
I. Priority Pollutant vs. Target Compound List
II. CLP Statement of Work Target Compound/Element Lists
III. 40 CFR Part 136, Table 2
CLP Holding Time and Preservation Requirements
IV. CLP Documentation
V. Quality Assurance Procedures for Field Analysis and Equipment
VI. Sample Bottle Repository Statement of Work
VII. Recommended Well Drilling Techniques
VIII.'The Effects of Grouts, Sealants, and Drilling Fluids on the Quality of Ground Water
Samples"
IX. Region II Standard Operating Procedure for Selecting Ground Water Well
Construction Material at CERCLA Sites
f
X. Comparison of Surface Water Collection Devices
XI. Comparison of Bottom Grab and Coring Devices
XII. Region II Data Validation Standard Operating Procedures
TECHNICAL PRESERVATION AND HOLDING TIME
REQUIREMENTS*
ANALYTE MATRIX HOLDING TIME v
Volatile Organics Solid 10 days cool to 4
-------
Extractable Organics Solid
7 days to extraction.cool to 4 degrees C
Cyanide
Solid
40 days to analysis
14 days cool, 4
Metals
Solid
6 months
none
Mercury
Solid
28 days
cool to 4
Dioxin (2,3,7,8-TCDD') Solid
6 months
none
-------
Volatile Organics Aqueous preserved: 14 days low/medium cone.:
HCItopH<2
cool 4
degrees C
unpreserved: 7 days cool 4
degre
esC
Extractable Organics Aqueous 7daystoextraction,low/mediumconc.:
40 days to analysis cool 4 degrees C
-------
Cyanide Aqueous 14 days cool 4
NaOH
acid if
oxidizer
high conc.:6
ozglass
present
cadmium
carbonate if
82- present
Metals Aqueous 6 months low/med cone.:
-------
cool 4
HNO3 pH<2
high cone.
none
Mercury Aqueous 28 days low/med
cool 4
-------
HNO3 pH<2
Dioxin (2,3,7,8-TCDD) Aqueous
high cone.
none
10 days to extraction, cool 4 .
40 days to analysis
* All holding times begin on date of sample collection. <
CLP CONTRACTUAL PRESERVATION AND HOLDING TIME
REQUIREMENTS*
ANALYTE
MATRIX
HOLDING TIME
Volatile Organics
Solid
10 days
cool to 4
-------
Extractable Organics Solid 10 days to extraction, cool to 4
40 days to analysis
Cyanide Solid 12 days cool, 4
Metals Solid 6 months none
Mercury Solid 26 days cool to 4
-------
Dioxin (2,3,7,8-TCDD) Solid
6 months
none
Volatile Organics Aqueous preserved: 10 days low/medium cone.:
HCI to pH <2
Extractable Organics Aqueous
cool 4
degrees C
high cone.: none
5 days to extraction.low/medium cone.
40 days to analysis cool 4 degrees C
-------
high cone.
Cyanide Aqueous 12 days cool 4
NaOH
acid if
-------
present
cadmium
carbonate if
82- present
Metals Aqueous 6 months low/med cone.:
cool 4
HN03 pH<2
high cone.:
none
Mercury Aqueous 26 days low/med
-------
cool 4
HNO3 pH<2
high cone.:
none
Dioxin (2,3,7,8-TCDD) Aqueous
10 days to extraction, cool 4
40 days to analysis
Contractual holding times being on date of Verified Time of Sample Receipt.
BIBLIOGRAPHY
-------
1. Characterization of Hazardous Waste Sites-A Methods Manual: Volume II. Available
Sampling Methods, Second Edition; EPA 600/4-84-076; December 1984.
2. Handbook for Sampling and Sample Preservation of Water and Wastewater; EPA
600/4-82-029; September 1982.
3. Samplers and Sampling Procedures for Hazardous Waste Streams; EPA 600/2";
80-018; January 1980.
4. Practical Guide to Ground Water Sampling: EPA 600/2-84/104; September 1984
5. RCRA Ground Water Monitoring Technical Enforcement Guidance Document,
September 1986. Office of Waste Programs Enforcement and Office of Solid Waste
and Emergency Response.
6. A Guide to the Selection of Materials for Monitoring Well Construction and Ground ~
Water Sampling. Barcelona, Gibb, Miller; Illinois State Water Survey, Champaign,
Illinois. January 1984.
7. Test Methods for Evaluating Solid Waste SW-846, Third Edition. Office of Solid
Waste and Emergency Response. November 1986.
8. Methods for Chemical Analysis of Water and Wastes. EPA 600/4-79-020. Revised
March 1983.
9. NEIC Policies and Procedures. United States Environmental Protection Agency
Office of Enforcement and Compliance Monitoring. EPA 330/9-78-001-R. May 1978,
revised May 1986. '
10. Ground Water and Wells. Driscoll, Fletcher G. Second Ed. 1986. Johnson Division,
St. Paul, Minnesota, 55112.
11. Summary Report:Transport of Contaminants in the Subsurface: The Role of
Organic and Inorganic Colloidal Particles, ISIS Seminar, USDOE, October 6-9,1986,
Manteo, NC.
12. The Nature and Properties of Soils, Nyle C. Brady, MacMillan Publishing Co., NY,
Ninth Edition, 1984.
13. User's Guide to the Contract Laboratory Program, USEPA Office of Emergency and
Remedial Response, December 1986.
------- |