'is/2pis
Chlorine Dioxide Fumigation of Subway
Materials Contaminated with
anthracis Surrogate Spores: Impact of
Environmental Conditions and
Presence of Dirt and Grime on
Decontamination Efficacy
United States
Environmental Protection
Agency
EPA/600/R-16/038 I June 2016
www.epa.gov/homeland-security-research
Office of Research and Development
National Homeland Security Research Center

-------
EPA/600/R-16/038
Chlorine Dioxide Fumigation of Subway Materials
Contaminated with B. anthracis Surrogate Spores:
Impact of Environmental Conditions and Presence of
Dirt and Grime on Decontamination Efficacy
National Homeland Security Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711

-------
Disclaimer
The U.S. Environmental Protection Agency, through its Office of Research and Development, managed
the research described here under contract EP-C-09-027, Work Assignments 5-12 and 6-12, with
ARCADIS U.S., Inc., and EP-C-15-008, Work Assignment 0-079 with Jacobs Technology, Inc. This study
was funded through the Underground Transport Restoration Program by the U.S. Department of
Homeland Security Science and Technology Directorate under interagency agreement (No. 70-
95866901).
This report has been subjected to the Agency's peer and administrative review and has been approved
for publication. Note that approval does not signify that the contents necessarily reflect the views of the
Agency. Mention of trade names, products, or services does not convey official EPA approval,
endorsement, or recommendation.
Questions concerning this document or its application should be addressed to:
Lukas Oudejans, Ph.D.
Decontamination and Consequence Management Division
National Homeland Security Research Center
U.S. Environmental Protection Agency (MD-E343-06)
Office of Research and Development
109 T.W. Alexander Drive
Research Triangle Park, NC 27711
Phone: 919-541-2973
Fax: 919-541-0496
E-mail: Oudeians.Lukas@epa.gov
i

-------
Acknowledgments
This effort was completed under U.S. EPA contract EP-C-15-008, Work Assignment 0-079 with Jacobs
Technology, Inc. The support and research efforts provided by ARCADIS, U.S., Inc. under EPA contract
EP-C-09-027, Work assignments 5-12 and 6-12, are acknowledged.
We also thank the following entities for their assistance:
MIT Lincoln Laboratory for providing subway building materials and Lawrence Livermore National
Laboratory for the collection of dirt and grime from the undercarriage of a Bay Area Rapid Transit (BART)
subway car.
This effort was managed by the principal investigator (PI) from EPA ORD's National Homeland Security
Research Center (NHSRC) with input from the following project team members:
Lukas Oudejans (PI)
Joseph Wood
Worth Calfee
Office of Research and Development, U.S. EPA
Research Triangle Park, NC 27711
Leroy Mickelsen and Shannon Serre
Office of Land and Emergency Management, U.S. EPA
Research Triangle Park, NC 27711
Additionally, the authors would like to thank EPA Quality Assurance reviewers Eletha Brady-Roberts and
Ramona Sherman for their quality assurance reviews of the QAPP and this Technical Report; and peer
reviewers Francisco Cruz and Larry Kaelin (EPA Office of Emergency Management) and Marshall Gray
(EPA Office of Research and Development) for their significant contributions.
ii

-------
Executive Summary
The Department of Homeland Security's (DHS's) Underground Transport Restoration (UTR) Program is
currently identifying potential methods for rapid characterization, cleanup, and clearance of biological
contamination an underground transit system. This includes physical structures (tunnels and stations) and
rolling stock (railcars). The UTR Project is expected to improve the capability for transit systems to
recover rapidly from a biological release event and thereby address a high-priority need expressed by the
Transportation Security Administration (TSA) and local transit systems. As part of this UTR Project, the
U.S. Environmental Protection Agency (EPA) is evaluating multiple methodologies for the
decontamination of a subway contaminated by a biological agent.
This project supports the U.S. EPA Office of Research and Development's (ORD's) Homeland Security
Research Program (HSRP) mission in that it is providing information relevant to remediation following a
wide area contamination incident. More specifically, this project focuses on the decontamination of a
transportation hub like a subway system, contaminated as a result of an act of terrorism. Remediation
might require the use of volumetric decontamination approaches such as fumigation with chlorine dioxide
(CIO2). While previous HSRP studies have shown this fumigant to be highly efficacious if applied under
specific environmental (temperature and relative humidity [RH]) conditions, it is unclear what the impact of
lower temperatures and the presence of dirt and grime on representative, real-world transportation
system building materials would be on the fumigant efficacy. Lower temperatures affect the absolute
humidity under otherwise constant RH, while the presence of dirt and grime might change the sporicidal
activity of the fumigant. Both scenarios would require changes in operational fumigation conditions to
reach remediation goals.
This report builds on previous research efforts that used relatively high (500-1500 parts per million
volume [ppmv]) CIO2 concentrations to assess the impact of dirt and grime on the decontamination
efficacy under room temperature. The primary objective of the research reported here was to evaluate the
impact of temperature, RH, and dirt and grime present on subway concrete and other subway
infrastructure materials on the fumigation efficacy as measured by reduction of surrogate spores on those
materials. Another objective included determining which sampling procedure provides the best recovery
of viable spores from grimed and cleaned concrete using a prescribed cleaning method from the New
York City Metropolitan Transportation Authority (MTA).
Two sampling techniques were evaluated to determine the most suitable technique for sampling from
subway type materials. The measured recovery of a sponge-stick and the polyester-rayon blend (PRB)
wipe surface sampling method from inoculated grimed subway materials (glazed ceramic tile and painted
steel) were evaluated in this study. The recovery from wetted PRB wipes was found similar (P-value
>0.05) to recovery from sponge-sticks. Since the laboratory methods for processing wetted PRB wipes
are less intensive than the laboratory methods for processing sponge-sticks, PRB wipes were used for
wiping rather than the sponge-stick wipe sampling approach.
The results of this evaluation, as summarized in Table ES-1, indicate that the fumigation conditions
(temperature, RH, CIO2 concentration, and fumigation time) all have a marked effect on the efficacy of the
CIO2 fumigant. A reduction in RH from 75% to 50% at 24 C reduced the efficacy from effective (better
than 6 log reduction [LR]) down to just 1 LR. Lower temperatures (11-13 C [52-55 F]), combined with
the RH near 75%, also reduced the effectiveness of the CIO2 decontaminant dramatically. At these
moderately lower temperatures, the kinetics of the spore deactivation seems to slow down, due to either

-------
the lower temperature and/or the lower (absolute) available amount of water vapor. These results suggest
that although the target RH might be easily attained in a subway system, CIO2 fumigation might not be the
best choice for subway system areas that cannot be heated to at least 24 C.
Table ES-1-1: Summary of CI02 Decontamination Efficacy Values under Various Environmental
Conditions
Test ID
Temperature
(C)
RH
(%)
Concentration
(ppmv)
Longest
Fumigation
Time
(hour (h))
Average Log
Reduction
Min - Max
Log Reduction
Across
Materials
4.1
24
75
1490
6
3.9
CM
LO
I
CO
CO
4.2B
27
76
230
12
6.3
OO
0
4.3
23
75
150
12
4.4
2.9-5.5
4.4
23
50
300
12
0.8
0
1
0
4.5
24
51
180
12
0.8
0
I
CD
O
4.6
11
71
240
12
1.8
1.4-2.3
4.7
11
79
210
24
1.8
1.3-2.1
4.8
27
76
3460
4
5.4
LO
I
CO
CO
4.9
13
68
3280
9
1.8
1.3-2.1
As observed in earlier studies, efficacy was material dependent as indicated by the range of LR values at
a specific fumigation condition. The impact of the type of grimed subway materials (concrete, ceramic tile,
and painted steel) on the CIO2 decontamination efficacy was assessed by a comparison of the number of
viable spores recovered from a grimed surface and from a grimed surface that was cleaned using a
method adapted from the New York MTA prior to the inoculation. The mean log reduction in viable spores
from cleaned surfaces was higher than from grimed surfaces for eight out of ten pairings. However, these
differences were mostly not statistically significant (P-values > 0.05). Only one of the eight direct
comparisons resulted in a statistically significant effect. The impact of the material itself was also less
noticeable than in previous studies where clean surfaces were utilized. This observation can be
explained by the dirt and grime forming a layer with which the CIO2 may interact without impact from the
underlying building material. The dirt and grime may also have reduced the surface porosity.
Implications of this study:
A six (6) log reduction in viable spores can be obtained for the subway infrastructure materials by CIO2
fumigation if the temperature is at or above 24 C combined with RH greater than 75%. These conditions
occur both for 12 h fumigation at 230 ppmv CIO2 or 4 h at 3500 ppmv CIO2. No six log reduction in viable
spores was observed at realistic (winter) temperatures in a subway environment (11-13 C and 70-80%
RH) for periods of fumigation that are otherwise efficacious at 24 C/ 75% RH. Extending the fumigation
time at this low temperature to 24 h at approximately 200 ppmv CIO2 or 9 h at 3300 ppmvCIC>2 did not
improve efficacy. Further research is recommended to identify whether efficacious CIO2 fumigation
conditions may occur at low temperatures through, e.g., a pre-wetting of building surfaces immediately
prior to fumigation.
Limitations:
The results reported here were obtained from a bench scale study. Extrapolation of these results to a full
scale subway station fumigation process should be made with caution. Additional fumigation testing on an
iv

-------
intermediate or large scale would assist in such extrapolation of results. Levels and composition of dirt
and grime may also vary significantly throughout a subway (tunnel) system and by geographical location.
v

-------
Contents
Disclaimer	i
Acknowledgments	ii
Executive Summary	iii
List of Figures	ix
List of Tables	ix
Acronyms and Abbreviations	xi
1	INTRODUCTION	1
1.1	Background	1
1.2	Project Description and Objectives	1
2	MATERIALS AND METHODS	3
2.1	Experimental Setup	3
2.2	General Approach and Test Matrix	4
2.2.1	Task 1: Coupon Preparation	5
2.2.1.1	Coupon Griming	5
2.2.1.2	Coupon Cleaning	6
2.2.1.3	Greased Stubs	7
2.2.2	Task 2: Determination of Surface Sampling Method	7
2.2.3	Task 3: Aerosol Deposition Method	7
2.2.3.1	Surrogate Spore Preparation	7
2.2.3.2	Coupon Inoculation	7
2.2.4	Task 4: Decontamination Tests	8
2.3	Sampling and Analytical Procedures	10
2.3.1	Sample Quantities and Frequency	10
2.3.2	Definition of Efficacy	11
2.3.3	Sampling Procedures	13
2.3.3.1	Wetted PRB wipe Sampling	13
2.3.3.2	Sponge-Stick Wipe Sampling	13
2.3.3.3	Extraction of Greasy Stubs	13
vi

-------
2.3.3.4 Wet Chemistry Samples	14
2.4	Sample Handling and Custody	14
2.4.1	Preventing Cross Contamination during Sampling	14
2.4.2	Sample Identification	14
2.5	Microbiological Methods	15
2.5.1	Wipe Sample Extraction	15
2.5.2	Colony Plating	15
2.5.2.1 Coupon Spore Enumeration	16
3	RESULTS AND DISCUSSION	17
3.1	Task 2: Determination of Surface Sampling Method	17
3.2	Task 3: Aerosol Deposition Method	18
3.3	Task 4: CIO2 Fumigation Testing	19
3.3.1	Fumigation Conditions	19
3.3.2	Efficacy Results	22
3.3.2.1	Effect of Temperature at Constant RH [75%]	22
3.3.2.2	Effect of (Relative) Humidity on Efficacy	24
3.3.2.3	Effect of CIO2 Concentration	25
3.3.2.4	Effect of Washing of Grimed Coupons	26
3.3.2.5	Efficacy for Greasy Stubs Compared to other Materials	27
4	Quality Assurance and Quality Control	28
4.1	Criteria for Critical Measurements	28
4.2	Quality Control Checks	29
4.2.1	Integrity of Samples and Supplies	29
4.2.2	Biocontaminant Laboratory Checks	29
4.3	QA/QC Sample Acceptance Criteria	30
4.3.1 QA/QC Test Results Validation	31
4.4	Calibration of Sampling/Monitoring Equipment	32
vii

-------
4.5	Data Quality Audits	33
4.6	Data Reduction	33
4.7	Data Reporting	34
5	SUMMARY AND RECOMMENDATIONS	35
6	REFERENCES	36
viii

-------
List of Figures
Figure 3-1. Recovery Using Various Sampling Methods. Error Bars Represent 95% CI	17
Figure 3-2. Recovery Using Various Settling Times	19
Figure 3-3. Effect of Temperature on Efficacy for 12-h and 24-h Exposure at 200 ppmv CIO2 and
75% RH. Error Bars Represent the 95% CI	22
Figure 3-4. Effect of Temperature on Efficacy for 12-h Exposure at 100 ppmvCIC>2 and 75% RH.
Error Bars Represent the 95% CI	23
Figure 3-5. Effect of Temperature on Efficacy for 4-h and 9-h Exposure at 3300 ppmv CIO2 and
75% RH. Error Bars Represent the 95% CI	23
Figure 3-5. Effect of Exposure Time on Efficacy at 200 ppmv CIO2 and 11 C/75% RH. Error Bars
Represent the 95% CI	24
Figure 3-6. Effect of Absolute Humidity on Decontamination Efficacy after 12 h at 200 ppmv CIO2.
Error Bars Represent the 95% CI	25
Figure 3-7. Effect of Chlorine Dioxide Concentration on Efficacy at 24 C / 75% and 4-h
Exposure. Error Bars Represent the 95% CI	26
Figure 3-8. The Effect of Washing on Decontamination Efficacy (Test 2B). Error Bars Represent
the 95% CI	27
Figure 3-9. Efficacy for Greased Stubs Compared to Other Materials	27
List of Tables
Table ES-1-1: Summary of CIO2 Decontamination Efficacy Values under Various Environmental
Conditions	iv
Table 2-1. Fumigation Test Matrix for Each Type of Coupon Material	9
Table 2-2. Sample Frequency	11
Table 2-3. Sample Identification	15
Table 3-1. Recovery from Grimed Tile Using Various Sampling Methods	17
Table 3-2. Recovery from Grimed Painted Steel Using Various Sampling Methods	17
Table 3-3. Recoveries at Various Settling Periods	18
Table 3-4. Student's t-test P-values (one-sided)	18
Table 3-5. Average Wet Chemistry and Photometer Readings	20
Table 3-6. Task 4 Fumigation CIO2 Concentration	20
Table 3-7. Task 4 Fumigation RH	21
Table 3-8. Task 4 Fumigation Temperature	21
Table 4-1. Critical Measurement Acceptance Criteria	29
Table 4-2. QA/QC Sample Acceptance Criteria	30
ix

-------
Table 4-3. Other QA/QC Acceptance Criteria	31
Table 4-4. Procedural Blanks and Negative Control QA/QC Results	32
Table 4-5. Sampling and Monitoring Equipment Calibration Requirements	33
Table 4-6. Analysis Equipment Calibration Frequency	33
x

-------
Acronyms and Abbreviations
ADA	aerosol deposition apparatus
AH	absolute humidity
ATCC	American Type Culture Collection
AWWA	American Water Works Association
BART	Bay Area Rapid Transit
CFU	colony-forming unit(s)
CI	confidence interval
CIO2	chlorine dioxide
COC	chain of custody
DAS	data acquisition system
DHS	Department of Homeland Security
DQI	data quality indicator
DTRL	Decontamination Technology Research Laboratory
EMS	Environmental Monitoring System
EPA	U.S. Environmental Protection Agency
h	hour(s)
H2O2	hydrogen peroxide
HSRP	Homeland Security Research Program
ID	identification
in	inch(es)
LR	log reduction
MDI	metered-dose inhaler
MOP	miscellaneous operating procedure
mSM	modified standard method
MTA	Metropolitan Transportation Authority
NA	not applicable
ND	non-detect
NHSRC	National Homeland Security Research Center
Nl	not included
NIST	National Institute of Standards and Technology
ORD	Office of Research and Development
oz	ounce(s)
pAB	pH-adjusted bleach
PBST	phosphate-buffered saline with 0.05% Tween 20
ppmv	parts per million by volume
PRB	polyester-rayon blend
xi

-------
QA	quality assurance
QAPP	quality assurance project plan
QC	quality control
QMP	quality management plan
RH	relative humidity
SD	standard deviation
SE	standard error
SM	Standard Method
TSA	Transportation Security Administration
UTR	Underground Transport Restoration
xii

-------
1 INTRODUCTION
1.1	Background
In the event of a biological incident in a transportation hub like a subway system, effective remediation
might require the use of volumetric decontamination approaches such as fumigation. Previous studies,
conducted under the US Environmental Protection Agency (EPA) Office of Research and Development's
(ORD) Homeland Security Research Program (HSRP), have shown that fumigants like chlorine dioxide
(CIO2) and hydrogen peroxide (H2O2) gases can be highly efficacious if applied under specific
environmental (temperature and relative humidity [RH]) conditions [1], Dirt and grime on the surface of
these real-world building materials, however, can potentially impact the effectiveness of these fumigation
methods. The presence of dirt and grime may possibly change the sporicidal activity of fumigants
requiring changes in operational fumigation conditions to reach remediation goals. Previous work [2] using
concrete commonly found in a subway system focused on CIO2 concentrations of 500 to 1500 parts per
million by volume [ppmv] and fumigation times up to 18 and 6h, respectively. Recent study results [3]
indicate that fumigation of concrete coupons at 100 or 200 ppmvCIC>2 concentrations with an 8 h or
longer contact time can also be effective (defined here as better than 6 log reduction (LR) in the number
of viable spores recovered from a surface). A demonstration of a > 6 log inactivation of viable spores of
an appropriate surrogate spore by a decontaminant is a requirement for product registration as a
sporicidal product/technology against spores of B. anthracis Ames. Decontamination tests conducted
within the HSRP have not addressed the impact of lower fumigation temperatures and/or lower RH. For
example, the average wintertime (December-February) temperature in a Boston subway station was
measured to be approximately 10 C. The effect of such a temperature on the decontamination efficacy
with CIO2 fumigation has not been established. In addition, the previous studies did not address the
impact of abundant dirt and grime on CIO2 fumigation efficacy.
1.2	Project Description and Objectives
The general process investigated in this project was the decontamination of surrogate Bacillus anthracis
(B. anthracis) spores from commonly encountered subway infrastructure surfaces to which grime was
added. CIO2 fumigation was conducted at temperatures and RH values encountered in subways and at
concentrations that may be sustainable in a subway environment. Various fumigation durations up to 24 h
were considered. The impact of the grime on the fumigation efficacy was determined through comparison
of results for grimed coupons that were not cleaned and grimed coupons that were cleaned prior to
inoculation of spores to the surface.
Here, concrete from the floor of the New York Metropolitan Transit Authority (MTA) Old South Ferry
station, glazed ceramic wall tiles from the MTA West 4th Street station, and painted steel surfaces
constructed as representative of painted steel structures in a (New York) subway system were
considered. Sections (coupons) of these materials were covered with grime, and a subset was cleaned
before the start of the fumigation test. Dirt and grime were obtained as scrapings from the undercarriage
of a Bay Area Rapid Transit (BART) subway car. Small coupons with a layer of subway car axle grease
were also part of this investigation.
Coupons were inoculated with Bacillus globigii (Bg) spores, a surrogate of B. anthracis, via aerosol
deposition using a metered-dose inhaler (MDI) inside an aerosol deposition apparatus (ADA) [4], A
1

-------
minimum amount of 1 x 107 colony forming units (CFU) was inoculated to be able to demonstrate a 6 log
reduction in viable spores through fumigation. The inoculation method using the ADA was originally
designed to include a settling time of 18 h for spores to settle onto a 12" x 12" surface. This scheme
would have required the construction and expense of essentially one ADA per coupon per test condition.
As part of this project, different (shorter) settling times were considered to determine whether the settling
time was a necessary component of the inoculation method, considering that the inoculated surface is
only 1.5" by 1.5".
Recovery of Bg spores from the fumigated coupons was compared to recovery from coupons that were
inoculated but not fumigated (positive control coupons) to determine the CIO2 fumigation efficacy on these
subway system materials. Quality control (QC) samples, such as fumigated coupons that were not
inoculated (procedural blank coupons) and coupons that were neither inoculated nor fumigated (negative
controls), were also included to monitor for cross contamination. All samples were analyzed for the
quantitative determination of viable spores.
Initial sampling methods included sponge-stick and wetted polyester-rayon blend (PRB) wipe sampling.
Both were evaluated for their ability to recover spores from grime-painted steel and grimed ceramic tiles.
A decision on which sampling method to use was based on a comparison of recovered spores for each
method. If both methods were found to be satisfactory, the ease of use in the extraction process
determined the selection of the sampling method.
2

-------
2 MATERIALS AND METHODS
2.1 Experimental Setup
All experimentation was conducted in an opaque chamber (830 Series glovebox, Plas-Labs, Inc.,
Lansing, Ml, USA) used to maintain and control a leak-free fumigation atmosphere and to allow for the
periodic addition and removal of sample material during fumigation. CIO2 was generated by a ClorDiSys-
GMP (ClorDiSys, Inc., Lebanon, NJ, USA) system, which passes 2 % chlorine gas in nitrogen through
sodium chlorite cartridges. The generator includes real-time feedback control of CIO2 concentration in the
chamber atmosphere via an internal Environmental Monitoring System (EMS) (ClorDiSys, Inc.)
photometric detector. The concentration of CIO2 in the test chamber was confirmed by using a modified
Standard Method (SM) 4500-CI02-B for Chlorine Dioxide [5] to collect hourly gas samples. A fan inside
the chamber provided internal vapor mixing. Pressure relief valves and check valves prevented over-
pressurization of the chamber.
Humidity within the chamber was controlled by a custom-built data acquisition system (DAS). A
RH/temperature sensor (Vaisala, Vantaa, Finland) was used in a feedback loop to control RH. When the
Vaisala RH sensor read lower than the RH set point, solenoid valves were opened to inject humid air from
a gas humidity bottle (Fuel Cell Technologies, Albuquerque, NM, USA) into the chamber. The gas
humidity bottle, heated to 60 C, passed compressed air through Nafion tubes surrounded by deionized
water, creating a warm air stream saturated with water vapor. Temperature was controlled by circulation
of cooling water from a refrigerating circulator (Isotemp 30165, Fisher Scientific, Waltham, MA, USA)
through radiators placed inside the chamber. HOBO U10 RH sensor/loggers (Onset Computer
Corporation, Bourne, MA, USA) were placed throughout the chamber to assess temperature and RH
spatial variability within the chamber. Figure 2.1 shows a schematic of the configuration used for the
tests.
3

-------
Isolation chamber
Airlock
DAS
Radiators
Cooling
water
RH/temp sensor
mSM-4500-B
Gas
humidity
bottle
EMS CIO2
concentration
CIO2 generator
Digital signal line Heated tubing for gas flow
Digital control line		 Cooling water line
Figure 2-1. CIO2 Fumigation System.
2.2 General Approach and Test Matrix
Four general tasks were associated with this testing:
Task 1: Coupon preparation. Coupons were either constructed or cut to size from existing materials.
Grime was applied to the coupons, and a subset of coupons was washed. Coupons were then inoculated
with the test organism.
Task 2: Determination of surface sampling method. New coupon types (subway tile and painted steel)
were sampled using two techniques (sponge-stick wiping and wetted PRB wiping) to determine suitable
sampling methods.
Task 3: Determination of settling time. A preliminary test was performed to determine if settling time is
predominant for a modified aerosol deposition apparatus (ADA-2R), or if most spores are deposited by
impaction.
4

-------
Task 4: Decontamination tests using techniques from Tasks 2 and 3. Subway material coupons were
fumigated and sampled to determine the fumigation efficacy.
2.2.1 Task 1: Coupon Preparation
This work was intended to explore the efficacy of CIO2 fumigant within subway systems. The concrete
and tile coupon materials were taken from an actual subway system. Concrete pieces from the floor of the
New York MTA Old South Ferry station and glazed ceramic wall tiles from the MTA West 4th Street station
were available for this research. Painted steel coupons (2" x2") were fabricated using materials typical of
steel support beams.
The concrete was previously tested and found not to contain substantial amounts of native grime.
Ceramic wall tiles for initial tests did contain native grime, but due to a finite supply, subsequent tests
were performed with subway-related grime added as discussed in the next section (Section 2.2.1.1).
Painted construction steel was represented by coupons of unpolished low-carbon steel 1/8 inches (in)
thick (P/N 8910K401, McMaster Carr, Atlanta, GA, USA) painted with Rust-Oleum Corporation (Vernon
Hills, IL, USA) primer (P/N 249330) and an acrylic enamel (P/N 249291). Additional stainless steel
coupons (2" x2") were used to determine the number of viable spores deposited onto the coupons and to
assess the stability of the MDI used to inoculate the coupons.
2.2.1.1 Coupon Griming
Coupons without native grime were grimed from material scraped or vacuumed from the undercarriage of
a BART subway car (Figure 2-2). The grime was applied to the sampling area of the coupons until either
the grime was clearly visible or the coupon gained 3 mg of weight. The griming process consisted of
applying the grime using a clean disposable brush, beginning with a circular motion, flattening and
pressing the grime within a 1.5" x 1.5" square drawn on the coupon. The griming process continued until
there was no visible change in the color or distribution of the grime with additional brushing, as illustrated
in Figure 2-3. During the application of grime, each coupon was given a unique alphanumeric
identification (ID). The ID was written on the side of the coupon using a permanent marker (e.g., black or
silver Sharpie). The stainless steel coupons (non-contaminated) were pre-labeled on the underside
using a black Sharpie.
Figure 2-2. Scraping (left) and Vacuuming (right) of Undercarriage of a Subway Car during
Maintenance.
5

-------
Figure 2-3. Concrete Coupon after Grime Application
2.2.1.2 Coupon Cleaning
With the exception of greasy stubs, a subset of coupons was cleaned using a method adapted from the
New York MTA (summarized below):
First, a cleaning solution was prepared. Tide institutional formula floor and all-purpose cleaner (Procter &
Gamble, Cincinnati, OH, USA) (4.2 g) was mixed into 1.5 liters of hot water to create a 0.28 % solution by
weight. This solution was used to clean the surfaces using the following procedure:
1.	Place material coupons flat in a sink with top surface facing up.
2.	Spray coupons with the 0.28 % Tide solution using a foaming spray applicator such as a trigger
sprayer (item # 3U603, Grainger, Lake Forest, IL, USA) connected to a 32-ounce (oz) spray
bottle (item # 3U593. Grainger).
3.	Scrub lightly with a 1,5-in soft pure-bristle paintbrush.
4.	Rinse each coupon well under flowing tap water.
5.	Stand each coupon on edge on a paper towel.
6.	Blow dry the coupons with dry nitrogen to remove surface water.
7.	Let each coupon dry for 24 hours (h) under normal laboratory conditions.
After cleaning, a 1,5-in x 1,5-in square was outlined on each coupon to frame the sampling area and align
the ADA for inoculation.
6

-------
2.2.1.3 Greased Stubs
Greasy stub coupons were manufactured by applying a drop of grease, collected as excessive grease
near an axle of a BART subway car, to the surface of 18-mm aluminum stubs (P/N 16119, Ted Pella, Inc.,
Redding, CA). The grease added was sufficient to cover the stub with approximately 1/8" thick layer of
grease. Greased stubs were included in five of the nine fumigation tests.
2.2.2	Task 2: Determination of Surface Sampling Method
To determine the best sampling method, six replicate grimed tile and grimed painted steel coupons were
inoculated (See Section 2.2.3.2). Two previously evaluated sampling methods, sponge-stick wiping and
wetted PRB wiping, were used to sample three coupons each. A successful sampling method would
collect at least 1 x 107 colony forming units (CFU) and have a precision of 0.5 log. Ease of use and cost
were also factors in determining the best sampling method to be used for the decontamination testing
under Task 4.
2.2.3	Task 3: Aerosol Deposition Method
The aerosol deposition method was required to deliver an average of at least 1 x 107 CFU and have a
precision of 0.5 log. Thus, the settling time for the deposited surrogate spore preparation targeted a
deposition of 1 x 107 CFU on a 1,5-in diameter portion of the subway surface. Four settling times were
tested: 18 h, 1 h, 1 minute, and rapid reuse (no settling time) of the same ADA. Three coupons were used
for each settling time. The coupons were sterilized before use and sampled with wetted PRB wipes.
2.2.3.1	Surrogate Spore Preparation
The test organism for this work was a powdered spore preparation of Bacillus globigii (Bg) (American
Type Culture Collection [ATCC] 9372) and silicon dioxide particles. The resulting powdered matrix
contained approximately 1 x 1011 viable spores per gram, which was prepared by dry blending and jet
milling the dried spores with the fumed silica particles (Degussa, Frankfurt am Main, Germany). This
preparation was loaded into MDIs according to a proprietary protocol. Control checks for each MDI were
included in the batches of coupons contaminated with a single MDI.
2.2.3.2	Coupon Inoculation
Coupons were inoculated (loaded) with the spore preparation to the targeted deposition of 1 x 107 CFU
using an MDI and a modified aerosol deposition apparatus (ADA-2R). The ADA-2R (Figures 2-4 and 2-5)
consists of a round ADA with an O-ring gasket to clamp the ADA to the surface of coupon for inoculation.
The MDI is attached to the top of the ADA and is activated though a slide on the top of the ADA. This
round ADA was designed for deposition of spores onto a 1,5-in x 1,5-in square area of a piece of subway
concrete. The 18-mm greasy stub coupons were inoculated using an MDI attached to an apparatus
described in Lee et al. [4],
Each MDI provides 150 discharges before degradation of concentration. The number of discharges per
MDI was tracked so that use did not exceed this value. Additionally, the weight of each MDI was
determined after completion of the contamination of each coupon. If an MDI weighed less than 10.5 g at
the start of the contamination procedure, the MDI was retired and a new MDI was used. For quality
control of the MDIs, an inoculation control coupon was run as the first, middle, and last coupon inoculated
with a single MDI in a single test.
7

-------
A log was maintained for each set of coupons that were dosed. Each record in this log recorded the
unique coupon identifier, the MDI unique identifier, the date, the operator, the weight of the MDI before
dissemination into the coupon dosing device, the weight of the MDI after dissemination, and the
difference between these two weights.
O-ring
3" stainless steel flange
O-ring
Hose Barb for Vent
Figure 2-4. Schematic of Round ADA.

Figure 2-5. Round ADA with O-ring Gasket.
2.2.4 Task 4: Decontamination Tests
Nine fumigation scenarios were used to decontaminate triplicate coupons of two types, cleaned and not
cleaned. Log reduction was calculated by comparing recovery from fumigated coupons to recovery from
coupons that were inoculated but not fumigated (positive controls). The goal was to provide a 6-log
reduction in CFU under fumigation conditions obtainable in the field. The test matrix is shown in Table 2-
1. The initial focus was on fumigation at low (100 or 200 ppmv) CIO2 fumigation concentrations at 75%
RH and 50% RH (Tests 2/2B - Test 5). Outcomes were compared against a medium high (1500 ppmv)
CIO2 concentration fumigation test (Test 1). Subsequent Tests 6 and 7 investigated the impact of lower
8

-------
temperature with up to 24 h fumigation (RH 75%). Tests 8 and 9 were conducted based on the outcomes
of Tests 6 and 7.
The lower temperature fumigation conditions were selected based on the average temperature in
wintertime (December-February) in a Boston subway station
Table 2-1. Fumigation Test Matrix for Each Type of Coupon Material
Test ID
Concentration
(ppmv)
Exposure Times
(h)
RH
(%)
Temperature
(C)
4.1
1500
0, 2, 4, and 6
75
24
4.2*
200
0, 4, 8, and 12
75
24
4.2B
200
0, 4, 8, and 12
75
24
4.3
100
0, 4, 8, and 12
75
24
4.4
200
0, 4, 8, and 12
50
24
4.5
100
0, 4, 8, and 12
50
24
4.6
200
0, 4, 8, and 12
75
12
4.7
200
0, 12, 18, and 24
75
12
4.8
3300
0,2, 3, and 4
75
24
4.9
3300
0, 3, 6, and 9
75
12
*: RH conditions were outside target range; test was repeated (Test 4.2B)
Testing was conducted in a modified glovebox (Figure 2-6) and proceeded as follows:
1.	Three test coupons per exposure time point and coupon type and one procedural blank coupon
per coupon type (negative control) were loaded into the glovebox.
2.	Coupons were fumigated with CIO2 using the ClorDiSys GMP system according to the
parameters in Table 2-1.
3.	After the exposure time was reached, the coupons were transferred to the airlock where they
were aerated before removal.
4.	Coupons were immediately sampled after removal from the airlock.
5.	Logical groups of samples along with chain of custody (COC) documentation were transferred to
the National Homeland Security Research Center (NHSRC) on-site biocontaminant laboratory in
sterile primary packaging within sterile secondary containment.
9

-------
Figure 2-6. Modified Glovebox - CIO2 Fumigation Chamber.
All test activities were fully documented with narratives in laboratory notebooks, digital photography, and
video. Where appropriate, documentation also included information such as the record of time required
for each decontamination step or procedure, any deviations from the test plan, and physical impact on the
materials.
2.3 Sampling and Analytical Procedures
Wipe samples were collected by sampling within a 1,5-in x 1,5-in sampling template pre-printed on the
coupons prior to inoculation with approximately 1 x 107 spores. For each inoculation event, additional
control samples were collected from stainless steel surfaces as MDI control samples. Within a single test,
surface sampling was completed first for ail blank coupons before sampling any inoculated coupons.
All materials needed for sampling were prepared using aseptic techniques prior to sampling. A sample
collection bin was used to transport samples to the biocontaminant laboratory. The exterior of the
transport container was decontaminated by wiping all surfaces with a bleach wipe ortowelette moistened
with a solution of pH-adjusted bleach (pAB) prior to transport from the sampling location to the NHSRC
biocontaminant laboratory.
2.3.1 Sample Quantities and Frequency
The sample quantities are outlined in Table 2.2. Concrete coupon quantities were limited by the finite
amount of unpainted subway concrete available at the time of this study. Table 2.2 also lists the
frequency of all samples for Task 4 fumigation tests.
10

-------
Table 2-2. Sample Frequency
Sample Type
Quantity
Frequency
Location
Purpose
Test coupon
3 per coupon type
and fumigation
condition
1 set per
fumigation time
Glovebox
(fumigated)
To determine the number of
viable spores after
fumigation
Negative control
coupon
1 per coupon type
1 per fumigation
test
Outside
glovebox; same
room (not
fumigated)
To determine extent of
cross contamination and/or
sterility of coupons
Procedural blank
coupon
1 per coupon type
1 per fumigation
time
Glovebox
(fumigated)
To determine extent of
cross contamination
Positive control
coupon
3 per coupon type;
inoculated as the
first, middle, and
last coupons
1 per inoculation
Outside
glovebox; same
room (not
fumigated)
To determine the number of
viable spores recoverable
from the coupons
MDI control coupons
(stainless steel)
3 per inoculation
event; inoculated
immediately
before each
positive control
coupon
1 per inoculation
Outside
glovebox; same
room (not
fumigated)
To determine the number of
viable spores deposited
onto the coupons and to
assess the stability of the
MDI
NHSRC
biocontaminant
laboratory material
blanks
1 per material
Once per use of
material
Biocontaminant
laboratory
To demonstrate sterility of
extraction and plating
materials
CIO2 monitor
1
Real-time during
CIO2 fumigations
Glovebox
To determine coupon
exposure
mSM* 4500-B CIO2
wet chemistry
samples
Duration
dependent
Once every 60
minutes
Glovebox
To validate operation of
CIO2 real-time monitors
RH/Temperature
1
Logged every 10
seconds
Glovebox
To determine
environmental conditions
during fumigations
NA = not applicable
*mSM = modified standard method
2.3.2 Definition of Efficacy
Efficacy is defined as the extent, by log reduction (LR), to which the agent extracted from the coupons
after the treatment with the decontamination procedure is reduced below that extracted from positive
control areas (not exposed to the decontamination procedure). Efficacy was calculated for each material
within each combination of decontamination procedure (/) and test material (J) as follows:
LR, = Y.^CFUJiNK-Y.to&CFUmVNm	<2~
c=1	k=1
where:
LR = the average log reduction of spores on a specific material surface.
11

-------
T.iz(CFU,>'Nc
the average of the logarithm of the number of viable spores
(determined by CFU) recovered on the control coupons (c = control,
j = coupon number, and Nc is the number of coupons [1,7]).
I'ogOWM
the average of the logarithm of the number of viable spores
(determined by CFU) recovered from the surface of a
decontaminated coupon (S = decontaminated coupon, k = coupon
number, and Nt is the number of coupons tested [1, k]).
The standard deviation of LRj is calculated by Eqn. 2-2:
SDn =1,-S=!	
1, f Ntt-1
(2-2)
where:
SD
j] = standard deviation of ijj.
LR y = the average log reduction of spores on a specific material surface.
_ the average of the log reduction of each from the surface of a decontaminated
''k coupon (Eqn. 2.3-3).
I {LwcFUpVNc-WCFU*))	<2-3>
v _ k C=1	
v'jk
N,
ijk
where:
the "mean of the logs"; the average of the logarithm transformed
^log (CFU )fN- - number of viable spores (determined by CFU) recovered on the
c	,JC ,3 control coupons (C = control, j = coupon number, and Nc is the
number of coupons [1,7])
CFU = number of CFU on the surface of the ktu decontaminated coupon.
The variances (i.e., the square of the standard deviation) of the log CFl/,iC and log CFL/V values were also
calculated for both the control and test coupons (i.e., Sv and S2ijk), and were used to calculate the pooled
standard error (SE) for the efficacy value calculated in Equation 2-1, as follows:
12

-------
[sT~s[
SEy-i-t-t
(2-4)
where the number 3 represents the number j of coupons in the positive control and test coupon data sets.
Each efficacy result is reported as a log reduction with an associated SE value.
The significance of differences in efficacy across different coupon materials and spore types was
assessed based on the 95% confidence interval (CI) of each efficacy result. The 95% confidence interval
(CI) is:
95% CI,, = LRij  (1.96 x SE,,)	(2-5)
Differences in log reduction/efficacy were judged to be significant if the 95% CIs of the two fumigation
results did not overlap.
2.3.3 Sampling Procedures
2.3.3.1	Wetted PRB wipe Sampling
The wetted PRB wipe sampling is typically used for small sample areas and is effective on nonporous
smooth surfaces such as ceramic, metal, and painted surfaces. The general approach is that a moistened
sterile PRB pad (Kendall Versalon Wipe, P/N 8042; Covidien, Minneapolis, MN, USA) is used to wipe a
specified area to recover bacteria, viruses, and biological toxins. The protocol that was used in this
project is adapted from that provided by Busher et al. [6] and Brown et al. [7], with an additional step of
squeezing out excess moisture from the pad before sampling. Normally, the moisture is needed when
sampling a larger surface area, but for smaller areas such as the 1,5-in x 1,5-in sample area used here,
the moisture either drips off or remains on the sampling area and reduces recovery. Wetted PRB wipe
samples were extracted in 20 ml_ of phosphate-buffered saline with 0.05% Tween20 (PBST, Sigma-
Aldrich, P/N P-3563), sonicated, vortexed, and subjected to serial 10-fold dilution and spread-plating.
2.3.3.2	Sponge-Stick Wipe Sampling
Sponge-stick wipe (3M P/N SSL10NB, Saint Paul, MN, USA) sampling was used for evaluation during
Task 2. Sponge-stick wipe samples were collected using two patterns: (1) Using the flat side of the
sponge-stick, the surface was sampled using one or two horizontal strokes covering the entire template
area. (2) The sponge-stick was flipped over to the opposite side to sample the surface in a vertical pattern
covering the entire template area with one or two vertical strokes. This procedure is an abbreviated
version of the method described in detail in Rose et al. [8], which was designed to sample a larger area
than the area used for this study. Sponge-stick samples were extracted in 90 ml_ PBST using a
stomacher (Seward Stomacher bags, P/N BA6041/CLR, Davie, FL, USA) and subjected to 10-fold serial
dilution and spread-plating.
2.3.3.3	Extraction of Greasy Stubs
The 18-mm aluminum stubs with grease were extracted in 10 ml_ PBST, sonicated, vortexed, and
subjected to serial 10-fold dilution and spread-plating.
13

-------
2.3.3.4 Wet Chemistry Samples
The concentration of CIO2 in the gas phase was determined using a modification of American Water
Works Association (AWWA) Standard Method 4500-B-CI02 [5], in which the CIO2 (and any Cb present)
react with potassium iodide in a buffered solution. The CIO2 extractive samples were collected hourly
during CIO2 fumigations. Generally, a dry gas meter box was used to pull (impinge) between 2 and 4 liters
of gas through 50 mL of buffered potassium iodide solution.
2.4 Sample Handling and Custody
2.4.1	Preventing Cross Contamination during Sampling
Sampling poses a significant opportunity for cross contamination of samples. In an effort to minimize the
potential for cross contamination, several management controls were included in the sampling protocol
summarized below:
	In accordance with aseptic technique, a sampling team was used, made up of a sampler, a
support person, and a sample handler.
	The sample handler was the only person to handle ADAs or material coupons during the
sampling event. The sampler handled only the sampling media, and the support person handled
all other supplies.
	Each wipe sample was placed in an individual sterile 50-mL conical tube for primary containment
and then into sterile sampling bags for secondary containment.
	All biological samples from a single test were then placed in a sterilized container for storage and
transport to the NHSRC biocontaminant laboratory. The container was wiped with a towelette
saturated with at least 5000 ppmv hypochlorite solution by weight.
Additionally, and equally important, the order of sampling was as follows: (1) all blank coupons,
(2) decontaminated coupons, and (3) positive control coupons. This order ensured that coupons were
handled in an order from lowest level of contamination to the highest.
2.4.2	Sample Identification
Each coupon was identified with a two-letter description of the material and a unique sample number.
Material codes for samples from inoculated coupons were preceded by an X, as shown in Table 2.3. The
sampling team maintained an explicit laboratory log that recorded each unique sample number and its
associated test number, contamination application, sampling method, and the date sampled. Each
coupon was marked with only the material descriptor and unique code number. Once samples were
transferred to the bio-contaminant laboratory for plate counts, each sample was additionally identified by
replicate number and dilution.
14

-------
Table 2-3. Sample Identification
Coupon Identification: WA-T-(X)MM-H-S-N
Category
Example
Code

WA
12
Work assignment number
T (Test ID)
5
Test ID from Table 2-1

X
Not inoculated

WC
Washed (Clean) Concrete

GC
Grimed Concrete
(X) MM
WT
Washed Tile
(Material)
GT
Unwashed Tile

SS
Stainless Steel (for QC purposes)

F
Field blank

L
Lab blank
H (duration of exposure)
#
Hours of exposure duration. Non-fumigated samples
will have a 0 in this place
S
P
PRB Wipe
(Sample Type)
M
Sponge-stick
N
(Sample Number)
N
Sequential numbers
DCMD Biocontaminant Laboratory Plate Identification: WA-T-(X)MM-H-S-N-R-d
WA-T-(X)MM-H-S-N
As above
R
(Replicate)
R
A-C
d(Dilution)
1
0 to 4, for 10E0 to 10E4
2.5 Microbiological Methods
2.5.1	Wipe Sample Extraction
Wipe kits were prepared with 2.5 mL of PBST in a conical tube with the aseptic addition of wipes (Curity
all-purpose sponges, Covedien, Minneapolis, MN, USA) to the top of the tube. This procedure kept the
wipes dry until they were vortexed immediately prior to sampling, thus wetting the wipe. The wipe was
then swept across the material surface in an "S" pattern to recover any bacterial spores that had adhered
to the surface. The wipes were extracted in 20 mL of PBST and vortexed for two minutes using 10-
second bursts to dislodge the spores into the solution.
2.5.2	Colony Plating
Once extracted, the wipe samples were dilution plated using a 1:10 ratio. The plates were completed in
triplicate at each dilution to determine the average CFU for each sample. The plates were incubated for
15

-------
18-24 h at 35 C. After incubation, the plates were manually enumerated, and dilution sets were recorded
for further analysis. The countable range of CFU was between 30 and 300 spores per sample. Any
samples not meeting this criterion were completed using filters to determine the lowest detection limit
possible.
2.5.2.1 Coupon Spore Enumeration
The NHSRC biocontaminant laboratory quantified the number of viable spores per sample. PBST was
used as the extraction buffer for all sample types. The extraction procedure used to recover spores varied
depending on the different matrices (PRB wipes or sponge-sticks). Details on the PBST wetting agent
and procedures for extraction of spores from PRB wipes and sponge-sticks can be found elsewhere
[9,10], Extraction of spores from the 18 mm stubs with grease followed methods described in Lee et al.
[4], After conducting the appropriate extraction procedure, the extract was subjected to a five-stage serial
dilution (101 to 105) in PBST, and a 0.1 ml volume was spread-plated onto tryptic soy agar in triplicate.
Plates were incubated overnight at 35  2 C and CFU were counted visually. Only plates containing
between 30 and 300 CFU were utilized for recovery estimates. Extracts were diluted and replated if none
of the 10-fold dilutions resulted in all three plates containing colony counts within the acceptable range.
Samples that had fewer than the reportable limit of 30 CFU/plate from the undiluted sample underwent
filter plating. In short, the liquid extract sample was filtered through a 0.2 jjm pore-size filter, which
retained the spores. The filter was then placed on medium where colony forming units were counted. All
extracts were stored at 4  2 C. While there are no EPA-approved methods for spore enumeration from
surfaces, the use of positive control samples as the baseline for log reduction calculations includes a
built-in verification of the deposition and enumeration methods.
16

-------
3 RESULTS AND DISCUSSION
3.1 Task 2: Determination of Surface Sampling Method
Task 2 measured the recovery of the sponge-stick wipe and the PRB wipe surface sampling methods
from inoculated grimed subway materials (tile and painted steel). The same recovery test was conducted
for grimed concrete as part of a previous study [3], The results for grimed tile and grimed painted steel
are displayed in Tables 3-1 and 3-2 and visualized in Figure 3-1. Stainless steel controls were sampled
using the sponge-stick wipe method.
Table 3-1. Recovery from Grimed Tile Using Various Sampling Methods
Sample
Average
Maximum
Minimum
RSD
Sponge-sticks
1.1 x 107
1.4 x 107
7.8 x 10
28%
Wetted PRB wipes
8.6 x 10
1.0 x 107
7.1 x 10
17%
Stainless steel controls
1.2 x 107
1.6 x 107
8.1 x 10
34%
Table 3-2. Recovery from Grimed Painted Steel Using Various Sampling Methods
Sample
Average
Maximum
Minimum
RSD
Sponge-sticks
1.8 x 107
2.3 x107
1.3 x 107
29%
Wetted PRB wipes
2.8 x 107
3.4 x107
2.1 x 107
25%
Stainless steel controls
1.5 x 107
1.9 x 107
1.1 x 107
29%
4.0x107-
3.5x107-
3.0x107-
~C5

1.5x107 -
1.0x107 -
5.0x106-
0.0
| Sponge-stick wipe
j PRB Wipe
CD
&
(D
W en
_Q) "o
"O
CD
E
6
0
E
O
Grimed Tile
Grimed Painted Steel
SS Reference
Figure 3-1. Recovery Using Various Sampling Methods. Error Bars Represent 95% CI.
17

-------
In this study, the recovery from wetted PRB wipes was found to be similar to the recovery from sponge-
sticks (Student's t-test P>0.05 for both grimed materials). The previous recovery study [3] resulted in the
use of the sponge-stick for wipe sampling of grimed concrete based on a lower relative standard deviation
compared to the sampling with the wetted PRB wipe. Since the laboratory methods from processing
wetted PRB wipes are less intensive than processing sponge-sticks, all surfaces were sampled using
wetted PRB wipes for the rest of this study.
3.2 Task 3: Aerosol Deposition Method
The inoculation method using the ADA-2R actuators was originally designed to include a settling time
(18 h) for spores settling into the surface. This approach would have required the construction and
expense of essentially one ADA-2R actuator per coupon per test condition. Task 3 was designed to
determine if the settling time was a necessary component of the inoculation method considering that the
inoculated surface is only 1.5 in by 1.5 in. Coupons were inoculated at different times so that sampling
would occur at the same time. Results of this test are shown in Table 3-3 and Figure 3-2. A comparison of
the different settling times was made using a Student's t-test with the null hypothesis that the two settling
times are the same (P<0.05). One-tail P values were calculated (Table 3-4) to determine whether the
mean recovery for the 18 h settling time was statistically different from the shorter settling time.
Table 3-3. Recoveries at Various Settling Periods
Replicate
Recovery (CFU)
18 Hour
1 Hour
1 Minute
Rapid
1
3.63E+07
3.41 E+07
2.50E+07
2.66E+07
2
4.64E+07
3.52E+07
4.99E+07
3.28E+07
3
4.22E+07
3.64E+07
3.72E+07
2.76E+07
able 3-4. Student's t-test P-values
(one-sided)

18 Hour
1 Hour
1 Minute
Rapid
18 Hour

0.08
0.31
0.01
1 Hour


0.40
0.04
1 Minute



0.18
Rapid




18

-------
Task 3 Recovery
| 60
Q.
0
oz 50
0
Q.
40
30
T3
0
0 20
o
O
0
a.
10
Z)
LL
O
18 Hour Settling Time
11 Hour Settling Time
1 Minute Settling Time
i Rapid Reuse of ADA
(No settling Time)
Replicate
Figure 3-2. Recovery Using Various Settling Times.
The results indicated that recovery following an 18-h settling time was not higher than recovery following
a 1-h settling time (Student's t-test value = 0.08), or a 1-minute settling time (Student's t-test value =
0.31). The use of no settling time led to a significantly lower number of spores deposited when compared
to the 18 h settling time (Student's t-test value = 0.01), supporting the hypothesis that the vast majority of
spores were deposited by impaction with a minimal, non-discernible, contribution by the settling of spores
from the space above the coupon. Consequently, a 1-minute settling time was used as part of the
inoculation procedures of these 1.5" by 1.5" surface areas. The reuse of the ADA did not lead to
accumulation biases and was adopted as the method for all inoculations as part of the fumigation efficacy
testing.
3.3 Task 4: CIO2 Fumigation Testing
3.3.1 Fumigation Conditions
As discussed in section 2.1, two measurement methods were used for determining the CIO2
concentration: a real-time photometer (EMS unit) and the periodic wet chemistry method based on
AWWA SM 4500-CI02 B [5], For the duration of the wet chemistry samples, photometer samples were
also collected. Table 3-5 shows an example ratio of wet chemistry to photometer readings for Task 4
tests. The temporal CIO2 concentrations, based on real-time photometer data, are listed in Table 3-6.
These measurements were standardized to the wet chemistry values using the ratios listed in Table 3-5.
The temporal average RH and temperatures conditions are shown in Tables 3-7 and 3-8, respectively.
19

-------
Table 3-5. Average Wet Chemistry and Photometer Readings
Test No.
Wet Chemistry
(mg/L)
EMS Reading
(mg/L)
Ratio
4.1
4.05
3.67
1.10
4.2*
0.63
0.72
0.87
4.2B
0.62
0.69
0.89
4.3
0.40
0.28
1.39
4.4
0.85
0.87
0.98
4.5
0.44
0.45
0.97
4.6
0.66
0.72
0.92
4.7
0.60
0.61
0.98
4.8
9.44
8.27
1.14
4.9
9.06
8.12
1.12
*: RH conditions were outside target range; test was repeated (Test 4.2B)
Table 3-6. Task 4 Fumigation CIO2 Concentration
Test No.
Target Condition
(ppmv)

Average (SD) CI02 (ppmv)

2h
4h
6h
4.1
1500
1482
41
1482
 56
1487
64


4h
8h
12h |
4.2*
200
258
27
272
 32
277
 32
4.2B
200
230
 33
230
 33
229
 33
4.3
100
124
20
141
 33
146
 32
4.4
200
284
 34
291
 34
298
 35
4.5
100
184
 14
184
 14
185
 14
4.6
200
251
 31
245
 32
236
 31


12h
18h
24h |
4.7
200
218
21
214
21
213
20 |


2h
3h
4h |
4.8
3000
3449
20
3454
 105
3456
 19 |


3h
6h
9h |
4.9
3000
3254
 32
3273
221
3278
42 |
*: RH conditions were outside target range; test was repeated (Test 4.2B)
20

-------
Table 3-7. Task 4 Fumigation RH
Test No.
Target Condition
Average (SD) RH (%)

(%)


4.1
75
75 0
75 0
75 0


4 h
8 h
12 h
4.2*
75
53
 8
65
 12
68
 11
4.2B
75
77
2
76
 1
76
2
4.3
75
75
 0
75
 0
75
 0
4.4
50
50
 0
50
 0
50
 0
4.5
50
53
2
52
2
51
 0
4.6
75
70
 0
70
 0
71
 1


12 h
18 h
24 h |
4.7
75
78
2
79
2
79
2


2 h
3 h
4 h |
4.8
75
76
 0
76
 0
76
0


3 h
6 h
9 h |
4.9
75
68
2
68
2
68
2
*: RH conditions were outside target range; test was repeated (Test 4.2B)
Table 3-8. Task 4 Fumigation Temperature
Test No.
Target Condition

Average (SD) Temperature (C)

(C)






4.1
24
24.1
 0.0
24.3
 0.1
24.4
0.3 |


4h
8h
12h |
4.2*
24
23.2
 0.2
23.6
 0.4
23.7
 0.4
4.2B
24
26.3
 0.2
26.9
 0.5
27.0
 0.6
4.3
24
22.9
 0.0
22.9
 0.0
23.0
 0.0
4.4
24
22.3
 0.1
22.5
 0.2
22.6
 0.0
4.5
24
23.8
 0.1
23.8
 0.1
23.7
 0.0
4.6
10
11.0
 0.2
11.0
 0.2
10.9
 0.2


12h
18h
24h |
4.7
10
11.5
 0.3
11.5
 0.3
11.2
0.5 |


2h
3h
4h |
4.8
24
26.7
 0.2
27.0
 0.3
27.1
0.3 |


3h
6h
9h |
4.9
10
12.5
 0.2
12.8
 0.4
12.9
0.4 |
*: RH conditions were outside target range; test was repeated (Test 4.2B)
The low temperature and high humidity conditions (Tests 4.6, 4.7, and 4.9) were difficult to reach and
maintain. The temperature of the cooling water and radiator was operating below the dew point of the
target condition, causing condensation on cooling lines. The impact of this condensation on the efficacy
test results was deemed minimal or non-existent as it only impacted the amount of water that was
required to maintain the RH condition.
21

-------
3.3.2 Efficacy Results
The aim of this study was to determine the efficacy of the different fumigation scenarios listed in Table 2-1
and evaluate the effect of each critical parameter that drives the efficacy in each of these scenarios. The
results presented in this section evaluate the effect of each critical parameter while maintaining the other
critical parameters constant.
3.3.2.1 Effect of Temperature at Constant RH [75%]
Reducing the temperature from 24C to 11 C dramatically reduced the efficacy of CIO2 fumigation
(Figures 3-3, 3-4, and 3-5). A 12-h exposure at 200 ppmv CIO2 (Test 4.2B) provided greater than 6 LR for
most materials, but at 11 C, the same exposure time led to only a 2 LR (Test 4.6). Extending the
exposure time to 24 h (Test 4.7) did not improve efficacy. Washed painted steel and washed tile were not
included in the 24 h exposure test based on limited availability of subway building materials.

Test 4.2b-24 C-12h
Test 4.6 -11 C - 12h
Test 4.7 -11 C - 24h






































I
1

1


I
1




T


T


















Grimed Concrete Washed Concrete Grimed Painted Washed Painted Grimed Tile Washed Tile
Steel	Steel
Figure 3-3. Effect of Temperature on Efficacy for 12-h and 24-h Exposure at
200 ppmv CIO2 and 75% RH. Error Bars Represent the 95% CI.
Similar results were seen at a 12-h exposure to 100 ppmvCIC>2 (Test 4.3 and 4.5). At this concentration,
CIO2 is not highly efficacious (3.0-5.5 LR range) at 24 C/75% RH for 12 h but still more efficacious than
at the tested 12 C/75% RH for the same 12 h fumigation period.
22

-------
Test 4.3 - 24 C - 12h
Test 4.5-12 C-12h
Grimed Concrete Washed Concrete Grimed Painted Washed Painted
Steel	Steel
Figure 3-4. Effect of Temperature on Efficacy for 12-h Exposure at 100 ppmv CIO2 and 75% RH.
Error Bars Represent the 95% CI.
At a much higher CIO2 concentration of 3300 ppmv (Tests 4.8 and 4.9), the differences are less
distinctive. Here, a 6 LR or greater was achieved for only two materials (grimed concrete and grimed tile).
Nevertheless, less than 2 LR was achieved for all materials at 12 C, even after a 9-h fumigation time.
Washed Concrete
Grimed Painted Steel
Grimed Tile
Figure 3-5. Effect of Temperature on Efficacy for4-h and 9-h Exposure at 3300 ppmv CIO2 and 75%
RH. Error Bars Represent the 95% CI.
23

-------
At the lower temperature, the kinetics of the spore deactivation appear to slow down. As shown in Figure
3.6 (Tests 4.6 and 4.7), a doubling of the exposure time from 12 to 24 h provided only an addition of one
log in LR at most. The upward trend in efficacy with fumigation time up to 18 h is noticeable but does not
continue to the longest fumigation time of 24 h. Linear extrapolations of linear fits through all data (with
zero LR offset at 0 h fumigation) suggest that a 6 LR would be achieved after 48-68 h fumigation,
depending on the material. Such an extended fumigation period was not considered in this study. This
time length may still be biased significantly low if the recorded LR for the 24 h fumigation time is indicative
of a leveling off in LR as function of time. In fact, as observed by Rastogi et al. [11] at 25 C, the
deactivation kinetics may not be linear.

~	Grimed Concrete
 Washed Concrete
A Grimed Painted Steel
	Grimed Tile







1
1


[ t
<
-
>
> i
|

1
: ]

[
:
1
i

1
1
1
0	4	8	12 16 20 24 28
Fumigation Time (h)
Figure 3-5. Effect of Exposure Time on Efficacy at 200 ppmv CIO2 and 11 C/75% RH. Error Bars
Represent the 95% CI.
3.3.2.2 Effect of (Relative) Humidity on Efficacy
Relative humidity is a measurement for the water vapor in air as a percentage of the total amount that
could be held at the identified temperature. A comparison of decontamination efficacy as a function of the
RH value at different temperatures is not useful. Instead, measured temperature and RH were used to
calculate the absolute humidity (AH) [12] which is a water vapor concentration independent of
temperature. The efficacies of the 12-h, 200 ppmv CIO2 fumigation data as a function of the calculated
absolute humidity (Tests 4.2B, 4.4, 4.6, and 4.7) are shown in Figure 3-6. Indicated areas in Figure 3.6
identify the associated temperatures. No strong correlation between absolute humidity and efficacy can
be derived from these data other than that there may be a minimum threshold of water vapor
concentration required to enhance the efficacy of the CIO2 fumigation. Such a threshold can be reached
through a combination of temperature and RH values above which the CIO2 fumigation would be
effective.
24

-------
9.0
Grimed Concrete
Washed Concrete
7.0
Grimed Painted Steel
24 C
Grimed Tile
6.0
c
o
5.0
4.0
3.0
11 C
2.0
1.0
0.0
0
4
8
12
16
20
24
Absolute Humidity (g/m3)
Figure 3-6. Effect of Absolute Humidity on Decontamination Efficacy after 12 h at 200 ppmv CI02.
Error Bars Represent the 95% CI.
3.3.2.3 Effect of CI02 Concentration
Not all materials showed an increase in efficacy with increasing CI02 concentration. Figure 3-7 shows the
average LR for a 4-h exposure at 75% RH at various CI02 concentrations. Grimed tile and grimed
concrete showed the strongest concentration dependence.
25

-------
9.0
Grimed Concrete
Washed Concrete
Grimed Painted Steel
7.0
Grimed Tile
6.0
4.0
3.0
2.0
1.0
0.0
0
1000
2000
3000
CI02 Concentration (ppmv)
Figure 3-7. Effect of Chlorine Dioxide Concentration on Efficacy at 24 C / 75% and 4-h Exposure.
Error Bars Represent the 95% CI.
3.3.2.4 Effect of Washing of Grimed Coupons
A grimed and then washed material (before inoculation) appears to increase the likelihood of a successful
CI02 fumigation as illustrated in Figure 3-8 for four test conditions that resulted in an at least 3 log LR. For
eight of the ten unwashed/washed material pairings, the LR is higher for the grimed and then washed
coupon than the grimed coupon without washing. The variances in recovered viable spores associated
with the low number of replicates (n = 3) per material are unfortunately too high to identify whether these
differences are statistically significant.
26

-------
9
Grimed Tile
Washed Tile
I Grimed Concrete
Washed Concrete
Grimed Painted Steel
Washed Painted Steel
24 C/75% RH - 24 C/75% RH-200 24 C/75% RH-100 24 C/75% RH -
1500ppm-6h ppm - 12h	ppm - 12h 3300ppm-4h
Figure 3-8. The Effect of Washing on Decontamination Efficacy (Test 2B). Error Bars Represent
the 95% CI.
3.3.2.5 Efficacy for Greasy Stubs Compared to other Materials
Greasy stubs were not always present in every fumigation test. Fumigation results for the greasy stubs
under moderately efficacious fumigation conditions are shown in Figure 3-9. The LR of spores on the
greased subs was in line with the LR observed for the other materials.
Test 4.1 24 C/75% RH -1500 ppm - 6h
Test 4.3 24 C/75% RH -100 ppm - 12h
ii j ii,, 11

wwwm
11 ii
1111
Grimed Washed Grimed Washed Grimed Tile Washed Tile Greasy
Concrete Concrete Painted Steel Painted Steel	Stub
Figure 3-9. Efficacy for Greased Stubs Compared to Other Materials.
27

-------
4 Quality Assurance and Quality Control
Quality Assurance (QA)/QC procedures were performed in accordance with the Quality Management
Plan (QMP) and the Quality Assurance Project Plan (QAPP). The QA/QC procedures and results are
summarized below.
All test activities were documented via narratives in laboratory notebooks and the use of digital
photography. The documentation included, but was not limited to, a record for each decontamination test
and any deviations from the QAPP. All tests were conducted in accordance with developed
Decontamination Technologies Research Laboratory (DTRL) and NHSRC Biocontaminant laboratory
Miscellaneous Operating Procedures (MOPs) to ensure repeatability and adherence to the data quality
validation criteria set for this research effort.
The primary objective of this research was to evaluate the impact that environmental conditions
(temperature and RH) and the dirt and grime on underground subway-related construction materials have
on the CIO2 fumigation efficacy. Secondary objectives included determining which sampling procedure
provides better recovery from grimed and cleaned steel and ceramic tile. This section discusses the
Criteria for Critical Measurements (Section 4.1), Quality Control (QC) checks (Section 4.2), and QA/QC
Sample Acceptance Criteria (Section 4.3) to accomplishing the project objectives.
4.1 Criteria for Critical Measurements
Critical measurements were identified to address the stated objectives and specify tolerable levels of
potential errors associated with simulating the prescribed decontamination environments. The following
measurements were deemed to be critical to accomplish part or all of the project objectives:
	Enumeration of spores/CFU on the surface of the subway material coupons;
	CIO2 concentration measurements to characterize the fumigation conditions;
	Measurement of environmental conditions during CIO2 fumigation (temperature and RH)
Data quality indicators (DQIs) for the critical measurements were used to determine if the collected data
met the quality assurance objectives. A list of these DQIs can be found in Table 4-1. Failure to provide a
measurement to meet these goals resulted in a rejection of results derived from the critical measurement.
For instance, if the plated volume of a sample was not known (i.e., was not 100% complete), then that
sample was declared invalid. If a collected sample was lost or did not meet the criteria for other reasons,
then another sample was collected to take its place.
Table 4-1 lists the quantitative acceptance criteria for these critical measurements. Failure to provide a
measurement method or device that met these goals would result in a rejection of results derived from the
critical measurement.
28

-------
Table 4-1. Critical Measurement Acceptance Criteria
Critical Measurement
Measurement Device
Accuracy
Precision
Detection Limit
Plated volume
Pipette
2%
 1 %
NA
CFU/plate
Hand counting
 10 % (between 2
counters)
 10%
1 CFU
CIO2 concentration
mSM 4500-CI02- B
 15% of
photometric value
5%
10 ppmv
Fumigation Time
Timer
 1 min
 1 min
NA
Temperature of Fumigation
Vaisala HMD40Y
5%
3%
NA
RH of Fumigation
Vaisala HMD40Y
5%
3%
NA
NA: Not Applicable
4.2 Quality Control Checks
Numerous QA/QC checks were used in this research effort to ensure that the data collected meet all the
critical measurements listed in Table 4-1. The measurement criteria were set at the most stringent level
that can routinely be achieved. The integrity of each sample during collection and analysis was evaluated.
Validated operating procedures using qualified, trained and experienced personnel were used to ensure
data collection consistency. When necessary, training sessions were conducted by knowledgeable
parties, and in-house practice runs were used to gain expertise and proficiency prior to initiating the
research. The quality control checks that were performed in this project are described in the following
sections.
4.2.1	Integrity of Samples and Supplies
Samples were maintained to ensure their integrity. Samples were stored away from biological standards
or other samples that could cross-contaminate them. While the size and shape of the concrete coupons
or ceramic tiles varied, the size of the inoculation and sampling area did not.
Supplies and consumables were acquired from reputable sources and were National Institute of
Standards and Technology (NIST)-traceable when possible. Upon receipt and prior to use, they were
examined for evidence of tampering or damage. Supplies and consumables that showed evidence of
tampering or damage were rejected and not used. All examinations were documented and supplies were
appropriately labeled. Project personnel checked supplies and consumables prior to use to verify that
they meet specified task quality objectives and did not exceed expiration dates. All pipettes were
calibrated yearly by an outside contractor (Calibrate, Inc., Carrboro NC), incubation temperature was
monitored using NIST-traceable thermometers, and balances were calibrated yearly by the EPA
Metrology Laboratory.
4.2.2	Biocontaminant Laboratory Checks
Quantitative standards do not exist for biological agents. Quantitative determinations of organisms in this
investigation did not involve the use of analytical measurement devices. Instead, CFU were enumerated
manually and recorded. If the CFU count for bacterial growth did not fall within the target range, the
sample was either filtered or re-plated. For each set of results (per test), a second count was performed
on 25 percent of the plates within the quantification range (plates with 30 - 300 CFU). All second counts
were found to be within 10 percent of the original count.
29

-------
4.3 QA/QC Sample Acceptance Criteria
Critical QC checks are shown in Table 4-2. The acceptance criteria were set at the most stringent level
that could be achieved routinely. Positive controls and procedural blanks were included along with the
test samples in the experiments so that well-controlled quantitative values were obtained. Background
checks were also included as part of the standard protocol. Replicate coupons were included for each set
of test conditions. The number of replicates was limited by the availability of the subway building
materials. Further QC samples were collected and analyzed to check the ability of the Biocontaminant
Laboratory to culture the test organism, as well as to demonstrate that materials used in this effort did not
themselves contain spores. The checks included:
	Negative control coupons: sterile coupons that underwent the same sampling process;
	Field blanks: transfer of sterile wipes from the sample tube to an empty conical tube at the
decontamination location;
	Laboratory blank coupons: sterile coupons not removed from the Biocontaminant Laboratory;
	Laboratory material coupons: includes all materials, individually, used by the Biocontaminant
Laboratory in sample analysis; and
	Stainless steel positive control coupons: coupons inoculated but not fumigated.
Additional QA/QC objectives are shown in Tables 4-2 and 4.3. These QA/QC objectives provide
assurances against cross contamination and other biases in the microbiological samples.
Table 4-2. QA/QC Sample Acceptance Criteria	
Sample Type	Purpose	Acceptance Criteria	Corrective Actions	Frequency
Negative Control
Coupons
Determine extent
of cross-
contamination
No detectable spores
Values on test coupons of
the same order of
magnitude will be
considered to have
resulted from cross-
contamination
1 per sample
type
Field Blank Coupons
Verify the process
of moving
coupons does not
introduce
contamination
No detectable spores
Determine source of
contamination and
remove
1 per test
Laboratory Blank
Coupons
Verify the sterility
of coupons
following
autoclaving
No detectable spores
Determine source of
contamination and
remove
3 per test
Laboratory Material
Coupons
Verify the sterility
of materials used
to analyze viable
spore count
No detectable spores
Determine source of
contamination and
remove
3 per
material per
test
Blank Tryptic Soy
Agar Sterility Control
(plate incubated, but
not inoculated)
Controls for
sterility of plates
No observed growth
following incubation
All plates are incubated
prior to use, so any
contaminated plates will
be discarded
Each plate
30

-------
Sample Type
Purpose
Acceptance Criteria
Corrective Actions
Frequency
Positive Control
Coupons
Used to
determine the
extent of
inoculation on the
coupons
1 x107 CFU 0.5 log
Outside target range:
discuss potential impact
on results and correct the
loading procedure for next
test and repeat depending
on decided impact
3 per coupon
type per test
Fumigation Extraction
Blank Samples
Validated
baseline of
extractive
techniques
Non-detect
Obtain new reagents
1 per test
Inoculation Control
Coupons
Used to
determine drift in
the MDI
The CFU recovered from
the first set of positive
controls must be within
0.5 log of the second set
of positive controls
Reject results and repeat
test
3 per
inoculation
Replicate Plating of
Diluted Microbiological
Samples
Used to
determine
variability in CFU
counts
The reportable CFU of
triplicate plates must be
within 100%. Reportable
CFU are between 30 and
300 CFU per plate
Replate sample
Each sample
Table 4-3. Other QA/QC Acceptance Criteria
Sample Type
Purpose
Acceptance Criteria
Corrective Actions
Frequency
mSM 4500-CI02-B
Wet Chemistry
Validate CIO2
concentration
measurements
15% of photometric
reading
Repeat or change CIO2
generator cartridges if
necessary
1 per hour
Post-test Calibration
of RH Sensors
(Vaisala, Helsinki,
Finland)
Used to validate
sensor operation
The post-test calibration
check readings must be
within 5% of target
reading
Reject results. Repeat
test as deemed
appropriate
1 per test
4.3.1 QA/QC Test Results Validation
Plated volume critical measurement goals were met. All pipettes are calibrated yearly.
The QA/QC control test results for the whole sampling campaign are shown in Table 4-4. All field blanks
and inoculum blanks were found to be non-detects (NDs).
Procedural blanks resulted in non-detectable viable spores under fumigation conditions that led to higher
(>4) LR conditions. Recovered spore amounts under less favorable fumigation conditions (e.g., Test 4.4,
4.5, and 4.6) were substantial (1E1 - 1E4 CFU/sample) which can be explained by the cross
contamination caused by the resuspension of spores in the fumigation chamber during fumigation without
deactivation. Under these conditions, recovered procedural blank amounts were still 1-2 orders of
magnitude below the amounts recovered from test coupons. Negative control coupons had detectable
spore amounts (3E0 - 3E3 CFU range) which can be attributed to the cross contamination of these
coupons during the inoculation process. The negative control amounts are insignificant compared to the
inoculated amount of spores (2E6-1E7 CFU range) on test and positive control coupons. The overall
impact of the presence of viable spores on the procedural blanks or negative controls is considered to be
minimal.
31

-------
Table 4-4. Procedural Blanks and Negative Control QA/QC Results

Procedural Blank Recovery (CFU)
Negative Control Recovery (CFU)

Material
Material

0
2
0
c
0
0
~o
<0
E
'C
0
Washed Concrete
Grimed Painted Steel
Washed Painted Steel
Grimed Tile
Washed Tile
Greasy Stub
Grimed Concrete
Washed Concrete
Grimed Painted Steel
Washed Painted Steel
Grimed Tile
Washed Tile
Greasy Stub
4.1
ND
ND
ND
ND
ND
ND
ND
640
17
17
79
10
45
Nl
4.2B
Nl
3
ND
ND
22
ND
-
Nl
860
1760
125
32
135
Nl
4.3
ND
ND
ND
ND
ND
ND
ND
42
42
104
13
785
12
Nl
4.4
11
ND
51
56
ND
8533
ND
413
17
8
2567
12
13
Nl
4.5
1880
6183
690
1350
101
9000
9
43
97
31
42
3
32
Nl
4.6
10E3
3
30
252
570
1
4820
148
42
223
123
607
63
Nl
4.7
ND
105
ND
-
ND
-
-
1320
1110
740
-
540
-
-
4.8
ND
ND
ND
-
ND
-
-
1220
3383
12
-
16
-
-
4.9
ND
ND
ND
-
ND
-
-
22
1500
560
-
25
-
-
ND: non detect (filter plating result of blank/control)
Nl: Not included
Not part of test matrix
The CIO2 photometer calibrations were checked prior to each test and were within the factory
specifications during each fumigation. The primary CIO2 measurements were the modified SM 4500-CI02-
B extractive samples. The accuracy and precision of the titration equipment were checked using a single-
point NIST-traceable standard solution.
4.4 Calibration of Sampling/Monitoring Equipment
Miscellaneous operating procedures for the maintenance and calibration of all laboratory and NHSRC
Biocontaminant Laboratory equipment were followed. All equipment was verified as either being certified
calibrated or having the calibration validated by EPA's Metrology Laboratory at the time of use. The
standard laboratory equipment such as balances, pH meters, biological safety cabinets and incubators
were routinely monitored for proper performance. Calibration of instruments was conducted at the
frequency shown in Tables 4-4 and 4-5. Any deficiencies were noted. The instrument was adjusted to
meet calibration tolerances and recalibrated within 24 h. If tolerances were not met after recalibration,
additional corrective action was taken, possibly including recalibration or/and replacement of the
equipment. All titrants have a certification of analysis with NIST-traceable concentration values.
32

-------
Table 4-5. Sampling and Monitoring Equipment Calibration Requirements
Equipment
Calibration/Certification
Expected Tolerance
Temperature Sensor
Compare to independent NIST
thermocouple annually
 2% full scale
RH Sensor
Compare to three calibration salts once a
week
 5%
Meter box
Volume of gas is compared to NIST-
traceable dry gas meter on an annual basis
 2%
Stopwatch/Clock
Compared against NIST Official time at
http://www.time.qov once everv 30 davs
 1 min/30 days
Buret
Annual gravimetric verification of volume
 1%
Table 4-6. Analysis Equipment Calibration Frequency
Equipment
Calibration Method
Calibration Frequency
Expected Tolerance
Scale
Compared to Class S weights
Before each use
 0.01% of target
Pipettes
Gravimetric
Annually
 1% target value
Burets
Gravimetric
Annually
 1% target value
Pressure Manometer
Compared to NIST-traceable Heiss gauge
Annually
 3% of reading
Incubator thermometer
Compared to NIST-traceable thermometer
Annually
 0.2 C
4.5	Data Quality Audits
This project was assigned QA Category III and did not require technical systems or performance
evaluation audits. Data quality problems, if encountered, usually required immediate, on-the-spot
corrective action. The nature of the problem and corrective steps taken were noted in the project
notebook of record.
4.6	Data Reduction
Data reduction for all tests performed included the total CFU recovered from each replicate coupon, the
average recovered CFU and standard deviation for each group of coupons, and LRs. For each
combination of test coupon material and sample type, the groups of coupons included the following:
	Positive control coupons (replicates, average, standard deviation)
	Test coupons (replicates, average, standard deviation)
	Procedural blank coupons.
Efficacy was defined as the extent (expressed as LR) to which the number of viable spores extracted
from the coupons after the fumigation was reduced in comparison to viable spores extracted from positive
control coupons (not fumigated). When no viable spores were detected, the detection limit of the sample
was used, and the efficacy was reported as greater than or equal to the value calculated by Eqn. 2-1. The
detection limit of a sample depended on the analysis method and could therefore vary. The detection limit
of a plate was assigned a value of 1 CFU, but the fraction of the sample plated varied. For instance, the
33

-------
detection limit of a 0.1 mL plating of a 20 mL sample suspension was 200 CFU (1 CFU/0.1 mL * 20 mL),
but if all 20 mL of the sample were filter-plated, the detection limit was 1 CFU.
4.7 Data Reporting
Data generated included notes recorded in a laboratory notebook (e.g., gravimetric records and
assessment of decontamination solutions) and electronic files created by digital camera. Written records
included observations, numerical data produced by any instrument that was not digitally recorded, and all
variables specific to any experiment. Photographs were taken of each procedure and protocol conducted
in general and of any unusual result. Digital files were maintained in their raw form on each of two
computers in the laboratory, on desk computers used by test personnel, and on the EPA local network for
backup. Processed data files were kept on desk computers and backed up on the EPA network on a
biweekly basis. Two laboratory notebooks at a time were maintained for this project, one in the laboratory
for notes related to the inoculation and sampling procedures, and another in the Biocontaminant
Laboratory for all notes related to biological sample analysis and coupon sterilization documentation.
34

-------
5 SUMMARY AND RECOMMENDATIONS
The primary objective of this investigation was to determine the effect of real-world (subway)
environmental conditions and the presence of dirt and grime on the efficacy of CIO2 to inactivate Bacillus
globigii (Bg) spores, a surrogate of Bacillus anthracis (B. anthracis) spores inoculated onto relevant
subway materials (concrete, ceramic tile, and painted steel).
Two sampling techniques were evaluated to determine the most suitable sampling technique from
subway type materials. The measured recovery of the sponge-stick and the PRB wipe surface sampling
methods from inoculated grimed subway materials (tile and painted steel) were evaluated in this study.
The recovery from wetted PRB wipes was found to be similar or superior to recovery from sponge-sticks.
Since the laboratory methods for processing wetted PRB wipes are less intensive than processing
sponge-sticks, wetted PRB wipes were used to sample all building material surfaces in this study.
The results of this evaluation suggest that not all CIO2 fumigation conditions may result in a 6 LR
decontamination efficacy for subway materials. A > 6 log inactivation of viable spores by a
decontaminant is a requirement for product registration as a sporicidal product/technology against spores
of B. anthracis Ames. A lower temperature of 10-12 C and/or a reduction in RH to 50% at 24 C were
found to reduce the effectiveness of the CIO2 fumigation dramatically. Losses in LR were not recovered
by doubling of the fumigation time (up to 24 h). At the lower temperature, the kinetics of the spore
inactivation seem to slow down, which suggests that, although the target RH might be more easily
reached in a subway system, these results indicate CIO2 fumigation might not be the best choice for
subway system areas that are not heated to at least 24 C or have lower RH. The impact of the type of
subway materials on the CIO2 decontamination efficacy seems to be not very significant, which can be
explained by the presence of the dirt and grime on the surface which essentially reduces the impact of the
actual underlying material on efficacy. Fumigation of a cleaned grimed surface (pre-inoculation) was
found to be only marginally more efficacious than fumigation of a grimed surface.
A six (6) log reduction in decontamination efficacy can be achieved for subway building materials if the
decontamination conditions of temperature greater than 24C with RH at 75% or higher and a required
CIO2 concentration/exposure time of 3000 ppmvfor 3 h or 200 ppmvfor 12 h are achieved.
35

-------
6 REFERENCES
1.	U.S. EPA. Bio-response Operational Testing and Evaluation (BOTE) Project. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-13/168, 2013.
2.	U.S. EPA. Decontamination of a Mock Office Using Chlorine Dioxide Gas. U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-14/208, 2014.
3.	U.S. EPA. Interactions of CIO2 and H2O2 Fumigants with Dirt and Grime on Subway Concrete .
U.S. Environmental Protection Agency, Washington, DC, EPA/600/R-14/226, 2014.
4.	Lee, S. D.; Ryan, S. P.; Snyder, E. G., Development of an aerosol surface inoculation method for
Bacillus spores. Applied and Environmental Microbiology 2011, 77 (5), 1638-1645.
5.	Standard Method 4500- CIO2 Method B. lodometric Method, in: A.D. Eaton, L.S. Clesceri, A.E.
Greenburg, M.A.H. Franson (Eds.), Standard Methods for the Examination of Water and Wastewater,
Nineteenth ed., American Public Health Association, Washington, DC, 1995.
6.	Busher, A. Noble-Wang, J.; Rose L., Surface sampling, in Sampling for Biological Agents in the
Environment. Emanuel, P.; Roos, J.W.; Niyogi, K., Eds. ASM Press: Washington, DC, 2008; p 95-131.
7.	Brown, G. S.; Betty, R. G.; Brockmann, J. E.; Lucero, D. A.; Souza, C. A.; Walsh, K. S.; Boucher,
R. M.; Tezak, M.; Wilson, M. C.; Rudolph, T., Evaluation of a wipe surface sample method for collection of
Bacillus spores from nonporous surfaces. Applied and Environmental Microbiology 2007, 73 (3), 706-710.
8.	Rose, L. J.; Hodges, L.; O'Connell, H.; Noble-Wang, J., National validation study of a cellulose
sponge wipe-processing method for use after sampling Bacillus anthracis spores from surfaces. Applied
and Environmental Microbiology 2011, 77 (23), 8355-8359.
9.	Calfee, M. W.; Ryan, S.P.; Wood, J.P.; Mickelsen, L.; Kempter, C.; Miller, L.; Colby, M.; Touati,
A.; Clayton, M.; Griffin-Gatchalian, N.; McDonald, S.; Delafield, R. Laboratory evaluation of large-scale
decontamination approaches. Journal of Applied Microbiology 2012 112 (5), 874-882.
10.	Tufts, J. A. M.; Meyer, K. M.; Calfee, M. W.; Lee, S. D. Composite Sampling of a Bacillus
anthracis Surrogate with Cellulose Sponge Surface Samplers from a Nonporous Surface. PLoS ONE
2114 9 (12): e114082.
11.	Rastogi, V. K.; Ryan, S. P.; Wallace, L.; Smith, L. S.; Shah, S. S.; Martin, G.B. Systematic
evaluation of the efficacy of chlorine dioxide in decontamination of building interior surfaces contaminated
with anthrax spores. Applied and Environmental Microbiology 2010, 76 (10), 3343-3351.
12.	Relative humidity to absolute humidity calculator, http://planetcalc.com/2167 Last accessed
February 11, 2016
36

-------
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

-------