EPA/600/R-17/212 August 2017
www.epa.gov/homeland-security-research
United States
Environmental Protection
Agency
oEPA
Field Application of Emerging
Composite Sampling Methods
2016-09-27 20:41:22
Office of Research and Development
National Homeland Security Research Center
-------
This page left intentionally blank
-------
EPA 600/R-17/212
Field Application of Emerging
Composite Sampling Methods
Technical Report
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
i
-------
Disclaimer
This document has been internally and externally peer-reviewed in accordance with U.S. Environmental
Protection Agency's (EPA's) Peer Review Handbook, 4th Edition 2015 (EPA/100/B15/001) and has been
approved for publication. Note that approval does not signify that the contents necessarily reflect the
views of the Agency. This procedure includes photographs of commercially available products. The
photographs are included for the purposes of illustration only. Any mention of trade names, products, or
services does not imply an endorsement by the U.S. government or EPA. The EPA does not endorse any
commercial products, services, or enterprises.
Questions concerning this document or its application should be addressed to the following individual:
Sang Don Lee, Ph.D.
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency (E343-06)
109 T.W. Alexander Drive
Research Triangle Park, NC 27711
Telephone No.: (919) 541-2973
E-mail Address: Lee.Sanadon@epa.gov
-------
Acknowledgments
The authors would like to thank the Department of Homeland Security (DHS) for their support of this work,
in particular, Dr. Donald Bansleben. Dr. Sang Don Lee is the project lead for this document. The
individuals and organizations listed below contributed to the development of this document.
EPA
M. Worth Calfee, National Homeland Security Research Center
Leroy Mickelsen, EPA Office of Emergency Management
John Archer, National Homeland Security Research Center
Timothy Boe, National Homeland Security Research Center
Oakridqe Research Institute for Science and Education
Douglas Hamilton
Jacobs Technology. Inc.
Laurie Brixey
Abderrahmane Touati
Jerome Gilberry
Ryan Stokes
-------
Contents
Disclaimer ii
Acknowledgments iii
Acronyms and Abbreviations x
Executive Summary 1
1 Introduction 1
1.1 Project Background 1
1.2 Project Description 2
1.2.1 Field Sampling Exercise 2
1.2.2 Laboratory Experiments 3
1.3 Project Objectives 4
2 Field Sampling Exercise Approach 5
2.1 Facility Description 5
2.2 Sampling Methods and Materials 7
2.2.1 Aggressive Air Sampling 7
2.2.1.1 Forced Aerosolization 7
2.2.1.2 Air Sampling 9
2.2.2 Robotic Floor Cleaner Sampling 10
2.2.3 Wet Vacuum Sampling 11
2.2.4 Ballast Coupon Preparation and Inoculation 11
2.2.5 Settling Plate Sampling 13
2.3 Composite Sampling Plan 13
2.3.1 Stage 1: RFC and Wet Vacuum Sampling, Round 1 15
2.3.2 Stage 2: AA Sampling, Round 1 15
2.3.3 Stage 3: AA Sampling with Hot Spots of Contamination, Round 2 16
2.3.4 Stage 4: Overnight Settling 16
2.3.5 Stage 5: RFC and Wet Vacuum Sampling, Round 2 17
3 Laboratory Experimental Approach 18
3.1 DFU Filter Loading Evaluation 18
3.2 NAM Prefilter Comparison 19
3.3 Forced Aerosolization Evaluation 20
3.3.1 Experimental Setup 20
3.3.2 Coupon Inoculation 22
3.3.3 Simulated Decontamination 24
4 Testing and Measurements 25
iv
-------
4.1 Sample Processing 25
4.1.1 DFU Filter Processing 25
4.1.2 NAM Prefilter Processing 25
4.1.2.1 Field Sampling Exercise NAM Prefilters 25
4.1.2.2 Laboratory Experiment NAM Prefilter Comparison 25
4.1.2.3 Laboratory Experiment Forced Aerosolization Evaluation NAM
Prefilters 26
4.1.3 RFC Sample Processing 26
4.1.4 Wet Vacuum Sample Processing 28
4.2 Analytical Procedures 29
4.2.1 Spore Analysis 29
4.2.2 Settling Plate Analysis 31
4.3 Flow Measurement 31
4.3.1 DFU Flow Measurement 31
4.3.2 NAM Flow Measurement 31
5 Results 33
5.1 Field Sampling Exercise Results 33
5.1.1 AA Sampling Results 33
5.1.1.1 DFU Sampling Results 33
5.1.1.2 NAM Prefilter Sampling Results 35
5.1.1.3 APS Sampling Results 36
5.1.1.4 Bioaerosol Button Sampling Results 37
5.1.1.5 Summary of AA Sampling Results for Stages 2 and 3 38
5.1.2 RFC and Wet Vacuum Sampling Results 38
5.1.3 Settling Plate Results 42
5.1.4 Resources Required for Field Sampling Exercise 44
5.2 Laboratory Experiment Results 45
5.2.1 DFU Filter Loading Evaluation Results 45
5.2.2 NAM Prefilter Comparison Results 47
5.2.3 Forced Aerosolization Results 47
6 Quality Assurance and Quality Control 50
6.1 Project Documentation 50
6.2 Integrity of Samples and Supplies 50
6.3 Instrument Calibrations 50
6.4 Critical Measurements 51
6.5 NHSRC Biolab Quality Checks 51
v
-------
7 Composite Sampling Method Conclusions and Recommendations 53
7.1 Field Test Operator and Observer Comments 54
7.1.1 PPE 54
7.1.2 Field Supplies 55
7.1.3 AA Sampling Procedure 55
7.1.4 RFC Sampling Procedure 56
7.1.5 Wet Vacuum Sampling Procedure 56
7.1.6 Settling Plate Sampling Procedure 56
7.2 Recommended AA Sampling Procedures 56
7.2.1 Future AA Sampling Recommendations 57
7.2.2 Recommended A A Sampling Deployment 58
References 59
Appendix A: Sampling Procedure Using Commercially Available Robotic Floor Cleaners for
Bacillus anthracis Spores - Neato® XV-21 Robotic Floor Cleaner
Appendix B: Sample Retrieval Procedure for Commercially Available Robotic Floor Cleaners for
Bacillus anthracis Spores
Appendix C: Sampling Procedure Using Commercially Available Wet Vacuum Cleaner for Bacillus
anthracis Spores
Appendix D: Coupon Inoculation Procedure for Spray-Dry Deposition
vi
-------
Tables
Table 1-1. FAPH Composite Sampling Stages 3
Table 2-1. Approximate Dimensions and Volume of FAPH Subway Tunnel 5
Table 2-2. Approximate Dimensions of Area Covered by Each Leaf Blower 8
Table 3-1. Test Matrix for Bench-Scale Forced Aerosolization Experiments 20
Table 3-2. Ingredients and Measurements for Spore Solution for Spray-Dry Deposition 23
Table 4-1. NAM Prefilter Comparison Testing and Processing 26
Table 5-1. DFU Filter Results from Stage 2 AA Sampling (without Hot Spot Contamination) 34
Table 5-2. DFU Filter Results from Stage 3 AA Sampling (with Hot Spot Contamination) 34
Table 5-3. NAM Prefilter Results from Stage 2 AA Sampling (without Hot Spot Contamination)
and Stage 3 AA Sampling (with Hot Spot Contamination) 35
Table 5-4. Bioaerosol Button Sampler Results from Stages 2 and 3 37
Table 5-5. Bioaerosol Button Sampler Results from Stage 5 38
Table 5-6. Average CFU Collected and Average Calculated Spore Air Concentration for Stages 2
and 3 38
Table 5-7. RFC Operating Time 38
Table 5-8. RFC and Wet Vacuum Sampling Results from Stage 1 (Post-decontamination) 40
Table 5-9. RFC and Wet Vacuum Sampling Results from Stage 5 (Post-AA Sampling with Hot
Spot Contamination) 41
Table 5-10. Debris Recovered from RFC Samples 41
Table 5-11. Sample Volume and Debris Recovered from Wet Vacuum Samples 41
Table 5-12. RFC and Wet Vacuum Sampling Results from Stage 1 (Post-decontamination) 42
Table 5-13. RFC and Wet Vacuum Sampling Results from Stage 5 (Post-AA Sampling with Hot
Spot Contamination) 42
Table 5-14. Kriging Prediction Error by Agar Plate Type and Location 43
Table 5-15. Settling Plate Comparison to Stage 5 Wet Vacuum Recovery 44
Table 5-16. Time and Area Sampled by Field Sampling Exercise Activity 45
Table 5-17. Estimated Field Sampling Exercise DFU Dust Load 46
Table 5-18. NAM Prefilter Bg Spore Collection Results 47
Table 5-19. Forced Aerosolization Results for fig-Inoculated Ballast Coupons 48
Table 5-20. Forced Aerosolization Results for Bf/<-lnoculated Ballast Coupons 48
Table 6-1. Instrument Calibration Methods and Frequencies 50
Table 6-2. DQIs and Acceptance Criteria for Critical Measurements 51
Table 6-3. Additional Quality Checks for Biological Measurements 52
vii
-------
Figures
Figure 2-1. Portion of FAPH Subway Tunnel Used for Study 5
Figure 2-2. FAPH Subway Tunnel Platform from North End Looking South (Subway Car Not
Present for Study) 6
Figure 2-3. FAPH Subway Tunnel Track (North of Platform) 6
Figure 2-4. Image from CCTV Camera on Ceiling of Tunnel at North End of Platform Looking
South 6
Figure 2-5. Black and Decker Inc. BV5600 Corded Electric Leaf Blower 7
Figure 2-6. AA Sampling Leaf Blower Zones in Subway Tunnel 8
Figure 2-7. Lane Divisions for Leaf Blower Operators in Subway Tunnel 8
Figure 2-8. Neato® XV-21 RFC 10
Figure 2-9. Hoover® Max Extract Cleaner Wet Vacuum 11
Figure 2-10. Ballast Coupons Showing Size and Weight in Pounds (left) and ADA Attachment
(right) 12
Figure 2-11. MDI and Actuator 12
Figure 2-12. Timeline of Relevant Activities in Subway Tunnel 14
Figure 2-13. RFC and Wet Vacuum Sampling Areas 15
Figure 2-14. AA Sampling Mixing Fan, Sampling Locations, and Coupon Locations 16
Figure 2-15. Settling Plate Placement Marked by Dots on Subway Platform 17
Figure 2-16. Settling Plate Placement Marked by Dots on Subway Tracks 17
Figure 3-1. DFU Filter Loading Test Setup (Arrows Indicate Air Flow Direction) 18
Figure 3-2. NAM Prefilter Comparison Test Setup (Arrows Indicate Air Flow Direction) 19
Figure 3-3. AA Sampling Wind Tunnel for Forced Aerosolization Evaluation 21
Figure 3-4. Leaf Blower Nozzle and Ballast Coupon Inside AA Sampling Wind Tunnel 21
Figure 3-5. Filtrete™ 1500 Filter Marked for Forced Aerosolization Evaluation 22
Figure 3-6. Spray-Dry Deposition Stack Modified for Ballast Coupon Deposition 24
Figure 4-1. Summary of Stage 1 RFC Sample Processing 27
Figure 4-2. Example Plates from Stage 5 RFC Samples 28
Figure 4-3. Summary of Stage 1 Wet Vacuum Sample Processing 28
Figure 4-4. Verification of Bg Colonies in Wet Vacuum Samples 29
Figure 4-5. Bacterial Colonies on Spiral-Plated Agar Plate 30
Figure 5-1. Starting and Ending DFU Flow Rates as Function of Sample Time for Stages 2 and 3 35
Figure 5-2. APS-Measured Total Particle Concentration vs. Time for Stages 2 and 3 36
Figure 5-3. APS-Measured Particle Size Distribution at Time 20:45 37
Figure 5-4. Stage 5 Sampling Path and Area for RFC Location 2 39
Figure 5-5. Heat Map of Spore Settling Generated from TSA Plate Counts 43
viii
-------
Figure 5-6. Heat Map of Spore Settling Generated from Brilliance™ Agar Plate Counts 43
Figure 5-7. DFU Filter Dust Load vs. DFU Flow Rate 46
ix
-------
Acronyms and Abbreviations
°c
Degree(s) Celsius
xg
times gravity
jjm
micrometer(s)
AA
aggressive air
ADA
aerosol deposition apparatus
ANOVA
analysis of variance
APS
Aerodynamic Particle Sizer
AREMA
American Railway Engineering and Mining Association
ATCC
American Type Culture Collection
AWTC
Asymmetric Warfare Training Center
Ba
Bacillus anthracis
Bg
Bacillus atrophaeus var. globigii
Biolab
ORD NHSRC Biocontaminant Laboratory
BSC
biosafety cabinet
Btk
Bacillus thuringiensis var. kurstaki
CBRN
Chemical, Biological, Radiological and Nuclear
CCTV
closed-circuit television
CDC
Centers for Disease Control and Prevention
CFM
cubic foot/feet per minute
CFU
colony-forming unit(s)
cm
centimeter(s)
CMAD
Consequence Management Advisory Division
CV
coefficient of variation
DCMD
Decontamination and Consequence Management Division
DFU
dry-filter unit
DHS
Department of Homeland Security
Dl
deionized
DOD
Department of Defense
DPG
Dugway Proving Ground
DQI
data quality indicator
ECBC
Edgewood Chemical Biological Center
EPA
U.S. Environmental Protection Agency
FAPH
Fort A.P. Hill
ft
foot/feet
ft2
square foot/feet
ft3
cubic foot/feet
g
gram(s)
GIS
geographic information system
GSD
geometric standard deviation
HEPA
high-efficiency particulate air
h
hour
HSRP
Homeland Security Research Program
in.
inch(es)
ISO
International Organization for Standardization
L
liter(s)
X
-------
LLNL
Lawrence Livermore National Laboratory
m
meter(s)
m3
cubic meter(s)
MDI
metered-dose inhaler
min
minute(s)
MULL
Massachusetts Institute of Technology Lincoln Laboratory
mL
milliliters)
mm
millimeters)
MMAD
mass median aerodynamic diameter
NAM
negative air machine
NHSRC
National Homeland Security Research Center
NIST
National Institute of Standards and Technology
OLEM
Office of Land and Emergency Management
ORD
Office of Research and Development
OTD
Operational Technology Demonstration
PAPR
powered air-purifying respirator
PBST
phosphate-buffered saline with 0.05% Tween®20
PNNL
Pacific Northwest National Laboratory
PPE
personal protective equipment
QA
quality assurance
QC
quality control
RFC
robotic floor cleaner
RH
relative humidity
rpm
rotation per minute
RTP
Research Triangle Park
SD
standard deviation
SS
stainless steel
TSA
tryptic soy agar
UTR
Underground Transport Restoration
xi
-------
Executive Summary
A release of hazardous biological material in an urban area would require decontamination of a wide
range of surfaces to protect the public. The U.S. Environmental Protection Agency (EPA) is responsible
for protecting human health and the environment from such contaminated areas. Accurate measurements
of residual contamination are needed to (1) select methods, locations, and other considerations required
for effective decontamination, and (2) inform decisions on re-entry and reoccupation of decontaminated
spaces. Traditional surface sampling methods (such as wipe, swab, and vacuum sock sampling) are used
for extent mapping, characterization, decontamination verification, and clearance sampling. For a wide-
area contamination incident, these traditional sampling methods can become a critical bottleneck in the
remediation process because they are time- and labor-intensive and may require large number of
samples to achieve reasonable confidence in the results.
Innovative composite sampling techniques may prove useful as an addition to currently used surface
sampling methods in a wide area biological incident. These composite sampling techniques include
aggressive air (AA) sampling as well as sampling using readily available surface cleaning technologies
such as robotic floor cleaners (RFCs) and wet vacuums. These methods will improve the sampling
capability in addition to the traditional surface sampling methods responding to a wide area incident. The
potential advantages of using these methods include the following:
• Reduced sampling time during a response
• Fewer samples requiring processing
• Detection of spore presence at unknown hot spots of contamination
• Improved detection of widespread contamination when concentrations are close to (or potentially
below) detection limits for traditional surface sampling methods
• Shortened timeline to recovery.
AA, RFC, and wet vacuum sampling are suitable for use in many building interiors and can allow rapid
sampling, requiring fewer personnel and fewer samples per unit area than current surface sampling
methods.
The study discussed in this report tested the effectiveness of AA, RFC, and wet vacuum composite
methods for sampling spores from a subway platform and rail surfaces. Specifically, this study consisted
of a field sampling exercise and laboratory experiments that are discussed separately in this report. The
field sampling exercise evaluated RFC, wet vacuum, and AA sampling. The field sampling exercise was
designed to evaluate the performance of these composite sampling methods for post-decontamination
sampling and sampling with the presence of multiple contamination hot spots. The separate laboratory
experiments evaluated AA sampling operational parameters and efficacy under controlled conditions.
For this project, the field sampling exercise was conducted in a mock subway system at Fort A.P. Hill
(FAPH) over a 24-hour (h) period. The AA, RFC, and wet vacuum sampling procedures were conducted
on the concrete platform and track (only AA sampling) in the subway system. Post-decontamination
sampling results showed that the wet vacuum and RFC methods can be used to sample areas containing
viable spores at concentrations as low as the single-digit range per square foot. The study showed that
ES-1
-------
the RFC and wet vacuum sample processing procedures require improvements in environments with
dusty surfaces. This study also showed that AA sampling methods require further development for large
volumes of air in dusty environments to avoid the overloading of filters. Section 7 of this report provides
specific recommendations, including the operational limitations of the sampling methods studied and
recommendations for improving these methods based on input by operators and observers of the field
sampling exercise and on the results of this study.
ES-2
-------
1 Introduction
The study discussed in this report tested the effectiveness of aggressive air (AA) sampling and other
composite methods for sampling spores from subway platform and rail surfaces. This research supports
the mission of the U.S. Environmental Protection Agency's (EPA's) National Homeland Security Research
Center (NHSRC) by providing information pertinent to the decontamination of areas contaminated
through an act of terrorism. This project supports the EPA's Homeland Security Research Program
(HSRP) and NHSRC's strategic goals as described in detail in the Homeland Security Strategic Research
Action Plan (EPA 2012a). This work is pertinent to Long-Term Goal 2, which states, "The Office of Land
and Emergency Management (OLEM) and other clients use HSRP products and expertise to improve the
capability to respond to terrorist attacks affecting buildings and the outdoor environments." This project
specifically addresses a need expressed by OLEM's Chemical, Biological, Radiological and Nuclear
(CBRN) Consequence Management Advisory Division (CMAD) to understand and optimize composite-
based sampling for a wide-area anthrax incident.
This project consisted of a field sampling exercise conducted in a mock subway system and laboratory
experiments to evaluate AA sampling operational parameters and efficacy under controlled conditions.
The field sampling exercise evaluated the use of three composite sampling methods for detecting spore
contamination: robotic floor cleaner (RFC), wet vacuum, and AA sampling. The field sampling exercise
was designed to evaluate the performance of these composite sampling methods for post-
decontamination sampling and sampling with the presence of hot spot contamination. The separate
laboratory tests evaluated AA sampling operational parameters and efficacy under controlled conditions.
The project background, description, and objectives are discussed below.
1.1 Project Background
A release of hazardous biological material in an urban area would require decontamination of a wide
range of surfaces to protect the public. EPA is responsible for protecting human health and the
environment from such contaminated areas. Accurate measurements of residual contamination are
needed to: (1) select methods, locations, and other considerations required for effective decontamination,
and (2) inform decisions on re-entry and reoccupation of decontaminated spaces. Traditional surface
sampling methods (such as wipe, swab, and vacuum sock sampling) are used for extent mapping,
characterization, decontamination verification, and clearance sampling. For a wide-area contamination
incident, these traditional sampling methods can become a critical bottleneck in the remediation process
because they are time- and labor-intensive and may require large numbers of samples to achieve
reasonable confidence.
Innovative composite sampling techniques may prove useful as an addition to currently used surface
sampling methods. These composite sampling techniques include AA sampling as well as sampling using
readily available surface cleaning technologies such as RFCs and wet vacuums. The potential
advantages of these methods include the following:
• Reduced sampling time during a response
• Fewer samples requiring processing
1
-------
• Detection of spore presence at unknown hot spots of contamination
• Improved detection of widespread contamination when concentrations are close to (or potentially
below) detection limits for traditional surface sampling methods
• Shortened timeline to recovery.
AA, RFC, and wet vacuum sampling are suitable for use in many building interiors and can allow rapid
sampling requiring fewer personnel and fewer samples per unit area than current surface sampling
methods.
EPA has conducted prior studies (Lee et al. 2013) to test commercial floor cleaning devices such as
RFCs and wet vacuum cleaners for sampling Bacillus spores. The devices were evaluated for their
usability on various surface types and under different contamination scenarios. Commercial floor cleaning
devices can sample a wider area per sampling event than traditional surface sampling methods, thereby
reducing labor. The sampling efficacy of the RFCs and wet vacuums used in this study is comparable to
currently used sampling methods such as wipe and vacuum sock sampling.
During this project, sampling procedures were developed for RFCs and wet vacuums to provide methods
for trained incident responders to collect environmental samples from flat, contiguous surfaces after a
biological contamination incident. Appendix A and Appendix B detail the sampling and sample retrieval
procedures for RFCs, respectively, and Appendix C details the sampling procedures for wet vacuums.
Data from the collected samples are intended to allow determination of the presence or absence of
pathogenic microorganisms and the contamination level after natural outbreaks and intentional or
accidental releases of pathogenic microorganisms.
1.2 Project Description
As indicated above, this project consisted of a field sampling exercise and laboratory experiments. Each
project component is discussed below.
1.2.1 Field Sampling Exercise
The field sampling exercise evaluated the RFC, wet vacuum, and AA sampling methods in a mock
subway system over a 24-hour (h) period during the Operational Technology Demonstration (OTD)
project, part of the Department of Homeland Security (DHS)-funded Underground Transport Restoration
(UTR) program.
The UTR-OTD project is an interagency effort involving the following federal agencies and National
Laboratories: EPA, DHS, Lawrence Livermore National Laboratory (LLNL), Massachusetts Institute of
Technology Lincoln Laboratory (MITLL), and Pacific Northwest National Laboratory (PNNL). The overall
purpose of the UTR-OTD project was to conduct and evaluate field-level mass transportation and
biological remediation of two decontamination technologies directed at the intentional release of a
biological agent such as Bacillus anthracis (Ba).
The OTD was conducted at a Department of Defense (DOD) mock subway tunnel at Fort A.P. Hill (FAPH)
in Bowling Green, VA. The UTR-OTD involved all aspects of remediation of a subway system
2
-------
contaminated with a biological agent, including pre-decontamination and post-decontamination
verification sampling and waste management. However, the aspects of the UTR-OTD pertinent to this
project are limited to the first dissemination of Bacillus atrophaeus var. globigii (Bg) spores in the tunnel
and decontamination of the tunnel by fogging with diluted bleach. The UTR-OTD report (EPA 2017)
discusses in detail the methods of spore dissemination, decontamination, sampling, and analysis
conducted before the activities described in this report.
The AA sampling method involves forced aerosolization of particles from a surface using a leaf blower,
collection of aerosolized particles by air sampling, followed by quantitative analysis of collected samples
for bacterial spores. During the field sampling exercise, the air samplers used were dry-filter units (DFUs)
and negative air machines (NAMs) equipped with prefilters. Traditional surface sampling techniques can
provide a measure of the surface contamination from a fraction of potentially contaminated surface area.
AA sampling can provide a collective measure of contamination in the impacted area regardless of
surface types with small number of samples compared to the traditional sampling methods.
Composite sampling was conducted on September 27 and 28, 2016, after the first round of
decontamination (by fogging) and post-decontamination sampling using conventional sampling methods.
Table 1-1 describes the composite sampling campaign, which was divided into five stages.
Table 1-1. FAPH Composite Sampling Stages
Stage
Details
Purpose
1
RFC and wet vacuum sampling on platform
Evaluate post-decontamination sampling
2
AA sampling
Evaluate post-decontamination sampling
3
AA sampling with hot spot Bg contamination
Assess detection of hot-spot contamination that could be
missed by traditional surface sampling methods
4
Overnight settling of Bg spores on agar
plates
Assess redistribution of hot spots from AA sampling
5
RFC and wet vacuum sampling on platform
Assess redistribution of hot spots from AA sampling
During Stage 1, RFCs and wet vacuums were used to conduct post-decontamination sampling in subway
platform floor areas. During Stage 2, post-decontamination AA sampling was conducted using leaf
blowers to aerosolize particles from all of the floor and track surfaces that could be reached. During Stage
3, trays of ballast rocks inoculated in the laboratory with Bg spores were placed in the subway tunnel to
create "hot spots" of concentrated spore contamination, and the AA sampling procedure was repeated.
Stage 4 consisted of an overnight settling period for the Bg spores using 200 agar plates distributed
throughout the subway tunnel. Finally, during Stage 5, RFCs and wet vacuums were used to sample the
same areas of the subway platform floor that were sampled during Stage 1.
1.2.2 Laboratory Experiments
The laboratory experiments included a DFU filter loading evaluation, NAM prefilter comparison, and
forced aerosol evaluation. These experiments were aimed to provide information that the NHSRC can
use to assess and improve the AA sampling method for Ba spores and to evaluate equipment and
methods used in the field sampling exercise.
3
-------
1.3 Project Objectives
The purpose of the field sampling exercise was to evaluate AA, RFC, and wet vacuum sampling
techniques for detecting spore contamination. The laboratory experiments were designed to evaluate AA
sampling operational parameters and efficacy under controlled conditions. The specific project objectives
included the following:
• During the field sampling exercise, characterize the aerosolization of surrogate spores from
ballast rock material used around train tracks in subway systems.
• During the field sampling exercise, test the hypothesis that using the AA sampling method in a
large area will distribute hot spot contamination throughout the entire area so that the
contamination is more likely to be detected by surface sampling methods.
• During the laboratory experiments, assess the impacts on spore aerosolization of certain
experimental conditions, including spore type, spore loading, and decontamination simulated by
spraying with deionized (Dl) water instead of a decontamination agent.
• During the laboratory experiments, conduct a limited evaluation of DFUs and NAMs for use in AA
sampling.
4
-------
2 Field Sampling Exercise Approach
This section discusses the general approach for the field sampling exercise, including the facility
description, sampling methods and materials, and the composite sampling plan.
2.1 Facility Description
The field sampling exercise was conducted at FAPH in Bowling Green, VA. FAPH is used to train active
and reserve troops of the U.S. Armed Forces as well as for training personnel from other government
agencies, including the Department of State, Department of the Interior, the U.S. Customs Service, and
federal, state, and local security and law enforcement agencies. FAPH's Asymmetric Warfare Training
Center (AWTC) is a 300-acre site consisting of a headquarters, barracks, administrative offices, training
and maintenance facilities, several training ranges, and an "urban area." The urban area includes a
subway station complete with subway cars, a train station with rail cars, and other urban buildings.
The Underground Transport Restoration (UTR)-Operational Technology Demonstration (OTD) project
was conducted in September and October 2016 in the subway station in the urban area of the FAPH
AWTC. The composite sampling was conducted in the subway tunnel on September 27 and 28, 2016.
Figure 2-1 shows a schematic drawing of the portion of the subway tunnel where sampling was
conducted.
South
North
Figure 2-1. Portion of FAPH Subway Tunnel Used for Study
Table 2-1 summarizes the approximate dimensions (in feet [ft]) and volume (in cubic feet [ft3]) of the
subway tunnel.
Table 2-1. Approximate Dimensions and Volume of FAPH Subway Tunnel
Tunnel Section
Length
Width
Height
Approx. Volume
(ft)
(ft)
(ft)
(ft3)
Track south of platform
53
22
19
22,000
Track center
162
16.5
19
51,000
Track north of platform
60
22
19
25,000
Platform
162
23
15
56,000
Kiosk (between staircases)
27
6.5
15
2,600
Total volume
157,000
5
-------
Figures 2-2 and 2-3 present photographs of the platform and track areas of the subway tunnel. Figure 2-4
is a still image from the closed-circuit television (CCTV) camera. No subway cars were used or present in
the tunnel for the study. The entrances to the subway tunnel were sealed to make an enclosed space,
and a non-pathogenic surrogate organism, Bg, was sprayed in the subway tunnel.
i
Figure 2-2. FAPH Subway Tunnel Platform from North End
Looking South (Subway Car Not Present for Study)
Figure 2-3. FAPH Subway
Tunnel Track (North of Platform)
Double-click to go to fullscreen, ctrl+ click to snap to video size
Figure 2-4. Image from CCTV Camera on Ceiling of Tunnel at North End of Platform Looking South
6
-------
2.2 Sampling Methods and Materials
Equipment and materials for the field sampling exercise were prepared in laboratories at EPA's Research
Triangle Park (RTP) campus in NC. Samples collected during the field sampling exercise were packed
and transported to the EPA RTP Office of Research and Development (ORD) NHSRC Biocontaminant
Laboratory (Biolab) for processing and analysis in accordance with culture-based microbiological assay
techniques developed by EPA and the Centers for Disease Control and Prevention (CDC). Except for the
settling plate samples, all sample results were quality checked by Biolab staff and results were delivered
electronically to project personnel. The settling plates were incubated and enumerated on site at FAPH.
The sampling methods included AA sampling, RFC sampling, wet vacuum sampling, ballast coupon
preparation and inoculation, and settling plate sampling, as discussed below.
2.2.1 Aggressive Air Sampling
The AA sampling procedure conducted in the subway tunnel used leaf blowers for forced aerosolization
of particles, mixing fans to help maintain particle suspension, and two different types of filter-based
samplers for particle collection, DFU samplers and an NAM duct with a prefilter. In brief, AA sampling in
the subway tunnel was executed as detailed below.
• Seven mixing fans were placed in the subway tunnel at prescribed locations on both the track and
platform areas.
• Nine DFU samplers were placed in the subway tunnel at prescribed locations on the track.
• One NAM outside the hot zone was fitted with flexible ductwork to locate the inlet on the platform
floor and connect a 14-inch (in.) by 20-in. prefilter for sample collection. (Two NAM prefilters were
planned, but the flange for connecting the NAM duct on the south stairway failed).
• Three leaf blowers were operated in a prescribed manner on all track and platform surfaces
within reach of the operators.
The equipment and procedures for forced aerosolization and air sampling are described in detail below.
2.2.1.1 Forced Aerosolization
To provide forced particle aerosolization, corded electric leaf blowers (BV5600, Black and Decker Inc.,
Towson, MD) were used on the platform and track of the mock subway tunnel (Figure 2-5).
Figure 2-5. Black and Decker Inc. BV5600 Corded Electric Leaf Blower
7
-------
For AA sampling during the field sampling exercise, the hot zone was divided into three areas: the
platform, the track from the middle of the platform to the edge of the hot zone, and the track from the
middle of the platform to the dead end. One operator was assigned to operate each of the three leaf
blowers in the following sections of the subway tunnel (shown in Figure 2-6):
• Leaf blower 1: Subway track south of center line of platform
• Leaf blower 2: Subway platform
• Leaf blower 3: Subway track north of center line of platform.
South
North
Figure 2-6. AA Sampling Leaf Blower Zones in Subway Tunnel
Table 2-2 summarizes the approximate dimensions in ft and square feet (ft2) of the area covered by each
leaf blower.
Table 2-2. Approximate Dimensions of Area Covered by Each Leaf Blower
Leaf
Blower
Width 1
(ft)
Depth 1
(ft)
Width 2
(ft)
Depth 2
(ft)
Approx. Area
(ft2)
Description
1
53
22
81
16,5
2,500
Track south
2
162
16,5
81
6.5
3,200
Platform
3
60
22
81
16,5
2,700
Track north
The operators were given approximately 30 minutes (min) to cover each area. Using leaf blowers, the
operators were required to sweep at a rate of 83 to 106 ft2/min for a single pass.
The track level was divided into lanes to guide the leaf blower operators. The sections of the tunnel north
and south of the platform are approximately 22 ft wide. These spaces were divided into five lanes as
shown in Figure 2-7.
Subway tunnel wall
Platform
Lane 5 - lowest level
Lane 4 - 4* sloping up to track
height
/MMmtmiwmiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiin
Lane 2 - sloping up to track height
Lane 1 - lowest level
Subway tunnel wall
Figure 2-7. Lane Divisions for Leaf Blower Operators in Subway Tunnel
8
-------
In the center section, the platform reduces the width of the track to 16.5 ft, and that space was divided
into four lanes. The field operators did not actually mark the lanes in the tunnel because the topography
of the subway tunnel provides clear delineations. The train tracks are approximately 2 ft higher than the
ballast closest to the tunnel walls, creating the following divisions:
• Lane 1: First 4 ft from far tunnel wall; lowest part of subway tunnel
• Lane 2: From edge of Lane 1 to rail farther from platform; sloping
• Lane 3: Between rails
• Lane 4: From rail nearer to platform to edge of platform or edge of Lane 5; sloping
• Lane 5: First 4 ft from tunnel wall on platform side; lowest part of subway tunnel.
Leaf blower 1 and leaf blower 3 operated at opposite ends of the tunnel, starting from the far ends of the
tunnel in Lane 1 and progressing toward the center of the platform. After completing Lane 1, leaf blowers
1 and 3 each returned to the end of the tunnel and completed a pass of Lane 2 and then each
subsequent lane until all surfaces were covered. Leaf blower 2 covered the platform area starting from
one side and progressing to the other. Each operator covered as much of his or her designated section
as possible during each AA sampling period. The operators attempted to maintain a 45° angle and 1
centimeter (cm) of clearance between the leaf blower nozzle and surface, covering all horizontal and
vertical surfaces within reach. Operators walked slowly forward through their designated areas, moving
the nozzle of the leaf blower from side to side in an action similar to that of blowing leaves.
Seven mixing fans operated during AA sampling to assist with mixing and keeping particles airborne. The
fans included oscillating pedestal fans (Model UP30BN-S, Airmaster Fan Company, Jackson, Ml) and 42-
in.-diameter barrel fans (Model HBPC4213, Triangle Engineering of Arkansas, Inc., Jacksonville, AR).
2.2.1.2 Air Sampling
DFU samplers (DFU-1000, Lockheed Martin Integrated Technology LLC, Gaithersburg, MD) were
deployed to collect aerosol samples on the track level. DFUs are high-volume air samplers that use 47-
millimeter (mm)-diameter polyester felt filters (DFU-P-24, Leidos, Reston, VA) with a 1-micrometer (jjm)
pore size for particle collection. The DHS BioWatch program and U.S. military use DFUs as samplers for
bioaerosol detection. Nine DFU samplers were placed along the centerline of the track approximately 30
ft apart.
The UTR-OTD health and safety plan dictates that at least one high-efficiency particulate air (HEPA)-
filtered NAM must be operating at all times to ventilate the subway tunnel. The nominal operating flow
rates for the type of NAM deployed in the tunnel (Omni-Aire 2200C, Omnitec Design, Inc., Mukilteo, WA)
is 1,000 cubic feet per minute (CFM) on the low setting and 1,800 CFM on the high setting. It was
decided to take advantage of these high flow rates for aerosol sampling by attaching prefilters to the inlets
of two NAMs. The initial plan for AA sampling included attaching 25-ft-long, 12-in.-diameter, non-insulated
aluminum flexible ducts (Part 3XK08, Grainger Industrial Supply, Lake Forest, IL) to two NAMs, one at
each stairwell landing, with a filter box (Part 62039, www.budaetheatina.com) and prefilter installed on the
inlet of each NAM duct for particle collection. However, only one NAM was deployed because the flange
for connecting the NAM duct on the south stairway failed. The NAM prefilters were 14-in. by 20-in.
9
-------
furnace filters (Filtrete™ MPR 2800, 3M, St. Paul, MN) rated to collect 97% of 1-jjm particles. The
prefilters were prepared by cutting away the metal mesh on the filter face and marking a 6-in. by 6-in.
area in the center of the filter face.
An Aerodynamic Particle Sizer (APS, Model 3321, TSI, Inc., Shoreview, MN) was deployed on a cart
located on the subway platform to measure the size distribution of particulates aerosolized during AA
sampling. The particle size data were saved on a laptop in the field, downloaded to the EPA network, and
summarized in a spreadsheet. The data also were used to produce graphs of the particle size distribution
and total concentration overtime during AA sampling.
Several Bioaerosol Button Samplers (SKC, Inc., Eighty Four, PA) were worn by personnel (personal
samples) and mounted to a cart (area sample) to collect localized, task-specific filter samples of inhalable
particles. Filters were retrieved from the button samplers, transported to the ORD NHSRC
Biocontaminant Laboratory (Biolab) and analyzed for viable spores.
2.2.2 Robotic Floor Cleaner Sampling
Neato®XV-21 RFCs (Neato® Robotics, Newark, CA) were charged, tested, recharged, and packed for
deployment in accordance with the procedure described in Appendix A. The Neato® XV-21 (Figure 2-8) is
equipped with mapping and navigation technologies and is capable of returning to its starting position
after covering the entire floor surface of an enclosed sampling area.
Figure 2-8. Neato® XV-21 RFC
An observer monitored the RFCs to ensure their proper operation and to notify sampling personnel if an
RFC malfunctioned. If an RFC malfunctioned, it was immediately removed and replaced by a backup
RFC. After sampling was complete, the RFC filter sample was retrieved from each deployed RFC in
accordance with the RFC sample retrieval procedure described in Appendix B.
10
-------
2.2.3 Wet Vacuum Sampling
Three areas on the subway platform were marked with magnetic strips for wet vacuum sampling. The wet
vacuum model used was the Hoover® Max Extract Cleaner (F7425-900 SteamVac Dual V with SpinScrub
Hand Tool, Hoover®, Glenwillow, OH). Appendix C describes the procedures used to prepare each wet
vacuum for sampling. The Hoover® Max Extract Cleaner has a clean water tank and a dirty water tank
(Figure 2-9). The clean tank was filled with 5 liters (L) of a sterile solution of 0.05% Tween® 20 surfactant
in Dl water.
Wet vacuum sampling was conducted concurrently with RFC sampling. The wet vacuums were operated
on all accessible floor space in the designated sampling areas with both "Rinse" and "Power Scrub"
modes turned on. The initial vacuuming stroke was made with the liquid dispensing trigger on, followed by
two vacuum-only strokes (liquid dispensing trigger off) covering the same area. The wet vacuum then was
moved over to cover an area consisting of 50% new area and 50% of the area just covered. A new initial
vacuum stroke was made with the liquid dispensing trigger on, followed by two vacuum-only strokes
covering the same area. Sampling proceeded in this manner (one wet stroke followed by two dry strokes)
until the entire sampling area was covered.
An observer monitored the wet vacuum sampling to ensure that the operators were conducting sampling
in accordance with the procedure described in Appendix C. If a wet vacuum malfunctioned, it was
immediately replaced by a backup wet vacuum. For each wet vacuum, the entire dirty water tank was
placed in a cooler and transported as one sample for microbiological analysis.
2.2.4 Ballast Coupon Preparation and Inoculation
During Stage 3, hot spots of contamination were brought into the subway tunnel in the form of ballast
rocks inoculated with Bg spores. Each coupon consisted of 28 pounds (± 0.5 pound) of ballast rocks in a
stainless-steel (SS) tray measuring approximately 12 in. by 12 in. by 3 in. deep, with a 1 -in. lip around the
perimeter suitable for clamping onto an aerosol deposition apparatus (ADA) as shown in Figure 2-10.
Spray Switch
Vacuum Switch
Clean Tank
Figure 2-9. Hoover® Max Extract Cleaner Wet Vacuum
11
-------
Figure 2-10. Ballast Coupons Showing Size and Weight in Pounds (left) and ADA Attachment
(right)
The ballast coupons were fabricated at the EPA RTP campus using rocks similar in type and morphology
to samples obtained from FAPH. The ballast used was 100% crushed granite meeting American Railway
Engineering and Mining Association (AREMA) #4 ballast specifications. The ballast source was checked
for microbiological contamination by agitating the rocks in a solution of sterile phosphate-buffered saline
with 0.05% Tween®20 (PBST) and plating samples of the solution in accordance with the procedure
described in Section 4.2.1. No evidence of contamination (no growth) was observed. Therefore, the
ballast coupons were not sterilized before spore inoculation. Before coupon assembly and inoculation, the
coupon trays were autoclaved, and the SS ADAs were sanitized using bleach wipes (Dispatch® 69150,
Clorox®, Oakland, CA).
The Sa-simulant spore used for hot spot coupon inoculation was a powdered spore preparation of Bg
American Type Culture Collection (ATCC) 9372 mixed with silicon dioxide particles. Bg is a gram-positive,
spore-forming, rod-shaped bacterium found in the environment, particularly in hay. This powdered spore
preparation was obtained from the U.S. Army Dugway Proving Ground (DPG) Life Sciences Division
(Dugway, UT). The procedure for preparing this preparation is described in Brown et al. 2007. Briefly,
after 80 to 90% sporulation, the suspension was centrifuged to generate a preparation of approximately
20% solids. The dried spores were dry-blended and jet-milled with fumed silica particles (Degussa,
Frankfurt am Main, Germany), resulting in a powdered matrix containing approximately 1 x 1011 viable
spores per gram (g). The powdered Bg spore preparation was loaded into metered-dose inhalers (MDis)
in accordance with a proprietary protocol. Figure 2-11 shows an MDI and its actuator.
Figure 2-11. MDI and Actuator
12
-------
Ballast coupons and 14-in.-square SS positive control coupons were inoculated with Bg spores using an
MDI (Calfee et al. 2013). In brief, each coupon was inoculated in a separate ADA (Figure 2-10) designed
to fit over a ballast coupon tray or one 14-in. by 14-in. coupon. A spore-loaded MDI was weighed, and
then the MDI and actuator were inserted into the inoculation opening in the center top of the pyramidal SS
ADA hood. The MDI was discharged once into the ADA and removed, and then the inoculation opening
cover was closed. The MDI was weighed again, and the weight change was calculated to verify a
successful inoculation (0.04-g to 0.07-g loss). This procedure was repeated until all the coupons were
inoculated. The first, middle, and last coupons inoculated were SS positive control coupons that were left
undisturbed for a minimum of 18 h and then sampled using a sponge stick (3M™ Sponge-Stick, St. Paul,
Minnesota; catalog number SSL-10NB) in accordance with the CDC-published procedure (CDC 2012).
Twenty-four ballast coupons, each with 1 -ft2 ballast area, were prepared and inoculated with Bg spores.
Each ballast coupon with its ADA still attached was double-bagged in polyethylene autoclave bags (01-
829F, Fisher Scientific, Waltham, MA), sealed in a rugged plastic tote (44066, Centrex Plastics LLC,
Findlay, OH), and transported to FAPH with the sampling equipment.
The total change in mass of the MDI over the inoculation of the 24 ballast coupons was 1.371 g.
Recovery from SS positive control coupons using the sponge-stick swab method (CDC 2012) averaged
8.9 x io8 colony-forming units (CFU) per g of MDI weight change, with a standard deviation (SD) of 8.4 *
107 CFU/g and a coefficient of variation (CV) of 9.5%. Therefore, the best estimate of the total Bg spores
(measured as CFU) contained in all of the 24 hot spot ballast coupons is 1.2 * 109 CFU total, or an
average of 5.1 * 107 CFU/ft2 of ballast coupon.
2.2.5 Settling Plate Sampling
During the overnight period after AA sampling with hot spots (Stage 3), significant settling of spores was
expected, with hot spot contamination. To quantify the overnight settling of spores, agar plates were
placed on the platform and track surfaces during Stage 4 after the completion Stage 3. Two types of agar
plates measuring 100-mm in diameter each were used: tryptic soy agar (TSA) plates (Difco 236050,
Becton Dickinson and Company, Franklin Lakes, NJ) and Brilliance™ B. cereus agar plates (Oxoid
CM1036, Thermo Fisher, Waltham, MA). Bg is easily identified and enumerated when cultured on both of
these types of agar plates. Bg forms orange-pigmented colonies on TSA plates and blue-pigmented
colonies on Brilliance™ agar plates.
One TSA plate and one Brilliance™ agar plate were collocated at each of the 100 settling plate locations
in the grid, for a total of 200 plates. Section 2.3.4 discusses the Stage 4 sampling locations and
procedures in more detail.
2.3 Composite Sampling Plan
After UTR-OTD post-decontamination surface sampling, this project's field sampling exercise investigated
the composite sampling methods of AA, RFC, and wet vacuum sampling. Briefly, the sampling plan
consisted of collecting one set of RFC and wet vacuum samples (Stage 1), and then conducting the AA
sampling procedure (Stage 2). Afterwards, hot spots of spore-contaminated ballast material were
uncovered in the subway tunnel and the AA sampling procedure was repeated (Stage 3), followed by an
overnight period of particle settling onto agar plates (Stage 4). The sampling concluded with collection of
13
-------
another set of RFC and wet vacuum samples (Stage 5). Figure 2-12 shows a timeline of the field
sampling exercise activities in the subway tunnel relevant to this project.
UTR-OTD first Bg release and decontamination by fogging with diluted bleach
I 1
Post-decon Post-decon AA sampling Overnight settling Post-AA sampling
RFC and wet AA sampling with hot spot RFC and wet
vacuum sampling contamination vacuum sampling
Figure 2-12. Timeline of Relevant Activities in Subway Tunnel
Possibly the greatest challenge in this study was the short period of time (less than 24 h) that the subway
tunnel was available for the field sampling exercise after UTR-OTD post-decontamination surface
sampling. The field sampling exercise was divided into the following five stages:
Stage 1: Post-decontamination RFC and wet vacuum sampling, Round 1
1. Delineation of areas on subway platform to be sampled using RFC and wet vacuum methods
2. RFC and wet vacuum sampling on the subway platform
3. Collection, labeling, and storage of RFC and wet vacuum samples
Stage 2: Post-decontamination AA sampling, Round 1
1. Setup of AA sampling mixing fans, DFU samplers, and NAM prefilters
2. Placement of sealed, spore-contaminated ballast coupons
3. AA sampling on platform and track levels
4. Collection, labeling, and storage of AA samples
Stage 3: AA sampling with hot spots of contamination, Round 2
1. Samplers loaded with clean filters
2. Unwrapping of pre-positioned, spore-contaminated ballast coupons in subway tunnel to create
hot spots
3. AA sampling on platform and track levels
4. Collection, labeling, and storage of AA samples
Stage 4: Overnight settling
1. Placement of pre-labeled TSA and Brilliance™ agar settling plates in tunnel to collect spores
during overnight settling period
2. Collection of settling plates and placement in incubator
14
-------
Stage 5: Post-AA sampling RFC arid wet vacuum sampling, Round 2
1. RFC and wet vacuum sampling on subway platform
2. Collection, labeling, and storage of RFC and wet vacuum samples
3. Enumeration of colonies on TSA and Brilliance™ agar settling plates
4. Packaging and transport of all samples to EPA RTP campus for processing and analysis
2.3.1 Stage 1: RFC and Wet Vacuum Sampling, Round 1
RFC sampling was conducted on the subway platform to assess post-decontamination residual spore
levels on the platform. Magnetic boundary markers (Neato® Robotics, Newark, CA) were used to
delineate the three areas for RFC sampling, RFC1, RFC2, and RFC3, shown in Figure 2-13.
. 1
South
North
RFC1 ^
RFC3 0
Robotic floor
cleaner
RFC#
Magnetic strips
WV# Wet vacuum
Figure 2-13. RFC and Wet Vacuum Sampling Areas
Three magnetic strips were used to continue the line formed by the walls enclosing the stairwells. These
strips divided the three RFC sampling areas from the three wet vacuum sampling areas. Two additional
magnetic strips divided the remaining area of the platform into thirds. One dividing line extended from the
wall enclosing the south stairwell to the support column south of center, then from the opposite side of the
column to the edge of the platform. The last dividing iine was in the corresponding location on the north
side of the platform. Each RFC sampling area measured approximately 900 ft2 (54 by 16.5 ft). One RFC
was placed and started in each of the three sampling areas.
Wet vacuum sampling was conducted on the subway platform at the same time as RFC sampling in the
three areas marked for wet vacuum sampling on Figure 2-13, WV1, WV2, and WV3. Each wet vacuum
sampling area measured approximately 175 ft2 (27 by 6.5 ft).
After sampling was complete, the RFC filter sample was retrieved from each deployed RFC in
accordance with the RFC sample retrieval procedure described in Appendix B. Wet vacuum samples
were collected as detailed in Appendix C.
2.3.2 Stage 2: AA Sampling, Round 1
Aggressive air sampling was conducted in the subway tunnel to assess residual spore levels in the
subway tunnel post-decontamination. As discussed in Section 2.2.1,1. three leaf blowers were used for
forced aerosolization during the Stage 2 AA sampling. Figure 2-14 shows the locations of the seven
mixing fans and the DFU and NAM samplers used during the AA sampling as discussed in Section 2.2.1.
15
-------
South
North
g Bg-containing
coupon hot spot
## DFU sampler
NAM duct with prefilter
Figure 2-14. AA Sampling Mixing Fan, Sampling, and Coupon Locations
During Stage 2, sampling was conducted concurrently with the leaf blowers. Nine DFU samples and one
NAM prefilter sample were collected from the locations shown in Figure 2-14. DFUs were placed on the
tracks equidistant from each other. In addition, seven mixing fans were placed at the following locations:
one at each end of the track facing the center of the tunnel (locations F1 and F5), two (F6 and F7) in the
corners of the platform at the bottom of each stairway and oriented facing the center of the platform, one
(F3) in the center of the tunnel wall facing the platform, and fans F2 and F4 along the tunnel wall at each
end of the platform facing the center of the platform. All of the fans were angled 45° upward and faced the
center of the zone. The fans operated for the entire time the operators were in the hot zone. Oscillating
pedestal fans were placed at locations F2, F3, F4, F6, and F7, and barrel fans were placed at locations
F1 and F5. After sampling was complete, the DFU and NAM filter samples were retrieved from each
deployed sampler and placed in labeled plastic bags.
In addition, ballast coupon hot spots inoculated with Bg spores were placed in the tunnel along the walls
at the locations shown in Figure 2-14 in preparation for Stage 3. The coupons remained sealed to prevent
cross-contamination of Stage 2 AA samples.
2.3.3 Stage 3: AA Sampling with Hot Spots of Contamination, Round 2
During Stage 3, new DFU and NAM prefilters were loaded, and the ballast material coupons inoculated
with Bg spores were uncovered. Figure 2-14 shows the coupon locations. Then, a second round of AA
sampling was conducted using the same procedures and locations used during Stage 2 as discussed
above.
2.3.4 Stage 4: Overnight Settling
After AA samples were collected, TSA and Brilliance™ agar plates were collocated at each location
shown in Figure 2-15 (platform) and Figure 2-16 (tracks).
16
-------
South 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 North
PA . . . . . . . * * . . .
PC*
Figure 2-15. Settling Plate Placement Marked by Dots on Subway Platform
BA
BB
BC
South
North
123456789
• • • • n I
10 11 12 13 14 15 16 17 18 19 20
| • • • •
• ••••••••
• ••••••••••
Figure 2-16. Settling Plate Placement Marked by Dots on Subway Tracks
The plates were labeled as follows: P (platform) or B (ballast), row (A, B, or C), and column number (for
example, PC13). The first settling plate was placed in the tunnel at 00:16, on September 28, 2016, and
the last was placed at 00:53 on the same day. The tunnel was left overnight (approximately 7.5 h) to allow
aerosolized particles to settle. The plates then were collected and placed in an incubator between 08:00
and 09:00 on September 28, 2016, when field study personnel returned to the site. The settling plate
sampling was conducted to test the hypothesis that using the AA sampling method in a large area will
distribute hot spot contamination throughout the entire area so that the contamination is more likely to be
detected by surface sampling methods.
2.3.5 Stage 5: RFC and Wet Vacuum Sampling, Round 2
After the settling plates were collected during Stage 4, the magnetic boundary markers were replaced and
a second round of RFC and wet vacuum sampling was conducted using the same procedures and
locations as during Stage 1. This second set of RFC and wet vacuum samples was collected to test the
hypothesis that using the AA sampling method in a large area will distribute hot spot contamination
throughout the entire area so that the contamination is more likely to be detected by surface sampling
methods.
17
-------
3 Laboratory Experimental Approach
This section discusses the approach used for the laboratory experiment component of the project. The
laboratory experiment included a DFU filter loading evaluation, NAM prefilter comparison, and forced
aerosol evaluation as discussed below.
3.1 DFU Filter Loading Evaluation
The DFU filter loading evaluation was conducted to gain an understanding of how the DFU sampler flow
rate is affected by dust loading of the filters. The question of how DFU sampler flow is impacted by filter
loading arose during the FAPH field exercise, when DFUs were exposed to very high particulate
concentrations during Stage 3, drastically reducing the flow/ rates during AA sampling.
The laboratory experiment was designed to simulate similar dust loading and to model the approximate
relationship between dust loading on the filter and flow rate. The experiments were conducted inside the
aerosol wind tunnel in laboratory BB005 on the EPA RTP campus. The wind tunnel controlled the
environmental conditions at 20 degrees Celsius (°C) ± 2 °C and relative humidity (RH) at 30% ± 2%
during all testing and contained the dust exhausted from the DFU. Figure 3-1 shows the test setup.
Dry, filtered
compressed
air source
Shutoff val\
Ve
vacuu ilet
DFU sampler
Figure 3-1. DFU Filter Loading Test Setup (Arrows Indicate Air Flow Direction)
18
-------
The dust injection setup was constructed using a Venturi vacuum pump (Model JS-90M, Vaccon
Company, Inc., Medway, MA). This pump uses compressed air and critical flow orifices to create a
vacuum. Compression fittings were used to create a small funnel to hold the dust injected by the Venturi
pump. The dust inlet was connected to a 2-in.-diameter Y-fitting and straight tubing to the inlet of the DFU
sampler. The Y-fitting allowed make-up air to be pulled into the DFU inlet because the dust injection
system can only operate at one pressure and therefore only one flow rate.
The DFU filter cassettes were preloaded with 47-mm polyester felt filters and weighed using an analytical
balance (Model GA200D, Ohaus Corp., Parsippany, NJ). Arizona test dust A2 Fine Grade (Powder
Technology, Inc., Arden Hills, MN) was selected for this experiment because this dust is a good match to
the size distribution data for particles obtained during Stage 3 of the field sampling exercise. After a small
amount of dust was injected into the DFU filter assembly, the DFU inlet velocity was measured, and then
the filters were removed and weighed.
3.2 NAM Prefilter Comparison
Five types of 14-in. by 20-in. furnace filters were tested to compare the relative spore collection of each
filter type under the same controlled circumstances. The chosen filters were all commercially available,
four from Filtrete™ (3M, St. Paul, MN): Filtrete™ Basic Flat Panel, Filtrete™ 1500 Ultra Allergen, Filtrete™
1900 Maximum Allergen, and Filtrete™ 2400 Elite Allergen Extra. The fifth filter was a WEB® Absorber
Electrostatic Carbon Filter (WEB Products Inc., Creola, AL). These filters were chosen specifically either
because the wire support on the filter face and back are easily separated from the pleated filter material
or because the filter had no wire components, allowing filter pieces to be processed using a Stomacher®
400 circulator without puncturing the plastic bag.
Each filter was prepared for testing by removing the protective plastic film, placing the filter on a
disinfected surface, and, if necessary, using disinfected diagonal wire snips to trim the wire support from
both the face and the back of the filter as close to the filter frame as possible. The wire was carefully
peeled off each face and discarded. A 6-in. by 6-in. square in the center of the filter face was then
measured using a disinfected ruler and marked using permanent marker. To test each filter, the filter was
fitted into a 14-in. by 20-in. filter box connected to the inlet of the NAM. A 12-in.-diameter, 30-in.-long duct
with several mixing baffles installed inside was connected to the face of the filter box. Figure 3-2 shows
the test setup.
Sg-containing
MDI
Duct with mixing baffles
Figure 3-2. NAM Prefilter Comparison Test Setup (Arrows Indicate Air Flow Direction)
19
-------
The NAM was switched on and set to high speed. An MDI containing Bg spores (described in Section
2.2.4') was used to dispense spores into the duct. The MDI was weighed, vortexed, placed in a one-piece
medical MDI actuator, and actuated twice into the inlet of the duct. The first actuation seated the MDI in
the actuator without any visible output, and the second actuation dispensed a visible puff of aerosol. The
NAM was turned off 1 minute (min) after the second MDI puff, and the filter was removed, bagged,
labeled, and transferred to the Biolab for processing. The filter was bagged in a 15-in. by 24-in. Twirl'em®
plastic bag (Labplas, Sainte-Julie, QC, Canada).
Five replicate tests were completed for the Filtrete™ 1500 Ultra Allergen, Filtrete™ 1900 Maximum
Allergen, and Filtrete™ 2400 Elite Allergen Extra filters. Two tests were conducted for the Filtrete™ Basic
Flat Panel, and one test was conducted for the WEB® Absorber Electrostatic Carbon Filter. Section
4.1.2.2 discusses the procedures for filter sample processing, and Section 4.2.1 discusses filter sample
analysis.
3.3 Forced Aerosolization Evaluation
The forced aerosolization evaluation tests were designed to evaluate the fraction of available spores that
would be aerosolized from subway surfaces using a leaf blower as described in Section 2.2.1.1 during the
field sampling exercise. Forced aerosolization testing was conducted in an environmentally controlled
chamber in laboratory B155A on the EPA RTP campus. The chamber controlled the environmental
conditions at 20 °C ± 2 °C and the RH at 30% ± 5% during all testing. Aerosolization experiments were
conducted on approximately 12-in.-square trays of ballast rock material inoculated with either Bg or Bacillus
thuringiensis var. kurstaki (Btk) spores. Table 3-1 summarizes the test matrix.
Table 3-1. Test Matrix for Bench-Scale Forced Aerosolization Experiments
Spore
Type
Target
Loading
(CFU/ft2)
Simulated
Decontamination?
Temperature
(°C)
Humidity
(% RH)
Number of
Replicates
Bg
104
No
22
30
5
107
No
22
30
5
107
Yes
22
30
5
Btk
104
No
22
30
5
107
No
22
30
5
107
Yes
22
30
5
Section 4.1.2.3 discusses the procedures for forced aerosolization filter processing, and Section 4.2.1
discusses sample analysis. The following sections discuss the forced aerosolization evaluation
experimental setup, coupon inoculation, and simulated decontamination.
3.3.1 Experimental Setup
A small wind tunnel designed to conduct bench-scale AA sampling experiments was placed inside the
environmentally controlled chamber in laboratory B155A on the EPA RTP campus. The AA sampling wind
tunnel used for testing primarily was constructed of SS and used a blower to pull air through the tunnel.
Figure 3-3 shows the AA sampling wind tunnel for the forced aerosolization evaluation.
20
-------
Figure 3-3. AA Sampling Wind Tunnel for Forced Aerosolization Evaluation
The tunnel was kept at a slightly negative pressure relative to the chamber to minimize spore
contamination of the work space in the chamber. Pressure was monitored using a Magnehelic® series
2000 (Dwyer®, Michigan City, IN) differential pressure gauge. The wind tunnel included an upstream
HEPA-filtered section where the ballast coupon was placed on a mechanical turntable in the center of the
bottom surface of the wind tunnel. Figure 3-4 shows a ballast coupon and the leaf blower nozzle inside
the AA sampling wind tunnel.
Figure 3-4. Leaf Blower Nozzle and Ballast Coupon Inside AA Sampling Wind Tunnel
21
-------
The upstream section of the AA sampling wind tunnel allows placement of the electric leaf blower nozzle
at a 45° angle from the vertical plane and at distances of 0 to 12 in. from the ballast coupon surface. The
forced aerosolization evaluation tests used the same leaf blower used during the field sampling exercise.
The leaf blower nozzle was placed approximately 2 in. from the ballast coupon surface. When the leaf
blower was activated, particles were removed from the coupon and carried downstream to a 14-in. by 20-
in. Filtrete™ 1500 filter connected to a NAM powered on high speed.
As for the NAM prefilter comparison, the wire support was removed from each filter but marked differently
for post-processing. A disinfected ruler and permanent marker were used to mark across the top of the
pleat closest, to the middle of the filter. Then the 1/3 and 2/3 points along each long side were measured
and marked, and the ruler was used to guide the permanent marker across the pleats, creating dotted
lines across the filter face. This marking resulted in six boxes on the filter face as shown in Figure 3-5.
Figure 3-5. Filtrete™ 1500 Filter Marked for Forced Aerosolization Evaluation
The filter was bagged in a 15-in. by 24-in. Twirl'em® plastic bag and labeled. One filter was retained and
processed as a blank.
3.3.2 Coupon Inoculation
Ballast coupons consisting of SS trays filled with rocks were prepared in the same manner as the
coupons for the field sampling exercise described in Section 2.2.4. The coupons were inoculated with two
types of Bacillus spores commonly used as surrogates for Ba, Bg and Btk. Dry powdered Bg spores were
obtained from the same source specified in Section 2.2.4. Like Bg, Btk also is a gram-positive, spore-
forming, rod-shaped bacterium found in soil. Btk produces an endotoxin protein during sporulation that is
commonly used as a pesticide. The bar-coded Btk used for this project is a genetically modified strain
developed by the Edgewood Chemical Biological Center (ECBC; Gunpowder, MD) that allows the spores
to be distinguished from naturally occurring Btk through polymerase chain reaction analysis. The bar-
coded Btk preparation was obtained from DPG.
22
-------
The coupons were inoculated with Bg spores in the same manner discussed in Section 2.2.4. The bar-
coded Btk cells were cultured by 10-L batch fermentation. After sporulation, the spores were concentrated
into a wet pellet, washed three times, and lyophilized. The lyophilized spores were a dry aggregate, not a
loose dry powder.
The Biolab prepared separate solutions of both Bg and Btk spore types. The spore solution required for
spray-dry deposition is a suspension of the spores in a 90% ethanol solution. Table 3-2 summarizes the
ingredients and measurements for preparing the spore solutions for spray-dry deposition.
Table 3-2. Ingredients and Measurements for Spore Solution for Spray-Dry Deposition
Ingredient
Amount Required to Prepare 30 mL Spore Stock
Solution (approx. 1 x 10® CFU/mL)
Lyophilized spores
Approx. 0.05 g
Sterile Dl water
0.85 mL
0.07% Tween® 20 in sterile Dl water
2.13 mL
Ethanol
26.8 mL
The spore and ethanol solution was prepared as summarized below.
1. Measure the required amount of lyophilized spores into a sterile, 1,5-milliliter (mL) microcentrifuge
tube (Thermo Scientific™ 3451, Waltham, MA).
2. Add sterile Dl water to the microcentrifuge tube, and vortex the tube for 1 min.
3. After letting the tube rest at 15 min at ambient temperature, vortex the tube again for 1 min, then
sonicate the tube for 1 min.
4. Repeat Step 3.
5. After 22 to 24 h, add 0.07% Tween®20 in sterile Dl water to a 50-mL conical tube. Transfer the
contents of the microcentrifuge tube (spores and water) to the 50-mL conical tube using a pipette.
Vortex the conical tube for 1 min, and then sonicate it for 1 min.
6. After 48 to 72 h, add 100% sterile ethanol to the conical tube.
7. Complete Steps 1 through 6 ten times.
8. Combine the contents of all 10 conical tubes in a glass jar, to yield 300 mL total.
9. Plate a sample of the final solution to perform a concentration check.
Spore solutions were plated to check the concentration. Solutions were diluted in 90% ethanol as needed
to obtain the target concentrations of 5 * 108 CFU/mL for the high loading level and 5 * 106 CFU/mL for
the low loading level.
Spore deposition for coupon inoculation was performed using the spray-dry deposition system and
method described in Appendix D. Briefly, this procedure involves using an ultrasonic nozzle (model Q060-
2-26-17-303-030, Sono-tek Corp., Milton, NY, USA) and a deposition stack (measuring 8 in. by 8 in. by 18
in. tall) to uniformly deposit dry spores from an alcohol-based suspension onto a surface. The deposition
stacks were designed to deposit onto 7.75-in.-square material coupons. Therefore, modifications were
necessary to accommodate deposition onto the ballast trays. For each deposition stack, a SS ballast tray
was turned upside down, a 7.75-in.-square hole was cut in the bottom of the tray, and a spray-dry
deposition stack was attached to the hole and sealed as shown in Figure 3-6.
23
-------
Figure 3-6. Spray-Dry Deposition Stack Modified for Ballast Coupon Deposition
Using the same spore solutions, 14-in.-square SS positive control coupons also were prepared using the
same equipment and methods as those used for the ballast coupons. Each SS coupon then was sampled
using a sponge stick in accordance with the CDC-published procedure (CDC 2012V Results from the SS
coupon swabs for each spore type and loading level were used as the measure of the number of spores
deposited onto the ballast coupon for aerosolization fraction calculations.
3.3,3 Simulated Decontamination
Simulated decontamination involved spraying an inoculated coupon with Dl water and allowing the ballast
to dry before testing. For simulated decontamination, a sanitized, gravity-fed, high-velocity, low-pressure
spray gun (ATD-6901, ATD tools inc. Wentzville, Missouri) was used to spray 180 mL of filter-sterilized Dl
water on the inoculated ballast tray. The water was applied uniformly to the ballast tray using an
overlapping "S"-pattern (modeled on the CDC sponge stick swabbing protocol [CDC 20121) until the entire
180-mL volume was dispensed. The uncovered tray then was left overnight to equilibrate in the test
chamber at 20 °C ± 2 °C and 30% ± 5% RH. The loading rate of 180 mL/ ft2 was chosen to mimic the
UTR-OTD decontamination loading of 400 gallons of decontaminant released over the approximately
8,900-ft2 site.
24
-------
4 Testing and Measurements
This section discusses the testing and measurement protocols for this project, including sampling
processing, analytical procedures, and flow measurement.
4.1 Sample Processing
All DFU, NAM, RFC, and wet vacuum samples collected during the field sampling exercise were packed
on ice in coolers and transported to the EPA RTP Biolab on September 29, 2016. Samples from the
laboratory experiment tests were delivered to the Biolab at the end of each day that the test was
completed. Spores were extracted from the samples, and the spores in these extracts were assayed by
growth on nutrient agar plates in the Biolab. The extraction processes for each sample type are described
below.
4.1.1 DFU Filter Processing
Each DFU filter was aseptically transferred to a separate, 50-mL conical centrifuge tube for extraction. To
each conical tube, 20 mL of sterile PBST was added, and then the tube was sonicated for 10 min and
vortexed continuously for 2 min. The extracted liquid was analyzed for spores as described in Section
4.2.1.
4.1.2 NAM Prefiiter Processing
The NAM prefiiter samples included samples from the field sampling exercise, the laboratory experiment
for the NAM prefiiter comparison, and the laboratory experiment for the forced aerosolization evaluation
as discussed below.
4.1.2.1 Field Sampling Exercise NAM Prefilters
The Biolab processed each NAM prefiiter collected during the field sampling exercise in a sterile biosafety
cabinet (BSC). The filter was removed from the bag, and the 6-in.-square area marked in the center was
carefully excised using sterile scissors and cut in half. Each filter piece was placed in a 7-in. by 12-in.
Stomacher® 400 circulator bag (BA6141/CLR, Seward Laboratory Systems Inc., Davie, FL), and 90 mL of
sterile PBST was added. Each filter piece was processed with the Stomacher® 400 circulator speed set at
230 rotations per min (rpm) for 2 min. The extracted liquid was analyzed for spores as described in
Section 4.2.1.
4.1.2.2 Laboratory Experiment NAM Prefiiter Comparison
The Biolab processed the filters collected during the laboratory experiment NAM prefiiter comparison
study described in Section 3.2 in a sterile BSC. Filters were cut one at a time, and the BSC was sterilized
between filters. Each filter was removed from the bag, and sterilized scissors were used to cut the 6-in.-
square area marked in the center of the filter. For the Filtrete™ 1500 Ultra Allergen, Filtrete™ 1900
Maximum Allergen, and Filtrete™ 2400 Elite Allergen Extra filters, the 6-in. by 6-in. filter section was cut in
half, and each filter piece was placed in a separate 7-in. by 12-in. Stomacher® circulator bag. For the
WEB® Absorber Electrostatic Carbon Filter and the Filtrete™ Basic Flat Panel filter, each 6-in. by 6-in.
25
-------
filter section was placed in a 7-in. by 12-in. Stomacher® 400 circulator bag. The second Filtrete™ Basic
Flat Panel filter was excised completely from the frame and placed in a 7-in. by 12-in. Stomacher® 400
circulator bag. Table 4-1 summarizes the filter types and how they were processed.
Table 4-1. NAM Prefilter Comparison Testing and Processing
Filter Type
Filter Section Processed
Number of
Replicates
Filtrete™ 1500 Ultra Allergen
Filtrete™ 1900 Maximum Allergen
Two 6-in. by 3-in. sections from center
5
Filtrete™ 2400 Elite Allergen Extra
WEB® Absorber Electrostatic Carbon Filter
6-in. by 6-in. center section
1
Filtrete™ Basic Flat Panel
Filtrete™ Basic Flat Panel
Entire filter
1
Each Stomacher® 400 circulator bag contained 90 mL of sterile PBST and was processed with the
circulator speed set at 230 rpm for 2 min. The extracted liquid was analyzed for spores as described in
Section 4.2.1.
4.1.2.3 Laboratory Experiment Forced Aerosolization Evaluation NAM Prefilters
The Biolab processed and analyzed all filters collected from the forced aerosolization evaluation
laboratory experiments in a sterile BSC. The filter was removed from the bag, and sterilized scissors were
used to cut the filter into six pieces along the marked lines. Each of the six filter sections then was cut in
half. Each of the 12 resulting filter pieces was placed in a separate, 7-in. by 12-in. Stomacher® 400
circulator bag, and 90 mL of sterile PBST was added to each bag. Each filter piece was processed with
the Stomacher® 400 circulator speed set at 260 rpm for 1 min. All 12 samples were then combined in a 1-
L container to make one sample that was sonicated for 10 min and analyzed for spores as described in
Section 4.2.1.
4.1.3 RFC Sample Processing
Each RFC sample was transported to the Biolab in a Stomacher® 400 circulator bag containing the filter
and dust collected from the dust bin. The filter was aseptically removed and placed in a separate
Stomacher® 400 circulator bag, while the dust bin contents remained in the original bag. To each bag,
180 mL of sterile PBST added. The bag then was placed into a 10-in. by 15-in. secondary containment
bag, which was put into an orbital shaker incubator (Model 3525, Barnstead, Melrose Park, IL) and
agitated at 300 rpm for 30 min. The samples were then aseptically combined into a sterile 1-L bottle.
Samples initially were plated in 1-mL aliquots, but background contamination prevented enumeration of
Bg colonies. Therefore, separate procedures were developed for processing the samples from Stage 1
and Stage 5, as discussed below.
Stage 1 samples posed a challenge because of the low levels of spores and large amount of debris. The
procedure developed to enumerate CFU per sample began by shaking the 1-L sample extract bottle, then
withdrawing four20-mL aliquots of "sludge" from the RFC sample extract and filtering each aliquot
through a 70-jjm cell strainer (352350, Thomas Scientific, Swedesboro, NJ) into a sterile conical tube to
reduce the amount of debris. The conical tubes of filtered aliquots then were heat-treated at 80 °C for 20
26
-------
min, vortexirig every 5 min, to reduce background contamination. The samples were allowed to cool to
room temperature and then centrifuged at 5,500 times gravity (x g) for 15 min at 4 °C to pellet the debris.
The supernatant was filter-plated, and the pelleted debris was resuspended in 10 mL of sterile PBST and
spread-plated on TSA plates in 1-rnL aliquots. Figure 4-1 summarizes the Stage 1 RFC sample
processing. The plates were incubated overnight at 35 °C ± 2 °C, and then the CFU were enumerated by
visual inspection.
1 filter per sample
Homogenize by shaking.
Withdraw 20 mL "sludge"
with a large-volume pipette.
10 spread plates
per sample
Withdraw the
clear
supernatant,
measure
volume, and
filter plate.
Weigh pellet/
debris, then
resuspend in
10 mL sterile
PBST Plate 1
mL on each of
10 TSA plates.
45
«0
»i
Let samples
cool to room
temperature,
»!
then centrifuge
j
at 5500 x g for
3 /
15 min at 4 °C
to pellet debris.
Filter the sludge
through a 70-pm cell
strainer into a sterile
conical tube.
1
Heat samples at 80 °C
for20 min, vortexing
every 5 min.
Figure 4-1. Summary of Stage 1 RFC Sample Processing
Stage 5 sample plates showed very high levels of background contamination, making enumeration
impossible. From the RFC sample extract, 10-mL aliquots were transferred into 50-mL sterile conical
tubes. The tubes were heat treated at 80 °C for 20 min, and then 0.01-mL, 0.02-mL, and 0.04-mL aliquots
were plated in triplicate. The plates were incubated at 35 °C ± 2 °C overnight and enumerated by visual
inspection. Contamination was present (large, white colonies) after heat treatment as shown in Figure 4-
2, but estimated Bg counts were performed (small orange colonies).
27
-------
Figure 4-2. Example Plates from Stage 5 RFC Samples
4.1.4 Wet Vacuum Sample Processing
Wet vacuum samples were received in the original wet vacuum containers and aseptically transferred to
sterile 1-gallon containers. A procedure was developed for processing the Stage 1 samples to address
the challenges of a large sample volume containing large amounts of debris and background
contamination while maintaining sensitivity to the low concentration of spores. The Stage 5 wet vacuum
samples were spread-plated on ISA plates in 0.1-rnL, 0.2-mL, and 0.4-mL aliquots. Because of the high
levels of spores in the samples, the CFU were easy to enumerate compared to the Stage 1 samples.
To process the Stage 1 samples, 100 ml. of the homogenized sample "sludge" was withdrawn and
divided evenly into four 50-mL sterile conical tubes. The samples were heat treated at 80 °C for 20 min,
vortexing every 5 min, to reduce background contamination. The samples were allowed to cool to room
temperature, then centrifuged at 5,500 x g for 15 min at 4 °C to pellet the debris. The clear supernatant
was filter-plated, and the pellet of debris was resuspended in 10 ml. of PBST and spread-plated on TSA
plates in 1-mL aliquots. Figure 4-3 summarizes the Stage 1 wet vacuum RFC sample processing. The
plates were incubated overnight at 35 °C ± 2 °C and then the CFU were enumerated by visual inspection
,
c
Wl
L
s i
J L
J
r
raL
L J ^ ^
4 filters per sample
Homogenize by shaking-
Withdraw 100 ml_ "sludge" with
a large-volume pipette and
distribute 25 ml_ into each of
four 50-mL conical tubes.
Approx. 40 spread
plates per sample
Withdraw the
clear
supernatant,
measure
volume, and
fitter plate-
Weigh pellet/
debris, then
resuspend in 10
mL sterile PBST
Plate 1 mL on
each of 10 to 12
TSA plates.
Let samples
cool to room
temperature,
then centrifuge
at 5500 x g for
15 min at 4 °C
to pellet debris.
1
Heat samples at 80 °C
for20 min, vortexing
every 5 min
Figure 4-3. Summary of Stage 1 Wet Vacuum Sample Processing
To confirm that colonies observed among wet vacuum spread-plate background contamination were
actually Bg, several colonies believed to be Bg were streak plated (T-streaked) for isolation onto fresh
TSA plates and compared to the Bg control as shown in Figure 4-4. When the T-streaks were prepared,
28
-------
occasionally cells from adjacent non-Sg colonies were inadvertently picked up in the loop as shown in
Figure 4-4 in the lower middle and lower right plates. The small, orange colonies were counted as Bg,
while the larger, white, doughnut-shaped colonies were not counted.
Bg ATCC 9372
positive control
Negative control
Examples of fresh TSA plates sub-cultured with colonies from
wet vacuum spread plates to confirm presence of Bg.
Figure 4-4. Verification of Bg Colonies in Wet Vacuum Samples
4.2 Analytical Procedures
Analytical procedures for this project included spore analysis and settling plate analysis as discussed
below.
4.2.1 Spore Analysis
Spores were extracted from the samples as described in Section 4.1, and the spores in these extracts
were assayed by growth on nutrient agar plates in the Biolab. The samples were analyzed quantitatively
for the number of viable spores recovered per sample (CFU).
All sponge stick samples were extracted in Stomacher® 400 circulator bags in 90 mL of sterile PBST for 1
min at 260 rpm. The solution then was pipetted into sterile specimen cups and sonicated for 10 min
before spiral plating. Reference tube samples collected during the spray-dry deposition process were
vortexed for 2 min and sonicated for 10 min before spiral plating.
All sample types were plated in triplicate using a spiral plater (Autoplate® spiral plating system, Advanced
Instruments Inc., Norwood, MA), which deposits a known volume of sample in three 10-fold serial
dilutions on each plate. Plates were incubated at 35 ± 2 °C for 16 to 24 h for Bg and 27 ± 2 °C for 16 to 19
h for Btk. During incubation, colonies develop along the lines where the liquid was deposited on the
rotating plate in decreasing amounts from the center to the edge of the rotating plate as shown in Figure
29
-------
4-5). The number of CFU was determined using a QCount® colony counter (Advanced Instruments Inc.,
Norwood, MA).
Figure 4-5. Bacterial Colonies on Spiral-Plated Agar Plate
Samples with unknown concentrations were plated with no dilution and with a 100-fold dilution. Samples
with known low concentrations were plated with no dilution. The QCount® instrument automatically
calculates the CFU/mL in a sample based on the dilution plated and the number of colonies that develop
on the plate. This information was recorded in a spreadsheet.
Only plates meeting the threshold of at least 30 CFU were used for spore recovery estimates. After
quantitation with the QCount® colony counter, samples with plate results below the 30-CFU threshold
were either re-spiral plated with a more concentrated sample aliquot or filter plated to achieve a lower
detection limit. The filter-plate volume was based on the CFU data from the QCount® results. The filters
were placed onto TSA plates and incubated at 35 ± 2 °C for 20 to 24 hours before manual enumeration.
Plates overgrown with indistinguishable colonies were re-spiral plated using a less concentrated aliquot.
When less than 30 CFU per plate were counted for a sample spiral plated with a neat (undiluted) aliquot,
then one of the following two methods was used:
1. Spread plate an undiluted aliquot with a larger volume (0.1, 0.2, and 0.4 mL, each in triplicate)
2. Filter plate an undiluted aliquot with a larger volume (such as 1, 2, or 10 mL)
Filter plating was performed using the Pall MicroFunnel unit with 0.45-|jm GN-6 Metricel white membrane
(P/N 4804, Pall Corporation, Port Washington, NY). The sample aliquot was added to 10-mL of Dl water,
which then was poured over the filter. The vacuum system was opened and the liquid funneled through
the filter, trapping the spores on the filter. The filter then was washed with another 10-mL aliquot of sterile
Dl water, removed from the plastic housing, and placed onto a TSA plate. Plates were incubated at 35 ± 2
°C for 16 to 24 h for Bg and 27 ± 2 °C for 16 to 19 h for Btk before manual enumeration.
30
-------
4.2.2 Settling Plate Analysis
An incubator (Model 1555, VWR, Radnor, PA) was transported to a field trailer used by UTR-OTD
personnel. The settling plates were incubated overnight at 37 °C ± 2 °C and enumerated on site. The
number of CFU observed on half of each plate was counted and recorded. To estimate total growth on
each plate, the recorded value was multiplied by 2. This approach was used to increase the efficiency of
CFU counting, as plates had high CFU counts, and CFU were spatially homogeneous across the plate
surface.
Settling plate counts, in CFU, were used to generate heat maps of spore settling after Stage 3 AA
sampling with hot spot contamination. Settling plate results were interpolated using a Kriging method to
estimate the distribution of contamination in the subway. Kriging is an interpolation technique used to
predict values for locations that lack sample data. Specifically, this method assumes that the distance or
direction between sampling points reflects a spatial correlation that can be used to explain variations in
the surface. This approach has great potential in identifying hotspots and aiding in the understanding of
wind flow patterns and decontamination efficacy for biological incidents.
4.3 Flow Measurement
DFU and NAM flow measurement procedures are discussed below.
4.3.1 DFU Flow Measurement
The flow rate of each DFU was measured in the laboratory as summarized below.
1. A tube with diameter identical to that of the DFU cartridge holder assembly was firmly attached to
the top of the DFU inlet.
2. A thermal anemometer (Series 471, Dwyer, Michigan City, IN) was inserted into the side of the
tube perpendicular to the flow direction.
3. The DFU was turned on, and velocity measurements were manually recorded.
4. The flow rate was calculated by multiplying the velocity by the cross-sectional area of the
sampling tube.
The flow rate of each DFU was measured in the field at FAPH as summarized below.
1. A rotating-vane anemometer (Model DA 410, Pacer Instruments, Keene, NH) was placed on top
of the DFU inlet.
2. The velocity measurement was recorded manually.
3. The flow rate was calculated by multiplying the velocity by the cross-sectional area of the DFU
inlet.
4.3.2 NAM Flow Measurement
When a NAM and prefilter are used in the field for AA sampling, the pressure drop will increase due to
particle loading of the filter, decreasing the sampling flow rate. Although it is not practical to measure the
31
-------
flow rate in the field, the pressure drop of the filter assembly can easily be measured by attaching a
pressure gauge to the filter holder assembly. Generating a pressure drop versus flow rate curve allows
direct correlation of the pressure drop recorded in the field to the flow rate.
The NAM flow rate was measured in the laboratory with a prefilter attached to generate a pressure drop
versus flow rate curve. The pressure drop of the filter assembly was measured by attaching a Magnehelic
gauge (Model 2010, Dwyer, Michigan City, IN) to the filter holder assembly downstream of the prefilter.
The velocity then was measured in the 12-in. round duct downstream of the filter using the Dwyer thermal
anemometer. The prefilter size was reduced incrementally by masking symmetrical portions of the filter
using corrugated plastic cutouts. The pressure drop and duct velocity were recorded for each size
opening. The data were entered into a spreadsheet, the flow rate was calculated by multiplying the
velocity by the duct cross-sectional diameter, and a curve was generated for the filter. The plan was to
record the pressure drop across the NAM prefilter in the field and use the curve to estimate the NAM flow
rate. However, this was not possible due to the failure of the plastic ductwork connecting the NAM to the
subway tunnel.
32
-------
5 Results
This section summarizes and discusses the field sampling exercise and laboratory experiment results.
5.1 Field Sampling Exercise Results
AA sampling was conducted in the mock subway tunnel during Stages 2 and 3 of the field sampling
exercise at FAPH. Composite surface sampling was performed on the subway platform during Stages 1
and 5, including RFC and wet vacuum sampling. Stage 4 of the field sampling exercise was the overnight
settling period when two types of agar plates (100 plates of each type) were placed on the platform and
track areas to assess particle settling after Stage 3 AA sampling. The following sections discuss the
results for each type of field sampling, followed by a discussion of the resources required to conduct the
field sampling exercise.
5.1.1 AA Sampling Results
Samples were collected by nine DFU samplers along the length of the track, one NAM on the north end of
the platform, an APS on a cart on the platform, and several Bioaerosol Button Samplers worn by
personnel (personal samples) and mounted to a cart on the platform (area sample). The following
sections discuss the results for each type of sampler, followed by a summary of AA sampling results.
5.1.1.1 DFU Sampling Results
Analysis of data from the DFU samplers was the primary means of quantifying the amount of viable Bg
spores aerosolized during each AA sampling stage. Table 5-1 summarizes the microbiological results and
operating conditions of each DFU sampler deployed for post-decontamination AA sampling during Stage
2. Table 5-2 summarizes the results for the DFU samplers deployed during Stage 3, simulating an area
with a large amount of contamination concentrated in distinct hot spots. The average air concentration of
spores in CFU per cubic meter (CFU/m3) was calculated by dividing the total CFU recovered from the two
filters by the volume of air sampled (average flow rate multiplied by sampling time). This average
concentration was calculated for each DFU unless there were missing data, and an overall average spore
air concentration was calculated for each stage.
Comparison of the total CFU recovered from the DFU filters and the average calculated air concentration
in CFU/m3 between Stages 2 and 3 shows that the Bg-inoculated hot spots significantly impacted DFU
spore recovery. The Stage 2 post-decontamination AA sampling overall spore air concentration was
estimated at 200 CFU/m3. The overall spore air concentration for Stage 3 AA sampling with hot spot
contamination was estimated at 1.6 * 105 CFU/m3, three orders of magnitude higher than the Stage 2
result. Based on the estimated spore load of the 24 inoculated hot spot coupons of 1.2 * 109 CFU and the
subway tunnel volume of approximately 4,450 m3 (157,000 ft3), the theoretical spore air concentration if
all spores in the hot spots were aerosolized and perfectly mixed in the tunnel would be 2.7 * 105 CFU/m3.
This result could lead one to estimate that the fraction of hot spot spores aerosolized during Stage 3 was
0.61. However, it is not realistic to assume that particles greater than 1 jjm in diameter would be well
mixed in such a large volume. Therefore, 0.61 is an upper limit of the fraction of spores aerosolized from
the hot spots. A more reasonable assumption is that during AA sampling, spores are well mixed within 2
meters (m) of the floor. Then, the effective volume would be approximately 2,000 m3. The theoretical
33
-------
spore air concentration if all spores in the hot spots were aerosolized and perfectly mixed in the lower 2 m
of the subway tunnel would be 6.1 * 105 CFU/m3, resulting in a more reasonable estimate of the fraction
of hot spot spores aerosolized in Stage 3 of 0.27.
Table 5-1. DFU Filter Results from Stage 2 AA Sampling (without Hot Spot Contamination)
Stage 2
DFU
Location*
Filter 1
(CFU/mL)
Filter 2
(CFU/mL)
Filter 1+Filter 2
(CFU/mL)
Total
Spores
(CFU)
Average
Flow Rate
(m3/min)
Sample
Time (min)
Average Air
Concentration
(CFU/m3)
DFU01
84
229
312
1.3 x 104
0.40
167
190
DFU02
194
110
304
1.2 x 104
0.43
164
170
DFU03
184
93
277
1.1 x 104
0.53
163
130
DFU04**
26
53
79
3.2 x 103
0.54
159
37
DFU05
10
6
16
6.6 x 102
0.49
156
8.7
DFU06
70
61
130
5.2 x 103
0.39
153
88
DFU07
174
81
255
1.0 x 104
0.54
150
130
DFU08
165
191
356
1.4 x 104
***
Not available
Not available
DFU09
899
851
1,750
7.0 x 104
0.57
143
860
Overall Average
157 min
200 CFU/m3
Notes:
* Locations listed south to north
**Unplugged for 1.5 min during sampling period
***Abnormality observed on video; total flow during sampling could not be determined
Table 5-2. DFU Filter Results from Stage 3 AA Sampling (with Hot Spot Contamination)
Stage 3
DFU
Location*
Filter 1
(CFU/mL)
Filter 2
(CFU/mL)
Filter 1+Filter 2
(CFU/mL)
Total Spores
(CFU)
Average
Flow Rate
(m3/min)
Sample
Time (min)
Average Air
Concentration
(CFU/m3)
DFU01
1.3 x 103
1.2 x 103
2.5 x 103
1.0 x 105
**
15
Not available
DFU02
5.5 x 104
5.2 x 104
1.1 x 105
4.3 x 10®
0.56
41
1.9 x 105
DFU03
3.7 x 104
4.2 x 104
7.9 x 104
3.1 x 106
0.52
42
1.4 x 105
DFU04
3.2 x 104
6.0 x 104
9.2 x 104
3.7 x 10®
0.50
42
1.8 x 105
DFU05***
3.0 x 104
2.5 x 104
5.6 x 104
2.2 x 10®
0.51
34
1.3 x 105
DFU06
3.1 x 104
4.0 x 104
7.1 x 104
2.8 x 10®
0.48
43
1.4 x 105
DFU07
2.8 x 104
3.3 x 104
6.1 x 104
2.4 x 10®
0.47
43
1.2 x 105
DFU08
5.8 x 104
5.9 x 104
1.2 x 105
4.7 x 10®
0.55
44
1.9 x 105
DFU09
5.2 x 104
4.1 x 104
9.3 x 104
3.7 x 10®
0.37
43
2.3 x 105
Overall Average
42 min
1.6 x105 CFU/m3
Notes:
* Locations listed south to north
**Unit lost power after 15 min; no second flow measurement
***Original DFU failed; inlet swapped to new DFU unit
34
-------
The inlet velocity of each DFU sampler was measured at the beginning and end of the sampling time, and
during Stage 2, an additional measurement was taken for most DFUs. The volumetric flow rate was
calculated by multiplying the inlet velocity measured in the field by the cross-sectional area of the DFU
inlet. Figure 5-1 shows a plot of measured flow rate vs. sampling time. The average of the beginning and
ending flow rates for each DFU (in Tables 5-1 and 5-2) was used to calculate the average air
concentration for each DFU sample.
1000
900
800
~ 700
E
d. 600
<1)
£ 500
&
400
D
U.
Q 300
200
100
0
—~— Stag e 2 —e— Stag e 3
0 20 40 60 80 100 120 140 160
Sampling time (min)
Figure 5-1. Starting and Ending DFU Flow Rates as Function of Sample Time for Stages 2 and 3
5.1.1.2 NAM Prefilter Sampling Results
Table 5-3 presents the limited amount of NAM prefilter data. The plastic duct connecting the subway
tunnel to the NAM collapsed partially during both Stage 2 and Stage 3 sampling. Therefore, the sampling
flow rate and time could not be determined for either stage.
Table 5-3. NAM Prefilter Results from Stage 2 AA Sampling (without Hot Spot Contamination) and
Stage 3 AA Sampling (with Hot Spot Contamination)
Stage
Sampling Time
(min)
Filter Piece 1
(CFU /mL)
Filter Piece 2
(CFU /mL)
Filter 1 + Filter 2
(CFU/mL)
Total Spores
(CFU)
Stage 2
128
364
328
691
1.2 x 105
Stage 3
20
1.3 x 104
2.0 x 104
3.2 x 104
5.8 x 10®
The NAMs used for the UTR-OTD were located on the upper level of the subway station, and a flexible
plastic duct was used to connect the inlet of each NAM to an opening in one of the barriers constructed
on the stairways to seal off the subway tunnel. However, the flexible plastic ducts are designed for use on
35
-------
the outlet of the NAM rather than the inlet and are prone to collapse under even a modest pressure drop
between the duct inlet and the NAM inlet. If the NAM is used for sampling in future work, the NAM unit
should be placed as close as possible to the sampling area, and all connections to the inlet should be
constructed of flexible aluminum ductwork. The Stage 3 NAM prefilter collection results are an order of
magnitude higher than the Stage 2 results. However, without flow rate information, no real comparison
can be made between the results for each stage.
5.1.1.3 APS Sampling Results
Figure 5-2 shows the Stage 2 and 3 results for the APS on a cart on the subway platform plotted as total
particle concentration vs. time. The data are color-coded by activity, and the beginning and end of each
AA sampling round are indicated.
1.E+9
E
§,
c
o
*2 1.E+8
_Q)
y
r
CO
31.E+7
ro
,o
1.E+6
oinomomoinoioomomomomotnoioomoiooinomomomomomomomomomom
OT-cO'5i-c>T-cwoT-cwoT-rt-?j-T-cN^nni-c\i^Mr>T-(\i'snnT-T-csj'tf-ir>T-cN'T-cN'snr>
t-t-t-i-(\I(NCJCNC\I(NOJ(N(\JCN(N04CMCM(NOJ
Time (hh:mm)
Figure 5-2. APS-Measured Total Particle Concentration vs. Time for Stages 2 and 3
APS only DFU on
Placement of settling plates All settling plates deployed
AAS start AAS stop
The APS-measured concentration increased rapidly at the start of each AA sampling round, then
decreased more slowly as particles were filtered out by samplers and ventilation NAMs and settled out of
the air. Figure 5-3 shows the average measured particle size distribution at 20:45 during Stage 2 AA
sampling, with a mass median aerodynamic diameter (MMAD) of 1.72 jjm and geometric standard
deviation (GSD) of 1.9. The average measured particle size distribution at 23:45 during Stage 3 AA
sampling was nearly identical, with an MMAD of 1.68 jjm and GSD of 1.9. The APS does not indicate the
nature of the particulates, so no conclusions can be drawn from these data regarding the number of
viable spores.
36
-------
3000
2500
1 2000
0
s
1 1500
ro
c
a)
o 1000
o
U
500
0
Figure 5-3. APS-Measured Particle Size Distribution at Time 20:45
5.1.1.4 Bioaerosol Button Sampling Results
Table 5-4 summarizes the Bioaerosol Button Sampler data from Stage 2 and Stage 3.
Table 5-4. Bioaerosol Button Sampler Results from Stages 2 and 3
Stage
Task
Spore Count
(CFU)
Air Volume Sampled
(m3)
Spore Air Concentration
(CFU/m3)
Cart
82
1.37
60
2
Leaf blower - platform
15
0.511
29
Leaf blower - tracks north
1
0.508
2
Leaf blower - tracks south
123
0.496
248
Cart
3.8 x 104
2.52
1.5 x 104
3
Leaf blower - platform
2.3 x 103
0.324
7.0 x 103
Leaf blower - tracks north
5.9 x 103
0.453
1.3 x 104
Leaf blower - tracks south
2.1 x 105
0.460
4.6 x 105
There was large variability in the number of spores recovered from samplers worn by the different leaf
blower operators and the sampler mounted on the cart. This variation is common for personal sampling
and not surprising given that the operators were moving around the tunnel while operating the leaf
blowers and wearing Level C personal protective equipment (PPE) (powered air-purifying respirators
[PAPRs], hooded chemical-resistant coveralls, and two layers of chemical-resistant gloves). Even though
the workers were conducting the same general task, the work practices, turbulent air movement, and
potential clothing interference likely led to the high variability.
Table 5-5 summarizes the Bioaerosol Button Sampler data from Stage 5 during Stage 5 RFC and wet
vacuum sampling. These results are comparable to the Stage 2 and 3 results presented in Table 5-4. The
similarity in results demonstrates that most of the spores aerosolized during Stage 3 AA sampling settled
out of the breathing zone during the Stage 4 overnight settling period.
ill
III
Hill
lllllllll -
OncONCDNNCMCDCOCDCDO)CO(DT-COO)^'tNNNOJ(DCO
CNCONNOlrtfflCOfflt^^^^CONtOJCONNCOajCDO)^
uouoc£>r--cooT-cOLncOT--r}-cocNh-cooc»r^r^cDQ^jc^L£-jo;j
ooooo-r^^-^^-^csicsicNcocO'^iriiricbr^cDT-T-T-T-T-
v
Aerodynamic diameter (urn)
37
-------
Table 5-5. Bioaerosol Button Sampler Results from Stage 5
Stage
Task
Spore Count
(CFU)
Air Volume Sampled
(m3)
Concentration Sampled
(CFU/m3)
5
RFC and wet vacuum
29
0.48
60
135
0.48
281
5.1.1.5 Summary of AA Sampling Results for Stages 2 and 3
Table 5-6 summarizes the AA sampling results for Stages 2 and 3. The average spore collection and
calculated air concentration results for each sampling method increased by orders of magnitude from
Stage 2 to Stage 3 due to the presence of hot spot contamination. It is interesting that no Stage 2 post-
decontamination AA sampling results are non-detect for Bg spores. Even after the tunnel was
decontaminated by fogging with bleach, significant numbers of viable Bg spores remained on the
surfaces. This Stage 2 AA sampling result is consistent with the RFC and wet vacuum sampling results in
terms of detecting viable spores. However, there is a possibility that the prepositioned hotspots, even with
multiple covers, may have introduced the spores during Stage 2 AA sampling. The hotspots were secured
with three different layers and introduced right before Stage 2. Leaf blowing directly to the hotspot
containers may have caused resuspension of spores. The results are inconclusive whether the spores
were from the incomplete decontamination or the cross contamination from hotspots.
Table 5-6. Average CFU Collected and Average Calculated Spore Air Concentration for Stages 2
and 3
Total Collection (CFU)
Spore Air Concentration (CFU/m3)
Sampler
Stage 2
Stage 3
Stage 2
Stage 3
Average
SD
Average
SD
Average
SD
Average
SD
DFU*
1.6 x 104
2.2 x 104
3.4 x 10®
8.7 x 105
1.3
1.9
4.0 x 103
8.4 x 102
NAM**
1.2 x 105
5.8 x 10®
Not available
Not available
NAM**
1.2 x 105
5.8 x 10®
Not available
Button***
55
57
6.4 x 104
9.9 x 104
85
110
1.2 x 105
2.2 x 105
* DFU results are summarized from Tables 5-1 and 5-2. DFU02 in Stage 2 and DFU01 in Stage 3 experienced
failures during sampling, and associated results are excluded from this data summary. The averages are for eight
DFU samples per stage.
** The NAM duct collapsed partially during sampling. Therefore, flow rate and sampling time could not be
determined. The averages are for one sample per stage.
*** Button sampler results are summarized from Table 5-4. The averages are for four samples per stage.
5.1.2 RFC and Wet Vacuum Sampling Results
To better understand the operational parameters of the RFCs in the field sampling exercise, operating
times and surface area data were collected. Table 5-7 summarizes the operating times by unit and round.
Table 5-7. RFC Operating Time
Stage
RFC Location
Start Time
End Time
Sample Duration (min)
1
17:16
Not available*
Not available
1
2
17:22
Not available*
Not available
3
17:25
Not available*
Not available
1
19:57
20:19
22
5
2
19:59
20:24
25
3
20:01
20:50
49
38
-------
*End time not recorded because personnel on required rest break and RFCs not in view of CCTV
cameras
A geographic information system (GIS) was used to estimate the RFC travel distance and sampled
surface area. The RFC path was traced by referencing on-site CCTV footage and was projected onto a
two-dimensional diagram representing the subway platform. A buffer approximately 1 ft wide was applied
to the path to determine the sampled surface area. Figure 5-4 shows the start and end location, path, and
sampled surface area for RFC Location 2 during Stage 5. The unit sampled approximately half of the
study area (460 of 1,000 ft2) in 23 min, traveling 8,000 ft. Because of the complexity of this task, only the
path and sampling area of the RFC at Location 2 during Stage 5 were simulated.
• Stilt Lo o> lio ii
• End Location
Robot Pnlli
Sampled Are#
CD Kiosk
Figure 5-4. Stage 5 Sampling Path and Area for RFC Location 2
Recoveries from Stage 1 sampling were non-detect for spread plating but were re-evaluated through
aliquot processing as described in Sections 4.1.3 and 4.1.4. Bg spores were detected from all 6 RFC and
wet vacuum platform samples. The recoveries (Table 5-8) of these sample aliquots were 3 - 4
CFUs per ft2 and 2 - 24 CFUs per ft2 for RFC and wet vacuum samples, respectively. The total coverage
of sampling using RFC and wet vacuum was approximately 1750 ft2 with 6 samples. The traditional
surface sampling using 37 mm cassettes was conducted on the platform prior to the Stage 1 composite
sampling. A total of 49 samples were taken and 3 samples came back positive with 4-6 CUFs per ft2
(EPA 2017). The total sampling coverage using 37 mm cassettes was approximately 49 ft2 with 49
samples. This result shows how new composite sampling methods can improve the detection capability
especially post decontamination sampling and reduce the number of samples, time, and labor compared
to the traditional sampling methods during a wide area response.
The detected spore levels from Stage 1 post-decontamination composite sampling (Table 5-8) were
negligible compared to levels from Stage 5 sampling after AA sampling with hot spot contamination
(Table 5-9). This confirms that Stage 3 AA sampling distributed spores from hotspots throughout the test
area. Spore recoveries from Stage 5 were three orders of magnitude greater than recoveries from post-
decontamination sampling because of the aggressive agitation of the hot spots with leaf blowers and
subsequent distribution of spores throughout the subway tunnel.
39
-------
Table 5-8. RFC and Wet Vacuum Sampling Results from Stage 1 (Post-decontamination)
Stage
Sampling
Technique
Location
Sampled
CFU Count per
Sampled Area
Location
Area (ft2)
Estimated Area
Sampled (ft2)
Sampling
Time (min)
Recovery
(CFU/ft2)
1
1.9 x 103
900
450
Not available*
4
RFC
2
1.4 x 103
900
450
Not available*
3
1
3
1.8 x 103
900
450
Not available*
4
1
3.8 x 103
175
156
8
24
Wet Vacuum
2
2.2 x 102
175
90
5
2
3
3.7 x 103
175
175
11
21
* Run time for Stage 1
RFCs could not be determined.
40
-------
Table 5-9. RFC and Wet Vacuum Sampling Results from Stage 5 (Post-AA Sampling with Hot Spot
Contamination)
Stage
Sampling
Technique
Location
Sampled
CFU Count per
Sampled Area
Location
Area (ft2)
Estimated Area
Sampled (ft2)
Sampling
Time (min)
Recovery
(CFU/ft2)
1
2.1 x 106
900
450
29
4.7 x 103
RFC
2
2.7 x 106
900
450
25
6.0 x 103
5
3
2.9 x 106
900
450
48
6.4 x 103
1
4.2 x 106
175
156
12
2.7 x 104
Wet Vacuum
2
1.5 x 106
175
90
4
1.7 x 104
3
3.2 x 106
175
175
10
1.8 x 104
As Table 5-10 shows, much more debris was collected by the RFCs during Stage 1 than Stage 5, most
likely because the leaf blowers operating during Stages 2 and 3 had swept a large amount of the debris
toward the walls of the platform and onto the track. In addition, vacuum-based devices such as RFCs are
known to cause a small but detectable amount of dust resuspension, presumably due to the presence of
surface agitation brushes, which also may have contributed to the RFCs collecting less debris during
Stage 5.
Table 5-10. Debris Recovered from RFC Samples
Debris Weight
Stage 1
Stage 5
RFC1
RFC2
RFC3
RFC1
RFC2
RFC3
Debris weight* (g)
219.5
129.2
176.5
43.5
81.1
57.6
*Debris was saturated, and some debris may have been lost during sample processing and transfer.
The same pattern of debris collection was not seen with the wet vacuums (Table 5-11).
Table 5-11. Sample Volume and Debris Recovered from Wet Vacuum Samples
Sample Volume and
Stage 1
Stage 5
Debris Weight
WV1
WV2
WV3
WV1
WV2
WV3
Sample volume* (mL)
3,045
2,150
2,587
2,760
1,806
2,258
Debris weight (g)
467
308
319
311
183
332
*Sample was weighed; assumed 1 g = 1 mL.
The amount of debris collected by the wet vacuums was nearly identical for Stages 1 and 5. The wet
vacuum sampling area was at the back of the subway platform, and the leaf blowers may have swept part
of the debris towards the back. Because the wet vacuums used a surfactant liquid to sample the floor,
dust was more easily suppressed and vacuumed compared to the dry-vacuum based RFCs. This
situation is evident in the recovery efficiencies summarized in Tables 5-12 and 5-13, which show that the
wet-vacuum cleaners recovered, on average, 360% more spores than the RFCs.
41
-------
Table 5-12. RFC and Wet Vacuum Sampling Results from Stage 1 (Post-decontamination)
Stage
Sampling
Technique
Location
Sampled
CFU Count per
Sampled Area
Location
Area (ft2)
Estimated Area
Sampled (ft2)
Sampling
Time (min)
Recovery
(CFU/ft2)
1
1.9 x 103
900
450
Not available*
4
RFC
2
1.4 x 103
900
450
Not available*
3
1
3
1.8 x 103
900
450
Not available*
4
1
3.8 x 103
175
156
8
24
Wet Vacuum
2
2.2 x 102
175
90
5
2
3
3.7 x 103
175
175
11
21
* Run time for Stage 1
RFCs could not be determined.
Table 5-13. RFC and Wet Vacuum Sampling Results from Stage 5 (Post-AA Sampling with Hot
Spot Contamination)
Stage
Sampling
Technique
Location
Sampled
CFU Count per
Sampled Area
Location
Area (ft2)
Estimated Area
Sampled (ft2)
Sampling
Time (min)
Recovery
(CFU/ft2)
1
2.1 x 106
900
450
29
4.7 x 103
RFC
2
2.7 x 10®
900
450
25
6.0 x 103
5
3
2.9 x 10®
900
450
48
6.4 x 103
1
4.2 x 10®
175
156
12
2.7 x 104
Wet Vacuum
2
1.5 x 10®
175
90
4
1.7 x 104
3
3.2 x 10®
175
175
10
1.8 x 104
The results for this study show that currently available wet vacuum cleaners and RFCs can systematically
sample large contaminated areas. Two benefits of using wet vacuum cleaners for wide area sampling
instead of the currently used sampling methods include (1) collection of fewer samples because one
sample is generated per deployment and (2) less risk of personnel exposure to Ba spores because
wetting reduces spore aerosolization. In addition to the advantage of wide area sampling, the wet
vacuums have hand tools that can be deployed to sample areas where sampling is difficult, such as
staircases and between furniture and other obstacles. However, for real-world application, wet vacuums
require further evaluation with regard to various surfaces, spore concentrations, and environmental
conditions (such as RH, exposure duration, high amounts of debris with animal remains, background
contamination, etc.)
For the RFCs, the current test method focused only on the sampling mechanism of the individual RFCs
by limiting the sampling surface area. Varying the area cleaning logistics or the algorithms of the RFCs
was not part of this study. However, varying these parameters could increase the collection efficiency of
RFCs for wide area sampling.
5.1.3 Settling Plate Results
Figures 2-15 and 2-16 show the sampling locations of the TSA and Brilliance™ agar settling plates for the
platform and tracks, respectively. There were 52 settling plates of each type on the platform and 48 plates
of each type on the tracks, arranged in a square grid pattern with an approximate spacing of 1 ft between
locations. Separate heat maps of settled spores were generated for each plate type and sampling area
combination using a simple kriging method. Prediction errors (Table 5-14) show a relatively high
confidence in the performance of all four kriging models, evidenced by the root-mean-square
standardized error (ratio of root-mean-square error to average standard error) being close to 1.
42
-------
Table 5-14. Kriging Prediction Error by Agar Plate Type and Location
Location
Plate Type
Average Standard
Error
Root-Mea n-Sq u a re
Root-Mea n-Sq u a re
Standardized
Platform
TSA
85
63
0.7
Brilliance™
91
74
0.8
Ballast
TSA
41
44
1
Brilliance™
53
52
0.9
Figures 5-5 and 5-6 show the separate heat maps generated for the combined surface area (platform and
track) from the TSA and Brilliance™ agar settling plate counts, respectively. A total of 10 data bins were
manually defined, each representing 50 CFU. Although the kriging equations represent the best linear,
unbiased predictor for unsampled locations, the resulting data points are not bound to minimum or
maximum values, which can result in gradients that disagree with the input dataset. However, this is a
widely accepted anomaly associated with kriging.
North
Figure 5-5. Heat Map of Spore Settling Generated from TSA Plate Counts
South
North
& ir &
-------
A one-sample t-test of the difference between collocated agar plates (number of samples = 98) revealed
no statistically significant difference between the TSA and Brilliance™ agar settling plates (p-value of
0.067). They show a pattern of higher spore settling on the platform than the tracks, particularly in the
kiosk area and at the bottom of each staircase. The results also suggest a higher concentration of spores
at the north end of the track (to the right in Figures 5-5 and 5-6).
The settling plate heat map results support the theory that the leaf blowers swept a large amount of
debris towards the back of the platform where the wet vacuum cleaners were deployed. The wet vacuum
cleaners suppress dust more easily than the RFCs by dispensing surfactant on the sampling floor. This
sampling process is similar to the well-established wet-wipe surface sampling method because both
methods use a wetting agent to recover spores. The wet vacuum spore recoveries were comparable to
the extrapolated CFU/ft2 results from the settling plates for the same section of the platform as the results
summarized in Table 5-15. The spore counts from settling plates are lower than the ones from wet
vacuum samples. This might be because the settling plates were distributed a couple of hours after the
AA sampling as seen in Figure 5-2, which the significant portion of resuspended spores might have
settled prior to the settling plate distribution.
Table 5-15. Settling Plate Comparison to Stage 5 Wet Vacuum Recovery.
Location
Settling Plate
Type
Area
(ft2)
Spore Count
(CFU)
Average Recovery
(CFU/ft2)
Notes
TSA
189
1.9 x 10®
1.0 x 104
162
1.6 x 10®
1.0 x 104
Extrapolated point data
WV1
Brilliance™
189
1.9 x 10®
1.0 x 104
162
1.7 x 10®
1.0 x 104
Wet vacuum
156
4.2 x 10®
2.7 x 104
Sampled area estimated
TSA
189
2.0 x 10®
1.0 x 104
162
1.7 x 10®
1.0 x 104
Extrapolated point data
WV2
Brilliance™
189
2.0 x 10®
1.1 x 104
162
1.7 x 10®
1.1 x 104
Wet vacuum
175
3.2 x 10®
1.8 x 104
Sampled area estimated
5.1.4 Resources Required for Field Sampling Exercise
To determine the resources required to conduct composite sampling during the field sampling exercise,
each stage of the field exercise was broken up into activities, and then the time and number of people
required for each activity was estimated. Table 5-16 summarizes the results, along with estimates of the
area or volume sampled as appropriate. Because the flow rate of the NAM could not be determined
during Stages 2 and 3, it is not included in the estimates of sampled volume for AA sampling.
44
-------
Table 5-16. Time and Area Sampled by Field Sampling Exercise Activity
Stage
Description
Activity
Time
(h)
No. of
People
Man-hours
(h)
Estimated
Sampled Area
(ft2)
Estimated
Sampled
Volume* (ft3)
RFC and wet
Setup
3
2
6
1
vacuum sampling,
Sampling
2
2
4
1,771
Round 1
Close-out
0.5
2
1
Initial setup
3
6
18
AA sampling,
Round 1
Sample setup
0.5
4
2
2
Sampling
0.5
6
3
21,000
Extra sampling**
1.5
0
0
Close-out
0.5
4
2
AA sampling,
Round 2
Sample setup
0.5
4
2
3
Sampling
0.5
6
3
5,800
Close-out
0.5
3
1.5
Settling plate
sampling
Setup
0.5
2
1
4
Sampling***
8
0
0
17
Close-out
0.5
2
1
RFC and wet
Setup
1
2
2
5
vacuum sampling,
Sampling
1
2
2
1,771
Round 2
Close-out
1
2
2
Rest breaks
3.5
6
21
Total
72
3,559
27,200
*Sampled volume does not include NAM sampling
"""Samplers running while personnel took a rest break
***Settling plates left out overnight
5.2 Laboratory Experiment Results
This section discusses the DFU filter loading evaluation, NAM prefilter comparison, and forced
aerosolization evaluation results for the laboratory experiments.
5.2.1 DFU Filter Loading Evaluation Results
Measurements of DFU inlet velocity and filter weight from tests with nine different DFU samplers were
entered in a spreadsheet, and the DFU inlet cross-sectional area was used to convert the inlet velocity to
volumetric flow rate. When the change in DFU flow rate was plotted as a function of dust loading, there
appeared to be a linear relationship within the range of the data points. Therefore, regression analysis
was performed using Microsoft Excel. The regression analyses for the individual data sets reported R2
values above 0.90, p-values in the 10-8 range, and significance F-values in the 10 8 range, indicating that
for the individual data sets collected, the regressions were linear over the range of the data. The data
from the nine DFU samplers were combined, and the same regression analysis was performed. Figure 5-
7 shows the results, which provided a slope coefficient value of -428, meaning for every gram of dust
accumulated on the DFU filters, the flow rate reduced by 428 L/min. The R2 value was 0.93, the p-v
alues were in the 10 26 and 1019 range, and the significance F-values in the 1019 range, indicating that the
slope is non-zero and that there is confidence in the calculated regression values.
45
-------
o
5=
200
100
0
DFU flow = -428 x dust load + 782
DFU flow [L/min],
dust load [g]
~
\>
~
~
~ \
~ X
0.5
1.5
Filter loading (g)
Figure 5-7. DFU Filter Dust Load vs. DFU Flow Rate
This relationship between DFU dust load and flow rate can be applied to the field sampling data to
estimate the amount of dust collected by each DFU sampler during each stage of the field sampling
exercise. Table 5-17 summarizes the sampling duration, measured end flow rate of each DFU sampler,
and estimated dust load by stage.
Table 5-17. Estimated Field Sampling Exercise DFU Dust Load
Stage 2
Stage 3
Sampling
Location
Sampling
Duration
(min)
End Flow
Rate (L/min)
Estimated
Dust Load
(g)
Sample
Duration
(min)
End Flow
Rate
(L/min)
Estimated
Dust Load
(g)
DFU 1
167
233
1.3
15
N/A
N/A
DFU 2
164
201
1.4
41
265
1.2
DFU 3
163
176
1.4
42
176
1.4
DFU 4
159
109
1.6
42
165
1.4
DFU 5
156
129
1.5
34
97
1.6
DFU 6
153
154
1.5
43
143
1.5
DFU 7
150
126
1.5
43
165
1.4
DFU 8
Not available
Not available
Not available
44
280
1.2
DFU 9
143
302
1.1
43
176
1.4
Average
157
179
1.4
42
183
1.4
The average end flow rates and estimated dust loads for Stages 2 and 3 were not significantly different.
However, the average sampling duration in Stage 3 was approximately one-quarter the sampling duration
in Stage 2, indicating that site conditions should be considered when determining the duration of AA
sampling. Under very dusty conditions, the samplers may require change-out part way through the AA
sampling to prevent filters from becoming loaded to the point that the samplers do not function properly.
46
-------
5.2.2 NAM Prefilter Comparison Results
Five types of 14-in. by 20-in. furnace filters were tested to compare their relative spore collection
efficiencies. Table 5-18 summarizes the Bg spore collection results for each type of filter.
Table 5-18. NAM Prefilter Bg Spore Collection Results
Filter Type
Total Extracted
(CFU)
Average
(CFU)
SD
(CFU)
Coefficient of
Variation (%)
3.1 x 107
4.2 x 107
Filtrete™ 1500 Ultra Allergen
3.1 x 107
3.1 x 107
7.7 x 10®
25%
3.1 x 107
2.0 x 107
2.1 x 107
2.3 x 107
Filtrete™ 1900 Maximum Allergen
2.3 x 107
2.6 x 107
6.6 x 10®
25%
2.6 x 107
3.7 x 107
3.6 x 107
2.6 x 107
Filtrete™ 2400 Elite Allergen Extra
2.9 x 107
2.7 x 107
5.6 x 10®
21%
2.3 x 107
2.2 x 107
Filtrete™ Basic Flat Panel
1.9 x 10®
9.3 x 10®
Entire 14-in. by 20-in. filter processed
WEB® Absorber Electrostatic Carbon Filter
3.8 x 10®
The Filtrete™ 1500, 1900, and 2400 filters all performed comparably, with a one-way analysis of variance
(ANOVA) p-value of 0.49. The spore recoveries for the Filtrete™ Basic Flat Panel and WEB® Absorber
Electrostatic Carbon Filter were an order of magnitude lower than for the Filtrete™ 1500, 1900, and 2400
filters. Based on these data, Filtrete™ 14-in. by 20-in. furnace filters with a Filtrete™ rating of 1500 or
higher should be used as NAM prefilters for AA sampling.
5.2.3 Forced Aerosolization Results
Laboratory tests were conducted to assess the fraction of spores aerosolized from ballast coupons by a
leaf blower. These tests were conducted separately with Bg and Btk spores for three conditions: high
spore loading (1 * 108 CFU/ft2), low spore loading (1 * 106 CFU/ft2), and high spore loading with
simulated decontamination. Table 5-19 summarizes the results from forced aerosolization tests with Bg
spores. The average fraction of spores aerosolized from ballast under high loading conditions was 0.34,
which is comparable to the estimated 0.27 fraction aerosolized during Stage 3 of the field sampling
exercise. The results from the forced aerosolization tests with (relatively) low spore loading showed
noticeably more variability (43% coefficient of variation [CV]) than the high loading test results (12% CV).
The same trend was observed with the simulated decontamination results (50% CV). As expected, the
application of Dl water to simulate a wet decontamination method decreased the fraction of Bg spores
collected during the forced aerosolization tests. Although the fraction reaerosolized differed for the
47
-------
different conditions, they were all within one order of magnitude. Smaller scale aerosolization tests from
other surface types conducted in the same laboratory (EPA 2012b') have shown aerosolization fraction
results that spanned several orders of magnitude.
Table 5-19. Forced Aerosolization Results for Bg-lnoculated Ballast Coupons
Test
Recovered
(CFU)
Recovered
- Background
(CFU)
Average
(CFU)
SD
(CFU)
cv
(%)
Amount
Deposited
(CFU)
Fraction
Aerosolized
5.8 x 107
5.8 x 107
5.4 x 107
5.4 x 107
High Loading
5.1 x 107
5.1 x 107
5.9 x 107
7.4 x 10®
12%
1.8 x 108
0.34
6.6 x 107
6.6 x 107
6.8 x 107
6.7 x 107
8.2 x 105
8.1 x 105
8.2 x 105
8.0 x 105
Low Loading
1.8 x 10®
1.8 x 10®
1.0 x 10®
4.4 x 105
43%
1.6 x 10®
0.65
7.7 x 105
7.6 x 105
9.0 x 105
8.9 x 105
4.8 x 10®
4.8 x 10®
Simulated
Decontamination
1.1 x 107
1.1 x 107
1.5 x 107
1.5 x 107
1.2 x 107
6.1 x 106
50%
1.8 x 108
0.07
o
X
c\i
o
X
c\i
1.0 x 107
1.0 x 107
Table 5-20 summarizes the results from forced aerosolization tests with Btk spores.
Table 5-20. Forced Aerosolization Results for Btfr-lnoculated Ballast Coupons
Test
Recovered
(CFU)
Recovered
- Background
(CFU)
Average
(CFU)
SD (CFU)
CV
(%)
Amount
Deposited
(CFU)
Fraction
Aerosolized
O
X
CO
3.0 x 107
High Loading
4.7 x 107
4.7 x 107
2.9 x 107
1.4 x 107
48%
1.0 x 108
0.28
1.4 x 107
1.3 x 107
2.6 x 107
2.6 x 107
8.7 x 105
8.5 x 105
9.5 x 105
9.3 x 105
Low Loading*
8.8 x 105
8.6 x 105
8.8 x 105
5.0 x 104
6%
N/A*
N/A*
9.6 x 105
9.4 x 105
8.5 x 105
8.3 x 105
2.3 x 107
2.3 x 107
Simulated
Decontamination
2.7 x 107
2.7 x 107
2.4 x 107
2.4 x 107
o
X
CO
8.6 x 106
28%
1.0 x 108
0.30
3.6 x 107
3.6 x 107
4.3 x 107
4.3 x 107
The results from tests with high loading of Btk spores averaged 0.28 fraction aerosolized, which is not
significantly different from the Bg result of 0.34 fraction aerosolized (unpaired t-test p-value of 0.40). The
result from the aerosolization tests with Btk and simulated decontamination of 0.30 fraction aerosolized
48
-------
was not significantly different from the Btk high spore loading results (unpaired t-test p-value of 0.86).
Table 5-20 also presents the results from the low Btk spore loading tests. However, the amount deposited
is not available. The SS positive control coupon swab results were one order of magnitude lower than
expected from the positive control samples of the inoculum, although it is not known at this time what
went wrong with these tests. The resulting reaerosolized fraction calculated from the low Btk spore
loading data was an impossibly high 3.6.
49
-------
6 Quality Assurance and Quality Control
The following sections discuss quality assurance (QA) and quality control (QC) for the project, including
project documentation, the integrity of samples and supplies, instrument calibrations, critical
measurements, and NHSRC Biolab quality checks.
6.1 Project Documentation
This project was performed under two separate Category B quality assurance project plans approved in
August and September 2016. All test activities were documented through narratives in laboratory
notebooks and using digital video and photographs. All tests were conducted in accordance with
established operating procedures to ensure repeatability and adherence to the data quality validation
criteria set for this project.
6.2 Integrity of Samples and Supplies
Samples were carefully maintained and preserved to ensure their integrity. Samples were stored away
from standards or other samples that could cause cross-contamination. Supplies and consumables were
acquired from reputable sources and were National Institute of Standards and Technology (NIST)-
traceable when possible. Supplies and consumables were examined for evidence of tampering or
damage before use. Supplies and consumables showing evidence of tampering or damage were
discarded.
6.3 Instrument Calibrations
The project used established operating procedures for the maintenance and calibration of all laboratory
equipment. All laboratory measurement devices used in this project were certified as having been
recently calibrated or were calibrated by the on-site EPA Metrology Laboratory at the time of use. Table
6-1 summarizes the calibration frequency for instruments used during this project.
Table 6-1. Instrument Calibration Methods and Frequencies
Equipment
Calibration or Certification Method and Frequency
Expected
Tolerance
Thermometer
Compare to independent NIST thermometer (a thermometer
recertified annually by either NIST or an International Organization for
Standardization (ISO)-17025 facility) value once per quarter
±1 °C
Temperature sensor
(chamber)
Compare to independent NIST thermometer value once per year.
±1 °C
RH sensor (chamber)
Compare to calibration salts once per year
±5%
Thermal anemometer
Compare to NIST-traceable anemometer once per year
±5%
Stopwatch
Compare to official U.S. time at time.aov everv 30 davs
± 1 min/30 days
Micropipettes
Certified as calibrated at time of use; recalibrated by gravimetric
evaluation of performance to manufacturer's specifications every year
±5%
Scale
Calibrate annually to Class 1 weights; compare reading to Class 2
weights every day
± 1 %
Any deficiencies were noted and the instrument replaced to meet calibration tolerances.
50
-------
6.4 Critical Measurements
The following measurements were deemed critical to accomplish project objectives:
• Volume
• Counts of CFU per plate
• Plated volume
• Temperature of incubation chamber
Data quality indicators (DQIs) were used to determine if the collected data met the QA objectives.
Decisions to accept or reject test results were based on engineering judgment used to assess the likely
impact of the failed criterion on conclusions drawn from the data. The acceptance criteria were set at the
most stringent levels routinely achievable. Table 6-2 lists the DQIs and acceptance criteria for the critical
measurements.
Table 6-2. DQIs and Acceptance Criteria for Critical Measurements
Measurement Parameter
Analysis Method
Accuracy
Acceptance
Criterion
Mean Value
Pass/Fail
Volume
Serological pipette tips
0.1 mL
± 10% of target value
Pass
Counts of CFU per plate
QCount®
1.82 x 104< QC
Plate < 2.3 x104
Within range of QC
plate
Pass
Plated volume (liquid)
Pipette
2%
± 1%
Pass
Temperature of incubation
chamber
NIST-traceable
thermometer (daily)
O
o
CM
+l
O
o
CM
+l
Pass
Results for all the DQIs were within the target acceptance criteria set for this project.
Several QC checks were used for measurement instruments to ensure that the data collected met the
criteria listed in Table 6-2. The integrity of the samples during collection and analysis was evaluated.
Validated operating procedures conducted by qualified, trained, and experienced personnel ensured data
collection consistency. When necessary, knowledgeable parties conducted training sessions, and in-
house practice runs were conducted to gain expertise and proficiency before research began. The QC
checks performed during this project are detailed in Section 6.5.
In addition to the measurement instrument checks, positive control samples and procedural blanks were
included along with the test samples so that optimal spore recovery and unintentional contamination of
test coupons could be assessed. Replicate coupons were included for each set of test conditions to
assess the variability of each test procedure.
6.5 NHSRC Biolab Quality Checks
Quantitative standards do not exist for biological agents. An Advanced Instruments QCount® system was
used to count viable spores. CFU counts greater than 300 or less than 30 were considered outside the
targeted range. If the CFU count for bacterial growth did not fall within the target range, the sample was
51
-------
either re-spiral plated at a different dilution, filter plated, or manually replated. Filter plate counts and
manual replate counts were enumerated manually.
A QC plate was analyzed before each batch of plates was enumerated using the QCount®, and the result
was verified to be within the range indicated on the back of the QC plate. As the plates were being
counted, a visual inspection of colony counts made by the QCount® software was performed. Obvious
count errors made by the software were corrected by adjusting the settings (such as colony size, light,
and field of view) and recounting or by manually removing or adding colonies as needed.
The acceptance criteria for the critical CFU counts were set at the most stringent level routinely
achievable. Positive controls were included along with the test samples so that spore recovery from the
different surface types could be assessed. Background checks also were included as part of the standard
protocol to check for unanticipated contamination. Replicate coupons were included for each set of test
conditions to characterize the variability of the test procedures.
Further QC samples were collected and analyzed to check the ability of the NHSRC Biolab to culture the
test organism as well as to demonstrate that materials used in this effort did not themselves contain
spores. The checks included the following:
• Field blank samples: Filters and liquid samples transported to the field site but not used for
sampling
• Procedural blank samples: filter samples collected in the same fashion as test samples but
without a contaminated test coupon in place (laboratory tests only)
• SS positive control coupons: Coupons inoculated in tandem with the test coupons and meant
to demonstrate the highest level of contamination recoverable from a particular inoculation event.
Table 6-3 summarizes the additional QC checks for NHSRC Biolab procedures. These checks provide
assurances against cross-contamination and other biases in microbiological samples.
Table 6-3. Additional Quality Checks for Biological Measurements
Blank TSA sterility
control: plate incubated
but not inoculated
Each plate
No observed growth after
incubation
Controls for sterility of
plates
All plates incubated before
use, so contaminated
plates discarded before use
Replicate plating of
diluted microbiological
samples
Each sample
Reportable CFU of
triplicate plates must be
within 100%
Used to determine the
precision of the replicate
plating
Replate sample
Reportable CFU between
30 and 300 CFU per plate
Unexposed field blank
samples
One per test
Non-detect
Level of contamination
present during sampling
Clean up environment;
sterilize sampling materials
before use
The results from all QC sterility control and blank samples were non-detect for this project.
52
-------
7 Composite Sampling Method Conclusions and
Recommendations
The current field and laboratory study results demonstrated the field application of innovative composite
sampling methods using RFC, AA, and Wet Vacuum. All 6 RFC (3) and wet vacuum (3) samples from
Stage 1 detected contamination on the platform after decontamination. However, the traditional sampling
results using 37 mm cassettes on the platform showed 3 positives out of 49 samples. In addition, the
recoveries (CFU per ft2) of the innovative sampling methods were similar (with RFC) or better (~5 times
more sensitive with wet vacuum) compared to the traditional sampling methods. The estimated total
sampled area using 3 RFC and 3 wet vacuum for each stage (6 total samples in each of Stages 1 and 5)
was approximately 1771 ft2 out of approximately 3000 ft2 total platform area (~59%). The traditional
surface sampling method (37 mm cassettes) covered approximately 49 ft2 with 49 samples (~1.6%). The
total estimated person minutes to collect samples (preparation and collection) were 36 for 3 RFC samples
(robot running time was ~30 minutes per robot), 84 minutes for 3 wet vacuum samples and 970 minutes
for 49 cassette samples. The estimated costs per sample (labor, material, and waste) were $267 for RFC,
$220 for Wet Vacuum, and $395 for 37 mm cassette. The estimated total costs were $19,355 ($395 per
ft2) for the traditional method and $1,461 ($0.82 per ft2) for RFC and Wet Vacuum together. For AA
sampling, the estimated total sampling time (preparation and collection) was approximately 25 hours for
Stage 2 and 6.5 hours for Stage 3. The AA sampling required extensive set up time for air filtration, air
mixing, and leaf blowing. In addition, the planning team spent most of time to plan out the AA sampling
including safety, electricity requirement, NAM pre-filter preparation, etc.
The RFC and wet vacuum results clearly showed that these composite sampling methods provide the
benefits of reduced sampling time during a response, fewer samples requiring processing, detection of
spore presence at unknown hot spots of contamination, improved detection of widespread contamination
when concentrations are close to (or potentially below) detection limits for traditional surface sampling
methods, and shortened timeline to recovery. AA sampling results could not be compared to the surface
sampling methods (both innovative and traditional methods) due to the uncertainty of spore contamination
from pre-deployed hotspots during the Stage 2 AA sampling operation.
The laboratory and field study results confirmed the following cautions for using these methods:
1. The tested composite sampling methods generated a large quantity of debris/dust in the sampling
media (water and filter). It is necessary to develop efficient sample processing procedures prior to
analysis either at the site or at the laboratory.
2. Wet Vacuum sampling generates aqueous samples that necessitate a more secured approach to
contain the samples during transport.
3. For RFC, it is difficult to assess the actual sampled area due to the unpredictable movement of
robots. RFC may prematurely terminate sampling due to high filter pressure drop on dusty
surfaces. It is recommended that magnetic strips be used to pre-define the discrete sampling
area (~100-200 ft2, dependent upon the amount of floor debris present).
53
-------
4. The AA sampling method for dusty environments requires airtight containment of the site and
improvements on the particle collection system such as the use of NAM pre-filter and/or cyclone
before filter sampling for dusty environments.
5. AA sampling method has high electricity requirements for operation due to the use of air
samplers, leaf blowers, and mixing fans. Careful site assessment will be necessary to determine
whether AA sampling is a viable option for the given site. Battery powered blowers and collectors
should be identified and evaluated to ease response operations.
6. Efficacy of AA sampling method may be impacted by the site conditions after the initial release
such as high humidity, decontamination, precipitation, etc., which may decrease the spore
resuspension potential and the overall AA sampling efficacy.
In summary, RFC and wet vacuum are likely useful composite sampling methods in addition to the
traditional discrete surface sampling methods. The AA sampling method will need thorough site
assessment for application and the current approach from asbestos abatement may need modifications to
be applied for anthrax site sampling. The AA sampling plan should be developed depending on the site
situations to properly address safety, containment, and effective sampling. Hence it is highly
recommended that AA sampling be planned and executed at the sites with thorough planning
incorporating input from a group of experts from industrial hygiene, aerosol science, and mechanical
engineering.
The following section provides composite sampling method recommendations based on field test operator
and observer comments during the field sampling exercise and recommended AA sampling procedures.
7.1 Field Test Operator and Observer Comments
The following sections provide input and suggestions for further improvement of field methods made by
operators and observers involved in AA, RFC, wet vacuum, and settling plate sampling during the field
sampling exercise.
7.1.1 PPE
• Operators preferred wearing PAPRs instead of full-face respirators.
• Operators preferred wearing Tyvek® suits instead of Tychem® suits.
• Because of the amount of physical activity and subsequent perspiration, outer gloves were
slipping away from the suit-glove interface taped at the wrist. An improved taping method or
waterproof adhesive may be needed.
• Because of the amount of physical activity and subsequent perspiration, operators should bring a
dry change of clothes to change into during each rest period.
• Operators recommended wearing ice vests underneath Tyvek suits.
• Glove change-outs were inconsistent. A system should be used to remind operators when to
change gloves.
54
-------
7.1.2
Field Supplies
• Operators required backup PAPR batteries.
• Containers should be packed with a single type of item (for example, separate storage for gloves,
data sheets, and plastic bags). Although all containers were labeled, access was not as easy as
anticipated.
• More or larger carts should be used because the few on-site carts were overfilled and cluttered.
Another option is a utility cart with more built-in storage.
• Clear containers should be used for easy identification of supplies.
• Data sheets and clipboards became contaminated because they were used in the hot zone.
Instead, each data sheet could be placed in a bag, and the bag could be decontaminated.
• During future tests, laminated sheets and permanent markers for note-taking could be used as
well as electronic tablets for data entry.
• Pictures of the completed data sheets should be taken for backup.
• In future tests, a portable refrigerator could be provided to store samples.
• Operators suggested wiping down all potentially contaminated equipment before it is brought on
site.
7.1.3 AA Sampling Procedure
• Operators suggested shoulder straps for the leaf blowers.
• Operators mentioned trip hazards from long cords and suggested retractable cord reels.
• A planned deployment of all extension cords was suggested to minimize tangles and overlaps in
the space.
• Hand signals should be devised because it was not possible to hear others when the leaf blowers
were operating and because of the need for hearing protection.
• Operators suggested over-the-ear hearing protection worn over the PAPR hood.
• A guide mounted to the end of the leaf blower nozzle was suggested to ensure the appropriate
distance from the surface.
• A time keeper or visible timer could be used to allow leaf blower operators to manage time more
effectively.
• Portable, smaller, lighter equipment was suggested, including lithium battery-powered fans and
blowers and a portable battery-powered wet/dry vacuum as an alternative to the DFU and NAM.
• Operators commented that the NAM filters required larger sample bags and that the wire backing
had sharp edges.
• Operators commented that the bags used to store DFU filters were too large and reaching into
the bags could create cross-contamination.
• Observers and operators recognized the need to constantly monitor the flow rates of the
samplers to know when the filters were fully loaded.
• The NAM could be placed in the hot zone and the filter directly mounted to the NAM to prevent
collapse of the flexible duct. Alternatively, only a metal duct should be used.
• All flanges for the NAMs should be mechanically secured. For this project, the flanges were
secured using tape.
55
-------
• Thicker, larger, or more layers of bags should be used for filter containment because some bags
tore during transport.
• Operators suggested eye protection under PAPR hoods during active leaf blowing activities.
• Observers suggested investigating the possibility of using the leaf blower intake as a sampler in
future tests by employing a filter or cyclone sampler.
• A two-handed, double leaf blower operation would shorten the application duration.
7.1.4 RFC Sampling Procedure
• Future work should use RFCs powered by lithium batteries.
• Larger Stomacher® 400 circulator bags were suggested for collection of the vacuum debris.
• Operators indicated difficulty emptying the dust bin into the Stomacher® 400 circulator bag and
suggested placing the entire bin in the sample bag.
• Use of the filter as the sample was suggested, although it is not known if the filter would provide a
representative sample of the dust collected.
7.1.5 Wet Vacuum Sampling Procedure
• Operators suggested transferring the wet vacuum liquid sample to a chemically resistant, airtight
Nalgene bucket in the field.
• Large Twirl'em® bags should be used instead of zip lock bags as secondary containment for the
dirty reservoirs because the zip lock bags had a tendency to tear when opened.
• A third person may be needed to take notes because the support person was occupied at all
times in assisting the lead sampler.
• A third person may be needed for cord management with the wet vacuum because the cord
drags on the contaminated surface and likely contaminates the support person. A third person,
designated note taker, and clean handler would be beneficial.
• Observers and operators questioned the need for following the wet vacuum procedure of
overlapping strokes and suggested that the effectiveness of this method be studied in the
laboratory.
• A portable, battery-powered wet/dry vacuum was suggested as an alternative to the wet vacuum.
7.1.6 Settling Plate Sampling Procedure
• Operators suggested using more samplers and locating samples in a grid-like pattern if heat
mapping is desired.
• Operators suggested an adequate number of personnel available to count plates.
• A large amount of waste was generated on site. Operators suggested consideration of disposal
requirements and Department of Transportation regulations.
7.2 Recommended AA Sampling Procedures
This section discusses future AA sampling recommendations and a recommended AA sampling
deployment based on the setup of the field sampling exercise. The recommended AA sampling
56
-------
procedures can inform decisions related to AA sampling in a space comparable to the FAPH mock
subway tunnel.
7.2.1 Future AA Sampling Recommendations
• The area designated for AA sampling should be airtight. Plastic sheeting should cover any
opening to outside of the contaminated area. Pass-throughs with 12-in. flanges should be
installed on the sheeting to allow NAMs to be installed inside the hot zone. The flanges allow
venting of the NAM outside of the contained area.
• Determine the total surface area to be sampled to decide the number of operators, leaf blowers,
mixing fans, and samplers required.
• Determine if electricity is available, the load capacity of circuits, and locations of the receptacles.
If electrical load is an issue, determine the best ways to reduce load (such as using fewer
samplers, fewer fans, or more battery-powered equipment [with backup batteries]).
• Limit the sampling area to 2,500 ft2 per operator or less.
• Determine if NAMs can be deployed inside the sampling area. If so, they should be used as high-
volume samplers. The installed HEPA filter can be used as the filter medium. However, the
installed HEPA filter may be a deep-box filter that would require some disassembly of the NAM to
replace. Additionally, HEPA filters are cumbersome to store and process. The recommended
procedure for using a NAM as a sampler is outlined below. This method allows quick change-out
of the filter and does not require cycling power on the NAM.
Acquire a piece of sheet metal the size of the NAM intake, a 12-in. diameter flange, 12-in.-
diameter flexible metal duct, a 14-in. by 20-in. metal filter box, and 14-in. by 20-in.
household filters with easily-removable support wire (such as Filtrete™ MPR 1500 or
higher).
Cut a hole in the sheet metal, and fasten the flange to the hole. Insert the sheet metal with
the flange onto the face of the NAM using the slots on the NAM.
Connect flexible metal duct to the flange. Position the open end near the location that will be
sampled.
Connect the 14-in. by 20-in. metal filter box to the open end of the duct.
If needed, carefully cut away and remove the metal support mesh from the furnace filter.
Install the furnace filter into the filter housing.
• If the volume of the contaminated zone is large, the reaerosolized particles (especially large size)
may not be suspended long enough to be collected by the centralized air filtration system. DFUs
or other portable samplers may be beneficial to sample the localized resuspended particles
before particle loss due to the gravitation settling. The samplers should be spaced at regular
intervals or in a grid throughout the zone. The goal should be to maximize the volume of air
sampled within the AA sampling period. During the field sampling exercise for this project, the
combination of one NAM and nine DFUs yielded approximately 2,100 CFM of air sampled. Over
the 160,000-ft3 space, 76 minutes would be required to sample the entire contaminated zone
volume (one air exchange). Use of more NAMs and DFUs will decrease the time required for AA
sampling and increase the amount of air sampled.
57
-------
• During the field sampling exercise for this project, passing the leaf blowers close to the surface
was effective at removing much of the particulate from the surface in one or two passes. Atypical
leaf blowing motion may be sufficient to remove particulate in the path of the blower. However,
known or suspected hotspot areas in the contained area should receive additional blowing time
during AA sampling.
• Large mixing fans should be used. Seven were used during the field sampling exercise, but fewer
may be acceptable in a smaller space.
• Supply one corded leaf blower per operator. Backup blowers should be on hand in case of failure.
• Supply heavy-duty power cords long enough to traverse the hot zone.
• Check flow rates frequently, and change out filters during AA sampling, especially in dusty
environments. The furnace filters used for this project had a maximum rated dust load of 15 g,
and the DFU filters were rated for 1 g of dust. In extremely dusty environments, filters may require
change out before the end of AA sampling.
• In dusty environments, the addition of a cyclone separator upstream of the filter assembly may be
useful to increase the life of the filter. A cyclone designed to operate efficiently at the flow rate of
the sampler can be plumbed upstream of the sample filter. For example, the commercially-
available Dust Deputy (Oneida Air Systems, Syracuse, NY) cyclone would be suitable for use
with the DFU, and the Super Dust Deputy (Oneida Air Systems, Syracuse, NY) cyclone would be
suitable for use with the NAM.
.2 Recommended AA Sampling Deployment
• Set up mixing fans in the corners of the hot zone.
• Install the NAM sampling filter at or near the middle of the hot zone.
• Install DFU and other samplers in a gridded pattern, if possible.
• Decide how to subdivide the area and the route for sweeping the entire area.
• Run power cords and extension cords to the mixing fans, DFUs and other samplers, and leaf
blowers.
• Turn on the fans.
• Load the filters for the DFUs and other samplers.
• Turn on the DFUs and other samplers.
• Load the NAM furnace filter. If the NAM is not already running, turn it on.
• Turn on the leaf blowers.
• Hold the leaf blower at or near 45° to the surface as close to the surface as possible without
touching it.
• Progress through the zone, making sweeping motions that cover a 4-ft-wide path.
• When the area has been covered by AA sampling, turn off the leaf blower, the DFU samplers,
and the fans.
-------
References
Brown, G. S., ; R.G. Betty, J.E. Brockmann,D.A. Lucero, C.A. Souza, K.S. Walsh, R.M. Boucher,M.
Tezak, M.C. Wilson, and T. Rudolph. 2007. "Evaluation of a wipe surface sample method for
collection of Bacillus spores from nonporous surfaces." Appl Environ Microbiol. 73 (3), 706
through 710.
Calfee, M.W.; S.D. Lee, and S.P. Ryan. 2013. "A rapid and repeatable method to deposit bioaerosols on
material surfaces." J Microbiol Meth. 92 (3), 375 through 380.
Centers for Disease Control and Prevention (CDC). 2012. "Surface sampling procedures for Bacillus
anthracis spores from smooth, non-porous surfaces." Accessed on March 3, 2017. On-line
Address: https://www.cdc.aov/niosh/topics/emres/surface-samplina-bacillus-anthracis.html
Lee, S.D; M.W. Calfee, L. Mickelsen, S. Wolfe, J. Griffin. M. Clayton, N. Griffin-Gatchalian, and A. Touati.
2013. "Evaluation of surface sampling for Bacillus spores using commercially available cleaning
robots." Environ. Sci. Technol. 47 (6), 2595-260, D0l:10.1021/es4000356.
EPA. 2012a. Homeland Security: Strategic Research Action Plan 2012-2016. EPA 601/R-12/008. Office
of Research and Development. Washington, DC.
EPA. 2012b. "Determination of the Difference in Reaerosolization of Spores off Outdoor Materials."
Technical Report, EPA/600/R-14/259. Research Triangle Park, NC.
EPA. 2017."Underground Transport Restoration (UTR) Operational Technology Demonstration (OTD)."
Technical Report, EPA/600/R-17/272. Research Triangle Park, NC.
59
-------
Appendix A: Sampling Procedure Using Commercially Available
Robotic Floor Cleaners for Bacillus anthracis Spores - Neato® XV-21
July 2017
Revision 0.0
National Homeland Security Research Center
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
A-1
-------
A1. Scope and Applicability
The U.S. Environmental Protection Agency's (EPA's) Homeland Security Research Program within
EPA's Office of Research and Development, and the Chemical, Biological, Radiological and Nuclear
Consequence Management Advisory Division within EPA's Office of Land and Emergency
Management, jointly developed this sampling procedure. This procedure is intended to provide a
method for trained incident responders to collect environmental samples after a biological contamination
incident. This procedure specifically applies to the collection of surface-bound particulates and
microorganisms using off-the-shelf robotic floor cleaners (RFC). The purpose of this procedure is to
guide the process of preparation, deployment, and collection using RFCs for sampling surfaces in a
specified area. The results from the collected samples can be used to determine the presence or
absence of contamination and the contamination level after natural outbreaks and after intentional or
accidental releases of pathogenic microorganisms and biotoxins.
At the time of publication, this sampling procedure has not been validated. At the date of this
publication, the RFC sampling procedure has been partially characterized for deployment feasibility and
collection performance for bacterial spores. This procedure will be updated or replaced with a fully
characterized and validated procedure upon availability. During emergencies, the use of non-validated
methods may be warranted when validated methods are not available. EPA's use of non-validated
methods must adhere to the EPA's Forum on Environmental Measurement (FEM) policy directive on
method validation (EPA 2010). Further information on method validation is presented in Validation of
U.S. Environmental Protection Agency Environmental Sampling Techniques that Support the Detection
and Recovery of Microorganisms (EPA 2012).
A2. Summary
This sampling procedure is for the sampling of a horizontal surface (such as a floor) using an RFC. After
sampling, the RFC is recovered and processed to determine the presence or absence of potential
surface contamination. This procedure provides a step-by-step sampling procedure for the following
RFC:
• Neato® XV-21 (Neato Robotics, Inc., Newark, CA): vacuum cleaner used to sample both
porous surfaces (such as carpet, wood, and bare concrete) and nonporous surfaces
(such as vinyl, tile, laminate, coated wood, and coated concrete)
The sections below discuss the following:
• Definitions (Section 3)
• Health and safety (Section 4)
• Waste management (Section 5)
• Equipment and supplies (Section 6)
• Deployment procedure (Section 7)
• Sample collection (Section 8)
• Post-deployment sample handling (Section 9)
• Documentation (Section 10)
Section 11 lists the references used to prepare this procedure.
A3. Definitions
Biological agent contamination: contamination that can be attributed to natural outbreaks, and
intentional or accidental releases, of pathogenic microorganisms and biotoxins.
Biotoxin: a poisonous substance either produced by or extracted from living or dead organisms
Conventional sampling method: a currently recommended surface sampling method (such as the use
A-1
-------
of swabs, wipes, and vacuums fitted with filter-type collection media for biological agents) typically used
on small, discrete areas; some conventional methods (such as the use of sponge wipes, swabs, and 37-
millimeter [mm] vacuum cassettes) have associated multi-laboratory-verified or -validated analytical
procedures
Enclosed facilities and objects: facilities and objects that typically have surface areas with clearly
defined boundaries such as walls and that are isolated from exterior environments, including
commercial and residential buildings and transportation vehicles
Method of dissemination: the means by which biological agents are dispersed; dispersal can occur
over large areas in wet and dry forms through aerosol generation and spreading devices that
contaminate indoor and outdoor sites or the food or supply chain
Outdoor areas and objects: building exteriors and other outdoor areas such as streets, parks, and
other open spaces with no clearly defined boundaries
Pathogen: a disease-causing agent that invades a host and replicates, including viruses, bacteria, and
fungi
Robotic floor cleaner (RFC): a commercially available, autonomous, floor-cleaning robot
Sampling area: area expected to be similar to the Neato® manufacturer's estimation of approximately
2,000 to 3,000 square feet (ft2) on one cycle, depending on environment, flooring, furniture, and other
factors
A4. Health and Safety
Laboratory testing of RFCs in contaminated areas indicates that the RFCs can cause resuspension of
spores through physical surface agitation. Therefore, the site of an RFC sampling event could pose an
exposure risk to the operator or support personnel. Operators should take precautions when setting up
and deploying the RFCs, including the use of proper personal protective equipment (PPE), including
(but not limited to) the use of a full-face powered air-purifying respirator (PAPR) with P100 cartridges
and full body covering including built-in hood and foot covers. RFC deployment using delayed
activation, if available, would reduce exposure risk.
Before exiting the exclusion zone (hot zone), sampling personnel should follow the standard operating
guideline that provides guidance to EPA and its contractors on decontamination (decon) for personnel
conducting long-term responses to biological contamination per EPA's CMAD decon line procedure.
This standard operating guideline was developed specifically for biological responses using Level C
PPE with a full-face PAPR or full-face air-purifying respirator (APR). Level C PPE is appropriate for most
incidents involving biological agents and is required when the concentration and type of airborne
substances is known and the Occupational Safety and Health Administration (OSHA) criteria for using
APRs are met (OSHA 1999). Level C PPE includes a protective coverall with integral hood and booties,
an APR (preferably a PAPR), inner and outer nitrile gloves, hard hat (optional), and disposable outer
boot covers. All use of respirators must comply with the OSHA Respiratory Protection Standard (29
CFR 1910.134).
There may be additional health and safety concerns associated with these sampling procedures. If the
RFCs are operated on a raised subway platform, the operator should be cognizant of a potential fall
hazard because the platform will not have a guardrail in place. Operators should maintain a safe
distance from the edge. Additionally, heat stress may be a factor depending on the ambient
temperature and length of time spent in modified Level C PPE. Appropriate work/rest regimens should
be adhered to based on the wet bulb globe temperature (WGBT) and other factors such as
acclimatization, hydration and fitness.
A5. Waste Management
The maximum usage for an RFC is one sampling event. After a contaminated area is sampled, the
sampling components of the RFC are collected for analysis. The rest of the RFC can be left in the hot
zone for retrieval for sterilization and then can be either stored or disposed of.
A-2
-------
A6. Equipment and Supplies
Vacuum-based RFCs consist of two components: (1) the robotic body and (2) the dirt collection bin with
a particle filter. The Neato®XV-21 RFC is a common design consisting of a semi-circular autonomous
body and a pop-out bin with filter. The filter is the RFC component that physically collects samples and
therefore is the component that is isolated and secured for sample analysis.
The RFC is operational straight from the box and can be implemented for a field sampling event after its
batteries are charged. The materials listed below are additional materials required to prepare the Neato®
XV-21 RFC for sampling and then securing the RFC for shipment, storage, or analysis:
• Sterile, labeled, 10-in. (inch) by 15-in. sealable, transparent (if available) plastic bags (such as
Twirl'em® [Labplas Inc., Quebec, Canada] bags) as primary, secondary, and tertiary containment
• Pre-printed labels
• Fourth containment container such as a large, clear, plastic bin
• Logbook
• Tychem® [Labplas Inc., Quebec, Canada] suits with hoods
• Boots and boot covers
• Sterile nitrile gloves
• Respirators (APRs or PAPRs)
• Sampling kit containers for pre- and post-sampling events
• Wetted bleach wipes (in a canister)
• Sampling cart
• National Institute of Standards and Technology (NIST)-calibrated timers (verify time at time.aov)
• Video camera (optional)
• Appropriate disposal bags
• Neato® XV-21 RFC with fully charged batteries
• Pens and permanent markers
• Checklist and chain-of-custody (COC) forms
A7. Deployment Procedure
The deployment procedures include pre-deployment preparation, the packaging of RFCs and supplies
for deployment, and surface area assessment and RFC deployment.
7.1 Pre-deployment Preparation
Pre-deployment preparation should occur in the support zone (uncontaminated area where the
sampling team and the sampling materials are not exposed to contaminants). The RFCs and the
sampling equipment must be prepared before the sampling process. The steps below are required for
an effective RFC sampling sequence before deployment.
1. Remove the RFCs from their boxes and place them on a table.
2. For each RFC, ensure that the sampling components (filter and dirt collection bin) are connected
and installed in a functional manner and that the filter is in its proper position.
3. Label each RFC using a pre-set labelling scheme.
4. Using the supplied cable, charge the batteries of each RFC to full capacity. The RFC batteries
should be fully charged before a cleaning cycle begins in accordance with the manufacturer's
instructions. Refer to the user's manuals for battery installation and removal.
A-3
-------
5. RFC is ready to use when the status light turns solid green. The manufacturer recommends
charge the batteries overnight before use for the first time. After that, run through three complete
cycles of charging and cleaning-until-recharge to get the most capacity from the batteries.
6. For each RFC, ensure that the RFC menu is in the right setting so that the sampler will have to
press the start button once only to deploy the RFC.
7.2 Packaging of RFCs and Supplies for Deployment
The RFCs, sampling kits, and ancillary supplies should be placed in three separate bins. These bins
can be large plastic bins with lids (such as Tupperware®[Tupperware Brands Corp., Orlando, FL] bins)
that will be placed on a cart for easy transport between deployment zones. Attach a trash bag to the cart
for disposal of used gloves and boot covers. The contents of each bin are summarized below.
1. Container 1 holds a pre-charged Neato® XV-21 RFC and pre-labeled, post-deployment sampling
kits that include (1) one pre-labeled 7-in. by 12-in. sterile, sealable, Stomacher® 400 (Seward
Ltd., West Sussex, UK) circulator bag for primary containment and (2) two 10-in. by 15-in.
Twirl'em® bags for secondary and tertiary containment).
2. Container 2 holds supplies such as extra gloves and boot covers, extra sterile sampling bags,
wetted bleach wipes, checklists, COC forms, pens, and timers.
3. Container 3 is used to transport the collected samples.
7.3 Surface Area Assessment and RFC Deployment
The surface area requiring sampling should be determined. Based on this area, the sampling team
should at a minimum consist of a lead sampler and a support person. The two-person team will wear the
required site-specific level of PPE and work jointly to handle the RFCs and sampling supplies. A
backup team should be prepared to relieve the sampling team at any time (two in, two out). At each
predetermined sampling location, the team will deploy one RFC in accordance with the procedures
specified in this sampling procedure. The duties of each person are described below.
The lead sampler shall perform the tasks below.
1. Check that the contaminated area is suitable for RFC deployment (no wires, liquid hazards, etc.).
Set up physical boundary markers if needed.
2. Verify the RFC label with the support person, and then place the RFC near the entrance or exit of
the sampling area floor.
3. Press the START button of the RFC, and check that the RFC begins the sampling process.
4. Visually verify that the RFC is functioning properly.
5. Don a new pair of gloves to deploy each additional RFC.
The support person shall perform the tasks below.
1. Verify that all items on the checklist are present on the cart.
2. Take notes in a logbook on the deployment procedure, such as RFC labels, start times, and other
comments.
3. Don a new pair of gloves to deploy each additional RFC.
Notes: If the lead sampler or support person moves to another sampling area contiguous with the first
sampling area, he or she does not need to don a new pair of gloves between RFC deployments. If the
sampling areas are not contiguous, the team member must doff PPE (gloves and boot covers) outside
A-4
-------
of the sampling area and don new gloves and boot covers before entering the new sampling area.
Gloves must be changed after contact with any suspected contaminated surface or item.
Depending on the size of the area to be sampled, the two-person sampling team may leave the
enclosed area and reenter (following appropriate requirements in the health and safety plan) it after
autonomous sampling is completed. Many RFCs are equipped with mapping and navigation
technologies and can return to their starting position after covering the entire floor surface of an
enclosed sampling area.
A8. Sample Collection
After the RFC ceases operation (either completes sampling, ceases because of a dead battery, or
ceases because it is immobilized or trapped), the lead sampler and support person will collect samples.
Each team member's sample collection duties are summarized below.
The lead sampler will collect the sample (filter only) from the Neato® XV-21 RFC dirt bin using the steps
below.
1. Don a new pair of sterile gloves.
2. Retrieve the RFC from its stopped piace in the sampling area using the handle on top of
the RFC.
3. Communicate relevant information to the support person, such as error messages,
recovery location of the RFC, visible material in collection bin, and other information.
4. Dislodge the dirt collection bin from the body of the Neato® XV-21 RFC by lifting the
lever.
5. Remove the dirt collection bin from the body of the Neato® XV-21 RFC.
6. Remove the filter from the dirt bin with the dirt bin turned upward by pulling the filter
handle down and out to separate the filter from the dust bin (Figure 1). Do not disturb the
filter surfaces.
Figure 1: Filter Removal from Neato® XV-21 RFC
7. Put the filter in a pre-labeled, 7-in. by 12-in. Stomacher® 400 circulator bag (opened by
the support person).
A-5
-------
8. Turn the dirt bin upside down, and pour the dirt bin sample contents into the same
Stomacher® 400 circulator bag used for the filter. Carefully empty the dust in the dirt bin
to the bag. Place the dust bin on top of the RFC. Hold the bag upright, and wait for a few
minutes for the dust to settle. Gently squeeze the bag to release the air and seal the bag.
9. Place this sample in its primary containment bag (Stomacher® 400 circulator bag) into a
secondary pre-labeled containment bag measuring 10 in. by 15 in. (opened by the
support person).
10. Doff gloves and don new gloves.
11. Disinfect the exterior of the secondary containment bag using wetted bleach wipes.
12. Place the secondary containment bag into a tertiary containment bag measuring 10 in. by
15-in., and disinfect the tertiary containment bag using wetted bleach wipes.
13. Place the tertiary containment bag in the transportation container (Container 3), and
disinfect Container 3 using wetted bleach wipes.
The support person shall perform the tasks below.
1. When the lead sampler starts the RFC, record the start time, end time, start location, end
location, any significant events that occur during the sampling event, and other important
and relevant information.
2. Record the location where the RFC stopped, time of collection, and comments (such as
error messages, description of the sampling area, RFC conditions, alerting sounds, and
information from the lead sampler).
3. Assist the lead sampler as needed by opening sampling bags, managing power cords,
and performing other required tasks.
4. Verify that the label on the RFC matches the labels on the sampling kits.
5. On the COC form, check the samples as complete.
6. Don new gloves to handle the sample bags.
A9. Post-Deployment Sample Handling
Post-deployment sample handling requires sample preservation, identification, COC, and archiving as
discussed below.
9.1 Sample Preservation
Biological samples in the transportation container should be shipped in insulated containers with cold
packs and appropriate biohazard label, refrigerated (2 to 8 degrees Celsius [°C]) until analysis, and be
archived at 4 ± 2 °C. Samples should be processed and analyzed within 48 hours. Further sample
transportation instructions are available at the Centers for Disease Control and Prevention website at
http://www.cdc.aov/niosh/topics/emres/syrface-samplina-bacillys-anth racis.html.
9.2 Sample Identification
Each sample will be identified using descriptors of the sampled materials and unique sample numbers.
The sampling team will maintain an explicit laboratory log that includes records of each unique sample
number. Each sample will be identified by a material descriptor and a sampling location number. After
transfer of the samples to the microbiology laboratory for analysis, each sample will be additionally
identified (and photographed if available) by replicate number and dilution.
A-6
-------
9.3 Sample Chain of Custody
Careful coordination with the microbiology laboratory is required for successful transfer of
uncompromised samples in a timely manner for analysis. Test schedules will be confirmed with the
laboratory before the start of each test. To ensure the integrity of samples and to maintain a timely and
traceable transfer of samples, an established and proven COC or possession procedure is mandatory.
Accurate records must be maintained whenever samples are created, transferred, stored, analyzed, or
destroyed. The primary objective of these procedures is to create an accurate written record that can be
used to trace the possession of the sample from the moment of its creation through the reporting of its
results. A sample is in custody if it is in any one of the following states:
• In actual physical possession
• In view after being in physical possession
• In physical possession and locked up so it cannot be tampered with
• In a secured area that is restricted except to authorized personnel
• In transit
In the transfer of custody, each custodian will sign, record, and date the transfer. Sample transfer can
be on a sample-by-sample basis or on a bulk basis. The protocol below will be followed for all samples
as they are collected and prepared for distribution.
• A COC record will accompany the samples. When turning over possession of samples, the
transferor and recipient will sign, date, and note the time on the COC record, which allows
transfer of custody of a group of samples from the test site to the microbiology laboratory.
Samples will be carefully packed and shipped as hazardous material (hazmat) samples to the
microbiology laboratory or will be hand carried between on-site laboratories.
9.4 Sample Archiving
Each sample will be archived by maintaining the primary extract at 4 ± 2 °C in a sealed extraction tube
until the data set has undergone quality control checks and the sample has been released for disposal.
Any deviations from sampling protocols must be documented in the laboratory logbook. Sampling
duration, time of day, and observations also will be recorded in the laboratory logbook.
A10. Documentation
All observations and experimental details will be recorded in a scientific logbook. Entries must meet
quality assurance requirements (such as the use of indelible ink, corrections made using lineout
deletions, witnessed signatures, and other requirements). The logbook will include information for all
deviations from project procedures, including spills, deviations from the aseptic technique, and faulty
RFC function. In addition, if possible, the entire sampling procedure should be recorded on video. A
video camera can be mounted above the sampling area. A team member should ensure that the entire
sampling area is in the frame of the camera. Alternatively, a camera can be mounted to the operator if
he or she remains in the enclosed area for the duration of the sampling event. Finally, the COC form
ensures the integrity of samples and allows timely and traceable transfer of sample possession.
A11. References
Occupational Safety and Health Administration (OSHA). 1999. OSHA's Respiratory Protection Standard.
Title 29 of the Code of Federal Regulations (CFR) 1910.134.
A-7
-------
U.S. Environmental Protection Agency. 2010. Ensuring the validity of agency methods validation and peer
review guidelines: Methods of analysis developed for emergency response situations. Washington DC:
U.S. Environmental Protection Agency. Agency Policy Directive Number FEM-2010-01.
U.S. Environmental Protection Agency. 2012. Validation of U.S. Environmental Protection Agency
environmental sampling techniques that support the detection and recovery of microorganisms. Prepared
by: The FEM Method Validation Team, U.S. Environmental Protection Agency. Washington DC: U.S.
Environmental Protection Agency. FEM 2012-01.
A-8
-------
Appendix B: Sample Retrieval Procedure for Commercially Available
Robotic Floor Cleaners for Bacillus anthracis Spores
July 2017
Revision 0.0
National Homeland Security Research Center
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
This sample retrieval procedure is for handling and analyzing samples contaminated with Bacillus
anthracis (Ba) from robotic floor cleaners (RFC). The sections below discuss the following:
• Laboratory operations (Section BP
• Sample processing (Section B2)
• Sample recovery (Section B3)
• Sample analysis (Section B4)
B1. Laboratory Operations
Ba is a National Institutes of Health (NIH) Risk Group 2 bacterial agent associated with serious or lethal
human disease for which preventive or therapeutic interventions may be available (high individual risk but
low community risk). It is also a select agent requiring registration with CDC and/or USDA for
possession, use, storage and/or transfer. Samples contaminated with Ba should be handled in a
Biosafety Level (BSL) 3 laboratory that requires special personal protective equipment (PPE). Risk
Groups correlate with but do not equate to biosafety levels. Laboratory protective clothing must not be
worn outside the laboratory. Facilities for washing and changing clothing after work should be available at
the laboratory. All laboratory manipulations of samples must be performed in a Class Nor Class III
Biological Safety Cabinet (BSC). Efforts should be made to avoid production of aerosols by working in a
BSC. In addition, all centrifugation should be done using aerosol-tight rotors that are opened within the
BSC after each run. Additional BSL 3 requirements can be found in CDC's Biosafety in Microbiological
and Biomedical Laboratories (BMBL). The required PPE and other materials are listed below.
B1.1 Personal Protective Equipment
• Sterile, disposable, long-cuffed, nitrile gloves for outer and inner gloves
• Safety glasses
• Disposable coveralls with hood and solid front
• Disposable boot covers (booties)
• Respiratory protection from particulate hazards if necessary. PPE selection should be consistent
with individual BSL3 facility guidance
B1.2 Other Required Materials
• Dispatch® hospital cleaner disinfectant towels with bleach (canister)
• 70% denatured ethanol wipes (canister)
• RFC samples (in a clear plastic container), accompanied by a relinquished chain-of-custody
(COC) form
• National Institute of Standards and Technology (NIST)-certified timers
• Laboratory notebook
• Pre-labeled biohazard trash bags and bins
• Kimwipes® or equivalent low-lint paper wipes
B-1
-------
• Seward™ Stomacher® 400 circulator bag racks or equivalent
• Sterile Seward™ Stomacher® 400 circulator bags
• Sonicator
• Orbital shaker incubator (OSI)
• Centrifuge
• 50- and 25-milliliter (mL) sterile serological pipettes
• Sterile 50-mL conical or centrifuge tubes
• Electronic serological pipetter, pre-charged
• 90-mL aliquots of phosphate-buffered saline with Tween® 20 (PBST) in sterile specimen cups
• Pens or pre-printed labels
• Tryptic soy agar (TSA) or equivalent agar plates
• Spiral plater
• Spiral plate spore counter
• 0.2- to 0.45-micron (jjm) pore-size disposable analytical filter units
• Sterile 1,000-microliter (|jL) pipette tips
• Sterile forceps
• Sterile deionized (Dl) water (in about 10-mL aliquots)
• Calibrated top-loading balance (320 grams [g] x 0.001 g) and calibration weights
• 50-mL conical tube holders
• Vortexer
• Biological Safety Cabinet (BSC)
Sample Processing
1. Verify that each COC form is complete and signed by authorized personnel at the package
shipping/receiving dock.
2. Review each COC form to ensure that all of the samples are complete and that there is no
notable variation in the sample identification (ID) labels compared to the IDs listed in the COC
form. If variation has occurred, note it in the laboratory notebook.
3. Don appropriate PPE in accordance with BSL3 facility guidance..
4. Gather all necessary required materials (see Section B1.2). and place them on a clean cart
beside the BSC within arm's reach so that sample processing may be performed without
interruptions.
5. Clean the BSC using Dispatch® hospital cleaner disinfectant bleach towels. Wait for 5 minutes
before spraying the cabinet with Dl water. Wipe clean using Kimwipes®.
6. Wipe the workspace using the 70% denatured ethanol wipes. Dry the workspace using clean
Kimwipes®. Discard outer gloves, and replace them with a new pair. Waste should be disposed
of in accordance with BSL3 facility procedures.
B-2
-------
7. Using Dispatch® bleach wipes, thoroughly disinfect the outside of the clear plastic container
containing the RFC samples that may contain Ba spores. Discard outer gloves, and replace them
with a new pair.
8. Inside the BSC, one at a time, remove the 10-inch (in.) by 15-in. Twirl'em® bags containing each
RFC sample from the plastic container. Each bag contains an inner 10-in. by 15-in. Twirl'em® bag
that in turn contains the 10-in. by 15-in. Stomacher® 400 circulator bag with the RFC filter.
9. Verify that the sample ID label on the outside of each Twirl'em® bag matches the samples listed
on the COC form. Using Dispatch® bleach wipes, disinfect each outer bag, and then wipe the bag
using clean Kimwipes® until it is dry. Discard the used bleach wipes and Kimwipes®.
10. Verify that the label on the inner Twirl'em® bag matches the label on the outer bag. If the inner
bag sample ID does not match the outer bag ID, quarantine the sample for further analysis.
11. If no label discrepancies are observed, retrieve the inner Twirl'em® bag from the outer bag and
discard the outer bag.
12. Disinfect the inner Twirl'em® bag with Dispatch® bleach wipes, and then wipe the bag using clean
Kimwipes® until the bag is dry.
13. Retrieve the inner Stomacher® 400 circulator bag from the outer Twirl'em® bag, and discard the
Twirl'em® bag. Take care not to let the outer Twirl'em® bag contact any surfaces that have not
been disinfected until disposal.
14. Disinfect the Stomacher® 400 circulator bag with Dispatch® bleach wipes, and then wipe the bag
using clean Kimwipes® until the bag is dry. Take care not to let the Stomacher® 400 circulator bag
contact any surfaces that have not been disinfected.
15. Place each inner Stomacher® 400 circulator bag containing the RFC filter in a Seward™
Stomacher® 400 circulator bag rack or equivalent. Set each bag upright, and allow the filter dust
to settle completely in the Stomacher® 400 circulator bag before opening the bag.
Sample Recovery
1. Inside a BSC, aseptically add two pre-measured specimen cups containing 90 mL of sterile PBST
to each Stomacher® 400 circulator bag containing a filter, resulting in a total of 180 mL of PBST
added to ach sample. Place each sample in a new, secondary Stomacher® 400 circulator bag to
prevent leakage.
2. Place each sample containing 180 mL of PBST lying flat into the OSI using flask clamps to hold
each sample securely. Make certain that the sample bags will not tip over or become unsecured.
3. Agitate the samples in the OSI at 300 rotations per minute (rpm) for 30 minutes (min) at ambient
room temperature. After agitation, remove the samples from the OSI, and transfer them to the
BSC to be split before centrifugation.
4. Aseptically transfer the liquid from each sample bag to four individual, sterile, pre-labeled 50-mL
conical tubes, making certain that the amount of liquid is evenly distributed. Once each sample
has been split, measure each tube for weight using a calibrated balance. The weight of each tube
should be ± 0.5 g of each other so that the centrifuge will be balanced.
5. Place the tubes from each sample into the centrifuge with sealed rotor and containment cups.
Weights of tubes positioned opposite from one another should be ± 0.5 g. Centrifuge the tubes at
B-3
-------
3,500 x gravity (* g) for 15 min. Do not use the brake option (if applicable) on the centrifuge to
slow the rotor because as resuspension of pellets may occur.
6. After centrifugation, transfer the tubes to the BSC. From each tube, carefully remove all but 5 mL
of supernatant using a 50-mL pipette and discard the removed supernatant. The pellet may be
easily disturbed and not visible, so place the pipette tip away from the tube bottom or side.
7. Vortex and sonicate each tube using the steps below inside the BSC.
a. Set the vortex mixer to the highest level.
b. Turn on the sonicator water bath.
c. Vortex each tube for 30 seconds (sec).
d. Transfer tubes to sonicator bath, and sonicate them for 30 sec.
e. Repeat the vortex and sonication cycle two additional times (three times per each sample
tube).
8. Remove the suspension from one tube using a sterile 25-mL pipette and place it in one of the
other tubes of the same sample. Repeat this process for the other two tubes of the same sample,
resulting in one tube containing approximately 20 mL of the sample. This combined sample is the
final elution suspension.
9. Measure the volume of the final suspension using a sterile 25-mL pipette, and record the volume
in the laboratory notebook.
B4. Sample Analysis
1. Don appropriate PPE in accordance with BSL3 facility guidance.
2. In a BSC, Proceed to serially dilute and plate all elution suspension samples on TSA plates.
3. If the samples are turbid, wide-orifice pipette tips may be used to prevent clogging of pipette tips.
4. If dilution plating yields colony forming unit (CFU) counts below the quantification range, filter
plate the samples.
5. Place all plates in an incubator set at 35 ± 2 degrees Celsius (°C) for a maximum of 2 days.
Plates should be examined within 18 to 24 hours after the start of incubation. Manually
enumerate CFU counts for the target organism, and record the data. Re-examine the plates after
48 hours to check for additional CFU.
B-4
-------
Appendix C: Sampling Procedure Using Commercially Available Wet
Vacuum Cleaner for Bacillus anthracis Spores
July 2017
Revision 0.0
National Homeland Security Research Center
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
C1. Scope and Applicability
The U.S. Environmental Protection Agency's (EPA's) Homeland Security Research Program within EPA's
Office of Research and Development, and the Chemical, Biological, Radiological and Nuclear
Consequence Management Advisory Division within EPA's Office of Land and Emergency Management,
jointly developed this sampling procedure. This procedure is intended to provide a method for trained
incident responders to collect environmental samples after a biological contamination incident. This
procedure specifically applies to the collection of surface-bound particulates and microorganisms using
off-the-shelf robotic floor cleaners (RFC). The purpose of this procedure is to guide the process of
preparation, deployment, and collection using RFCs for sampling surfaces in a specified area. The results
from the collected samples can be used to determine the presence or absence of contamination and the
contamination level after natural outbreaks and after intentional or accidental releases of pathogenic
microorganisms and biotoxins.
At the time of publication, this sampling procedure has not been validated. At the date of this publication,
the RFC sampling procedure has been partially characterized for deployment feasibility and collection
performance for bacterial spores. This procedure will be updated or replaced with a fully characterized
and validated procedure upon availability. During emergencies, the use of non-validated methods may be
warranted when validated methods are not available. EPA's use of non-validated methods must adhere to
the EPA's Forum on Environmental Measurement (FEM) policy directive on method validation (EPA
2010). Further information on method validation is presented in Validation of U.S. Environmental
Protection Agency Environmental Sampling Techniques that Support the Detection and Recovery of
Microorganisms (EPA 2012).
C2. Summary
This SP is for the sampling of a horizontal surface (such as a floor) using a wet vacuum cleaner. After
sampling, the wet vacuum cleaner is recovered and processed to determine the presence or absence of
potential surface contamination. This SP provides a step-by-step sampling procedure for the following wet
vacuum cleaner:
• Hoover Max Extract® Steam Vac Dual V® Cleaner (F7425-900 with SpinScrub Hand Tool;
Hoover Company, North Canton, OH): wet vacuum cleaner used to sample both porous
surfaces (such as carpet, wood, and bare concrete) and nonporous surfaces (such as vinyl, tile,
laminate, coated wood, and coated concrete)
The sections below discuss the following:
• Definitions (Section C3)
• Health and safety (Section C4)
• Waste management (Section C5)
• Equipment and supplies (Section C6)
• Deployment procedure (Section C7)
• Sample collection (Section C8)
• Post-deployment sample handling (Section C9)
C-1
-------
• Documentation (Section C10)
Section C11 lists the references used to prepare this SP.
C3. Definitions
Biological agent contamination: contamination that can be attributed to both natural outbreaks and
intentional or accidental releases of pathogenic microorganisms and biotoxins.
Biotoxin: a poisonous substance either produced by or extracted from living or dead organisms
Enclosed facilities and objects: facilities and objects that typically have surface areas with clearly
defined boundaries such as walls and that are isolated from exterior environments, including commercial
and residential buildings and transportation vehicles
Method of dissemination: the means by which biological agents are dispersed; dispersal can occur over
large areas in wet and dry forms through aerosol generation and spreading devices that contaminate
indoor and outdoor sites or the food or supply chain
Outdoor areas and objects: building exteriors and other outdoor areas such as streets, parks, and other
open spaces with no clearly defined boundaries
Pathogen: a disease-causing agent that invades a host and replicates, including viruses, bacteria, and
fungi
Sampling area: area expected to be similar to Hoover Company manufacturer's estimation of
approximately 200 ft2 per liquid container, depending on environment, flooring, furniture, and other site-
specific factors
Wet vacuum cleaner: a commercially available, upright vacuum cleaner that dispenses and retrieves
liquid cleaning agent
C4. Health and Safety
Laboratory testing of RFCs in contaminated areas indicates that the RFCs can cause resuspension of
spores through physical surface agitation. Therefore, the site of an RFC sampling event could pose an
exposure risk to the operator or support personnel. Operators should take precautions when setting up
and deploying the RFCs, including the use of proper personal protective equipment (PPE), including (but
not limited to) the use of a full-face powered air-purifying respirator (PAPR) with P100 cartridges and full
body covering including built-in hood and foot covers. RFC deployment using delayed activation, if
available, would reduce exposure risk.
Before exiting the exclusion zone (hot zone), sampling personnel should follow the standard operating
guideline that provides guidance to EPA and its contractors on decontamination (decon) for personnel
conducting long-term responses to biological contamination per EPA's CMAD decon line procedure. This
standard operating guideline was developed specifically for biological responses using Level C PPE with
a full-face PAPR or full-face air-purifying respirator (APR). Level C PPE is appropriate for most incidents
involving biological agents and is required when the concentration and type of airborne substances is
C-2
-------
known and the Occupational Safety and Health Administration (OSHA) criteria for using APRs are met
(OSHA 1999). Level C PPE includes a protective coverall with integral hood and booties, an APR
(preferably a PAPR), inner and outer nitrile gloves, hard hat (optional), and disposable outer boot covers.
All use of respirators must comply with the OSHA Respiratory Protection Standard (29 CFR 1910.134).
There may be additional health and safety concerns associated with these sampling procedures. If the
RFCs are operated on a raised subway platform, the operator should be cognizant of a potential fall
hazard because the platform will not have a guardrail in place. Operators should maintain a safe distance
from the edge. Additionally, heat stress may be a factor depending on the ambient temperature and
length of time spent in modified Level C PPE. Appropriate work/rest regimens should be adhered to
based on the wet bulb globe temperature (WGBT) and other factors such as acclimatization, hydration
and fitness.
C5. Waste Management
The maximum usage for a wet vacuum cleaner is one sampling event. After a contaminated area is
sampled, the sampling components of the wet vacuum cleaner are collected for analysis. The rest of the
wet vacuum cleaner can be left in the hot zone for retrieval for sterilization and then can be either stored
or disposed of.
C6. Equipment and Supplies
The Hoover F7452-900 cleaner has brushes and two nozzles to deliver equal suction power across the
width of the nozzle. The cleaning nozzle is approximately 13 inches (in.) wide. The wet vacuum has
separate clean and dirty liquid tanks as well as hand tools for cleaning hard-to-reach areas.
The dirty tank is isolated from the clean tank. A designated aliquot of about 100 milliliters (mL) is obtained
from the dirty tank liquid for analysis. Before use during a field sampling event, the wet vacuum cleaner
must be assembled from its store packaging. The user's manual provides assembly instructions. The
materials listed below are additional materials required to prepare the Hoover F7452-900 cleaner for
sampling and then securing the cleaner for shipment, storage, or analysis:
• Primary sample container, the dirty liquid tank
• Secondary sample containers consisting of extra-large food storage bags (such as Ziploc® XL HD
Big Bags measuring 2 feet (ft) by 20 in.
• 0.05% Tween® 20 solution in a clean tank prepared using 5 liters (L) of deionized (Dl) water with
2.5 mL Tween® 20, with 5-L level marked on the tank
• Sterilite 40-gallon wheeled industrial tote or equivalent (Item. No. 553504223, Walmart,
Bentonville, AR)
Note: The 0.05% Tween® 20 solution will be pre-loaded in the clean tank and stored in 40-gallon
wheeled industrial totes until wet vacuum cleaner deployment.
• Pre-printed labels
• Logbook
• Sterile gloves
C-3
-------
• Tychem suits with hood
• Boots and boot covers
• Chem tape
• PAPR or equivalent respirator
• Nitrile gloves
• Clear, labeled sampling kit containers for pre- and post-sampling events
• Wetted bleach wipes (in a canister)
• Sampling cart
• National Institute of Standards and Technology (NIST)-calibrated timers (verify time at time.gov)
• Appropriate disposal bags
• Pre-assembled Hoover F7452-900 cleaner stored in a sterile bag (such as a 42-in. by 42-in.
Tyvek drawstring bag, General Econopak, Philadelphia, PA)
• Extension cords if needed
• 14-in. by 14-in., pre-cut sheets of Bond paper (in a sterile bag)
• Pens and permanent markers
• Tables and chairs
• Checklist and chain-of-custody (COC) forms
• Cooler (such as Igloo Model H-1353 industrial 5-gallon water cooler, Katy, TX)
C7. Deployment Procedure
The deployment procedures include pre-deployment preparation and the surface area assessment and
wet vacuum cleaner deployment.
C7.1 Pre-deployment Preparation
Pre-deployment preparation should occur in the support zone (uncontaminated area where the sampling
team and the sampling materials are not exposed to contaminants). The wet vacuum cleaners should
each be labeled using a pre-set labelling scheme. The sampling kits and ancillary supplies should be
placed in four separate bins. These bins can be large plastic bins with lids (such as Tupperware®bins)
that will be placed on a cart for easy transport between deployment zones. Attach a trash bag to the cart
for disposal of used gloves and boot covers. The contents of each bin are summarized below.
1. Container 1 holds pre-labeled, extra-large food storage sampling bags containing the double-
bagged dirty tank.
2. Container 2 holds supplies such as extra gloves and boot covers, extra sampling bags, bond
paper, wetted bleach wipes, boundary markers, serological pipettes, checklists, COC forms,
pens, and timers.
3. Container 3 contains the pre-loaded clean tank loaded with 0.05% Tween® 20.
C-4
-------
C7.2 Surface Area Assessment and Wet Vacuum Cleaner Deployment
A sampling plan should be developed as part of the overall site decontamination plan and site safety plan.
The sampling plan includes the following information:
• Surface area and layout of the sampling site
• Number of wet vacuum samplers required for each sampling site
• Sampling procedures
Based on this sampling site area, the sampling team should at a minimum consist of a lead sampler and
a support person. The two-person team will wear the required site-specific level of PPE and work jointly to
transport the wet vacuum cleaners and sampling supplies into the exclusion zone where the samples will
be collected. A backup team should be prepared to relieve the sampling team at any time (two in, two
out). At each predetermined sampling location, the team will deploy one wet vacuum cleaner in
accordance with the procedures specified in this SP. The duties of each person are described below.
The support person shall perform the tasks below.
1. Verify that the items from the checklist are present on the cart.
2. Remove each Hoover F7452-900 wet vacuum cleaner from its bag, and place each wet vacuum
cleaner individually just outside the sampling site.
3. Check and record pre-labeled information for each vacuum cleaner.
4. Attach the pre-loaded clean liquid tank to each wet vacuum cleaner (see Figure C-1).
OR
m\
Figure C-1. Hoover F7452-900 Clean Liquid Tank
5. Using the supplied cord (and an extension cord if necessary), plug in the vacuum cleaner.
6. Make notes of the deployment procedure in the logbook, including the start time, end time, start
location, end location, label information, and any significant event during the sampling event.
Take photographs if possible.
7. Don a new pair of outer gloves whenever deploying an additional wet vacuum cleaner.
Note: Change outer gloves after coming into contact with any suspected contaminated surface or item.
C-5
-------
The lead sampler shall perform the tasks below.
1. Check that the contaminated area is suitable for vacuum cleaner deployment (no obstacles, trip
hazards, etc.).
2. Verify the wet vacuum cleaner label with the support person, and then place the vacuum cleaner
in a corner of the sampling area.
3. Ensure that the clean tank pre-loaded with 0.05% Tween® 20 and the dirty tank are properly
seated in the vacuum cleaner.
4. Set the vacuum cleaner to "Wash Auto Rinse" mode, and slide the vacuum switch to "ON"
position to start the wet vacuum sampling process (see Figure C-2).
Note: If moving between sampling areas, don new outer gloves at the new sampling location.
Vacuum On/Off Switch
Solution Spray Switch
Wash Auto Rinse Mode
Dirty Tank
Clean Tank
Figure C-2. Hoover F7452-900 Max Extract Cleaner
Each wet vacuum cleaner will be moved back and forth in a specified pattern on the designated sampling
area using the steps summarized below.
C-6
-------
1. Divide the width of the total sampling area by half of the nozzle width area ,and round the result to
the nearest larger whole number, N. N will be the number of sampling strip passes. Each
sampling strip will consist of half the nozzle width except for the last sampling strip, which will
have a width of one nozzle (see Figure C-3).
2. Place the vacuum cleaner nozzle on the sampling area so that the front edge of the vacuum
cleaner nozzle lip coincides with the line defining the beginning of the sampling area and the side
of the nozzle coincides with the one side boundary of the second strip as shown in Figure C-3.
3. Complete Stroke 1, a backward stroke starting at the end of the sampling area.
4. Complete Stroke 2, a forward stroke.
5. Move the vacuum cleaner horizontally by half of the nozzle width area.
6. Repeat Steps 2 through 5 for each subsequent sampling strip to the end of the sampling area.
Nozzle width Nozzle start locations
hirst stroke
spraying/vacuuming
~
Second stroke
vacuuming
~
2
1
2
u
Wet vacuum starting position
Figure C-3. Wet Vacuum Cleaner Sampling Pattern
C8. Sample Collection
After the vacuuming is completed, the samples will be collected from the wet vacuum cleaner. Each team
member's sample collection duties are summarized below.
The support person shall perform the steps below.
1. Don a new pair of outer gloves.
2. Assist the lead sampler by handing out wetted bleach wipes as necessary.
3. Verify that the labels on the primary containers match the labels on the secondary containers and
coolers, and that the lead sampler is handling samples correctly.
4. Check the samples on the COC form as complete.
C-7
-------
5. Once all samples have been collected, assist the lead sampler in disinfecting Containers 1
through 3 and the sample cart.
6. Step out of the sampling area (exclusion zone) and proceed to the decon line (contamination
reduction zone).
The lead sampler shall collect the liquid samples from the wet vacuum cleaner using the steps below.
1. Verify with the support person that the labels on the dirty tanks and the coolers match.
2. Retrieve the dirty tank from the wet vacuum cleaner (see Figure C-4), and disinfect the exterior of
the tank using a wetted bleach wipe.
Figure C-4. Ziploc® Extra Large Bag and Hoover F7452-900 Dirty Tank
3. Double bag the dirty tank in pre-labeled, large food storage zipper bags (such as the Ziploc® bag
shown in Figure C-4) for secondary sample containment. Disinfect the exterior of the outer bag
using a wetted bleach wipe.
4. Place the bags with the dirty tank in a cooler such as the Igloo 5-gallon cooler shown in Figure C-
5, and ciose the lid. Verify that the label on the dirty tank matches the cooler label. Add ice to the
cooler.
mmmuns warM
Figure C-5. igloo Industrial 5-Gallon Water Cooler
5. Disinfect the exterior of the cooler using a wetted bleach wipe, and place the entire cooler in
Container 1. Disinfect the outside of Container 1 using a wetted bleach wipe.
C-8
-------
6. Return the used wet vacuum cleaner to its bag, and disinfect the outside of the bag using a
wetted bleach wipe.
7. Once all samples have been collected, assist the lead sampler in disinfecting Containers 1
through 3 and the sample cart.
8. Step out of the sampling area (exclusion zone) and proceed to the decon line (contamination
reduction zone).
C9. Post-Deployment Sample Handling
Post-deployment sample handling requires sample preservation, identification, COC, and archiving as
discussed below.
C9.1 Sample Preservation
Biological samples should be shipped ground in insulated containers with cold packs and appropriate
biohazard label, refrigerated (2 to 8 degrees Celsius [°C]) until analysis, and be archived at 4 ± 2 °C.
Samples should be processed and analyzed within 48 hours. Further sample transportation instructions
are available at the Centers for Disease Control and Prevention website at
http://www.cdc.aov/niosh/topics/emres/surface-samplina-bacillys-anth raeis.html.
C9.2 Sample Identification
Each sample will be identified using unique sample numbers. The sampling team will maintain an explicit
laboratory log that includes records of each unique sample number. After transfer of the samples to the
microbiology laboratory for analysis, each sample will be additionally identified (and photographed if
available) by replicate number and dilution.
C9.3 Sample Chain of Custody
Careful coordination with the microbiology laboratory is required for successful transfer of uncompromised
samples in a timely manner for analysis. Test schedules will be confirmed with the laboratory before the
start of each test. To ensure the integrity of samples and to maintain a timely and traceable transfer of
samples, an established and proven COC or possession procedure is mandatory. Accurate records must
be maintained whenever samples are created, transferred, stored, analyzed, or destroyed. The primary
objective of these procedures is to create an accurate written record that can be used to trace the
possession of the sample from the moment of its creation through the reporting of its results. A sample is
in custody if it is in any one of the following states:
• In actual physical possession
• In view after being in physical possession
• In physical possession and locked up so it cannot be tampered with
• In a secured area that is restricted except to authorized personnel
• In transit
C-9
-------
The sampling team members will receive copies of test plans prior to each test. Pre-study briefings will
then be held to apprise participants of the objectives, test protocols, and COC procedures to be followed.
These protocols must mesh with any protocols established by EPA.
In the transfer of custody, each custodian will sign, record, and date the transfer. Sample transfer can be
on a sample-by-sample basis or on a bulk basis. The protocol below will be followed for all samples as
they are collected and prepared for distribution.
• A COC record will accompany the samples. When turning over possession of samples, the
transferor and recipient will sign, date, and note the time on the COC record, which allows
transfer of custody of a group of samples from the test site to the microbiology laboratory.
Samples will be carefully packed and shipped as hazardous material (hazmat) samples to the
microbiology laboratory or will be hand carried between on-site laboratories.
C9.4 Sample Archiving Requirements
Each sample will be archived by maintaining each sample at 4 ± 2 °C in a sealed extraction tube until the
data set has undergone quality control checks and the sample has been released for disposal. Any
deviations from sampling protocols must be documented in the laboratory logbook. Sampling duration,
time of day, and observations also will be recorded in the laboratory logbook.
C10. Documentation
All observations and experimental details will be recorded in a scientific logbook. Entries must meet
quality assurance requirements (such as the use of indelible ink, corrections made using lineout
deletions, witnessed signatures, and other requirements). The logbook will include information for all
deviations from project procedures, including spills, deviations from the aseptic technique, and faulty wet
vacuum function. In addition, if possible, the entire sampling procedure should be recorded on video. A
video camera can be mounted above the sampling area. A team member should ensure that the entire
sampling area is in the frame of the camera. Alternatively, a camera can be mounted to the operator if he
or she remains in the enclosed area for the duration of the sampling event. Finally, the COC form ensures
the integrity of samples and allows timely and traceable transfer of sample possession.
C11. References
Occupational Safety and Health Administration (OSHA). 1999. OSHA's Respiratory Protection Standard.
Title 29 of the Code of Federal Regulations (CFR) 1910.134.
U.S. Environmental Protection Agency. 2010. Ensuring the validity of agency methods validation and peer
review guidelines: Methods of analysis developed for emergency response situations. Washington DC:
U.S. Environmental Protection Agency. Agency Policy Directive Number FEM-2010-01.
U.S. Environmental Protection Agency. 2012. Validation of U.S. Environmental Protection Agency
environmental sampling techniques that support the detection and recovery of microorganisms. Prepared
by: The FEM Method Validation Team, U.S. Environmental Protection Agency. Washington DC: U.S.
Environmental Protection Agency. FEM 2012-01.
C-10
-------
Appendix D: Coupon Inoculation Procedure for Spray-Dry Deposition
May 2017
National Homeland Security Research Center
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
SCOPE: Procedure for loading dry spores onto a test surface
PURPOSE: To provide consistent loading of spores onto test coupons for reaerosolization testing
MATERIALS:
• Deposition suspension in 50-milliliter (mL) conical tube
• 10% pH-amended household bleach solution
• 70% isopropanol solution
• Deionized (Dl) water
• Coupon transfer case
• Disposable gloves
• Laboratory coat
• Safety glasses
• Test coupon
• New Era Pump Systems Inc. Multi-Phaser™ Model NE-1000 Syringe Pump
• 3-mL BD Luer lock syringe
• Deposition chamber
• Reference sample conical tube (sterile 50-mL conical tube with 10 mL of sterile phosphate-
buffered saline with Tween® 20 [PBST])
• 50-mL conical tubes labeled for waste collection
• Vortex mixer
• Sonication bath
• Hospital-grade bleach disinfectant wipes (such as Dispatch®)
• Test coupons and medium-density fiberboard (MDF) coupon risers
• Spray adhesive
PROCEDURE:
Laboratory personnel should wear appropriate personal protective equipment (PPE), including a
laboratory coat, safety glasses, and disposable gloves. All personnel handling the samples should be
trained in the proper deposition procedure by an experienced staff member.
Spray-Dry Deposition
1. Ensure that all components of the deposition chamber are functional and correctly set up, as
shown in Figure D-1 (with nozzle on deposition test stand) and Figure D-2 (with nozzle on spray-
dry deposition chamber). Note: A clean, 50-mL conical tube labeled "waste" should be placed
below the nozzle head when the nozzle is on the test stand to collect excess liquid.
2. Expel any sporistatic alcohol left in the deposition nozzle tubing and syringe from the
decontamination process (see "Decontamination of the System" at the end of this procedure.
3. After expelling the alcohol, replace the syringe with a clean, 3-mL syringe.
D-1
-------
Figure D-1. Deposition Test Stand to Hold Nozzle for Collecting Reference Samples and between
Depositions
Syringe
Pump
Two spray-dry
deposition stacks
Syringe pump
Nozzle
Deposition
suspension
Figure D-2. Complete Spray-Dry Deposition Setup
4. Sonicate the deposition suspension for 30 seconds (sec), and then vortex the suspension for an
additional 30 sec.
5. Place the deposition suspension tube in the tube rack, and screw on the cap with tube to connect
the three-way valve.
D-2
-------
Initial Programming of Syringe Pump
Refer to the user's manual if needed (New Era Pump Systems Inc. 2009).
1. Set the diameter, which is the internal diameter of the syringe to be used.
2. Set the rate to 200 microliters per minute (jjL/min).
3. Set the dispensed volume to 200 |jl_.
Collection of a Positive Control Sample
1. Place the positive control tube in the test stand clamp.
2. Actuate the three-way valve to direct flow from the syringe to the deposition suspension,
vigorously fill the syringe, and then expel its contents. Repeat this step 10 times.
3. Slowly fill the syringe, making sure there are no air bubbles.
4. Actuate the three-way valve to direct flow to the ultrasonic nozzle. Fully dispense the suspension
from the syringe into the tubing leading to the nozzle.
5. Repeat Steps 2 through 4 until the desired volume is sampled.
Collection of a Reference Sample
1. Verify that a waste collection tube is in place under the ultrasonic nozzle, with the waste tube
seated against the body of the nozzle.
2. Turn on lighting if needed.
3. Actuate the three-way valve to direct flow from the syringe to the deposition suspension,
vigorously fill the syringe, and then expel its contents. Repeat this step 10 times.
4. Slowly fill the syringe, making sure there are no air bubbles.
5. Actuate the three-way valve to direct flow to the ultrasonic nozzle. Slowly dispense the
suspension from the syringe into the tubing leading to the nozzle until the tubing is void of air.
6. Load the syringe into the syringe pump. Refer to the user's manual if needed (New Era Pump
Systems Inc. 2009).
7. Turn on the syringe pump, and push any button to stop the LEDs from flashing.
8. Turn on the nozzle power generator.
9. Turn off the nozzle power generator.
10. Push the Start/Stop button on the syringe pump to dispense the deposition suspension at the
programmed rate (200 jjL/min) and volume (200 |jl_), and visually confirm that liquid is collecting
on the nozzle.
11. Turn on the nozzle power generator, and verify that it is set to 2.5 watts (W).
12. Once a continuous mist is observed, continue dispensing another 10 |jl_.
13. Press the Start/Stop button on the syringe pump to stop the spray.
14. After approximately 3 sec, turn off the nozzle power generator and the syringe pump.
15. Remove, cap, and set aside the waste tube.
D-3
-------
16. Place a reference sample collection tube under the nozzle, with the tube seated against the body
of the nozzle.
17. Turn on the syringe pump, and push any button to stop the LEDs from flashing.
18. Turn on the nozzle power generator, and verify that it is set to 2.5 W.
19. Push the Start/Stop button on the syringe pump to dispense the suspension at the programmed
rate (200 jjL/min) and volume (200 |jl_).
20. When the pump has stopped dispensing, turn off the nozzle power generator and the syringe
pump.
21. Allow the reference sample tube to settle for 2 min, and then carefully remove, cap, label, and
vortex the sample for 30 sec.
22. Place the waste tube back into the tube clamp under the nozzle on the deposition test stand.
23. Remove the syringe from the pump and retract it slowly to remove all suspension from the tubing.
24. Dispense the suspension from the syringe into the tubing leading to the nozzle until the tubing is
void of air.
25. Retract the syringe slowly to remove all suspension from the tubing.
Deposition onto a Test Coupon
1. Verify that a waste collection tube is in place under the ultrasonic nozzle, with the waste tube
seated against the body of the nozzle.
2. Turn on the lighting if needed.
3. Load the coupon onto the transfer enclosure, place the deposition chamber over the coupon, and
place the lid on the chamber.
4. Connect the mixing fans to the power supply.
5. Ground the chamber and lid.
6. Actuate the three-way valve to direct flow from the syringe to the deposition suspension,
vigorously fill the syringe, and then expel its contents. Repeat this step 10 times.
7. Slowly fill the syringe, making sure there are no air bubbles.
8. Actuate the three-way valve to direct flow to the ultrasonic nozzle, and slowly dispense
suspension from the syringe into the tubing leading to the nozzle until the tubing is void of air.
9. Load the syringe into the syringe pump. Refer to the user's manual if needed (New Era Pump
Systems Inc. 2009).
10. Turn on the syringe pump, and push any button to stop the LEDs from flashing.
11. Turn on the nozzle power generator.
12. Turn off the nozzle power generator.
13. Push the Start/Stop button on the syringe pump to dispense the deposition suspension at the
programmed rate (200 jjL/min) and volume (200 |jL), and visually confirm that liquid is collecting
on the nozzle.
14. Turn on the nozzle power generator, and verify that it is set to 2.5 W.
15. Once a continuous mist is observed, continue dispensing another 10 |jL.
16. Press the Start/Stop button on the syringe pump to stop the spray.
D-4
-------
17. After approximately 3 sec, turn off the nozzle power generator and the syringe pump.
18. Turn on the syringe pump, and push any button to stop the LEDs from flashing.
19. Turn on the mixing fans.
20. Transfer the nozzle to the lid, and ground the lid.
21. Turn on the nozzle power generator, and verify that it is set to 2.5 W.
22. Push the Start/Stop button on the syringe pump to dispense the suspension at the programmed
rate (200 jjL/min) and volume (200 |jl_).
23. When the pump has stopped dispensing, turn off the nozzle power generator and the syringe
pump.
24. Turn off the mixing fans.
25. Allow the deposition chamber to settle for 2 min.
26. Remove the nozzle, and slide the nozzle seat cover into place.
27. Disconnect all wires.
28. Remove the syringe from the pump and retract it slowly to remove all suspension from the tubing.
29. Dispense suspension from the syringe into the tubing leading to the nozzle until tubing is void of
air.
30. Retract the syringe slowly to remove all suspension from the tubing.
31. Place the entire deposition chamber and transfer enclosure into the environmental chamber to
dry and equilibrate. The minimum post-deposition equilibration time is 3 hours for all coupons.
Decontamination of the System
1. Retract all deposition suspension from the tubing, and dispense it back into the original container.
2. Actuate the three-way valve to direct flow from the syringe to the deposition suspension,
vigorously fill the syringe, and then expel its contents. Repeat this step 10 times.
3. Remove the cap with tubing from the deposition suspension, and place the cap onto a 50-mL
conical tube containing 10% pH-amended household bleach. Cap the deposition suspension tube
with the original cap.
4. Fill the syringe with bleach solution, and dispense it into the nozzle tubing until approximately 1
mL has been ejected from the nozzle into the waste tube.
5. Refill the syringe.
6. Wait at least 2 min, and then expel all syringe and tubing contents into the waste tube.
7. Remove the cap with the tubing from the conical tube containing bleach solution, and place the
cap into a conical tube containing Dl water.
8. Fill the syringe with Dl water, dispense the syringe into the nozzle tubing, and then fill the syringe
with air and dispense it into the nozzle tubing to expel all liquid. Repeat this step four times.
9. Remove the cap with tubing from the conical tube containing Dl water, and place the cap on the
conical tube containing sporistatic alcohol.
10. Fill the syringe with sporistatic alcohol, dispense the syringe into the nozzle tubing, and then fill
the syringe with air and dispense it into the nozzle tubing to expel all liquid. Repeat this step two
times.
D-5
-------
11. Fill the syringe with sporistatic alcohol, and dispense the syringe into the nozzle tubing until
approximately 1 mL is expelled from the nozzle.
12. Refill the syringe with sporistatic alcohol.
13. Cap and dispose of the waste collection tube.
14. Place a new waste collection conical tube in the clamp under the nozzle.
15. Wipe all contaminated surfaces of the work area, the deposition chamber, and all experimental
equipment using Dispatch® wipes or 10% pH-amended household bleach solution.
16. Wait at least 2 min.
17. Wipe all decontaminated surfaces of the work area, deposition chamber, and all experimental
equipment with Dl water.
18. Wipe all decontaminated surfaces of the work area, deposition chamber, and all experimental
equipment with 70% isopropanol solution.
REFERENCE:
New Era Pump Systems Inc. (2009) Multi-Phaser™ Model: NE-1000 Syringe Pump User Manual,
Publication #1200-01, Revision 15, V3.74, 02/10/09.
D-6
-------
vvEPA
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
------- |