EPA 600/R-11/122 | October 2011 | www.epa.gov/ord
United States
Environmental Protection
Agency
A Performance-Based
Approach to the Use
of Swipe Samples in
Response to a Radiological
or Nuclear Incident
>. \
« V »
Office of Research and Development
National Homeland Security Research Center
Office of Radiation and Indoor Air
National Air and Radiation Environmental Laboratory
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
EPA600/R-11/122
October 2011
A Performance-Based Approach to the
Use of Swipe Samples in Response to a
Radiological or Nuclear Incident
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
This report was prepared for the National Air and Radiation Environmental Laboratory of the Office of
Radiation and Indoor Air and the National Homeland Security Research Center of the Office of Research
and Development, United States Environmental Protection Agency. It was prepared by Environmental
Management Support, Inc., of Silver Spring, Maryland, under contract 68-W-03-038, work assignment 50,
and EP-W-07-037, Work Assignments B-41 and I-33, all managed by David Carman. Mention of trade
names or specific applications does not imply endorsement or acceptance by EPA.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
PREFACE
This document describes the various swipe techniques that may be used to sample surfaces
contaminated by radioactive materials following an incident such as the detonation of an
improvised nuclear device (IND) or a radiological dispersal device (RDD) ("dirty bomb"). While
simple in concept, procedures used to take a swipe sample may vary considerably in practice. A
standard method or technique for taking swipe samples does not exist. This means the fraction of
the total removable radioactive surface contamination transferred to the swipe will also vary
depending on the technique used. It is anticipated that a large number of swipes will be taken, so
it is essential that the data generated are accurate so that they will be useful for the decisions that
need to be made. While some may be counted in the field, others will be sent to laboratories for
analysis. This document was developed to provide guidance to those radioanalytical laboratories
that will support EPA's response and recovery actions following a radiological or nuclear
incident.
The need to ensure an adequate laboratory infrastructure to support response and recovery
actions following a major radiological or nuclear incident has been recognized by a number of
federal agencies. The Integrated Consortium of Laboratory Networks (ICLN), created in 2005 by
10 federal agencies,1 consists of existing and emerging laboratory networks across the Federal
Government. ICLN is designed to provide a national infrastructure with a coordinated and opera-
tional system of laboratory networks that will provide timely, high quality, and interpretable
results for early detection and effective consequence management of acts of terrorism and other
events requiring an integrated laboratory response. It also designates responsible federal agencies
(RFAs) to provide laboratory support across response phases for chemical, biological, and
radiological agents. To meet its RFA responsibilities, EPA established the Environmental
Response Laboratory Network (ERLN) to address chemical, biological, and radiological threats
during nationally significant incidents (www.epa.gov/erln/). EPA is the RFA for monitoring,
surveillance, and remediation of radiological agents. EPA will share responsibility for overall
incident response with the U.S. Department of Energy (DOE).
EPA's responsibilities, as outlined in the National Response Framework., include response and
recovery actions to detect and identify radioactive substances and to coordinate federal
radiological monitoring and assessment activities.
As with any technical endeavor, actual radioanalytical projects may require particular methods or
techniques to meet specific measurement quality objectives (MQOs). Uncertainties associated
with swipe samples can be extremely large. Understanding the components of uncertainty,
beginning with the removal factors associated with various combinations of swipe materials,
surfaces, chemical forms of radionuclides, and sampling techniques, will provide Incident
Commanders and other decisionmakers with a better estimate of the uncertainties in swipe-
sample analytical results, how these can affect decisionmaking, and an appreciation of how
uncertainties may be reduced if necessary. However, it is clear that additional research is needed
to develop better sampling techniques, to quantify the limitations of the various techniques, and
to assist in the selection of the technique that is most appropriate to the circumstances.
1 Departments of Agriculture, Commerce, Defense, Energy, Health and Human Services, Homeland Security,
Interior, Justice, and State, and the U.S. Environmental Protection Agency.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Detailed guidance on recommended radioanalytical and site survey practices can be found in the
Multi-Agency Radiological Laboratory Analytical Protocols (MARLAP) Manual and the Multi-
Agency Radiation Survey and Site Investigation Manual (MARSSIM). MARLAP provides
detailed radioanalytical guidance for project planners, managers, and radioanalytical personnel
based on project-specific requirements. MARSSIM provides information for planning, conduc-
ting, evaluating, and documenting building surface and surface-soil final status radiological
surveys for demonstrating compliance with dose or risk-based regulations or standards. These
documents are available at www.epa.gov/radiation/programs.html. Familiarity with Chapter 10
of MARLAP will be of significant benefit to users of this guide.
This document is one in a planned series designed to present radioanalytical laboratory person-
nel, Incident Commanders (and their designees), and other field response personnel with key
laboratory operational considerations and likely radioanalytical requirements, decision paths, and
default data quality and measurement quality objectives for samples taken after a radiological or
nuclear incident, including incidents caused by a terrorist attack. Documents currently completed
or in preparation include:
• Radiological Laboratory Sample Analysis Guide for Incidents of National Significance -
Radionuclides in Water (EPA 402-R-07-007, January 2008)
• Radiological Laboratory Sample Analysis Guide for Incidents of National Significance -
Radionuclides in Air (EPA 402-R-09-007, June 2009)
• Radiological Laboratory Sample Screening Analysis Guide for Incidents of National
Significance (EPA 402-R-09-008, June 2009)
• Method Validation Guide for Qualifying Methods Used by Radiological Laboratories
Participating in Incident Response Activities (EPA 402-R-09-006, June 2009)
• Guide for Laboratories -Identification, Preparation, and Implementation of Core
Operations for Radiological or Nuclear Incident Response (EPA 402-R-10-002, June 2010)
• Uses of Field and Laboratory Measurements During a Radiological or Nuclear Incident (in
preparation)
• Guide for Radiological Laboratories for the Control of Radioactive Contamination and
Radiation Exposure (in preparation)
• Radiological Laboratory Sample Analysis Guide for Radiological or Nuclear Incidents -
Radionuclides in Soil (in preparation)
Comments on this document, or suggestions for future editions, should be addressed to:
Kathleen M. H all Dr. John Griggs
U.S. Environmental Protection Agency U.S. Environmental Protection Agency
Office of Research and Development Office of Radiation and Indoor Air
National Homeland Security Research Center National Air and Radiation Environmental
Laboratory
26 West Martin Luther King Drive 540 South Morris Avenue
Cincinnati OH 45268 Montgomery, AL 36115-2601
(513)379-5260 (334)270-3450
hall. kathy@epa. gov Griggs.John@epa.gov
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
ACKNOWLEDGMENTS
This document was developed by the National Air and Radiation Environmental Laboratory
(NAREL) of EPA's Office of Radiation and Indoor Air in cooperation with and funding from the
National Homeland Security Research Center (NHSRC) of the Office of Research and
Development. Dr. John Griggs was the project lead for this document. Several individuals
provided valuable support and input to this document throughout its development. Special
acknowledgment and appreciation are extended to Ms. Kathy Hall, of NHSRC, and Mr. Daniel
Mackney and Mr. David Garman, both of ORIA/NAREL. We also wish to acknowledge the
external peer reviews conducted by Dr. Eric L. Darois, Mr. Lindley J. Davis, Mr. Edward E.
Walker, Mr. Todd Baker, and Mr. Scott Hudson whose insightful comments contributed greatly
to the understanding and quality of the report. Numerous other individuals, both inside and
outside of EPA, provided peer review of this document, and their suggestions contributed greatly
to the quality and consistency of the final document. Technical support was provided by Dr. Carl
V. Gogolak, Dr. N. Jay Bassin, Dr. Anna Berne, Mr. David Burns, Dr. Robert Litman, Dr. David
McCurdy, Mr. Robert Shannon, and Ms. Leca Buchan of Environmental Management Support,
Inc. EPA wishes to acknowledge Mr. Lawrence Warren for his permission to use his data on
removal factors and strippable coatings (Figures 2, 3, and 4 and Tables 2, 5, 6, and 7) and
Springer Science and Business Media for its kind permission to reprint portions of Sansone
(1987, Table 2) in Appendix B.
in
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Contents
Preface i
Acknowledgments iii
Acronyms, Abbreviations, Units, and Symbols vi
Radiometric and General Unit Conversions viii
1. Introduction 1
1.1 Scope 1
1.2 Background 2
1.3 Current Practice 4
1.4 A Performance-Based Approach 6
2. Measurements of Surface Contamination 7
2.1 Determination of Instrument Efficiency 9
2.2 The Determination of Source Efficiency 9
2.3 Indirect Measurements of Surface Contamination on Swipe Samples 10
2.4 Direct Measurements of Surface Contamination 11
2.5 Comparing Direct and Indirect Measurements of Surface Contamination 12
2.6 Sampling Surface Contamination with Swipes 12
3. Sources Of Uncertainty In Swipe Procedures 13
3.1 Removal Factor 13
3.2 Swipe Surfaces and Materials 16
3.3 Sampling Method 19
4. Radionuclides and Chemical Form 20
5. Counting Methods 21
6. Planning Measurements of Surface Activity Concentration By Means of Swipe Samples .. 25
6.1 Measurement Quality Objectives and the Data quality Objectives Process 25
6.2 Example Scenario 32
6.3 Uncertainty Budget for Example Scenario: Field Total Alpha 33
6.4 Uncertainty Budget for Example Scenario: Laboratory Alpha Spectrometry 36
7. Conclusions 38
8. Recommendations 39
9. References 40
9.1 Sources Cited 40
9.2 Other Sources 43
Appendix A: Review of Regulatory Approach and Applications for Swipes in Radiation
Protection Activities 45
A.I Introduction and Definition of Concepts 45
A.2 Performance of Surveys - Procedures, Typical Sensitivity, and Appropriate
Instrumentation 46
A.3 Applications to Which Removable Contamination Surveys Apply 49
A.4 Conclusions 50
Appendix B: Measurements of Transferable Surface Contamination Reported in the
Literature 52
Appendix C: Examples of Recommended Swipe Procedures 55
IV
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Figures
Figure 1 - Is removable contamination present? A negative result does not necessarily mean
there is no removable contamination present 5
Figure 2 -Surface Activities After Swiping with Demineralized Water 15
Figure 3 -Removal Factors Using "Exhaustive Swipe Sampling" Method 15
Figure 4 - Swipe Activity (10-Second Counts) Versus the Contact Pressure as Measured by
Placing the Linoleum on a Scale While Being Swiped 20
Figure 5 - Example Illustrating Case (a) (see Table 9C, Step 6.1). Baseline Condition (null
hypothesis): Parameter Exceeds the AAL 29
Figure 6 - Example Illustrating Case (b) (see Table 9C, Step 6.1). Baseline Condition (null
hypothesis): Parameter Does Not Exceed the AAL 29
Tables
Table 1 - Surface Efficiencies by Type of Particle and Energy Range 12
Table 2 -Removal Factors Using "Exhaustive Swipe Sampling" Method 16
Table 3 - Swipable Surfaces and Swiping Techniques 16
Table 4- Suggested Swipe Material to Use on a Surface 17
Table 5 - Strippable Coating on Stainless Steel Disks 18
Table 6 - PENTEK 604 Strippable Coating on Contaminated Lead Bricks 18
Table 7 -PENTEK 604 Strippable Coating on Stainless Steel Criticality Barriers 18
Table 8 - Common Chemical Forms of Some Radionuclides that May Be in anRDD 21
Table 9 - Organization of Analytical Techniques into Non-Destructive or Destructive and Gross
or Radionuclide-Specific 23
Table 10A- The DQO Process Applied to a Decision Point 26
Table 10B -Possible Decision Errors 27
Table IOC - The DQO Process Applied to a Decision Point 27
Table 11 - Example Table for Establishing DQOs and MQOs 32
Table 12 - Uncertainty Budget for Gross Alpha Count 35
Table 13 - Uncertainty Budget for Alpha Spectrometry Count 37
Table A-l - Table 1 from NRC RG 1.86 - Acceptable Surface Contamination Levels 47
v
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
ACRONYMS, ABBREVIATIONS, UNITS, AND SYMBOLS
(Excluding chemical symbols and formulas)
a alpha particle
a probability of a Type I decision error
AAL analytical action level
ADL analytical decision level
AL action level
ALARA as low as reasonably achievable
AS alpha spectrometry
AR activity removed by the first swipe sample
AT: total removable activity
P beta particle
ft probability of a Type II decision error
Bq becquerel (1 dps)
CFR Code of Federal Regulations
cm centimeter
cpm counts per minute
d day
DL discrimination limit
DOE U.S. Department of Energy
DOT U.S. Department of Transportation
dpm disintegrations per minute
dps disintegrations per second
DQO data quality objective
Epmax maximum energy of the beta-particle emission
si instrument efficiency for radiation type (alpha or beta radiation)
es source efficiency
EPA U.S. Environmental Protection Agency
ERLN Environmental Response Laboratory Network
F removal factor
y gamma ray
g gram
Ge germanium [semiconductor]
G-M Geiger-Muller [detector]
GP gas proportional
GPC gas proportional counting/counter
GS gamma spectrometry
GUM Guide to the Expression of Uncertainty in Measurement
Gy gray
h hour
HO null hypothesis
HI alternate hypothesis
HPGe high-purity germanium [detector]
1C Incident Commander (or designee)
ICLN Integrated Consortium of Laboratory Networks
ICP-MS inductively coupled plasma mass spectrometry
VI
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
IND improvised nuclear device (i.e., a nuclear bomb)
ISO International Organization for Standardization
keV kilo (thousand) electron volts
KPA kinetic phosphorescence analysis
L liter
LBGR lower bound of the gray region
LS liquid scintillation
LSC liquid scintillation counter/counting
MARLAP Multi-Agency Radiological Laboratory Analytical Protocols [Manual]
MARSAME Multi-Agency Radiation Survey and Assessment of Materials and Equipment
[Manual]
MARS SIM Multi-Agency Radiation Survey and Site Investigation Manual
MDA minimum detectable activity
MDC minimum detectable concentration
MeV mega (million) electron volts
mg milligram (10'3 g)
min minute
mL milliliter (10'3 L)
MQO measurement quality obj ective
mrem millirem (10'3 rem)
ug microgram (10'6 g)
Nal(Tl) thallium-activated sodium iodide detector
«5 background count
NIST National Institute of Standards and Technology
NORM naturally occurring radioactive materials
NRC U.S. Nuclear Regulatory Commission
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
RADIOMETRIC AND GENERAL UNIT CONVERSIONS
To Convert
years (y)
disintegrations per
second (dps)
Bq
Bq/kg
Bq/m3
Bq/m3
microcuries per
milliliter ((iCi/mL)
disintegrations per
minute (dpm)
cubic feet (ft3)
gallons (gal)
gray (Gy)
roentgen equivalent
man (rem)
To
seconds (s)
minutes (min)
hours (h)
days (d)
becquerels (Bq)
picocuries (pCi)
pCi/g
pCi/L
Bq/L
pCi/L
jiCi
pCi
cubic meters (m3)
liters (L)
rad
sievert (Sv)
Multiply by
3.16xl07
5.26xl05
8.77xl03
3.65xl02
1
27.0
2.70xl(T2
2.70xl(T2
1(T3
109
4.50xlO~7
4.50X10"1
2.83xlO~2
3.78
102
io-2
To Convert
s
min
h
d
Bq
pCi
pCi/g
pCi/L
Bq/L
pCi/L
pCi
jiCi
m3
L
rad
Sv
To
y
dps
Bq
Bq/kg
Bq/m3
Bq/m3
(iCi/mL
dpm
ft3
gal
Gy
rem
Multiply by
3.17xl(T8
1.90xl(T6
1.14x10^
2.74xlO~3
1
3.70xl(T2
37.0
37.0
IO3
io-9
2.22
2.22xl06
35.3
0.264
io-2
IO2
NOTE: Traditional units are used throughout this document instead of International System of
Units (SI) units. Protective action guides (PAGs) and their derived concentrations appear in
official documents in the traditional units and are in common usage. Conversion to SI units will
be aided by the unit conversions in this table.
Vlll
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
1. INTRODUCTION
If a radiological or nuclear device were detonated in the United States, a large number of
samples would need to be analyzed to assess and control the spread of fine particulate matter
contaminated by radionuclides. This document deals with the analysis of swipe2 samples from
areas that may have been contaminated as the result of a radiological or nuclear event, such as a
radiological dispersal device (RDD), improvised nuclear device (IND), or intentional release of
radioactive materials. In the event of a major incident that releases radioactive materials to the
environment, EPA will turn to radioanalytical laboratories to support its response and recovery
activities. In order to expedite sample analyses and data feedback, the laboratories will need
guidance on the EPA's expectations. Three response phases are defined by EPA:
• The early phase, also known as the emergency phase, is the initial reaction to the emergency
and can last for a few hours or up to a few days. During this phase, single-point swipes at
both random and targeted locations are taken. This information is used to establish contam-
ination controls for areas, structures, and components; to assess decontamination and dose
assessment requirements for affected members of the public; and to assess protection factors
and respiratory protection requirements for initial responders and for recovery planning.
• The intermediate phase initiates when the immediate emergency situation is under control
and reliable environmental measurements are available for use as the basis of additional
protective actions. This phase may overlap with the other two phases and can last from weeks
to months. Swipe data are used to assess the adequacy of—and ongoing requirements for—
radiation protection and controls for recovery personnel and for assessing progress of
decontaminati on.
• The late phase, also known as the recovery phase, begins with recovery actions. Recovery
actions are designed to reduce radiation levels in the environment to levels acceptable for
unrestricted use. Swipe data are used to assess final radiological conditions with respect to
incident goals and limits.
1.1 SCOPE
This document focuses primarily on the intermediate and late phases. Many of the procedures are
designed to detect and characterize radiation in potentially contaminated areas. However, as
discussed in Section 1.3, "Current Practice," the limitations and potential sources of uncertainty
introduced by these procedures will impact the decisionmaking processes in each phase. For
example:
• Intermediate phase: False negatives may lead to inadequate worker protection, inadequate
decontamination, potential release of contaminated surfaces or components, adverse public
and political reactions, and additional costs to recover from each. False positives lead to
over-protection for workers and/or excessive decontamination; both resulting in adverse costs
and schedule issues.
The terms "swipe," "wipe," and "smear" are often used interchangeably in the literature. In this document, the term
"swipes" is used, unless quoting directly from another document
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
• Late phase: False positives result in excessive decontamination, extended schedule, and cost
impacts, while false negatives again result in potential release of contamination with
attendant public/political reactions and recovery costs.
During the long-term recovery phase of incidents involving RDDs or INDs, EPA will lead the
radiological environmental characterization and will manage the federal radiological cleanup
activities. While field detection capabilities can quickly be used to take action following an
incident, more extensive and time-consuming fixed-laboratory analyses will be needed to assess
whether the public can resume normal use of the affected areas. Such assessments are
intrinsically different from those used, for example at nuclear facilities, to monitor and control
the spread of contamination within radiologically controlled areas. A site-specific optimization
process that incorporates local needs, health risks, costs, technical feasibility, and other factors
will be used to establish cleanup levels. This process is designed to be transparent and involve
representative stakeholders in the decisionmaking process. The data upon which these decisions
will rely must therefore meet the same quality criteria required in other EPA cleanup activities.
In some cases, swipe samples may be a key source of information for these decisions. However,
current practice uses swipes often as a tool for qualitatively assessing the presence of removable
activity on surfaces. Following an RDD or IND, more quantitative assessments will be needed
for recovery phase decisions, and the data will need to be reported with realistic uncertainty
evaluations. This new application will present a paradigm shift in how swipes are taken, how the
data are analyzed, and how the results are interpreted.
This document is meant for those with a background in ionizing radiation and radioactivity
measurements. It is intended to assist in the planning and implementation of surveys using swipe
samples to evaluate the amount of radioactivity on surfaces that might be separated from a
surface under normal or light abrasive contact. This document provides recommendations for the
process of taking swipes and evaluating the results. The objectives of this document are to:
• Review the current practice in obtaining swipe samples;
• Compare direct and indirect methods for assessing surface contamination;
• Provide a framework for evaluating the results;
• Provide general recommendations on the process of taking swipes; and
• Make specific recommendations concerning how swipe samples should be taken and
analyzed following an RDD or IND.
1.2 BACKGROUND
Fixed contamination refers to the portion of contamination that remains attached to a surface
after reasonable attempts to clean or decontaminate that surface. Contamination that is fixed to
the surface would not be transferred to the body and is usually of concern only as a source of
external exposure, unless it becomes loose and is redistributed. However, it is the removable
contamination that is transferrable to humans by contact, inhalation, or ingestion that poses a
hazard of internal radiation exposure. The amount of removable contamination is usually
determined by obtaining swipe samples. Usually, only a portion of the removable activity is
collected on the first swipe, and several swipes are needed to assess the total removable activity.
The interpretation of swipe sample results quantitatively is generally difficult because the
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
sampling and measurement uncertainties are often not adequately evaluated. However, for an
RDD there are some details of the circumstances that will be known soon after the event. These
will aid in the interpretation of the results and will reduce the overall uncertainty. The situation
for INDs may be more complex due to the increased number of radionuclides likely to be
present.
The levels of removable contamination that are of concern may be much lower than those for
fixed contamination. The intensity and effective duration of the exposure to radiation from
internally deposited radionuclides can result in a committed dose of radiation. These internal
doses may be much greater than the external dose that may be received from sources of radiation
outside the body. These external doses are delivered only when in the presence of those sources.
The amount of removable surface contamination transferred to a swipe sample will vary
according to the:
• Type of swipe material,
• Method used,
• Physical and chemical nature of the contaminated material,
• Surface roughness of the material swiped, and
• Physical and chemical nature of the radionuclide contaminant(s).
In order to determine the extent of surface contamination of materials and the effectiveness of
the decontamination processes reliably, the swipe removal factor for the contaminant must be
determined for the various materials swiped. The removal factor is the ratio of the activity of the
radionuclides removed from the surface by one initial swipe sample to the total removable
activity. This swipe removal factor must then be applied when evaluating the radioanalytical
results of the swipe samples. Among the many types of swipe materials used are dry swipes that
use various dry absorbent materials such as glass and cellulose fibers, and wet swipes used by
application of various solvents to the dry swipe to enhance the amount of material removed from
the surface.
A recent review of swipe sampling methods for chemical and biological agents (EPA, 2007)
documented that, in most cases, cotton or gauze was used as the swipe material. Usually paper,
cloth or glass fiber filter materials are used for radioactivity swipes. Although swipes of
radioactivity were not covered, some of the conclusions are relevant:
[I]t is clear that there is not an overwhelming consensus on how to take a [s]wipe sample... from
surfaces. Different methods, media, and wetting solvents have been recommended and used by
various groups and studies. Many of the compounds... do not even have a specific [s]wipe
sampling methodology for their collection. If the goal is to establish a [s]wipe sampling method
(or methods) for the compounds discussed in this report, then the next steps in this process must
be researched] to investigate and fill in the gaps in [s]wipe sampling knowledge that exist,
followed by method validation to optimize the methods. (EPA, 2007)
One can come to many of the same conclusions for swipes used to sample surface radioactivity.
As an RDD or IND event unfolds, many specifics about the radionuclides, chemical form, and
surfaces involved will become available. Applying a consistent swipe sampling method using
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
this information will greatly aid in reducing the uncertainty of the results.
References used in this document are listed in Section 9.1. Other sources, including discussions
with subject matter experts and additional Internet searches, are listed in Section 9.2. Although
there is universal agreement that there are large uncertainties in the results of swipe sampling
that may complicate their interpretation, there does not appear to be a lot of active experimental
research going on in this field. Many studies exist, but they are not readily generalized to
contaminants, solvents, surfaces, and detection methods other than those considered in the
specific experiments reported. Appendix A contains a review of Department of Energy (DOE),
Nuclear Regulatory Commission (NRC), and Department of Transportation (DOT) regulations
and regulatory guidance documents that identify various uses of swipes that might be applicable
to incident response activities, especially in the intermediate and recovery phases.
1.3 CURRENT PRACTICE
Despite their high variability and poor detection limits, [s]wipe tests remain one of the universally
accepted techniques for detecting removable radioactive contamination on surfaces. It is often a
stipulation of radioactive materials licenses and is widely used by laboratory personnel to monitor
their work areas, especially for low-energy radioisotopes that are otherwise difficult to detect with
hand-held survey instruments. (Klein et al., 1992, Klein et al., 1997, Campbell et al., 1993).
Swipe samples may be taken over a designated surface area of interior and exterior building
surfaces and public/private transportation vehicles, sidewalks, streets, ventilation systems, and
various equipment and objects.
For compliance measurements, swipes are taken in a prescribed manner, and the measured
activity on the swipe is compared to a limit specified by the cognizant regulatory agency. This in
effect defines "removable" as what is transferred to a swipe. The removal efficiency or other
sampling and analysis factors that may affect the uncertainty in the swipe result do not appear to
be examined routinely. However, it is known that generally not all removable activity will be
removed on a single swipe. The sum of an exhaustive3 series of swipe samples should be used.
Since this is not often practical, such a series of samples is done only occasionally in order to
define a removal factor that is the fraction of the removable activity removed by the first swipe
sample.
A good summary of the current state of the art of interpreting swipe samples is given by Frame
and Abelquist (1999). Although there appears to be a consensus in the literature that swipe
samples provide important information about human exposures to radioactivity in the
environment that often cannot be obtained otherwise, there is also agreement that swipe sample
results for removable contamination may have a very high uncertainty. Despite this recognition
that the uncertainties can be large, the results of swipe sampling are seldom reported with their
associated uncertainties. The lack of this information can adversely affect the data usability for
decisionmaking. Thus, the procedures for swipe sampling and interpretation may benefit from a
reassessment and reevaluation in the context of modern data quality objectives.
3 "Exhaustive" in this context means swipe sampling until 10% or less of the first swipe's activity is obtained.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Frame and Abelquist (1999) observe that,
At a minimum, it may be concluded that removable activity is present when smear [swipe] results
exceed the critical level of the smear [swipe] analysis procedure. On the other hand, if a smear
[swipe] result indicates that activity does not exceed background levels, it might be inappropriate
to conclude that removable activity is not present since the collection efficiency could be near
zero.
In essence, if a swipe sample shows contamination, it confirms that there is removable
contamination present. However, if a swipe sample does not show contamination, it cannot
automatically be concluded that there is no removable contamination present. Note that directly
measuring the surface using a field instrument will give an estimate of the sum of the fixed-plus-
removable activity. It is also true that this measurement may be influenced by the self-absorption
of the material due to surface characteristics. A different concern exists for physical samples of
the surface taken for radiochemical analysis. Physical samples taken from the surface for
analysis by radiochemical separation also will not provide information regarding the removable
versus fixed content of the surface. It should be noted that each approach to the determination of
surface contamination (direct measurements, swipes, or physical samples) has its benefits and
limitations.
If removable contamination levels are present at low levels, they may not be detectable by use of
swipes due to low transfer efficiency. Additionally, such low levels may not be detectable with
portable survey instruments where a direct surface measurement is made, due to interference
from background activity. The only technique for assessing low levels with a greater degree of
certainty would be by use of "grab" sampling4 of the material with subsequent destructive
analysis at a laboratory. Thus, the use of swipes, field monitoring equipment, and grab sampling
for laboratory analysis are all important components of assessing the state of radioactive
materials contamination of a surface (Figure 1). Field measurements and grab samples generally
will result in an estimate of the sum of fixed and removable activity, whereas swipes generally
will result only in an estimate of removable activity.
Positive
Swipe Result
Negative
Swipe Result
Removable Activity Present
• No Removable Activity
• Removal Factor Low
•Low Energy p or a
c
Take Grab Sample
Figure 1 - Is removable contamination present? A negative result does not necessarily
For purposes of this document, a "grab sample" means a single physical sample collected at a particular time and
place that represents the composition of the surface under study (www.epa.gov/ocepaterms/gterms.html').
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
mean there is no removable contamination present.
1.4 A PERFORMANCE-BASED APPROACH
As indicated in the previous section, a key issue is how to interpret swipe measurements that
result in a decision that no removable contamination is detected, i.e., the measurement is a "non-
detect." Without a statement of the level of removable activity that could have been measured
had it been there, such a result contains little useful information. The minimum detectable
activity (MDA) is a measurement quality objective (MQO) that specifies the level of removable
contamination that can be detected with high probability, if present, with the measurement
method being used. It is not enough to report "not detected;" one also must specify how thorough
the sampling process must be. Doing this requires developing MQOs for the detection of
contamination of a specified type on a specific surface using a particular sampling and analysis
protocol. To calculate the MDA, there must be some estimate of the uncertainty in a
measurement taken near background. Indeed, the use of any measurement without an
accompanying statement of the uncertainty in that measurement is virtually useless for
decisionmaking. This is a point that was stressed in the performance-based approach contained in
MARLAP, which includes a method for setting MQOs based on project-specific data quality
objectives (DQOs). This topic is covered in greater detail in Section 6.1. The connection between
DQOs and MQOs is important not just to laboratory measurements, but to any measurement,
including those made with field survey instruments. The MARLAP (2004) approach to DQOs
and MQOs was extended to cover such measurements in MARSAME (2009), where specific
procedures and examples are given.
The fundamental MQO in MARLAP is the required method uncertainty at the action level. In
each case, an action level should be specified for the scenario so that it can be determined that
the sampling and analysis processes used have the required method uncertainty or sensitivity
(e.g., minimum detectable surface activity) to be useful for the decision that must be made on the
basis of the data. To advance to a more quantitative interpretation of swipe results, it is necessary
to examine the various sources of uncertainty associated with the measurement. Several factors
should be considered, including:
• Radionuclide(s) of concern and their chemical form,
• Removal technique (including human factors),
• Removal factor of the surface contamination with the initial swipe, and
• Debris matrix that the specific radionuclide is encapsulated in as part of the swipe.
Applying the uncertainty to the decisionmaking process, including conclusions to be drawn from
"non-detects," requires consideration of the:
• Action level,
• Analytical decision level,
• Discrimination level,
• Acceptable error rates, and
• Detection capability.
Specific sources of uncertainty in swipe measurements are discussed in greater detail in Section
3. The subject of specifying requirements for limiting uncertainty is covered in Section 6.1.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Examples of how to evaluate the uncertainty in a swipe sample measurement are given in
Sections 6.2-6.4.
The components of uncertainty for each part of the swipe sampling methodology will generally
be large. However, when a positive indication of activity is found, the result of that measure-
ment, even though not known with a high degree of accuracy, can provide useful information
regarding the type of radioactive materials and the ability of removal techniques to reduce the
quantities on the surface. For a specific scenario, such as the case of an RDD scenario with a
source term with defined radionuclides and chemical form, it may be possible to produce an
uncertainty budget (see Sections 6.3 and 6.4) that can identify the dominant sources of
uncertainty for that radionuclide and for specific surfaces. From this, one can determine those
sources of uncertainty that may be easily reduced, minimized, or eliminated. Uncertainty budgets
may be more complicated for INDs for several reasons, including the increased number of
radionuclides likely to be present.
MARLAP (2004) identifies two kinds of questions one commonly asks about samples of
radioactivity: (1) is there something there? and (2) how much is there? For swipes, the amount of
detail necessary to answer the second question with any confidence is often not available, or not
used, even in controlled, non-emergency situations. This has consequences for the usability of
this data for making decisions. A major objective of this document is to suggest improvements to
the current approach for taking and analyzing swipe data, so that the results can be used with
greater confidence for the decisions being made.
The result of a swipe sample analysis generally is used to make a simple detection decision, i.e.,
MARLAP question (1), is there something there? Answering that question requires the
determination of the major sources of uncertainty both near background and at the action level,
so that an appropriate MDA can be established for the swipe samples. The first step would be to
list all of the potential sources of uncertainty in the swipe measurement, in some way estimate
their contribution to the total, and assemble an uncertainty budget as illustrated in the Guide to
the Expression of Uncertainty in Measurement (GUM; ISO 1995). A swipe sample not showing
contamination does not aid the decisionmaking process unless it is accompanied by the
uncertainty associated with the measurement. Otherwise, it cannot be concluded that there is no
removable contamination present at the action level. A quantitative interpretation of a swipe
result, i.e., an answer to MARLAP question (2)—how much is there?—also requires that both
the measurement result and its uncertainty be reported. Only then can an objective statement be
made about the probability that a given amount of removable contamination is present. This then
allows the decisionmaker to assess the chance of a decision error.
2. MEASUREMENTS OF SURFACE CONTAMINATION
As previously mentioned, there are generally considered to be two types of surface contamina-
tion: fixed and removable. Removable surface contamination is contamination that may be
removed by normal, non-destructive contact with surfaces (a swipe sample is intended to mimic
this type of contact), or by reasonable attempts at cleaning and decontaminating. Fixed contam-
ination refers to the portion of contamination that remains.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
There are also two ways to measure surface contamination: directly and indirectly. Direct
measurements of surface contamination are made by means of a field survey meter or monitor. A
direct measurement estimates the combined fixed-plus-removable contamination, but may also
include contributions from interfering radioactivity inherently present in the surface being
measured or from ambient sources in the area adjacent to the measurement. An indirect
evaluation of removable surface contamination is made by taking and subsequently analyzing a
swipe sample or a grab sample.
The contrast between direct and indirect measurements in the context of the evaluation of surface
contamination shares some common characteristics with the more general contrast between field
and laboratory measurements. A companion EPA report in this series, Uses of Field and
Laboratory Measurements During a Radiological or Nuclear Incident (in preparation) discusses
some of these characteristics.
A swipe is obtained by taking a sample of removable activity by rubbing the surface with dry or
wet material. A single swipe will not normally remove all potentially removable contamination.5
The removal factor is the ratio of the activity of the radionuclides removed from the surface by
one swipe sample to the total removable activity of the surface prior to this sampling. The
removal factor, F, is defined by the following relationship:
F=A1/ASum (1)
where:
AI is the activity removed by the initial swipe sample.
AJ is the activity removed by the /'* swipe sample.
ASumis the total activity of the removable surface contamination prior to taking the first
swipe sample. It is estimated by summing a set of repetitive swipes, ASum =
The removal factor should be determined experimentally for each set of measurement conditions
encountered, using the method of exhaustive removal by repetitive swipe sampling.6 The sum of
the activities removed by repetitive swipes yields an estimate of the total removable activity AT.
This can then be compared to the activity removed by the initial swipe sample, AI, to yield the
removal factor. If it is not possible to determine the removal factor experimentally, a value of F
= 0.1 is sometimes assumed (see for example 49 CFR 173.443(a)(l)). Depending on what, if any,
prior knowledge of the surface characteristics is available, the use of this 0.1 value as a default
may introduce a fairly large uncertainty into the result that may not be reflected in the
uncertainty reported with the result.
5 It should be noted that under the influence of natural physical and chemical interactions in the environment, fixed
contamination may become removable, or removable contamination may become fixed. The total surface activity
may also be reduced by such processes.
6 If the removal efficiency is very low and the first swipe is the only one that shows significant activity, the removal
factor may be erroneously overestimated as 100%.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
2.1 DETERMINATION OF INSTRUMENT EFFICIENCY
The overall efficiency for measurements of surface activity is theoretically considered to be the
product of two factors: the instrument efficiency (EJ ) and the source efficiency (es). Thus, the
overall efficiency is £j es.
The instrument efficiency is defined as the ratio of the response of an instrument relative to the
surface emission rate of a source in a specified measurement geometry. ISO 7503 (1988) defines
surface emission rate as the number of particles of a given type above a given energy emerging
from the front face of the source per unit time. The maximum instrument efficiency for a surface
measurement is 1. Quantitative measurements of radioactivity are not possible given instrument
efficiency alone; rather, they require separate determinations of source efficiency. Considerations
impacting determination of source efficiency are addressed in the subsequent sections below.
The instrument efficiency for a radiation detector must be determined for each geometry in which
it will be used. The instrument efficiency is measured by exposing the detector to a wide-area or
a planchet calibration source with a NIST-traceable surface emission rate7 to a detector in a
geometry that matches that of measurements for which the detector will be used.
This instrument efficiency, Sj, for the radiation type (alpha or beta radiation) is calculated
according to:
c =nlt-nBltB ^
Vz*
where:
£j = instrument efficiency for radiation type (alpha or beta radiation);
n = measured total count from the reference source;
nB = background count;
t = total count time of the reference source [s];
fe = background count time [s]; and
q2ji = surface emission rate of the calibration source for radiation type (alpha or beta
radiation) incident on the detector face [s *].
The surface emission rate incident on the detector face will depend on the area of the source
subtended by the detector probe, and so will vary with source-detector distance. This effect is
accommodated by determining the instrument efficiency in the same geometry used for field or
sample measurements.
2.2 THE DETERMINATION OF SOURCE EFFICIENCY
As stated above, the overall efficiency is expressed as the product of the instrument efficiency
and the source efficiency, e. es. The determination of the instrument efficiency was addressed in
7 It is important to distinguish sources calibrated for surface emission rate from those calibrated in terms of total
source activity. Total source activity is generally expressed in units of absolute radioactivity, such as Bq, dpm, or Ci
per source, and is usually determined by the manufacturer of the source by correcting the measured surface emission
rate of the source to reflect the presumed backscatter and self-absorption properties of the source.
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
the previous section. The source efficiency is defined as the ratio of the number of particles or
photons of a given type emerging from the front face of a source and the number of particles of
the same type created or released within the source per unit time. Thus, the source efficiency is
the ratio between the emission rate of the source (or surface) and the activity contained in the
source. The source efficiency takes into account the increased particle emission due to
backscatter effects, as well as the decreased particle emission due to surface roughness and self-
absorption losses. These factors are usually unimportant for gamma radiations on a swipe
sample, so the source efficiency can be estimated to be 0.5. Also, for an ideal source (i.e., no
backscatter or self-absorption), the value of the source efficiency is 0.5. Many real sources will
exhibit values less than 0.5, although values greater than 0.5 are possible, depending on the
relative importance of the absorption and backscatter processes. The source efficiency for swipe
samples of alpha and beta radiation will not vary as much as the source efficiency for surface
materials encountered in the field. Values for s can be determined for a specific swiping
protocol in advance, thus reducing the uncertainty in quantifying swipes measurements.
Otherwise, the values given in Table 1 in the next section could be used as defaults. For installed
instruments with 4-rc counting geometry (i.e., liquid scintillation counters), the source efficiency
is essentially one.
2.3 INDIRECT MEASUREMENTS OF SURFACE CONTAMINATION ON SWIPE
SAMPLES
For instruments with a 2-rc counting geometry (i.e., gas-flow proportional counters, scintillation
counters, portable survey meters, etc.), the activity per unit area, ^R, of the removable con-
tamination of the swiped surface, expressed in Bq cm" , is given by the equation:
R e,-F-S-£s
where:
r\
AR = activity of the swipe sample per unit area [Bq cm" ];
n = measured total count from the swipe sample;
TIB = background count;
t = total count time of the swipe sample [s];
IB = background count time [s];
S = surface area [cm ];
F = removal factor;
Sj = instrument efficiency for radiation type (alpha or beta radiation); and
e = source efficiency.
s
In some cases, it is possible to determine empirically the overall detection efficiency for a swipe
geometry by measuring the response of the detector to a swipe spiked with a known (i.e., NIST
traceable) activity of the radionuclide of concern. Thus, the measured swipe count rate can be
converted directly to swipe activity without separate determinations of instrument and source
efficiency. When this method is possible, it is important that the conditions of the calibration be
the same as those used for the measurement of the swipe sample. These conditions include,
among other things, the source-detector geometry, the distribution for activity on the surface, the
depth of penetration of the activity, and the type of material and surface roughness of the swipe
material.
10
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
2.4 DIRECT MEASUREMENTS OF SURFACE CONTAMINATION
The determination of an overall efficiency, as described above for swipes, has largely been
abandoned for direct measurements of contamination due to the difficulty of obtaining
representative NIST-traceable standards that will match the distribution for activity, the depth of
penetration of the activity, and the type of material and surface roughness of surfaces
encountered under field conditions.
The total surface activity per unit area ATotal of fixed-plus-removable contamination on the
surface being checked, expressed in s"1 cm"2, in relation to the measured count rate n/t, is given
by the equation:
nlt-nKltK ...
where:
r\
W= detector total window area (including screens) [cm ] (the other factors are as defined in
Equation 3).
Determining the source efficiency used to estimate the level of fixed-plus-removable surface
contamination in Bq cm"2 from direct measurements is a difficult task, considering sources of
uncertainty entering into this determination. An ideal assumption would be that under ideal
conditions, 50% of the emitted decay particles are incident on the detector because radioactive
material emits particles into 4-7T radians and only half of these particles can be incident on the
detector face. Source-detector geometry (e.g., distance, relative area) can cause variations from
this.
Source efficiencies should be determined experimentally whenever possible. Documents such as
ISO-7503-1 (ISO 1988) and NUREG-1507 (NRC 1995) provide suggested default surface
efficiency values for beta- and alpha-emitters (see Table 1, for example). These values do not
have a strong technical basis and may not always be conservative. They do provide, however, a
sense of the relative impact of different surfaces on detector response. For gamma-emitters and
medium-to-high-energy beta-emitters (Epmax > 0.4 MeV) on smooth clean surfaces, es is
estimated as 0.5. ASTM E1893-08 also includes data and references for determining source
efficiency factors.
Radiation from low-energy beta-emitters (0.15 MeV < Epmax < 0.4 MeV) and alpha-emitters can
be blocked by very thin materials such as grease, moisture, dust, and paint. Measurements of
beta activities for energies <0.15 MeV, using gas-proportional or G-M tube counting techniques,
will not provide very reliable data due to self-absorption and low detector efficiency.
It is very difficult to make a direct estimate of the surface contamination under these
circumstances, and any such estimate will have a high degree of uncertainty associated with it.
Consequently, indirect methods may provide the only practical means for obtaining adequate
data for low-energy alpha- and beta-emitters. In these cases, sampling by continuous swipe tests
or even physical abrasion of the surface may be the only reasonable sampling approach. This
11
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
would be followed by indirect measurements of these samples by liquid scintillation counting
(LSC), which may provide a much better means of analysis for low-energy beta-emitters (see
subsequent discussion).
Table 1 - Surface Efficiencies by Type of Particle and Energy Range
Type of particle and energy range (MeV)
Gamma
Beta (Epaac> 0.4)
Beta (0.15 <£,»„««< 0.4)
Beta (Epmax< 0.15)
Alpha
«i
0.5
0.5
0.25
See note
0.25
NOTE: For low-energy emissions, the absolute efficiency may be so low, and the
associated uncertainties so high, that measurements of these are of very limited utility.
2.5 COMPARING DIRECT AND INDIRECT MEASUREMENTS OF SURFACE
CONTAMINATION
A direct measurement of surface contamination is a combined estimate of fixed-plus-removable
contamination, without distinguishing between the two. An indirect measurement of surface
contamination is an estimate of removable contamination.
In terms of uncertainty, one can observe that the input factors for calculating both of these
quantities involve many of the same factors. Typically, for direct measurements, the source
efficiency, es, will be a major source of uncertainty. At low activity levels, the uncertainty
associated with background may be the predominant source of uncertainty. For low-energy beta
and alpha radiation, the source efficiency may become vanishingly small. The same is true for
low-energy gamma-emitters and X-ray-only-emitters. In this case, both indirect and "grab"
sampling methods will be necessary. Also, indirect measurements may be the only way to
estimate the removable contamination level. For indirect measurements, the uncertainty of the
source efficiency will generally be smaller. Since the swipe material is always the same, it may
be possible to estimate ss experimentally for a given type of radiation. To do the same for direct
measurements, a sample from every surface followed by radiochemical analysis would be
needed. Indirect swipe-sample measurements of removable activity have a key source of
uncertainty that does not appear for direct measurements of combined removable-plus-fixed
activity: the removal factor, F.
The removal factor is the ratio of the activity of the radionuclides removed from the surface by
one swipe sample to the total removable activity. The removal factor can easily vary as much as
the source efficiency, although for different reasons. The removal factor is discussed in more
detail in Section 3.1.
2.6 SAMPLING SURFACE CONTAMINATION WITH SWIPES
There appears to be no consensus on what actually constitutes a swipe. There are many materials
and procedures for taking swipe samples. The guidance is often very general, e.g., a swipe
sample is taken by rubbing, with slight pressure, a piece of soft filter paper over a representative
12
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
type of surface area. Different facilities following the same guidance may have very different
standard operating procedures for carrying out the actual sampling; thus, results tend not to be
comparable. Yet the results are often compared to the same set of limiting values. While the
general description of a swipe procedure is similar in many cases, the particulars may vary
considerably. The pressure used will vary considerably among those individuals doing the
sampling. The area sampled is often specified to be 100 cm2. However, this is usually done
freehand, so the area sampled can vary substantially. The pattern of wiping can also vary (e.g.,
circular, serpentine, etc.). Appendix C gives examples from EPA, ISO, and 10 CFR 835 on how
to take swipe samples. Materials commonly used are Whatman #41 filter papers or glass-fiber
filters. However, these can be wet or dry, and the efficiency of the solvent used on wet swipes
depends on how well it dissolves the chemical form of the contamination being removed.
Project- and surface-specific calibrations together with careful estimates of uncertainty are
currently the best ways to address and to minimize these issues. An uncertainty budget analysis
can pinpoint those input parameters most needing additional study.
3. SOURCES OF UNCERTAINTY IN SWIPE PROCEDURES
Understanding the potential uncertainties in a sampling and measurement process is essential to a
performance-based MARLAP process. DQOs and MQOs cannot be established without this
understanding. Some of the more important sources of uncertainty specific to swipe sampling
and analysis are considered below.
3.1 REMOVAL FACTOR
One fact that becomes immediately apparent is that the removal factor is an important parameter
in interpreting swipe results. It seems to be universally accepted, although not commonly
acknowledged in practice, that only a fraction of the removable contamination is sampled with
the first swipe. Some of the parameters that affect the removal factor are the chemical form of
the contaminant, and how it interacts or is absorbed on the surface under consideration. This can
affect the fixed component, but the chemical form of the contamination will also affect the
removal factor depending on whether dry or wet swipes are used.
There are large differences even for different dry swipe media. Hogue (2002) states:
Regulatory requirements for smear [swipe] materials are vague. The data demonstrate that the
difference in sensitivity of smear [swipe] materials could lead to a large difference in reported results
that are subsequently used for meeting shipping regulations or evaluating workplace contamination
levels.. .Available data on the sensitivity of smear [swipe] material are scarce.
The removal factor will depend on whether the contaminant has a greater affinity for the surface
being sampled or the swipe material. Campbell et al. (1993) investigated the difference in tritium
swipe efficiencies using dry swipes and wet swipes with either polar or non-polar solvents and
found that the removal factors were directly related to the solubility of the tritium compounds in
the swipe solution.
Sansone (1987) reviewed measurements of "transferrable" surface contamination and concluded
13
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
that among the many variables affecting the determination of such contamination were:
... [T]he specific chemical compound involved, its chemical state, the manner in which it is applied,
the particle size of the material and the degree to which the material adheres to the surface.
The surface itself, its composition, and its roughness are also factors. The results of his extensive
review are contained in Appendix B.
Royster and Fish (1967) looked at the removal factor for thorium dioxide dust particles using
adhesive paper, swipes with Whatman #50 paper, and a device called a "smair," which dislodges
material by an air flow and captures it on a filter on surfaces as smooth as glass and as rough as
unsealed concrete. The percent removed varied from 1.32% to 78%. The highest removal tended
to be for adhesive paper on smooth surfaces. The smair sample removal was much smaller. The
typical swipe sample had removal factors of about 25% to 75%.
Jung et al. (2001) found it took up to 10 consecutive swipes to obtain a good estimate of the total
removable activity, even on relatively smooth surfaces, with the initial swipe having a removal
factor of 10% to 20%. The variability in swipe results, even from portions of surfaces that would
be expected to give consistent results, was greater by an order of magnitude or more than the
counting statistical errors. ISO 11929-7 (2005) contains an example calculation for the uncertain-
ty and limit of detection for swipe samples. It models the removal factor with a rectangular
o
distribution from 0.06 to 0.62 with a mean of 0.34, based on the results of previous experiments.
Yu et al. (2003) reviewed the literature and concluded by modeling the removable fraction as a
triangular distribution9 from 0 to 1 with a mode (most likely value) at 10%. The removal factor
can be determined experimentally using the ISO 7503 method of "exhaustive removal by
repetitive [s]wipe tests." The step-by-step addition of the removable activities leads to a good
approximation of the total removable activity (AT) to which the activity removed by the initial
swipe test (AR) can then be related to yield the removal factor.
Warren (2007)10 examined removal factors using 32P as orthophosphate in dilute hydrochloric
acid on various surfaces swiped with Whatman GF/A 60-mm smear paper folded twice in half to
form a four-leaved triangle. The swipe was obtained by holding the middle two leaves of the
folded paper with tongs. Figure 2 shows Warren's results with demineralized water wet swipes
on various materials. The surfaces included linoleum, aluminum, stainless steel, ceramic tile, and
Perspex.11 Stainless steel and aluminum show a high level of fixed contamination even after 11
swipes. As a rule of thumb, 10 swipes are probably enough to be considered exhaustive, but
fewer may be adequate as long as the surface activity has leveled off sufficiently to determine a
removal factor. The other surfaces showed both lower fixed contamination and higher removal
factors for the first swipe. Figure 3 shows the removal factors he obtained using the "Exhaustive
8 A random variable with a rectangular distribution is equally likely to take on any value between its endpoints, so
the distribution is flat like a rectangle.
9 A triangular distribution rises linearly from its lower bound to its most likely value, then decreases linearly to the
upper bound, so the distribution is peaked like a triangle.
10 EPA wishes to acknowledge the kind permission of Lawrence Warren to reproduce his data in Figures 1, 2, and 3
and Tables 2, 5, 6, and 7.
11 Perspex is a hard, transparent plastic, an acrylic resin similar to Plexiglas.
14
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Swipe Sampling" method for different materials and for dry and wet swipes with different
wetting agents, 10% DECON 90,12 demineralized water, and methanol. The bar chart makes it
easy to see that the variation with dry and wet swipes with different solvents seems to have as
much of an effect as the type of surface. The data are shown numerically in Table 2.
1400 :
-
1200 -
-1000 :
•— onn "
u -
$ 600 :
Z 400 :
200 '-
0 4
x
x
\ ^
l^V *""* -
1 ^^ """^
w
1
V*
f »
i 1
I 1
1 \
y^^rr:
1 23456789 10 11
Numbei of Me.isiiieinents
Linoleum
Aluminum
Ceramic Tile
Peispex
Redraw from Warren
(2007), by permission
Figure 2 -Surface Activities After Swiping with Demineralized Water
1.00 -
0.90 -
0.80 -
0.70 -
0.60 -
0.50 -
0.40 -
0.30 -
0.20 -
0.10 -
Onn
r
E
E
E
j
E
1 r
1
S^- ''^?*
\K> '£*
'&* iS* &
J
!
= _
E
E
!
E_
E
l_l
__n
___[l___ri __.
a Dry
• 1 0% Decon 90
n Demineralized Water
Q Methanol
Redrawn from Warren (2007), by permission
Figure 3 - Removal Factors Using "Exhaustive Swipe Sampling" Method
Jabroc" is a non-impregnated, dense wood laminate. DECON 90 is a surface active cleaning agent, and/or
radioactive decontaminant, for laboratory, medical and specialized industrial applications.
15
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Table 2 - Removal Factors Using "Exhaustive Swipe Sampling" Method
Surface
Linoleum
Aluminum
Stainless Steel
Ceramic Tile
Perspex
Jabroc
Concrete
Brick
Mica
Dry
0.05
0.04
0.2
0.31
0.31
0.36
0.54
0.39
0.36
10%
DECON
90
0.48
0.62
0.62
0.88
0.82
0.16
0.4
0.14
0.27
Demineralized
Water
0.72
0.16
0.60
0.83
0.73
Methanol
0.69
0.31
0.43
0.51
0.1
Because the removal factor can vary so substantially, the sampling procedure following an RDD
or IND should include multiple swipes from at least one location for each combination of surface
type and swipe material. This should be repeated periodically as part of the overall quality
assurance (QA) program.
3.2 SWIPE SURFACES AND MATERIALS
Swipe surfaces fall into four categories: rough non-porous, rough porous, smooth porous, and
smooth non-porous. Swipes may be used on a variety of porous and non-porous surfaces, and
may be used wet or dry. Many different materials are used for swipes. Table 3 lists some of the
materials that fall into each of the four categories with suggested swiping materials designated by
the diamond symbol.
Table 3 - Swipable Surfaces and Swiping Techniques
\ Surface
\ Examples
\
\
\
Swipe \
Technique \
Strippable coating
Wetted swipe
Wetted swipe, with
blotting
Dry swipe
Surfaces
Rough Non-
porous
Unpainted or
unpolished
metals
+
+
Rough Porous
Roof tile
Concrete
Unfinished wood
Stucco
Ceiling tile
Fabric, wallpaper
Limestone, slate
Ventilation filters
+
Smooth Porous
Asphalt
Finished limestone
Finished wood
Plaster
Unglazed ceramic tile
+
+
Smooth Non-porous
Vehicle exteriors
Linoleum
Formica
Painted wood/
wall board
Painted/plated metal
Glass, hard plastics
Polished metal
Glazed ceramic tiles
Finished furniture
Vinyl siding
+
•
+
•
16
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Urban surface contamination can be influenced by many factors, including the presence of dirt
and dust, migration into pores and cracks, pH, humidity, chemical interactions of the
contamination with the surface, and weathering. The depth of contamination will vary with the
radionuclide, its chemical form, type of surface, and the time since deposition. Penetration of
contamination into the surface does not necessarily imply that it is not removable.
Table 4 shows different types of surfaces and the suggested swipe material to use on each. Many
of the swipe materials in Table 4 may be used wet or dry. If the swipe is wet, the solvent can
vary. Various solvents may or may not be a good match for the chemical form of the
radionuclide, and thus, may not always increase transfer efficiency. However, following an
RDD, the chemical form may be known or ascertained quickly, providing an opportunity for a
good match to the wetting solution.
Typically, loose surface dirt is removed much better with a wet swipe. If the surface is coated
with a fine layer of dust and the contamination falls on top of the dust, removing the dust with a
wet swipe would be very effective.
Table 4 - Suggested Swipe Material to Use on a Surface
Swipe or Removal
Medium
Cotton
Glass Fiber
Paper
Masslin
Strippable Coatings
Strippable Gels
Wood
(finished)
R
D
D
W
D
Concrete
W
N
N
N
D
Wood
(unfinished)
N
N
N
N
N
Tile or
Linoleum
(non-porous)
R
D
R
W
D
Asphalt
W
D
R
W
N
Stainless or
Painted
Metal
R
D
R
R
D
Unpainted
Metal
N
N
N
W
D
Shows promise for all surfaces
W = wet with solvent, D = dry, R = either wet or dry, N = not useful
Strippable coatings have not been commonly used as "swipe" materials. However, they are often
used for decontamination. Their use for swipes will require some experimental studies on the
measurement of activity in this medium, especially for beta and alpha. Strippable gels have been
developed that appear to work well on all surfaces and with high removal factors. Drying times
depend on the thickness of the gel application and environmental conditions, and vary from
about an hour to over 24 hours. These gels may be counted without further preparation for
gamma radiation. Some gels are able to be rehydrated, and thus also amenable to rapid
preparation and liquid scintillation counting.
Archibald and Demmer (1999) conducted tests with Strippable coating using simulated
contamination (SEVICON) of Cs and Zr salts dried (SEVICON I) or dried and then baked at
700 °C for 24 hours (SEVICON II) on 1-inch stainless steel disks. SIMCON I was used to
simulate removable contamination, and SIMCON II was used to simulate fixed contamination.
Three types of stripcoats were tested on the SIMCON coupons. TLC Stripcoat applied and
peeled very easily. ALARA 1146 was difficult to remove if not sprayed on to a thickness of at
least 1 mil (0.0254 mm). PENTEK 604 self-stripping coating was better at decontaminating
SIMCON I, simulating removable contamination, than the other Strippable coatings. For
SIMCON II, simulating fixed contamination, the performance of all three was similar. The
results are shown in Table 5.
17
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Table 5 - Strippable Coating on Stainless Steel Disks
Stripcoat Material
TLC Stripcoat
ALARM 146
PENTEK 604
SIMCON I Cs
(% Removal)
87
83
96
SIMCON I Zr
(% Removal)
66
76
90
SIMCON II Cs
(% Removal)
42
45
57
SIMCON II Zr
(% Removal)
73
76
75
For decontamination of lead bricks, the strippable coating was applied three times by Archibald
and Demmer (1999), allowing one day for drying between applications. The first two coats were
applied with a small brush. The third and final coat was applied by being poured onto the surface
of the brick. This third coating was not uniform and took 12 hours longer to dry. The strippable
coating stuck to areas with deep pores or cracks, but was removed with a small hand scraper. The
initial contamination on the lead brick was reduced by a factor of 10, as shown in the last two
columns of Table 6. Lead may subsequently oxidize, converting some fixed activity to
removable. However, lead is not expected to be a commonly encountered surface following an
RDD or IND. Thus, this data should be used only as an indication of performance on a surface
other than stainless steel. Generally, any surface can degrade overtime.
Table 6- PENTEK 604 Strippable Coating on Contaminated Lead Bricks
Applications
First
Second
Third
Before Fixed Beta-
gamma (cpm)
*
*
15,000
After Fixed Beta-
gamma (cpm)
*
15,000
6000
Before Swipable
Beta-gamma/
Alpha (cpm)
4715/26
376/0
150/0
After Swipable
Beta -gamma/
Alpha (cpm)
376/0
150/0
**
* = unknown, ** = well below free release criteria (<200 dpm Beta-gamma, <10 dpm alpha)
Five stainless steel criticality barriers13 were tested by Archibald and Demmer (1999) using
PENTEK 604. The coating was applied to both sides of the barrier using a paintbrush and then
left to dry overnight. PENTEK 604 was able to remove the swipable contamination from the lids.
Some fixed contamination remained. The results are shown in Table 7. Again, this is an
illustration only of potential removal factors that may be achieved using strippable coatings as
swipe material. Actual surfaces likely to be encountered may be quite different.
Table 7 - PENTEK 604 Strippable Coating on Stainless Steel Criticality Barriers
Item
Criticality Barrier Lid
#1
Criticality Barrier Lid
#2
Criticality Barrier Lid
Swipable Before
Beta/Gamma
(dpm/100 cm2)
Front 43944 Back
132681
Front 13594
Back 20931
Front 51 856
Swipable Before
Alpha
(dpm/100 cm2)
Front 348
Back 755
Front 237
Back 200
Front 452
Swipable After
Beta/Gamma
(dpm/100 cm2)
Front <1 000
Back 1331
Front <1 000
Back < 1000
Front <1 000
Swipable After
Alpha
(dpm/100 cm2)
Front <20
Back 22
Front <20
Back <20
Front <20
Criticality barriers are stainless steel plates used in fuel pools to separate spent fuel during storage.
18
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Item
#3
Criticality Barrier Lid
#4
Criticality Barrier Lid
#5
Swipable Before
Beta/Gamma
(dpm/100 cm2)
Back 12362
Front 2251 3
Back 10013
Front 16294
Back 67075
Swipable Before
Alpha
(dpm/100 cm2)
Back 104
Front 192
Back 104
Front 126
Back 489
Swipable After
Beta/Gamma
(dpm/100 cm2)
Back < 1000
Front <1 000
Back < 1000
Front <1 000
Back < 1000
Swipable After
Alpha
(dpm/100 cm2)
Back <20
Front <20
Back <20
Front <20
Back <20
As seen in Tables 5-7, PENTEK 604 appears to be quite effective at removing swipable
contamination, although procedures for using strippable coatings as swipe materials have not
been fully developed.
Strippable gels such as Decon Gel 1101, which are relatively new materials, have a high removal
factor (up to 90-100%), even on porous surfaces such as concrete (Holt 2007, Sutton et al. 2008,
and VanHorne-Sealy 2008). The narrower range in removal factors should reduce the uncertainty
associated with this factor. However, repeated sampling should be performed to verify this as
both a quality control (QC) and validation measure. The removal factor for the strippable gel
does not seem to depend on the chemical form of the contaminant. This could minimize the
uncertainty due to the combination of material and solvent chosen for traditional swipes.
Strippable gels can be applied over an area larger than the desired sample area. The exact area
needed can be cut from the stripped gel, provided it can be removed as a single large piece. This
eliminates uncertainties in the area due to freehand swiping. Pressure applied would not appear
to enter as a factor.
The gel can be counted directly if it has been previously calibrated for the radionuclide of
concern. Variations in thickness may contribute to the uncertainty due to self-absorption. These
gels may be rehydrated and the encapsulated radionuclides separated for analysis, for example,
by liquid scintillation counting.
Further research is needed to determine the effectiveness of using strippable gels as swipe
material, especially for recovery of the encapsulated contaminants. This is a relatively new
technology, but preliminary results are encouraging. Ideally, this new material has the potential
for making swipe measurements a more realistically quantitative determination of removable
surface activity with detection limits that may be lower than current practice can achieve.
3.3 SAMPLING METHOD
A standard method or technique for taking swipe samples does not exist. As indicated in the ISO
guidance and in other references for swipes taken freehand with moderate pressure, no quantita-
tive measure of pressure is identified. The uncertainty of the area swiped can be minimized by
using a template with a defined area and sampling within a template. Variability among sampling
personnel might be reduced by inserting the swipe material in a standard mount so that the entire
swipe is in contact with the surface. Swiping pressure might be regulated, either by using
weights, or incorporating a pressure sensor in the block. The effect of swiping pressure on
linoleum was examined by Warren (2007). Figure 4 shows that the effect of pressure seems to
19
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
impact wet swipes more than dry.
•
200 :
? '•
E1i
£
= 100 ;
S
*^
5 en _
tfi
o -
; 10o
/ Dry
i
i
/""''
/
J
f~
X
f~
* ^ ~ ^^ /
.,.,''" .
s~ ^
b Decon90
0 20 40 60 80 100 120 140 160 Redrawn from Warren
Contact Pressme on Linoleum
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
weathering.
Table 8- Common Chemical Forms of Some Radionuclides that May Be in an ROD
Alpha-Emitters
Radio-
nuclide
241Am
242Cm [31
243Cm [3,
244Cm
237Np
210Po
238pu
239Pu
240Pu
252Cf
227A(; [2,
Chemical
Form
Am203,
Am metal
141 Am ,0, /Be
Cm, oxide,
silicate or
aluminate
Np03
Po04,
Po Metal
Pu oxide
(mixed
oxidation
states)
[41PuOxide/Be
Cf203 in a Pd
metal matrix,
Cf202S04in
an Al powder
matrix
141 Ac/Be
Radio-
nuclide
226Ra
223^ [2,
230Th [3,
232Tn H
234u
235u
238u
U-Nat
Chemical
Form
RaS04
141 R a/Be
Th02
U308
Yellowcake,
U308,
U02(N03)2,
U metal
Beta /Gamma-Emitters
Radio-
nuclide
141Ce
144Cg [3]
57Co
60Co [1]
134Cs[3]
137Cs
154Eu [3,
3H
125,
129 1 [3]
131,
192|r [1,
Chemical
Form
CeCI4
Cyano-
cobalamin
(Vitamin B12)
CoO, Co
metal
Cs*
CsCI
Eu203
H-0-3H,
Organo-
tritium
compounds
Agl (within a
titanium
capsule), This
salt would be
carrier free.
iodide, iodate
or iodine
(based on
level of
oxygen)
Nal, Kl,
Sodium
lodohippurate
Ir metal
Radio-
nuclide
"Mo
32p
lOSpj [1]
241 pu H
223Rg [2,
103Ru [3,
106Ru [3,
75Se
89Sr
90Sr
99Tc
99mTc
Chemical
Form
Mo042~
Organo-
phosphorus
compounds,
P043~
Pd metal
Pu oxide
(mixed
oxidation
states),
[41PuOxide/Be
RaC03,
RaS04, RaC,9
Ru02
Ru02
Organo-
selenium
compounds
SrCI2
SrTi03, SrF2
TcO;,
Organo-
technetium
compounds
1' This group of radionuclides is used frequently in the metallic form in short, needle-like shapes.
121 These radionuclides are naturally occurring and have little or no commercial use.
131 These radionuclides result from fission reactions (either activation of transuranics or fission products) that do not
have any routine commercial use. Their chemical forms after an event would be the result of environmental
weathering.
141 The specific combination of these alpha-emitters with beryllium oxide is referred to as a "neutron source."
5. COUNTING METHODS
A variety of analytical methods are available for measuring the activity on a swipe sample. Some
of the available methods are discussed in this section. Each method has advantages and
disadvantages that depend on the specificity and sensitivity of the technique. The selection of a
particular method will depend on the radionuclides involved and the MQOs, especially the
required method uncertainty. The establishment of MQOs is discussed in detail in Section 6.
21
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
For clarity, the various types of swipe analysis methods can be organized into two groups: non-
destructive techniques, and techniques that may require destruction of the swipe material prior to
analysis.
• Non-destructive measurement techniques, including field screening techniques and
laboratory techniques, measure the activity on the swipe by directly exposing the swipe to the
radiation detector. In some cases, direct measurements allow for rapid delivery of analytical
data, and swipe sample integrity to be maintained so that other analytical methods may be
performed, if necessary. The direct measurement of swipes requires that the instruments used
be calibrated in an appropriate counting geometry that is equivalent to the swipes being
measured.
• Destructive measurement techniques are generally performed in the laboratory and include
those analytical techniques that require the destruction of the swipe sample in order to
perform the analysis. Some techniques require dissolution of the sample material and
chemical separation of the radionuclides of interest to minimize interference or to achieve the
required MQOs.
In some cases, swipes simply may not conform to a standard counting geometry and may need to
be leached or digested in order to facilitate what would otherwise be a direct analysis.
Whether the measurement technique is non-destructive or destructive, consideration also should
be given to whether the technique is a "gross" measurement, or a "radionuclide-specific"
measurement.
• Gross measurements, including field screening measurements, provide an estimate of the
general type of radiation being measured, such as total alpha activity or gross gamma
activity. When used as an estimate of activity from a specific radionuclide, these techniques
may be subject to bias from interfering constituents. In addition, the instrument calibration
may not account for the potentially vast array of radionuclides that might be encountered,
and their various calibration factors. For that reason, gross measurements may be regarded as
having relatively high levels of uncertainty.
• Radionuclide-specific measurements provide measurements of particular radionuclides and,
in most cases, are not subject to significant interference or elevated uncertainty. The results
are generally considered to be more accurate and reliable, but the preparation and analysis
processes may be more lengthy and labor-intensive in some cases.
A simplified organization of analytical techniques into non-destructive or destructive, and gross
vs. radionuclide-specific is presented in Table 9, with additional discussion following.14
14 A more detailed comparison of the various field and laboratory instrumentation can be found in the companion
document, Uses of Field and Lab oratory Measurements During a Radiological or Nuclear Incident (in preparation).
More information on these and other gross and radionuclide-specific techniques may be found in the Inventory of
Radiological Methodologies for Sites Contaminated with Radioactive Materials (EPA 2006c).
22
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Table 9- Organization of Analytical Techniques into Non-Destructive or Destructive and
Gross or Radionuclide-Specific
Destructive
Non-Destructive
Gross
• G-M Probe
• Gas Proportional Counting
• Liquid Scintillation Counting
G-M Probe
Alpha scintillation
Gas Proportional Counting
Beta scintillation
Beta spectrometry
Radionuclide-specific
• Gamma Spectrometry
• Alpha Spectrometry
• Gas Proportional Counting
• Liquid Scintillation Counting
Gamma Spectrometry
ICP-MS and others
Geiger-Muller (G-M) probes may be useful, particularly for hand-held field measurements, when
only an estimate of the gross sample activity is needed. G-M probes are designed for the
measurement of gamma and beta radiation, but under certain conditions may also respond to
higher levels of alpha radiation.
Gamma spectrometry, including thallium-activated sodium iodide [Nal(Tl)] and high-purity
germanium (HPGe) detection systems, may be used to count the sample directly or after
chemical digestion and preparation. Gamma spectrometry is used to identify specific radionuc-
lides and can quantify the radionuclide activity with high precision under the appropriate
conditions.
Gamma analyses by a G-M, a Nal(Tl), or an HPGe detector can be performed with little
preparation aside from ensuring that the samples are properly contained to avoid detector
contamination. A 47-mm diameter round swipe, for example, is commonly used. These swipes
may be left in their glassine sleeve or wax envelope, which may then be placed inside a
protective sealable plastic bag. The bag can then be placed directly on the detector. A variety of
sizes and shapes of swipe material can be analyzed directly for gamma activity, with appropriate
consideration for the counting geometry.
Large swipes, such as shop towels or absorbent pads, may require a geometry that accommodates
the physical constraints of the detector system and shield assembly. For example, the laboratory
may choose to create a counting geometry that consists of a large swipe that is either folded or
packed into a container that is typically used for solid materials. In any case, the instrument
calibration sources should conform as closely as possible to the actual samples. Swipes that
require gamma analyses may need to be leached or digested and the digestate solution analyzed
in a geometry intended for liquid samples. This approach preserves the digestate for further
testing, if necessary.
Alpha scintillation detectors, which may be found in hand-held field measurement detectors as
well as laboratory screening detectors, may be useful for the estimation of elevated levels of
gross alpha activity. These devices may be useful when it is necessary to discriminate between
gross alpha activity and gross beta/gamma activity at higher activity levels.
23
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Gas proportional counting (GPC) is used to measure both alpha and beta activity. These
measurements may be done simultaneously, with good alpha/beta discrimination. Some GPC
detectors operate with very low background count rates and relatively high counting efficiencies,
allowing for the detection of low levels of alpha and beta activity.
Properly sized swipes may be analyzed directly for alpha and/or beta activity by counting on an
alpha scintillation detector, GPC, or other appropriate survey instrument. This requires the direct
exposure of the instrument to the sample, and care should be taken to minimize the risk of
detector contamination and to monitor frequently for such contamination. In some cases, friable
residue or other loose material that presents an increased risk of detector contamination may be
fixed to the swipe surface prior to analysis with commercial hair spray or some other fixative
agent. The laboratory should determine whether this practice will have an effect on the analysis,
for example by increasing attenuation, and make appropriate corrections during calibration. For
example, using fixatives may not be appropriate for swipe samples of low (<200keV) beta-
emitters. Pure beta-emitters, such as 90Sr, might be analyzed by beta spectrometry or using beta
scintillation counters. The use of combined alpha-beta scintillation detectors with appropriate
cross-talk corrections may also be considered.
Liquid scintillation counting is primarily used for the measurement of lower energy beta
emissions that may not be detected easily with a GPC detector, and the measurement of volatile
radionuclides, such as 3H and 14C, that may not be amenable to analysis by other techniques,
though LSC counting may also be useful for the analysis of routine alpha and beta activity as
well. A consideration in LSC counting is that the sample must be immersed in scintillation
cocktail during analysis, which may interfere with future analyses.
Care must be taken in the direct analysis of swipes by LSC to ensure that analytical interference
from particulates, chemicals, or other "quenching agents" is properly addressed. The loading of
dirt or other interferents in field swipes can be notoriously inconsistent, and the laboratory may
need to create calibration "quench curves" to address these issues.
Alpha spectrometry may be a valuable technique when the isotopic speciation and/or low-level
quantification of specific alpha-emitting radionuclides are necessary. Alpha spectrometry
generally requires rigorous chemical separation techniques and may employ long sample count
times, which should be considered when the rapid delivery of analytical data is requested.
Inductively coupled plasma mass spectrometry (ICP-MS) may be a useful technique for the
7^S 7^R 9^7 7^9 QQ
analysis of very long-lived radionuclides (e.g., U, U, Np, Th, Tc, etc.), after sample
dissolution. In some cases, the instrument can achieve very low detection limits very quickly,
which can be useful when large numbers of samples must be analyzed.
Kinetic phosphorescence analysis (KPA) may be used as a fast and relatively inexpensive
approach for the analysis of uranium, and X-ray fluorescence (XRF) may still be employed
under certain circumstances for the rapid measurement of high concentrations of certain analytes
under proper conditions. The list of analytical methods shown above is not intended to be
exhaustive, but is meant to highlight the most common techniques.
24
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
In some cases, the swipe material may require wet- or dry-ashing or even chemical fusion prior
to dissolution. In all cases, care should be taken to ensure that the additional processing of the
sample does not compromise the analytical methods that are used. For example, the dry-ashing
of a sample would invalidate any subsequent analyses for volatile radionuclides.
While there seem to be few analytical methods specifically developed for swipes, swipe samples
are amenable to most radionuclide-specific analyses, and analytical methods that require the
dissolution of a solid sample.
6. PLANNING MEASUREMENTS OF SURFACE ACTIVITY CONCENTRATION BY
MEANS OF SWIPE SAMPLES
Acceptable uncertainty depends on the anticipated use of the data. Screening measurements may
have different requirements than radionuclide-specific measurements. The action level and
tolerable decision-error rate should be established and considered in selecting measurement
methods. Example scenarios should be developed in advance, as there will be little time for this
in an emergency.
Given the complexity of interpreting swipe measurements in general, these examples should
focus on particular applications for assessing removable radioactivity following a radiological or
nuclear event. Specific sources of uncertainty in the interpretation of swipe results should be
considered. Specific cases can then be considered for developing an uncertainty budget.
6.1 MEASURE ME NT QUALITY OBJ ECTIVES AND THE DATA QUALITY
OBJECTIVES PROCESS
MQOs are statements of performance objectives or requirements for a particular analytical
method performance characteristic. Examples include:
• Method uncertainty,
• Detection capability,
• Quantification capability,
• Ruggedness,
• Specificity, and
• Range.
The most important MQO is the required method uncertainty at a specified concentration (the
analytical action level), WMR. The calculation of UMR'IS discussed in Table IOC, Step 6.3.
MQOs need to be developed separately for each radionuclide and each phase of the incident to
ensure that the data quality will be sufficient for the decisions to be made. Generally, the value of
the required method uncertainty15 (MMR) should decrease from the earlier to later phases,
15 Method uncertainty, UM, refers to the predicted uncertainty of the result that would be measured if the method were
applied to a hypothetical laboratory sample with a specified analyte concentration. Although individual measure-
ment uncertainties will vary from one measured result to another, the required method uncertainty, u-lSBi, is a target
value for the individual measurement uncertainties, and is an estimate of uncertainty (of measurement) before the
sample is actually measured.
25
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
reflecting the fact that the action level will be decreasing. Even in the early phases of an incident,
MQOs may be very different for highly affected areas near ground zero as opposed to less
affected boundary areas because the analytical action levels (AALs) may differ. In addition, the
input quantities that enter the model equation for the measurement result may differ. An
uncertainty budget is a tool that can be used to understand sources of uncertainty and the
influences of changes in the measurement process on the overall measurement uncertainty. Thus,
an uncertainty budget can be used to understand the measurement process and identify the
factors that may most influence the ability of the measurement process to meet the MQOs. The
development and use of an uncertainty budget are discussed further in a later section of this
report.
The DQO process may be applied to all programs involving the collection of environmental data
with objectives that cover decisionmaking activities. When the goal of the study is to support
decisionmaking, the DQO process applies systematic planning and statistical hypothesis testing
methodology to decide between alternatives. Data quality objectives can be developed using the
Guidance in EPA (2006a) Guidance on Systematic Planning Using the Data Quality Objectives
Process (EPA QA/G-4).
Table 10A summarizes the DQO process. From this process, MQOs can be established using the
guidance in MARLAP. The information in this table should be sufficient to enable the
decisionmaker to determine the appropriate MQOs. The output should include an AAL,
discrimination limit (DL), gray region, null hypothesis, analytical decision level (referred to in
MARLAP as "critical level"), and required method uncertainty at the AAL. A table summarizing
the DQO process for each decision point can be prepared in advance and summarized in Table
10 A.
Table 10A - The DQO Process Applied to a Decision Point
STEP
Step 1 . Define the problem
Step 2. Identify the decision
Step 3. Identify information needed
for the decision
Step 4. Define the boundaries of
the study
Step 5. Develop a decision rule
This defines the decision point
OUTPUT
... with a preliminary determination of the type of data needed and how they
will be used; identify decisionmaker.
...among alternative outcomes or actions and a list of decision statements that
address the problem.
Analytical action levels that will resolve the decision and potential sources for
these; information on the number of variables that will need to be collected; the
type of information needed to meet performance or acceptance criteria;
information on the performance of appropriate sampling and analysis methods.
Definition of the target population with detailed descriptions of geographic
limits (spatial boundaries), detailed descriptions of what constitutes a sampling
unit time frame appropriate for collecting data and making the decision or
estimate, together with any practical constraints that may interfere with data
collection, and the appropriate scale for decisionmaking or estimation.
Identification of the population parameters most relevant for making inferences
and conclusions on the target population; for decision problems, the "if
then... else..." theoretical decision rule based upon a chosen AAL.
26
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
The theoretical decision rule specified in Step 5 can be transformed into statistical hypothesis
tests that are applied to the data. Due to the inherent uncertainty with measurement data, there is
some likelihood that the outcome of statistical hypothesis tests will lead to an erroneous
conclusion, i.e., a decision error. This is illustrated in Table 10B.
Table 10B - Possible Decision Errors
Decision Made
Decide that the parameter of interest is
greater than the analytical action level
Decide that the parameter of interest is less
than the analytical action level
True Value of the pan
Greater than the ML
Correct decision
Decision Error
ameter of interest
Less than the ML
Decision Error
Correct decision
Data that are inconsistent with the null hypothesis will cause it to be rejected. The probability of
this happening in error (a Type I error) is more easily controlled during the statistical design. In
order to choose an appropriate null hypothesis (or baseline condition), consider which decision
error would have the greater consequences. Then choose the null hypothesis so that the Type I
error (false rejection) corresponds to the decision error with the greater consequence, i.e., would
cause the greatest harm.
Table IOC - The DQO Process Applied to a Decision Point
STEP
OUTPUT
Step 6. Specify limits on
decision errors
Step 6.1 Determine analytical
action level (ML) on the gray
region boundary and set
baseline condition (null
hypothesis, Ho)
Which is considered worse: decision error (a) deciding that the parameter of interest
is less than the ML when it actually is greater, or (b) deciding that the parameter of
interest is greater than the ML when it actually is less? Case (a) is usually
considered to be a conservative choice by regulatory authorities, but this may not be
appropriate in every case.
If (a), the ML defines the upper boundary of the gray region (UBGR). The null
hypothesis is that the sample activity is above the ML. (All samples will be
assumed to be above the ML unless the data are convincingly lower.) A desired
limit will be set on the probability (a) of incorrectly deciding the sample is below the
ML when the sample activity is actually equal to the ML (Figure 5).
If (b), the ML defines the lower boundary of the gray region (LBGR). The null
hypothesis is that the sample activity is below the ML. (All samples will be assumed
to be below the ML unless the data are convincingly higher.) A desired limit will be
set on the probability (a) of incorrectly deciding the sample is above the ML when
the sample activity is actually equal to the ML (Figure 6).
6.2 Define the discrimination
limit (DL)
If (a), the discrimination limit defines the lower boundary of the gray region.1 It will be
activity below the ML where the desired limit will be set on the probability (/J) of
incorrectly deciding the sample is above the ML. (see Figure 5)
If (b), the discrimination limit defines the upper boundary of the gray region.2 It will
be activity above the ML where the desired limit will be set on the probability (/J) of
incorrectly deciding the sample is below the ML.
27
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
STEP
6.3 Define the required
method uncertainty at the AAL
Step 7. Optimize the design
for obtaining data
OUTPUT
According to MARLAP Appendix C, under either case
recommended required method uncertainty is:
UBGR-LBGR A
(a) or case (b) above, the
MR -
7 -\- 7 7 -\- 7
^l-a T zl-/? zl-a T zl-/5
where zi-a and zi-p are the 1-a and 1-f$ quantiles of the standard normal
distribution function. 3
Iterate Steps 1-6 to define optimal values for each
measurement method required.
of the
parameters and
the
NOTES:
1 The DL is the point where it is important to be able to distinguish expected signal from the AAL. When one
expects background activity, then it might be zero. If one expects activity near the AAL, however, it might be at
90% of the AAL.
2 The DL is the point where it is important to be able to distinguish expected signal from the AAL. If the AAL is
near zero, the DL would define activity deemed to be too high to be undetected. Thus, the DL may be set equal to
the MDA. If one expects activity near the AAL, however, it might be at 110% of the AAL.
3 Values of Zj_a (or z^) for some commonly used values of a (or /?), taken from tables of the cumulative normal
distribution (EPA 2009), are:
a or/?
0.001
0.01
0.025
0.05
Zi-a&IZi-p)
3.090
2.326
1.960
1.645
a or/3
0.10
0.20
0.30
0.50
Zi-a&IZi-p)
1.282
0.842
0.524
0.000
Failing to detect a sample that exceeds the AAL could have consequences to public health, and
analyzing additional samples will slow the overall process and therefore, may also impact public
health. The probability that such decision errors occur is defined as the parameters a and ft in
Steps 6.1 and 6.2 in Table IOC. Values of alpha and beta should be set based on the
consequences of making an incorrect decision. How these are balanced will depend on the AAL,
sample loads, and other factors as specified by Table 9C.
The most commonly used values of alpha and beta are 5%, although this is by tradition and has
no sound technical basis. These values may be used as a default, but should be optimized in Step
7 of the DQO process according to the actual risk of the decision error being considered.
In this document, the analytical decision level (ADL) is the concentration or activity correspon-
ding to the critical value (see MARLAP Attachment 3B.2). The critical value is the minimum
measured value of the instrument signal required to give confidence that a positive (nonzero)
amount of analyte is present in the material analyzed. Thus, a measurement less than the critical
value would result in a decision that the analyte is not present. The critical value is sometimes
called the critical level.
28
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Oi
o
g
^
« a
* I
jj w
15J3
Q o
I1
£s
"«
h
• Altemaflve
*"!•*-
1
0.9
0.8
0.6
0.5
0.4
0.3
0.2
Gray Region
Relatively Large-
Decision Error Rates
are Considered
Tolerable
Tolerable False
Acceptance Decision
ErrorWes
1 I I I I L_
..^.-......— — •
Tolerable False
Rejection Decision
Error Rates
0 20 40 60 80 100 120 140 160 180 200
Action Level
True Value of the Parameter (Mean Concentration, ppm)
Figure 5 - Example Illustrating Case (a) (see Table 9C, Step 6.1).
Baseline Condition (null hypothesis): Parameter Exceeds the AAL.
Alternative-
I
0) ra
•8°
e -e
If
cs
o
£
Tolerable False
Rejection Decision
Erro/Wes
Tolerable False
Acceptance Decision
Err or Rates
Gray Region
Relatively Large
Decision Error Rates
are Considered
Tolerable
i i |
20 40 60 80 100 120 140 160 180 200
Action Level
True Value of the Parameter (Mean Concentration, ppm)
Figure 6 - Example Illustrating Case (b) (see Table 9C, Step 6.1). Baseline
Condition (r Figures 5 and 6 ^gn from EPA QA/G-4 (2006) J tne AAL•
29
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Figure 5 shows that in case (a), the ADL will be
UBGR - zi -a UM (6)
where UM is its combined standard uncertainty of the measurement result, x, and UBGR is the
upper boundary of the gray region. Only measurement results less than the ADL value will result
in rejecting the null hypothesis that the true activity is greater than the AAL.
Figure 6 shows that for case (b), the ADL will be
LBGR + zi -a MM, (7)
where UM is its combined standard uncertainty of the measurement result, x, and LBGR is the
lower boundary of the gray region. Only measurement results greater than the ADL will result in
rejecting the null hypothesis that the true activity is less than the AAL.
Decisions related to specific samples will be made by comparing the results of measurements to
ADLs. Whenever the measured analyte activity equals or exceeds the applicable ADL activity, it
will be concluded that the AAL (from a protective action guide or from an agreed upon cleanup
level) has been exceeded.
The equations for the ADL given above are consistent with the acceptable decision error rates
established during the DQO/MQO process. In this process, the MQO of greatest significance is
the required method uncertainty,
r\
For example, an ADL can be calculated for a hypothetical AAL of 180 dpm/100 cm and DL of
90 dpm/100 cm2 for 90Sr based on previously determined tolerable Type I and Type II error rates
of 5% and an assumed required method uncertainty, Z/MR, of 27 dpm/100 cm2:
ADL = UBGR - zia Z/M = AAL - zi o.os WM = 180- 1.645x27 = 136 dpm/100 cm2 (8)
Measurement Type AAL (dpm/100 cm2) ADL (dpm/100 cm2) UMR (dpm/100 cm2)
Radionuclide-specific 180 136 27
Only measurement results less than the ADL value will result in rejecting the null hypothesis that
the true activity is greater than the AAL.16 For example, a result of 100 dpm/100 cm2 would
result in the decision that the sample does not exceed the AAL, while a result of 160 dpm/100
cm2 would result in the decision that the sample does exceed the AAL. The fact that the measure-
ment result itself does not exceed the AAL is not relevant. It is the ADL that is used as the
decision criterion specifically to avoid making the wrong decision by calling a sample below the
AAL when it may actually be above it. The result may be below the AAL only by chance —
random variations in the observed concentration.
16 Note that these values are being used only for demonstration purposes, and have no regulatory basis. Specific
AALs will be provided by the Incident Commander (or designee) and likely will be different.
30
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Many types of measurements will be made during an incident, and the specified AAL may
change as the response moves from early to recovery phases. Because different DQOs and
MQOs are applicable to different types of measurements, different Z/MR and corresponding ADL
values will be required for each measurement. It is anticipated that in the case of an incident,
specific DQOs and MQOs may be developed by Agency personnel to reflect the specific nature
and concerns of the incident.
Developing the required method uncertainty, WMR, will be crucial in selecting measurement
methods for both direct and indirect measurements. Certain uncertainty factors may dominate for
direct measurements that are different from those for indirect measurements. For example, the
uncertainty in the source efficiency is likely to be larger for direct measurements than for indirect
measurements. The removal factor for swipes, in contrast, has an uncertainty that affects only
swipe samples. Establishing the appropriate uncertainties for the source efficiency for direct
measurements requires laboratory analysis of physical samples of the various surfaces. An
evaluation of the uncertainty in removal factors may be obtained by exhaustive swipe sampling.
Some sources of uncertainty that should be considered in developing the required method
uncertainty, MMR, and in reporting the uncertainty of individual activity measurements include:
• Instrument efficiency,
• Self-absorption in the surface being measured,
• Distance between surface and detector,
• Count/count rate,
• Time of measurement,
• Area of measurement,
• Effective area of detector,
• Ambient background radiation (especially for gamma),
• Variability of background with surface type, and
• Inherent detector background.
The ISO Guide to the Expression of Uncertainty in Measurement (GUM, ISO 1995) should be
used to determine measurement uncertainties. MARLAP Chapter 19 provides additional
examples. It is not sufficient to consider counting uncertainty alone.
It may be useful to use a tool such as Table 1 1 to assist in developing DQOs and MQOs. The
three distinct phases of emergency response are the early, intermediate, and recovery phases. The
1 17 on ois
radionuclides considered in this example are Cs, Sr, and Pu.
31
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Table ]] - Example Table for Establishing DQOs and MQOs
0)
OT
re
Q.
c
CD
•o
'o
c
Early
Early
Early
Intermediate
Intermediate
Intermediate
Recovery
Recovery
Recovery
Nuclide
137Cs
90Sr
238pu
137Cs
90Sr
238pu
137Cs
90Sr
238pu
>-
O
C£l
*
i/l
i/l
rt
c
M-
0
Ol
Q.
J
P
a
J
P
a
T
P
a
<; g
o
° v in
X ,_ ro
' — ' —
c ro ro
§00
see
TO — —
C _l _l
133
OVA
.<2 _i _i
o o o
. — .
o
1
* ^
•, "ro
CD o-
OQ
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
The desired required method uncertainty is:
AAL-DL
20
-1-
^
0.842 + 0.842
11.9 dpm/100 cm2,
or about 12 dpm/100 cm at and below the AAL. The required relative method uncertainty is
about 60% above the AAL. Swipe samples taken over a 100 cm area whose activity exceeds the
AAL when counted using field instrumentation are to be sent immediately to a radioanalytical
laboratory for confirmatory analysis using alpha spectrometry. The decision is based on whether
the measurement exceeds the ADL of AAL - z^a UM = 20 - (0.842)(11.9) = 10 dpm/100 cm2.
Confirmation that 241Am exceeds 20 dpm/100 cm2 will require that the surface be re-cleaned and
re-tested.
Below is a row from Table 11 that has been filled in for this example. The actual values are for
illustrative purposes only.
ident Phase
c
Recovery
AAL o
, case (a) or c
3 .C QJ
Z 0 .1:
a
^_^
CM
O
0
o
1
Q.
•^
<
"ro
o
^
TO
<
20
2" -o
Sis
> Ss
AAL in cas
.?i _i _i
o o o
0
o
o: 5
35
ofc 3
g sr
"1
<3 -2-
20
^
01
m
Li
£
OJ
[^
.20
^
i
E
Ol
8.
i^"
.20
s
•£
c
5
s
3
•o
o
"5
-o
£
'3
o-
£
12
•o
o
"o
o rt
TO
1 >•
T3 .E
2 ^
& 8
£ i
.60
§
Q 'e
<, 0
11
^ E
st
•ol
S^
s s
^^
ro O
5 S
10
ij
"o
1
&
Incident
Commander
The 100 cm2 area swipe samples were taken from interior building surfaces using a 47-mm
diameter cellulose-acetate filter paper. A representative sample of swipes from each batch was
weighed initially, and each swipe was re-weighed in the field after sampling but prior to
counting. The difference in the weight of the swipe before and after sampling provides an
indication of the particle loading that can be used to estimate the amount of self-absorption of
emitted radiation in the swipe material.
6.3 UNCERTAINTY BUDGET FOR EXAMPLE SCENARIO: FIELD TOTAL ALPHA
The field measurements were made using a ZnS(Ag) scintillator in a 50-mm drawer with a sealer
for recording alpha counts. Recall from Section 2.3 on indirect measurements of removable
surface activity using swipe samples:
For instruments with a 2rc counting geometry (i.e., gas-flow proportional counters, scintillation
counters, portable field survey meters, etc.), the activity per unit area ^R of the removable con-
tamination of the swiped surface, expressed in dpm/lOOcm , is given by the equation:
33
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
=60-(n/t-nB/tB) = 60-(n/t-nB/tB)
R
.
F-(S/lOO)-(£t-e
where:
AR = removable activity per unit area of the swiped surface [dpm/lOOcm2]
n = measured total count from the swipe sample
riB = background count
t = total count time of the swipe sample (s)
IB = background count time (s)
S = swiped surface area (cm )
F = removal factor
£i = instrument efficiency for radiation type (alpha or beta radiation)
ES = source efficiency
Based on empirical data for this particular detector, the product of these efficiencies can be
expressed as
Overall efficiency = £i£s =et-e WOpR (10)
where:
R = mean linear range for alpha particles or beta particles in the swiped matter, (e.g., 30
p = density of the swiped matter (g/cm )
wa = weight of filter with swiped matter (mg)
w\, = weight of clean filter (mg)
100 = represents the 100 cm surface area for deposition.
The specific values given for measured input quantities are purely for illustrative purposes and
are not to be construed as typical of current or possible future practice.
The uncertainty analysis for this sample is shown in Table 12. Note that there is an entry for each
input variable in the right hand side of the equation for^R. The name of the input quantity and its
symbol are given in the first two columns. The value is the best estimate of the input for this
sample. The standard uncertainty is determined using the GUM methodology. The distribution
assumed for the input quantity is also given. The component of uncertainty due to a given input
is its standard uncertainty multiplied by its sensitivity coefficient. The sensitivity coefficient is
obtained by evaluating the partial derivative of the equation for AR with respect to the input
variable, evaluated using the data for the particular sample. This is done to weight the
uncertainty contribution of each input according to the effect changes in that input have on the
output. The combined standard uncertainty of the output, AR, is the square root of the sum of the
squares of the components of uncertainty. The percent of the combined variance is the ratio of
the square of each input component of uncertainty to the sum of the squares of all the uncertainty
components. This is an indicator of how much each input contributes to the overall uncertainty.
From this uncertainty budget, it is clear that most of the uncertainty is due to the removal factor
(56%) and the efficiency factor (21%). The relative combined standard uncertainty in the
34
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
estimated activity per unit area is about 52%. Clearly, this would be improved by experimentally
determining the removal factor by the method of exhaustive swiping. If the uncertainty in the
removal factor were halved, its contribution to the overall variance would be reduced from 56%
to about 24%, and the overall relative uncertainty in the result would be reduced from 52% to
42%. Further improvement may be possible by reducing the uncertainty in the value 0.37 of the
efficiency factor. By using available software, it is possible, with little additional effort, to
determine the effect that reducing the uncertainty in particular input parameters would have on
the result.
If one were concerned only with the activity on the swipe, the factors S/100 and F would not be
used. In this case, the activity on the swipe would be 10.7 dpm with a combined standard
uncertainty of 3.6 dpm or 34%. The major contributors to the uncertainty budget are the
efficiency factor (55%) followed by the sample counts (34%) and the background counts (11%).
Nonetheless, the value found in Table 12, 53 dpm/100 cm2, clearly exceeds the threshold for
sending the sample to the laboratory for alpha spectrometric analysis, but the required relative
method uncertainty was met.
Table 12- Uncertainty Budget for Gross Alpha Count'
OS
* ^
«-3
!a
S- 3
£ 0
Sample counts
Time sample counted
Background counts
Time background counted
Area swiped
Efficiency factor
Removal factor
Weight of filter before swiping
Weight of filter after swiping
Range of alpha particles in swiped
matter
Density of swiped matter
"3
.c
>,
!/3
n
t
nB
IB
S
e
F
WB
WA
R
P
OJ
_s
"3
>
56
600s
18
600s
100 cm2
0.37
0.2
800 mg
825 mg
2.1
g/cm3
Standard
Uncertainty
7.5
0.2887
4.2
0.2887
12.247
0.0925
0.08165
10
1
0
0
Distribution
Poisson
Rectangular
Poisson
Rectangular
Triangular
Normal
Triangular
Normal
Normal
Constant
Constant
Component of
uncertainty
10.5
0.0379
5.97
0.0122
6.54
13.4
21.8
0.848
0.0848
Percent of
combined
variance
13.1
0.000
4.2
0.000
5.1
21.1
56.4
0.085
0.001
Results
Removable activity
Standard uncertainty in the
removable activity
AR
V(AR)
53 dpm/lOOcm2
29 dpm/lOOcm2
HI Calculations were performed using the software GUMCalc available at www.mccroan.com/sumcalc.htm.
35
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
6.4 UNCERTAINTY BUDGET FOR EXAMPLE SCENARIO: LABORATORY ALPHA
SPECTROMETRY
The swipe sample was analyzed in the laboratory for 241Am by alpha-particle spectrometry. The
9A^
swipe was spiked with the radioactive tracer Am and muffled at 600 °C, and the ash residue
was acid-digested. The acid digestate was gravimetrically split, with 50% directed to the 241Am
analysis and 50% reserved for re-analysis, if necessary. Americium was separated from the
analysis fraction by solid-phase ion-chromatography extraction, and the final sample test source
was prepared by microprecipitation onto a membrane filter. The sample test source was counted
for 600 minutes using an ion-implanted silicon detector, enabling spectrometric measurement of
the individual americium isotopes. Chemical yield of the separation process was determined by
the measured recovery of 243Am in the sample and that yield determination was applied to the
9/11
Am activity calculation.
The full mathematical model for this measurement might be given by
where:
9/11 9
AR = removable Am activity per unit area of the swiped surface [dpm/lOOcm ]
A/aS = sample count in the241 Am region of interest (ROI)
A/ab = blank count in the 241 Am ROI
9/T^
A/ts = sample count in the Am ROI
A/tb = blank count in the 243Am ROI
ts = sample count time [s]
tb = blank count time [s]
94^
ct = Am activity concentration of the tracer solution [pCi/L]
V\ = volume of tracer solution added to the sample aliquant [L]
A = correction factor for decay of 243 Am from the tracer reference date through counting
Pt = alpha emission probability for the 243Am ROI
V = volume of the sample aliquant analyzed [L]
Pd = total volume of solution after the filter is dissolved [L]
A = correction factor for decay of 241Am from sample collection through counting
Pa = alpha emission probability for the 241 Am ROI
S = swiped surface area [cm ]
F = removal factor
For simplicity in this example, the decay factors will be neglected since they are very close to 1.
The alpha emission probabilities are assumed to be exactly 1 (with no spillover outside each
region of interest (ROI). The resulting model equation is:
A = NJts-NBb/tb^ 2.22xctxFt
R Nts/ts-Nib/tb (V/Vd)xFx(S/\00)
Because the count times ts and tb have negligible uncertainty, only the uncertainty components
due to A/as, A/ab, A/ts, A/tb, ct, Ft, Fd V, S, and F will be considered.
36
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
The uncertainty analysis for this sample is shown in Table 13. The analysis assumes:
None of the input estimates are correlated with each other;
Dead time is negligible;
Peaks in the alpha spectrum are cleanly separated, and there is no spillover from either ROI.
Subsampling uncertainty is negligible for this sample;
• Historical QC data indicate no significant amount of 241Am contamination in method blank
samples; and
The decay-correction factors are negligible.
Table 1 3 - Uncertainty B udget for Alpha S pectrometry Count'11
3 ^
§t
1 s
Q- 3
.E 0
Sample count in the
241 Am region of interest
(ROI)
Blank count in the 241Am
ROI
Sample count in the
243Am ROI
Blank count in the 243Am
ROI
Sample count time (s)
Blank count time (s)
243Am activity
concentration of the
tracer solution
Volume of tracer solution
added to the sample
aliquant [L]
Volume of the sample
aliquant analyzed
Total volume of solution
after the filter is dissolved
Area swiped
Removal factor
0
-Q
>>
CO
Mas
Mab
Ms
Mb
ts
tb
Ct
Vt
V
Vd
S
F
0)
•3
£
810
1
1282
2
60 000 s
60 000 s
3360 pCi/L
1 ml, or
0.001 L
0.1500L
0.3000L
100
0.2
Standard
Uncertainty
28.478
1.414
35.819
1.732
50 pCi/L
0.000006 L
0.0013 L
0.0030L
12.247
0.08165
c
o
+3
_Q
i
<«
b
Poisson,
«(*J = J*I
Poisson (low level) I2',
«w,b)=>Kb+i
Poisson,
u(NJ = ^s
Poisson (low level),
u(NJ = jNtb+l
Negligible uncertainty
Negligible uncertainty
L/=100pCi/L
(k = 2)
u(Vi) = 0.006 ml, or 6 x
10-6L
u(V) = 0.0013 L
u(Vd) = 0.0030 L
Triangular
Triangular
Component of
Uncertainty
1.66
0.0824
1.32
0.0638
0.702
0.283
0.409
0.471
5.77
19.2
Percent of
Combined
Variance
0.673
0.002
.425
0.001
0.12
0.02
0.041
0.054
8.147
90.5
Results
Removable activity AR <
Standard uncertainty in the ,. ,
, . ,. ., ' U(AR)
removable activity v '
47dpm/100cm2
?0dpm/100cm2
[1] Calculations were performed using the software GUMCalc available at www. mccroan. com/gumcalc
[2] As recommended in MARLAP, when counts are low, N is replaced by N+1.
htm.
37
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
The output estimate (the activity concentration of 241Am) is calculated below.
A = NJts-Nab/tb^ 2.22xctxVt
NJts-Nib/tb (V/Vd)x(S/lOO)xF
810/ (60000s)-II(60000s) (2.22)(3360 pCi/L)x(0.001 L)
12827(60000 s)-2/(60000s) (0.150 L/0.300 L)x(100/100)x(0.2)
= 46.72dpm/100cm2
From Table 13, it can now be seen that the analytical uncertainty is dwarfed by the uncertainty in
the removal factor, which accounts for 90% of the variance. Including the uncertainty of the
swiped area (8%), these factors account for virtually all of the uncertainty in inferring surface
removable activity from swipe samples. If the uncertainty in the removal factor were halved, the
relative combined standard uncertainty would fall from 43% to 23%. The dominant contributors
to the field measurement uncertainty were the removal factor, the instrument efficiency, and the
number of sample counts. Clearly, there would be much to be gained from a better understanding
of removal factors for the initial swipe. It may be of benefit to examine two or three successive
swipe samples at each of several locations, with at least one exhaustive swipe series for each
different surface type.
Again, if one were concerned only with the activity on the swipe, the factors 5/100 and F would
not be used. In this case, the 241Am activity on the swipe would be 9.43 dpm with a combined
standard uncertainty of 0.48 dpm or 5%. The major contributors to the uncertainty budget are the
sample counts (50%) followed by the tracer counts (32%) and the background counts (11%), and
followed by the tracer activity (9%). The volume determinations each contribute a few percent.
Thus, the laboratory analysis of the activity on the swipe has a much smaller relative uncertainty
than the gross alpha count in the field. In addition, it is a radionuclide-specific measurement so
that one does not need to be concerned about any alpha-emitting background radionuclides that
may be present.
7. CONCLUSIONS
From the data and examples considered in this document, it would appear that determining the
activity on the swipe is not especially difficult, particularly if the swipe is dissolved and
radionuclide-specific measurements are made in a laboratory. The major issues concern inferring
from the activity on the swipe what the amount of removable contamination is on the surface.
The major sources of uncertainty in making this inference are primarily in the removal factor (F),
and in the area swiped (S). Direct counting of the swipe has the added uncertainty of estimating
the source efficiency. Direct counting of the surface itself will increase this contribution to the
uncertainty, and it is not a measure of removable activity alone, but of fixed-plus-removable
contamination.
Current practice for determining the removal factors often involves simply assuming that it is 1
(the swipe defines removability) or using 0.1 (a value that may be considered conservative).
Often, the value used is not reported. The actual value in a specific instance most likely lies
between these extremes. In the absence of any data, the use of a conservative value with a large
38
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
uncertainty is probably necessary in practice. As suggested in this document, the data from some
multiple swipes or occasional exhaustive swipes can reduce this uncertainty.
Much of the uncertainty in determining the removable activity from the activity on a swipe can
be reduced by simply improving the specified protocol for obtaining the swipe sample, and
making it more uniform. Current practices involving freehand swiping using hand pressure
introduce uncertainties that can be reduced by using templates for the area sampled and some
type of mounting jig for the swipe material. Specifying improved protocols to be used during the
recovery phase of an RDD or IND event is one possibility.
The uncertainty in both S and F could be reduced dramatically by the use of strippable gels as a
swipe material. In early studies, these have shown high removal efficiencies (over 90%) and the
desired area can be cut from the sample with much less uncertainty than freehand swiping.
Research would be needed to determine how well these materials perform as a surrogate swipe.
The application method drying time, ease of stripping an intact sample, the source efficiency for
direct counting, and the method of dissolution for radiochemical analysis all need further
investigation. However, this preliminary examination of early results indicates that this would be
well worthwhile.
When contamination consists of alpha or low-energy beta activity, grab sampling of the surface
followed by radiochemical analysis may be the only alternative to swipe sampling. The area
sampled would need to be carefully controlled. More importantly, how the sample is taken will
strongly affect what the result represents. If the surface is lightly abraded, it may be a
representative of removable activity. A deeper sample may more closely represent fixed-plus-
removable contamination.
If swipes are to be used as a quantitative measure of whether remediation meets required limits,
rather than simply as a method-based, prescriptive sample indicating the presence of removable
activity, it is clear that improvements in sampling, analysis, reporting, and interpretation are
necessary.
8. RECOMMENDATIONS
First and foremost, it is necessary that the uncertainty of swipe results should be reported along
with the results. In reporting the results, it should be clearly specified what the results represent
(e.g., activity on initial swipe, estimate of total removable contamination, etc.).
Each phase of the event will need a basic swipe investigation to determine the best combination
of swipe material and solvent to use for the radionuclides of interest, their chemical forms, and
the surfaces involved. The data needed will depend on the action levels and data quality
objectives for the swipe samples. Fortunately, the amount of information available from
laboratory analyses of the RDD material will increase for each successive event phase, as action
levels also decrease. Such additional information should help reduce the uncertainties in the
measurements due to the assumptions that may be necessary without this information. This will
help the laboratory meet more stringent MQOs (e.g., required method uncertainty, minimum
39
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
detectable activity, etc.) as the event unfolds. For example, optimizing the swipe material and
solvent should increase the removal factor and reduce its uncertainty.
A more specific protocol for obtaining swipes would reduce the overall uncertainty in the
removable activity. Existing swipe procedures (see Appendix C for examples) could be
improved by specific improvements in defining the area sampled. It is becoming more common
to use a template to define a reproducible area. Care would be needed to prevent cross-
contamination, and may require disposal after each use. A pad of paper templates may be a
practical solution. Strippable coatings can be applied, and then a coupon of the size desired can
be cut from the dried coating.
Table 2 (page 16) suggests that dry swipes may have higher removal factors on rough surfaces
and wet swipes may have higher removal factors on smoother surfaces. This should receive
further investigation.
Mounting the swipe material on a block would prevent the large inhomogeneities in activity
across the material caused by finger pressure. A weight or spring might be used to control the
pressure applied to the surface and reduce the uncertainty due to this variable. Strippable
coatings may reduce these effects, but reproducibility would need to be verified.
Removal factors should be determined periodically by means of exhaustive swiping. At a
minimum, each type of surface should be evaluated. This would include multiple applications of
Strippable coatings if they are used.
If there are large differences between direct and indirect measurements, the implication may be
that there is a large amount of fixed activity, or the source efficiency for the direct measurements
is incorrectly specified, or both. Periodic physical sampling and laboratory analysis of the
surface material for certain matrices should be performed to assess the accuracy of direct
measurements. This is analogous to repeated swipes being used to evaluate removal factors.
For analyses of swipe samples, the weight of the swipe material before and after use could be
used to estimate source efficiency for the swipe.
9. REFERENCES
References used in this document are listed in Section 9.1. Other sources, including discussions
with subject matter experts and additional Internet searches, are listed in Section 9.2.
9.1 SOURCES CITED
ASTM El893-08, Standard Guide for Selection and Use of Portable Radiological Survey
Instruments for Performing In Situ Radiological Assessments to Support Unrestricted
Release from Further Regulatory Controls. West Conshohocken, PA. Available for purchase
from: www.astm.org/Standards/E1893.htm.
40
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Archibald, Kip E., and Rick L. Demmer. August 1999. Tests Conducted with Strippable
Coatings. Idaho National Engineering and Environmental Laboratory. INEEL/EXT-99-
00791.
Batilson, Benjamin, et al. 1987. "Smear sampling apparatus." U.S. Patent # 4,848,165.
Campbell, Jerry L., Charles R. Santerre, Peter C. Farina, and Lowell A. Muse. 1993. Wipe
Testing For Surface Contamination By Tritiated Compounds. Health Physics 64(5): 540-544.
U.S. Environmental Protection Agency (EPA). 2006a. Guidance on Systematic Planning Using
the Data Quality Objectives Process (QA G-4), Washington, DC, EPA 240/B-06/001,
February. Available at: www.epa.gov/qualitv/qs-docs/g4-fmal.pdf.
U.S. Environmental Protection Agency (EPA). 2006b. Sample Collection Procedures for
Radiochemical Analytes in Environmental Matrices; Module II: Sampling Procedures - Site
Characterization and Remediation Phases. Office of Research and Development, National
Homeland Security Research Center. Cincinnati, OH. December. EPA/600/S-07/001.
Available at: www.epa.gov/NHSRC/pubs/600s07001.pdf.
U.S. Environmental Protection Agency (EPA). 2006c. Inventory of Radiological Methodologies
for Sites Contaminated With Radioactive Materials. Office of Air and Radiation, Washing-
ton, DC. EPA 402-R-06-007, June. Available at: www.epa.gov/narel/analysis.html.
U.S. Environmental Protection Agency (EPA). 2007. A Literature Review of Wipe Sampling
Methods for Chemical Warfare Agents and Toxic Industrial Chemicals. Office of Research
and Development. Washington, DC 20460. January. EPA/600/R-07/004. Available at:
www.epa.gov/NHSRC/pubs/600r07004.pdf
U.S. Environmental Protection Agency (EPA). 2009. Radiological Laboratory Sample Analysis
Guide for Incidents of National Significance-Radionuclides in Air. Revision 0. Office of Air
and Radiation, Washington, DC. EPA 402-R-09-007, June. Available at: www.epa.gov/narel/
incident_guides.html.
U.S. Environmental Protection Agency (EPA) (in preparation). Uses of Field and Laboratory
Measurements During a Radiological or Nuclear Incident. Office of Air and Radiation,
Washington, DC.
Fischer, Robert, and Brian Viani. 2007. Report on the 2007 Workshop on Decontamination,
Cleanup, and Associated Issues for Sites Contaminated with Chemical, Biological, or
Radiological Materials. U.S. Environmental Protection Agency (EPA), Washington, DC.
EPA/600/R-08/059. June. Available at: www.epa.gov/NHSRC/pubs/
600r08059.pdf.
Frame, Paul W., and Eric W. Abelquist. 1999. Use of Smears for Assessing Removable
Contamination. Health Physics 76 (Supplement 2): S57-S66.
Holt, Kathleen C., 2007. Results from decontamination testing using Decon Gel 1101 on
concrete, carbon steel, stainless steel, and Plexiglas coupons contaminated with Am-241, Pu-
239 and Cs-137. http://www.orau.gov/DDSC/ret/CBI-Testing%20Report-10-8-07.pdf
41
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Hogue, Mark G. 2002. Field Comparison of the Sampling Efficacy of Two Smear Media: Cotton
Fiber and Kraft Paper. Health Physics 83(Supplement 1): S45-S47.
International Organization for Standardization (ISO) Guide 98. 1995. Guide to the Expression of
Uncertainty in Measurement. Geneva, Switzerland. Available at: www.iso.org/iso/
catalogue detail.htm?csnumber=45315.
International Standard ISO 7503-1:1999 (E). Evaluation of surface contamination -Part 1: Beta-
emitters (maximum beta energy greater than 0.15 MeV) and alpha-emitters.
International Standard ISO 7503-2: 1988 (E). Evaluation of surface contamination - Part 2:
Tritium surface contamination.
International Standard ISO 7503-3: 1996. Evaluation of surface contamination - Part 3: Isomeric
transition and electron capture emitters, low energy beta-emitters (Epmax < 0.15 MeV).
International Standard ISO 11929-7: 2005. Determination of the detection limit and decision
threshold for ionizing radiation measurements, Part 7: Fundamentals and general applica-
tions.
Jung, Haijo, Jay F. Kunze, and James D. Nurrenbern. 2001. Consistency and Efficiency of
Standard Swipe Procedures Taken on Slightly Radioactive Contaminated Metal Surfaces.
Health Physics 80(Supplement 2): S80-S88.
Klein, Robert C., Ilona Linins and Edward L. Gershey. 1992. Detecting Removable Surface
Contamination. Health Physics 62(2): 186-189.
Klein, R. C., E. Party, and E. L. Gershey. 1997. Removable Surface Contamination at a
Biomedical Research Institution. Health Physics 72(2): 296-299.
MARLAP. 2004. Multi-Agency Radiological Laboratory Analytical Protocols Manual. Volumes
1 - 3. Washington, DC: EPA 402-B-04-001A-C, NUREG 1576, NTIS PB2004-105421. July.
Available at www.epa.gov/radiation/marlap/links.html.
MARSAME. 2009. Multi-Agency Radiation Survey and Assessment of Materials and Equipment
[Manual] (MARSAME), EPA 402-R-06-002. Available at www.epa.gov/radiation/marssim/
marsame.html.
National Research Council on Committee on Standards and Policies for Decontaminating Public
Facilities Affected by Exposure to Harmful Biological Agents: How Clean is Safe? 2005.
Reopening Public Facilities After a Biological Attack: A Decisionmaking Framework.
Washington, DC.
U.S. Nuclear Regulatory Commission (NRC). 1974. Termination of Operating Licenses for
Nuclear Reactors. Regulatory Guide 1.86. Washington, DC. Available at:
http://adamswebsearch2.nrc.gov/idmws/DocContent.dll?library=PU_ADAMSApbntad01&Lo
gonID=9afaddbdb07aafbfb947flb48aOdOca8&id=003957281.
U.S. Nuclear Regulatory Commission (NRC) 1995. Minimum Detectable Concentrations with
42
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Typical Radiation Survey Instruments for Various Contaminants and Field Conditions.
NUREG-1507. Office of Nuclear Regulatory Research, Washington, DC. Available at:
http://pbadupws.nrc.gov/docs/ML0036/ML003676046.pdf
U.S. Nuclear Regulatory Commission (NRC). 1999. Consolidated Guidance About Materials
Licenses: Program-Specific Guidance About Licenses of Broad Scope. Washington, DC.
NUREG-1556, Volume 11, April. Available at: www.nrc.gov/reading-rm/doc-collections/
nuregs/staff/sr 155 6/v 11 /.
U.S. Nuclear Regulatory Commission (NRC). 2006. Consolidated Decommissioning Guidance.
NUREG-1757. Office of Nuclear Material Safety and Safeguards, Washington, DC.
Available at: www.nrc.gov/reading-rm/doc-collections/nuregs/staff/srl757/.
Royster, G.W., Jr., and B.R. Fish. 1967. "Techniques for Assessing Removable Surface Con-
tamination," B.R. Fish (editor), Surface Contamination. Pergamon Press, New York, N.Y.
201-207.
Sansone, E.B. 1987. "Redispersion of Indoor Surface Contamination and Its Implications," in
Vol. 1, Treatise on Clean Surface Technology. Plenum Press, New York, N.Y.
Seo, Bum-Kyoung, Bong-Jae Lee, Kune-Woo Lee, and Jin-Ho Park. 2004. Development of an
Automatic Smear Sampler and Evaluation of Surface Contamination. Korea Atomic Energy
Research Institute. Daejeon, Korea. Radiation Protection Dosimetry, Vol. 112 No. 2, 287-
290.
Sutton, Mark, Robert P. Fischer, Mark Thoet, Mike O'Neill, and Garry Edgington. 2008.
Plutonium Decontamination Using CBI Decon Gel 1101 in Highly Contaminated and Unique
Areas at LLNL Lawrence Livermore National Laboratory Report LLNL-TR-404723
VanHorne-Sealy, Jama D., 2008. Evaluating the Efficiency of Decon Gel 1101 for Removal of
Cs-137, Co-60, and Eu-154 on Common Commercial Construction Materials, Master's
Thesis Oregon State University. Available at: http://ir.library.oregonstate.edu/dspace/
bitstream/1957/8933/l/MS%20Thesis%20-%20Jama%20VanHorne-Sealv.pdf
Warren, Laurence. March 2007. An investigation into the pick-up factors of wipes on various
laboratory-type surfaces. URN: 6010955. Department of Physics, University of Surrey,
United Kingdom.
Yu, C., D.J. LePoire, J.-J. Cheng, E. Gnanapragasam, S. Kamboj, J. Arnish, B.M. Biwer, A.J.
Zielen, W.A. Williams, A. Wallo III, and H.T. Peterson, Jr. 2003. User's Manual for
RESRAD-BUILD Version 3. ANL/EAD/03-1. Environmental Assessment Division, Argonne
National Laboratory, Illinois.
9.2 OTHER SOURCES
ANSI/HPS Nl3.49-2001. American National Standard: Performance and Documentation of
Radiological Surveys. August 6, 2001. American National Standards Institute, Inc.
43
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Brunskill, R.T. 1967. "The Relationship between Surface and Airborne Contamination," B.R.
Fish (editor), Surface Contamination. Pergamon Press, New York, N.Y. 93-105.
Caplan, KJ. February 1993. The Significance of Wipe Samples. American Industrial Hygiene
Association, Vol. 54 No. 2, 70-75.
Cooper, E.L., J.M. Cox and WJ. Workman. 1998. Analysis ofSr-90 and Alpha-Particle Emitters
on Air Filters and Swipe Samples Using a Liquid Scintillation Counter with Alpha/Beta
Discrimination. Radioactivity & Radiochemistry 9(3), 25-40.
Dun, Sarah, and Joseph Wood. 2005. Decontamination, Cleanup, and Associated Issues for
Sites Contaminated With Chemical, Biological, or Radiological Materials. U.S.
Environmental Protection Agency (EPA), Washington, DC. EPA/600/R-05/083. Available at:
www.epa.gov/nhsrc/pubs/600r05083.pdf.
Dun, Sarah and Joseph Wood. 2006. 2006 Workshop on Decontamination, Cleanup, and
Associated Issues for Sites Contaminated with Chemical, Biological, or Radiological
Materials. April 26-28, 2006. U.S. Environmental Protection Agency (EPA), Washington,
DC. EPA/600/R-06/121. Available at: www.epa.gov/nhsrc/pubs/600r06121 .pdf
Dun, Sarah and Joseph Wood. 2008. 2007 Workshop on Decontamination, Cleanup, and
Associated Issues for Sites Contaminated with Chemical, Biological, or Radiological
Materials. U.S. Environmental Protection Agency (EPA), Washington, DC. EPA/600/R-
08/059. Available at: www.epa.gov/NHSRC/pubs/600r08059.pdf
U.S. Environmental Protection Agency (EPA). 2010. Standardized Analytical Methods for
Environmental Restoration Following Homeland Security Events — SAM 2010 (Revision
6.0), Cincinnati, OH, EPA 600/R-10/122, October. Available at: www.epa.gov/sam/
index.htm.
Jones, IS. and S.F. Pond. 1967. "Some Experiments to Determine the Resuspension Factor of
Plutonium from Various Surfaces," B.R. Fish (editor), Surface Contamination., Pergamon
Press, New York, N.Y. 83-92.
Lichtenwalner, C.P. 1992. Evaluation of Wipe Sampling Procedures and Elemental Surface
Contamination. American Industrial Hygiene Association Journal, Vol. 53 No. 10, 657-659.
MARS SIM. 2000. Multi-Agency Radiation Survey and Site Investigation Manual, Revision 1.
NUREG-1575 Rev 1, EPA 402-R-97-016 Revl, DOE/EH-0624 Revl. August. Available at:
www.epa.gov/radiation/marssim/index.html.
Schuler, Christoph, Gernot Butterweck, Christian Wernli, Fra^ois Bochud, and Jean-Fran9ois
Valley. March 2007. Calibration and Verification of Surface Contamination Meters -
Procedures and Techniques. Paul Scherrer Institut Division for Radiation Safety and
Security, Switzerland. PSI Report 07-01. ISSN 1019-0643.
44
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
APPENDIX A:
REVIEW OF REGULATORY APPROACH AND APPLICATIONS FOR SWIPES IN
RADIATION PROTECTION ACTIVITIES
A.I INTRODUCTION AND DEFINITION OF CONCEPTS
If a radiological dispersal device (RDD or IND) were detonated in the United States, a large
number of samples would need to be analyzed to assess and control the spread of fine paniculate
matter contaminated with radionuclides. A review of Department of Energy (DOE), Nuclear
Regulatory Commission (NRC), and Department of Transportation (DOT) regulations and
regulatory guidance documents was conducted to identify various uses of swipes that might be
applicable to incident response activities, especially in the intermediate and recovery phases.
Other agencies, including the U.S. Environmental Protection Agency (EPA), Department of
Homeland Security (DHS), Centers for Disease Control and Prevention (CDC), Department of
Defense (DOD), and Food and Drug Administration (FDA), clearly have interests within the area
of radiation protection. To minimize redundancy, it is noteworthy and reassuring that different
agencies appear to approach this topic in a similar manner. This need not be surprising since
although DOE and NRC are now autonomous agencies, they used to be a single entity, and the
licensing and use of radioactive materials for all other entities in the United States are regulated
by the NRC. Thus, the following discussion focuses primarily on the approach presented in the
NRC documents reviewed.
Minimizing the dose of ionizing radiation to the public is a key goal of radiation protection
activities. In order to effectively protect members of the public, one must be able to measure the
sources of radiation and radioactive materials that lead to radiation exposures. Radiation
protection regulations differentiate between the doses caused by exposure to external and internal
sources of ionizing radiation. External exposures result from high levels of gamma- or X-ray
radiation (i.e., penetrating radiations) located outside the body. The absorbed dose that a person
receives from exposure to external or ambient sources of radiation is dependent upon the energy,
rate of exposure, and length of time that the individual spends in proximity to the radiation.
Exposure rates high enough to be of concern to health and safety generally will cause a response
in an appropriate radiation detector placed in very close proximity to the radioactive source and
are thus relatively easy to measure in surveys for ambient radiation. Internal exposures result
when radionuclides (especially alpha- and beta-emitters) are taken into the body (via inhalation,
ingestion, injection, or absorption). Once within the body, relatively small amounts of
radioactivity will cause much more damage than comparable sources of radioactivity outside the
body (especially alpha and beta particles). Additionally, since some radionuclides inside the
body are only slowly excreted from the body, they will lead to larger exposures since the
exposure can continue possibly for the rest of an individual's life.
Health physicists and regulatory bodies tasked with radiation protection recognize that there is a
substantial difference in risk between internal and external radiation doses and have developed
approaches to measure and control sources of radiation and radioactive materials that could lead
to radiation dose. The approaches taken by the NRC and DOE, while not strictly identical, share
substantial common direction in almost all respects. This paper discusses examples of how
45
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
regulatory entities approach the measurement and controls related to internal dose.
It is also important to distinguish between the concepts of radiation and radioactive
contamination. Radiation refers to gamma-rays, X-rays, and alpha and beta particles that are
emitted following the radioactive decay of radionuclide. On the one hand, gamma- and X-rays
are electromagnetic radiations that are emitted as radioactive atoms decay (e.g., 137Ce). Gamma-
and X-ray radiation is very penetrating and thus is a hazard to health and safety, even when its
source is outside or "external" to the body. However, this radiation has no substance or mass and
does not persist once the source of radiation is removed from the area. In contrast, radioactive
contamination refers to the physical presence of a substance (in this case, radioactive materials or
radionuclides) in a place where it constitutes a hazard to human health and safety. An example of
contamination would be the presence of radioisotopes of strontium (e.g., 90Sr) or cesium (137Cs)
on the floor in an entryway to a building. The contaminants are the atoms of radioactive
strontium or cesium.
In the discussion of swipes, the terms "removable contamination" and "fixed contamination"
also need to be defined. Removable contamination refers to the portion of radioactive
contamination that is transferred from a contaminated surface to the swipe. Swipe samples of
removable contamination (also referred to as "swipes," "wipes," or "smears") are used to
determine the presence of radioactive contamination, and to estimate the amount of
contamination that could be removable into the body and thus lead to an internal dose. Fixed
contamination refers to the portion of contamination that remains attached to a surface after
reasonable attempts to clean or decontaminate that surface. Since the contamination is fixed to
the surface, it will not be transferred to the body and is of concern only as a source of external
exposure. As mentioned in the above discussion, it is the contamination that is removable that
poses the greatest hazard and thus, is of greatest regulatory concern. The levels of removable
contamination that are of concern are much lower than those for fixed contamination because the
intensity and effective duration of the exposure to radiation caused by internally-deposited
radionuclides are much greater than sources of radiation outside the body.
A.2 PERFORMANCE OF SURVEYS - PROCEDURES, TYPICAL SENSITIVITY, AND
APPROPRIATE INSTRUMENTATION
NRC (1999) provides consolidated guidance on radioactive materials licenses and may be one of
the best sources of information on operations involving swipes. Section 8.10.7 of NUREG-1556
(NRC 1999) defines a survey as "an evaluation of the radiological conditions and potential
hazards incident to the production, use, transfer, release, disposal, or presence of radioactive
material or other sources of radiation." Among the types of surveys that this document addresses
are ambient "radiation surveys" and "contamination surveys." Contamination surveys measure
and evaluate the (unwanted) presence of radioactive materials to determine whether the levels of
radioactive contamination constitute a hazard to the health and safety of workers or the general
public. Fixed contamination is addressed separately from removable contamination. Since by
definition, swipes cannot measure fixed contamination, the subsequent discussion will restrict
itself to removable contamination surveys.
46
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
NRC takes a performance-based approach to a number of topics. Thus, as NUREG-1556 (NRC
1999) points out, 10 CFR Part 20 does not specify action limits for removable contamination, nor
does it prescribe a specific method for performing swipe surveys. Instead, the licensee must
propose an approach for approval by the NRC. Appendix S of NUREG1556 (NRC 1999),
however, does present examples of action limits, frequencies, and survey procedures that are
"acceptable to the NRC." Similarly, the footnotes to Table S.5 of Appendix S provide an
example of a method for performing swipes that would be "acceptable to the NRC":
A standardized method for smear [swipe] testing of a relatively uniform area should be used to
aid in comparing contamination at different times and places. A smear taken from an area of
about 100 cm2 is acceptable to indicate levels of removable contamination.
The amount of removable radioactive material per 100 cm2 of surface area should be determined
by wiping that area with filter or soft absorbent paper, applying moderate pressure, and assessing
the amount of radioactive material on the [s]wipe with an appropriate instrument of known
efficiency. When removable contamination on objects of less surface area is determined, the
pertinent levels should be reduced proportionally and the entire surface should be [s]wiped.
Table A-l - Table 1 from NRC RG 1.86 - Acceptable Surface Contamination Levels
NUCLIDE [1]
U-nat, U-235, U-238, and
associated decay products
Transuranics, Ra-226m Ra-
228, Th-230, Th-228, Pa-
23 l,Ac-227, 1-125. 1-129
Th-nat, Th-232, Sr-90, Ra-
223, Ra-224, U-232, 1-126,
1-131,1-133
Beta-gamma-emitters
(nuclides with decay modes
other than alpha emission
or spontaneous fission)
except Sr-90 and others
noted above.
AVERAGE12'31
5,000 dpm a/100 cm2
100 dpm/100 cm2
1,000 dpm/100 cm2
5,000 dpm p-y/100 cm2
MAXIMUM12'41
15,000 dpm a/100 cm2
300 dpm/100 cm2
3,000 dpm/100 cm2
15,000 dpm p-y/100 cm2
REMOVABLE12 51
1,000 dpm a/100 cm2
20 dpm/100 cm2
200 dpm/100 cm2
1,000 dpm p-y/100 cm2
[1] Where surface contamination by both alpha- and beta-gamma-emitting nuclides exists, the limits established
for alpha- and beta-gamma-emitting nuclides should apply independently.
[2] As used in this table, dpm means the rate of emission by radioactive material as determined by correcting the
counts per minute observed by an appropriate detector for background, efficiency, and geometric factors
associated with the instrumentation.
[3] Measurements of average contaminant should not be averaged over more than 1 square meter. For objects of
less surface area, the average should be derived for each object.
[4] The maximum contamination level applies to an area of not more than 100 cm2.
[5] The amount of removable radioactive material per 100 cm2 of surface area should be determined by wiping
that area with a dry filter or soft absorbent paper, applying moderate pressure, and assessing the amount of
radioactive material on the [s]wipe with an appropriate instrument of known efficiency. When removable
contamination on objects of less surface area is determined, the pertinent levels should be reduced
proportionally and the entire surface should be [s] wiped.
47
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Table S.5 in Appendix S of NUREG-1556 (NRC 1999) lists "acceptable surface contamination
levels" for removable and fixed contamination surveys for unrestricted areas.17 Table S.5 is
derived from Table 1 in NRC Regulatory Guide 1.86 (NRC 1974).18 It lists action limits for
removable contamination down to 20 dpm/100 cm2 for a number of alpha-, beta-, and photon-
emitting radionuclides, including several that are of incident response concern (e.g., radium-226,
all isotopes of plutonium, americium, curium, neptunium, iodine-125, iodine-131, and strontium-
90). It is not envisioned that these are the exact levels that would be applied during an incident
response. Rather, these values are provided as a starting point to consider detection levels that
might be applicable to measurements of removable contamination during an incident response
during the recovery phase. As pointed out in Footnote 17 on this page, the as-low-as-reasonably-
achievable (ALARA) principle would possibly drive detection sensitivity needs to levels lower
than those shown in Table 1.
Appendix S indicates that "appropriate instrumentation" should be used to perform measure-
ments of swipes and specifically lists low-background liquid scintillation counting, sodium
iodide or germanium gamma counting, or gas-proportional alpha/beta counting. Appendix S also
specifies that "to ensure achieving the required sensitivity of measurements, survey samples will
be analyzed in a low-background area" while Table S.5 footnotes continue, saying that the
instrument must be calibrated for "background, efficiency [implying a radionuclide-specific
calibration] and geometric factors associated with the instrumentation."
While the action limits shown in Table A-l above will vary case-by-case, they are helpful in
considering the sensitivity needed for instruments appropriate for use when measuring swipes.
Given these action levels, and considering ALARA, any instrument used to measure removable
contamination would have to dependably obtain minimum detectable activities below 20
dpm/100 cm2 levels for the nuclide(s) in question. Such measurements are generally not
problematic at fixed laboratories, where environmental conditions such as ambient background
and environmental conditions are well-known and carefully managed, and strong quality
assurance and quality controls are in place to demonstrate such.
If one wishes to count swipes outside of a fixed laboratory, however, it is critically important that
one look beyond theoretical capabilities of an instrument. It is vital to empirically assure that all
environmental conditions, but especially ambient background, are stable and well-known, and
that there are effective controls in place to identify and document any excursions from these
conditions. Otherwise, dependable measurements at very low activity levels will be of ques-
tionable quality and may lead to false detection and false non-detection of analyte in these
samples.
17 Any reference to "acceptable levels" for unrestricted release should be understood in the context of "ALARA"
(as-low-as-reasonably-achievable). Under ALARA, radioactively contaminated areas are decontaminated, where
possible, until there is no contamination present. When decontamination to zero levels is technically impossible or
economically unfeasible, "acceptable" levels of residual contamination are established. In all likelihood, these levels
will be lower than Table 1 values (shown in Table A-l above).
18 Table S-5, NUREG-1556, Volume 11. According to NUREG-1757, volume 1, revision 2, the values and the
swipe procedure presented in this table were originally derived from Table I in NRC RG 1.86 (shown in Table A-l
above).
48
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
A.3 APPLICATIONS TO WHICH REMOVABLE CONTAMINATION SURVEYS APPLY
NUREG-1556 (1999) addresses applications for which swipes are applicable and required of
radioactive materials license holders. The common theme for all of these is control of removable
contamination that could cause a risk to the health and safety of workers or members of the
general public. Surveys for removable contamination are conducted using swipes for:
• Routine contamination surveys, monitoring, and control in and adjacent to areas where
radioactive materials are used or stored.
• Verification and control of the integrity of radioactive sources to detect and thus minimize
contamination that would result from source leakage.
• Control of contamination during shipping and receiving of radioactive materials packages.
• Contamination surveys of equipment, facility, and infrastructure, and areas adjacent to the
above that could be subject to contamination during release of equipment from licensed
facilities or facility decommissioning of entire facilities prior to release for unrestricted (or
possibly restricted) use.19
NUREG-1556 (NRC 1999) also mentions the use of surveys complementary to those for
removable contamination. Surveys for ambient radiation are routinely required to control
external exposures. These surveys, primarily for gamma- and X-ray radiation, may also serve as
a first indicator of the presence of high levels of radioactive contamination in an area that should
trigger special radiation protection measures and controls to minimize personal exposures and
potential contamination of equipment.
Similarly, total contamination survey measurements are in-situ measurements of alpha, beta, and
photon emissions that reflect the total or combined radioactive contaminant activity (fixed and
removable). Depending on the contaminant, the stability of detector efficiency, ambient
background activity levels, and the quality and sensitivity of in-situ measurements may not be
adequate to measure the presence of removable contamination at low levels. Rather, a
measurement of total activity may provide an early indication of the possible presence of high
levels of removable contamination and allow use of measures to prevent personal exposures or
the contamination of sensitive low-background instrumentation that will be used to analyze the
swipes.
In 10 CFR Parts 30, 40, 70, and 72, the NRC addresses the termination of operating licenses
(i.e., decommissioning) for a number of different types of facilities. This situation perhaps most
closely resembles survey activities that might be encountered during the recovery phase of an
incident response. Paragraph 36, Section (j)(2)(i) of each of the above-mentioned parts stipulates
that the licensee must perform surveys and report the results documenting levels of fixed and
removable contamination:
19 None of the guidance documents consulted consider the question of transfer efficiency coefficients (the amount of
non-fixed activity actually transferred to a swipe relative to the total non-fixed contamination present) and the
advantage of modifying approaches to taking swipes to improve or at least account for transfer efficiency.
49
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
[LJevels of gamma radiation in units of millisieverts (microroentgen) per hour at one meter from surfaces,
and report levels of radioactivity, including alpha and beta, in units of megabecquerels (disintegrations per
minute or microcuries) per 100 square centimeters—removable and fixed—for surfaces...
Several additional NRC-related applications requiring the use of swipes in radiation protection
are noted in CFRs. These are less apropos in terms of their applicability to incident response and
environmental protection. For example:
• 10 CFR Parts 20, 31, 32, 34, 35, 39, and 70 require the use of swipes to test various types of
sources and equipment containing sealed sources for leakage that could lead to
contamination.
• In 10 CFR Parts 20 and 71, the NRC requires the use of swipes to determine removable
contamination when shipping and receiving Type A quantities of radioactive materials. Note
that the specifications are similar to those specified by DOT in 49 CFR 173.433. Specifically,
the shipper must perform dose rate and swipe surveys to determine whether removable
contamination may be present above acceptable levels of <0.5 mrem/h and <0.005 uCi/100
cm2, respectively.20
A.4 CONCLUSIONS
• Swipes will be applicable to incident response activities given the likelihood for spread of
fine particulate matter contaminated with radioactive materials following the detonation of a
radiological dispersal device.
• In the documents consulted, the approach to measuring removable contamination (i.e.,
radioactive contamination) was relatively consistent from agency to agency. The approach
used by the NRC in regulating licensees addressed all swipe-related applications encountered
from other agencies and was deemed as the best example to use for considering issues related
to use and measurement of swipes in the assessment of contamination.
• The NRC makes extensive use of swipe surveys, as well as other measurements, to identify,
estimate, and control the amount of removable radioactive contamination in a number of
different applications, including:
O Routine operational contamination survey and monitoring in and adjacent to areas where
radioactive materials are used,
o Leak-testing sources,
o Shipping and receiving of radioactive materials,
o Determining unrestricted release of (potentially) surface-contaminated objects, and
o Determining unrestricted release of facilities as part of decommissioning and radioactive
material license termination.
• NRC regulations consistently differentiate between, and thus require separate surveys for
"ambient radiation," "fixed contamination," and "removable contamination." Unique
controls and action limits apply to each of these.
• NRC does not specify limits for "acceptable" levels of removable (or fixed) contamination in
10 CFR regulations, or in its regulatory guidance documents.
20 DOE requirements in 10 CFR Parts 830 and 835 read very similarly to the corresponding NRC regulations.
Similarly, Department of Transportation requirements in 49 CFR Part 173 specify use of swipe and gamma dose rate
surveys when shipping radioactive materials.
50
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
91
• DOE, however, has established limits in Appendix D to 10 CFR 835. The corresponding
table, given in Appendix C of this document, provides a reasonable initial target when
considering the sensitivity that would be required for measurements of swipes. Detection
requirements below these established limit values may be influenced by the ability of the
detection system to distinguish "real" values above background but below these established
limits. ALARA considerations might then drive the detection requirements to values lower
than those listed in these tables.
• The non-mandatory acceptable activity levels of removable contamination for alpha- and
beta-emitters for unrestricted release shown in NUREG-1556 (NRC 1999), Table S-5, and
Regulatory Guide 1.86 (NRC 1974) Table 1 are low enough (especially for alpha and very
low energy beta-gamma activity) such that environmental conditions (e.g., variability of
ambient background in the field) will in many cases prevent reliably detecting these levels
using field instrumentation.
• The procedure for performing dry swipes as listed in NUREG-1556 (NRC 1999) Table S-5,
consistent with personal experience, is the most commonly applied approach to performing
removable contamination surveys.
• The NRC guidance consulted did not address several issues that may be of concern in using
swipes, including:
O The second most common technique in common use in the industry, based on personal
experience, of using a swipe wetted with a solvent to enhance transfer efficiency;
o Use of other procedures for performing swipes such as large area swipes with masslin
cloth, beyond noting that NRC requires licensees to propose their own procedures; and
o Applicability, use, and determination of material transfer efficiency.
21 These limits appear to be based on NRC's Regulatory Guide 1.86 (NRC 1974) Table 1 (reproduced as Table A-l
in this document).
51
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
APPENDIX B:
MEASUREMENTS OF TRANSFERRABLE SURFACE CONTAMINATION
REPORTED IN THE LITERATURE
The table is derived from data contained in Sansone (1987), adapted with the kind permission of
Springer Business Media. Data on swipes for microorganisms have been omitted, as they are
probably not relevant for radionuclides. Note that the author uses the term "wipe" rather than
"swipe." The original terminology has been retained.
Contaminant
204T1C1
Various
radioisotopes
U
Pu
a-Emitters
ThO2
PuO2
Pu(N03)4
a -Emitters
Ra
Surface
Resin tile
Waxed resin tile
Painted resin tile
PVC
Vinyl sheet
Glass
Waxed resin tile
Vinyl sheet
Smooth concrete
or embossed metal
plates
Plywood, Perspex,
PVC, stainless
steel, aluminum,
linoleum, waxed
protective paper
Granolithic
concrete
Cotton
Stainless steel
PVC
Waxed linoleum
Unwaxed
linoleum
Paper
PVC
Waxed linoleum
Granolithic
concrete
Not specified
Surface contamination
measurement
Wiped with 2. 5 -cm dia
quantitative filter paper
#5 using a mean
pressure of 1 kg
Wipe with Whatman # 1
paper over 100 cm
Wipe 100-1000 cm2
using 10-cm square
Whatman D.H.C. filter
paper
Rate meter
Wiped with dry filter
paper per over 100 cm
Wipe [1J
Adhesive paper PI
Smair [3]
Wipe, no details
Dry filter paper wipes
(6) over 900 cm2
Wipe
Removal efficiency
(%)
4.0 ±1.3
6.6 ±1.5
9.9 ±0.5
53.0 ±9.9
45.4 ±4.9
42.1 ±7.7
1.7-37.3
45.8-66.5
2-3
11-20
8-53 mean = 37
2-17
96, 86, 49
96, 100, 68
58,75,10
14
58
20
0.1,0.2
21,29,31
6
1-3
50-85
Remarks
1 mL of an aqueous
solution (pH — 5.4) was
used; data are the mean
and standard deviation for
five samples
137Cs, 90Sr-90Y, 32P, 60Co,
and U applied in HNO3
solutions
Wipe pressure about 30
g/cm
Surfaces sampled before
and after water wash
Data for 1.5-, 5-, and 10-
um settled particles
respectively; the 10-um
particles were
agglomerates; constant
area sampled
PuO2 applied in aqueous
suspension and dried
Pu(NO3)4 applied in HNO3
solution and dried
Water wash removed 25%;
subsequent detergent wash
43%
Calculated from reported
data
52
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Contaminant
[ 3H] Sodium
acetate
[ 3H] Paraffin
ThO2
Be
U
Various a- and
p-emitters
3H
[3H]
Thymidine
Various
radioisotopes
Pb
Pb
210Po and
241Am
Surface
Shellstone
Fiberglass
Shellstone
Fiberglass
Various
Wood
Preaflex
Stainless steel
Brass
PVC
Stainless steel
Glass
Wood
23 -mm dia
unwoven fabric on
adhesive tape
Wood, painted or
varnished
Formica
Not specified
PVC
Aluminum
Glass
Surface contamination
measurement
Wipe (wet or dry) with
Whatman # 1 papers
Wipe [1J
Adhesive paper PI
Smair [3]
Wipe (back and forth) 1
ft2 with 5" x 8"
Whatman #41
Smair [4]
Wipe-constant pressure
(8 g/cm2) over 100 cm2
Wipe over 100 cm2
using 5. 5 -cm dia filter
paper
Wipe, no details
Wiped with Whatman
#3 filter paper soaked
with glycerol
Wiped with 2.5 cm dia
filter paper 15 times
across a 5-cm square
Remove tape from
surface and count with
appropriate detector
Wiped "briskly" with a
paper towel over 1 ft
Wiped with a paper
towel impregnated with
20% denatured alcohol
and 1 :750 benzalkonium
chloride over 1 ft2
Wiped using 25 -mm dia,
30-mg/cm Toppan
paper over 100 cm
using 0.2 kg/cm
pressure
Removal efficiency
(%)
Dry Wet
5-10 7-19
20-30 26-32
18-23 12-17
15-30 5-6
24-75
44-86
1-33
3 of total, 20 of loose
for each of three
wipes over same area
0.06-2.6
20-67
42-70
21
26.1 ±4.4
28.2 ±4.8
86.1 ±4.6
70.4 ±5.4
4.2 ±1.0
No data
80-100
>95
77 ±2
Po Am
48. 1± 42. 3 ±
1.4 1.2
19.3 ± 20.4 ±
0.5 1.1
68.1 ± 69.8 ±
2.0 1.5
Remarks
Range for three replicates
Constant area sampled;
particle size ~1 um
Large portion of total
remained in wood after
washing with detergent
U as UO2(N03) 2 • 6 H2O in
HNO3, UO2 in C2H5OH,
and UO2 powder; standard
deviation usually < 15% of
mean
Successive wipes removed
9.4 ±1.1 and 7.2 ±1.7%;
ethylene glycol-soaked
paper gave similar results,
dry paper removed about
half as much, and
aluminum foil about a
quarter
Ten replicates; three
different papers used for
wipes showed no marked
differences
Correlation coefficient =
0.82 for tape-smear
comparison
Efficiencies calculated for
surfaces sampled before
and after scrubbing with
water and a brush
Second wipe removed the
remaining contamination
1 ml of nitrate dissolved in
0. 1 A^F£NO3 was spread
over 100 cm2 of the surface
and dried; mean and
standard deviation for six
samples
53
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
Contaminant
125J
Inorganic salts
PbO
Surface
Not specified
Zinc
Formica
Surface contamination
measurement
Wipe 100 cm2 using
swab soaked in 70%
isopropyl alcohol
1.27-cm square
Whatman 542 ashless,
hardened paper
moistened with distilled
water placed on surface
and removed after
drying
Wiped with moist
Whatman #42 paper, 10
x 10 cm, over 100 cm
Wiped with moist paper
towel, 10x10 cm, over
100 cm2
Removal efficiency
(%)
Assumed 10% of total
100
86-91
74-84
Remarks
Three successive samples
removed all chloride,
nitrate, sulfate, and
sodium, ammonium,
potassium, magnesium,
and calcium cations from
"normally contaminated"
surfaces; from surfaces
"heavily contaminated"
with smoke, three samples
removed all nitrate and
ammonium and Mg
cations; pH variation from
4.0-5.8 had no effect
Removal generally
increased with increasing
surface concentration from
64 to 730 ug/100 cm2
NOTES:
[1] Whatman #50 paper on #5 rubber stopper; rubbed over 5.8-cm diameter sample location.
[2] 3.8-cm square paper pressed with #10 stopper against sample location.
[3] 5-cm2 head held for six s on sample location (air flow = 30 m/s).
[4] 113-cm head held for five min on sample location (air flow = 30 m / h).
54
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
APPENDIX C: EXAMPLES OF RECOMMENDED SWIPE PROCEDURES
EPA
Note that EPA (2006b) has published Sample Collection Procedures for Radiochemical
Analytes in Environmental Matrices. This is excerpted from Module II: Sampling
Procedures - Site Characterization and Remediation Phases:
7.0 Collection of Surface Area Samples Using Swipes
NOTE: Appropriate swipe materials and sizes to be used for the collection of surface area
samples, along with the number of swipes that should be taken, are selected based on
requirements included in the Sample Collection Plan (SCP).
7.1. Dry Swipes
7.1.1. Measure or determine by observation the total surface area to be sampled, and
record the area on the Field Sample Logbook.
7.1.2. Using a large area swipe [e.g., at most 300 cm2 (48 inches2)], [s]wipe the
surface area in parallel strokes. Place the swipe into a glassine envelope or bag,
and place a sample label on the envelope or bag.
7.1.3. Using a small area swipe [e.g., 25 cm2 (4 inches2) disc or square], [s]wipe the
surface in one continuous stroke of approximately 40 cm in length (16 inches),
or a 10 x 10 cm (4x 4 inches) square area, so that an area of approximately 100
cm2 is sampled. An "S" pattern, or moving from one edge to the other without
overlap, is the preferred method. Place the swipe into a glassine envelope or
bag, and place a sample label on the envelope or bag.
7.1.4. Proceed with 7.4 (Swipe Handling).
7.2. Wet Swipes
7.2.1. Measure or determine by observation the total surface area to be sampled, and
record the area on the Field Sample Logbook.
7.2.2. Dampen either a large area or small area swipe with the solvent fluid prescribed
by the SCP. DO NOT soak the swipe. If necessary, allow the swipe to dry
slightly before use.
7.2.3. If a volatile solvent is used, proceed with speed to prevent evaporation of the
solvent.
7.2.4. [S]Wipe the area per the procedures described in Section 7.1 (Dry Swipes) for
either large area or small area swipes.
7.2.5. Proceed with 7.4 (Swipe Handling).
7.3. Tape Swipes
NOTE: Tape swipes are typically collected for field screening and are not intended for
transport to and analysis in the laboratory. When analyzed for radioactivity, the glue
side of the tape must face the detector, because the paper backing of the tape will
attenuate any alpha particles.
55
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
7.3.1. Measure or determine by observation the total surface area to be sampled, and
record the area on the Field Sample Logbook.
7.3.2. Create a tape swipe by laying successive strips of 5 cm (2 inches) duct tape
sufficient to collect an area of 100 cm2 (16 inches2) or less. The edges of the
tape should be folded over or covered with tape to prevent them from sticking
to the surface of the object. This will create a "picture frame" around the actual
sample.
7.3.3. Lay the tape swipe onto the surface to be sampled and press down over the
sample area.
7.3.4. Carefully remove the tape and cover the exposed area with a piece of plain
paper.
7.3.5. Place the swipe in a plastic bag or envelope. A sample label is to be placed on
the bag or envelope.
7.3.6. Proceed with Section 7.4 (Swipe Handling).
7.4. Swipe Handling
7.4.1. Exit the sampling area using proper techniques to minimize the spread of
contamination.
7.4.2. Record the required information on the Field Sample Logbook, Field Sample
Tracking Form, and the sample label(s). The following information is to be
included at a minimum:
• SIC (Sample Identification Code)
• Time and date sample collected
• Sample location
• Sample area collected
• Percent of total area (calculated from surface area recorded in the Field
Sample Logbook)
• Sampler's initials
7.4.3. Place a sample label on the container.
7.4.4. Once outside of the area and back at an appropriate location, process the
sample for direct reading by Radiation Protection Personnel or, if required in
the SCP, for transport per the requirements of Module I, Section 7.0 (Sample
Packaging and Transport).
56
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
ISO
ISO 7503 contains the following guidelines on swipe sampling:
Detection and evaluation of surface contamination can be carried out using one or more dry or
wet smear [swipe] samples. When taking smear [swipe] samples from large areas, the
following points shall be taken into consideration to determine the distribution of
contamination:
r\
a) If possible, the area to be smeared [swiped] shall measure 100 cm ;
b) Where regulations permit the averaging of the surface contamination over larger
areas, such areas may be used for sampling and shall be included in the calculation
of the result;
c) The smear [swipe] material should be chosen to suit the surface to be checked (for
example, filter paper for smooth surfaces, cotton textile for rough surfaces);
d) If a wetting agent is used for moistening the smear [swipe] material, this wetting
agent should not exude from the material; WARNING: since the contamination
may be absorbed into the structure of the smear [swipe] material or may be covered
by residual moisture, the use of a wetting agent may lead to a significant
underestimation of the contamination in the case of alpha-emitters;
e) The smear [swipe] should be pressed moderately against the surface to be checked,
using fingertips or, preferably, by means of a holder which is designed to ensure
uniform and constant pressure;
f) The entire area of 100 cm2 shall be smeared [swiped];
g) If possible, circular filter papers should be used as the smear [swipe] material;
h) The contaminated area of the smear [swipe] sample shall be smaller than or equal
to the sensitive area of the probe;
i) After sampling, the smear [swipe] material shall be carefully dried in such a way
that loss of activity is prevented.
57
-------
Performance-Based Approach to the Use of Swipe Samples in a Radiological or Nuclear Incident
CFR
10 CFR 835 Appendix D—Surface Contamination Values—contains the following:
The data presented in Appendix D are to be used in identifying and posting contamination and
high-contamination areas 2 in accordance with § 835.603(e) and (f) and identifying the need for
surface contamination monitoring and control in accordance with § 835.1101 and 1102.
Surface Contamination Values"1
Radionuclide
U-nat, U-235, U-238, and associated decay products
Transuranics, Ra-226, Ra-228, Th-230, Th-228, Pa-231, Ac-227,
1-125, 1-129
Th-nat, Th-232, Sr-90, Ra-223, Ra-224, U-232, 1-126, 1-131, 1-133
Beta-gamma-emitters (nuclides with decay modes other than alpha
emission or spontaneous fission) except Sr-90 and others noted
above [5]
Tritium and STCs [6]
Removable
(dpm/100 cm2)
[2,4]
1,000 [7]
20
200
1,000
10,000
Total (Fixed +
Removable)
(dpm/100 cm2)
[2,3]
5,000 [7]
500
1,000
5,000
[6]
[1] The values in this appendix, with the exception noted in [note] 6 below, apply to radioactive contamination
deposited on, but not incorporated into the interior or matrix of, the contaminated item. Where surface
contamination by both alpha- and beta-gamma-emitting nuclides exists, the limits established for alpha- and
beta-gamma-emitting nuclides apply independently.
[2] As used in this table, dpm (disintegrations per minute) means the rate of emission by radioactive material as
determined by correcting the counts per minute observed by an appropriate detector for background, efficiency,
and geometric factors associated with the instrumentation.
[3] The levels may be averaged over one square meter provided the maximum surface activity in any area of 100
cm2 is less than three times the value specified. For purposes of averaging, any square meter of surface shall be
considered to be above the surface contamination value if: (1) from measurements of a representative number of
sections it is determined that the average contamination level exceeds the applicable value; or (2) it is determined
that the sum of the activity of all isolated spots or particles in any 100 cm2 area exceeds three times the
applicable value.
[4] The amount of removable radioactive material per 100 cm2 of surface area should be determined by swiping the
area with dry filter or soft absorbent paper, applying moderate pressure, and then assessing the amount of
radioactive material on the swipe with an appropriate instrument of known efficiency. (Note - The use of dry
material may not be appropriate for tritium.) When removable contamination on objects of surface area less than
100 cm2 is determined, the activity per unit area shall be based on the actual area and the entire surface shall be
[s]wiped. It is not necessary to use swiping techniques to measure removable contamination levels if direct scan
surveys indicate that the total residual surface contamination levels are within the limits for removable
contamination.
[5] This category of radionuclides includes mixed fission products, including the Sr-90 which is present in them. It
does not apply to Sr-90 that has been separated from the other fission products or mixtures where the Sr-90 has
been enriched.
[6] Tritium contamination may diffuse into the volume or matrix of materials. Evaluation of surface contamination
shall consider the extent to which such contamination may migrate to the surface in order to ensure the surface
contamination value provided in this appendix is not exceeded. Once this contamination migrates to the surface,
it may be removable not fixed; therefore, a "Total" value does not apply. In certain cases, a "Total" value of
10,000 dpm/100 cm may be applicable either to metals of the types from which insoluble special tritium
compounds are formed, that have been exposed to tritium, or to bulk materials to which insoluble special tritium
compound particles are fixed to a surface.
[7] These limits apply only to the alpha-emitters within the respective decay series.
22 Contamination area means any area, accessible to individuals, where removable surface contamination levels
exceed or are likely to exceed the removable surface contamination values specified in Appendix D of this part, but
do not exceed 100 times those values. High contamination area means any area, accessible to individuals, where
removable surface contamination levels exceed or are likely to exceed 100 times the removable surface
contamination values specified in Appendix D of this part. (10 CFR 835.2)
58
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |