EPA/600/R-18/158 | August 2018
www.epa.gov/homelarid-security-research
United States
Environmental Protection
Agency
&EPA
Evaluation of Commercial Wet
Vacuums for Bac Spore
Sampling on Surfaces
Office of Research and Development
Homeland Security Research Program
-------
Disclaimer
The United States Environmental Protection Agency (EPA), through its Office of
Research and Development's National Homeland Security Research Center, funded and
managed this investigation through contract number EP-C-15-008, WA (0-1)- 66 with Jacobs
Technology. Data from a previous study (WAs 4-26 and 5-26 under contract No. EP-C-09-027
with ARCADIS U.S. Inc.) were used to measure the efficacy of sample collection performed
under this study. This report was peer and administratively reviewed and has been approved for
publication as an Environmental Protection Agency document. It does not necessarily reflect the
views of the Environmental Protection Agency. No official endorsement should be inferred. This
report includes photographs of commercially-available products. The photographs are included
for purposes of illustration only and are not intended to imply that EPA approves or endorses
the product or its manufacturer. EPA does not endorse the purchase or sale of any commercial
products or services.
Questions concerning this document or its application should be addressed to the
following individual:
Sang Don Lee Ph.D.
Decontamination and Consequence Management Division (DCMD)
National Homeland Security Research Center (NHSRC)
U.S. Environmental Protection Agency (MD-E343-06)
Office of Research and Development
109 T.W. Alexander Drive
Research Triangle Park, NC 27711
Phone: 919-541-4531
Fax: 919-541-0496
E-mail Address: lee.sanqdon@epa.gov
-------
Acknowledgments
U.S. EPA Project Team
Sang Don Lee, Principal Investigator; National Homeland Security Research Center
M. Worth Calfee; National Homeland Security Research Center
Leroy Mickelsen; Office of Land and Emergency Management
U.S. EPA Quality Assurance
Ramona Sherman, National Homeland Security Research Center
Jacobs Technology Inc. Project Team
Abderrahmane Touati, Project Manager
Brian Sechrest
Denise Aslett
Ahmed Abdel-Hady
Kathleen May
Lee Brush
Steve Terll
Zora Drake-Richman
Matt Allen
-------
Executive Summary
This project supports the mission of the U.S. Environmental Protection Agency's (EPA)
Office of Research and Development's (ORD) Homeland Security Research Program (HSRP).
EPA's National Homeland Security Research Center (NHSRC) strives to accomplish the HSRP
mission by providing information, expertise and products that can be widely used to prevent,
prepare for, and recover from public health and environmental emergencies arising from
terrorist threats and incidents.
The existing surface sampling strategy for Bacillus anthracis (B. anthracis) spore attack
from a post-terror incident requires the use of various sampling methods, depending on the
surface type (porous or nonporous). The established comparative surface sampling methods
include wet wipes (for smooth nonporous surfaces), dry vacuuming (for rough and porous
surfaces), and wet swabs (for small and/or hard to sample areas such as keyboards). The
existing methods may be labor-intensive, costly, and time-consuming for wide area incident
response.
The objective of this work was to develop and optimize a wet vacuum cleaner-based
sampling method so that this widely-available commercial device could be used for sampling
spores on both porous and nonporous surface types. Such a sampling device would use a
liquid sampling medium that could be analyzed directly without an extraction step (spore
recovery from the sampling medium) required for other surface sampling methods. This direct
analysis could potentially increase recovery efficiency while reducing the sample analysis
turnaround time and cost.
The main objective of this project was to assess an alternative cost-effective, reliable,
commercially available (or built with off-the-shelf materials) wet vacuum cleaner that could be
used for sampling Bacillus spores (i.e., surrogates of B. anthracis) on both porous and
nonporous surfaces. The technical approach for this study involved bench-scale research, as
part of Phase I tests, on the effectiveness of sampling liquids and operational parameters (such
as elapsed time and liquid volume) to sample spores from different surfaces.
A custom-made vacuum sampling device (built with off-the-shelf materials) was used to
sample spores from representative flooring surfaces. As part of Phase II tests, four classes of
commercially-available wet vacuum, also known as carpet cleaners (portable, residential,
commercial and wet/dry cleaners) were evaluated for spore sampling efficiency on nonporous
surfaces (vinyl flooring) as well as porous surfaces (concrete and carpet). The evaluation criteria
for down-selection of the vacuum cleaners to test included vacuum efficiency, availability, ease of
use, ability to access remote areas and cost.
Phase I: Evaluation of Wet Vacuum Cleaner Operational Parameters
The Phase I study evaluated the sampling efficiency of wet vacuum cleaners as a
function of liquid agent, elapsed time between liquid application and suction, and liquid volume
used to perform the needed wetting process. Three material types were used for this evaluation:
carpet, concrete, and laminated wood, inoculated with Bacillus spores (surrogates for Bacillus
anthracis) of 2 x 106 colony forming units (CFU)/ft2. The Phase I study results showed that
-------
deionized (Dl) water amended with Tween® liquid achieved the highest recovery among the
tested liquid/material combinations with an average recovery of 53% on laminated wood
coupons, while Dl water resulted in the lowest average recovery of 9.3% when used with
concrete coupons. The spore recoveries were found to be dependent on the recovered
volumetric fraction of liquid sprayed onto the coupons. The highest recovered liquid volume was
from laminated wood (72-80%), followed by carpet (39-49%), and lastly concrete (16-19%).
Tests were conducted to determine optimal elapsed time between liquid application and
suction. The results showed that the elapsed time between the liquid spray application and wet
vacuum sampling of the target material had little or no effect on the wet vacuum sampling spore
recovery for both laminated wood (29 ± 3.9%), and carpet materials (31 ± 4.9%) over a range of
1 to 300 seconds. Due to rapid liquid absorption into the concrete paver surface, the elapsed
time could not be varied for sampling efficiency. Concrete surface tests were conducted by
dividing the sampling area per coupon into one (no division), two and four sections. The test
results showed the increased spore recovery efficiency with more divided sampling area
(quicker liquid retrieval). The spore recovery increased from an average of 15% for one section,
29% for two sections, to 59% for four sections per coupon.
The total volume of liquid sprayed onto carpet material had a strong effect on the overall
spore recovery for a constant elapsed time between the time the liquid medium was sprayed
and the time the surface was sampled. For carpet, the spore recovery increased from 3.4% to
31% when the liquid application volume was increased from 44 to 111 ml_/ft2. For laminated
wood, the effect of the volume sprayed for the spore recovery seemed to be negligible, with an
average recovery varying between 31 to 38% when the liquid volume was increased from 8.9 to
22 ml_/ ft2. For concrete, the liquid volume seems to have a negligible effect, if any, on the
overall spore recovery with spore recoveries approximately 37-38% when the liquid volume was
increased from 70 to 100 ml_/ft2.
The Phase I study concluded that the optimal wet vacuum sampling conditions were Dl
water amended with Tween® liquid, a short elapsed-time (less than 20 seconds) between liquid
application and suction, and a minimum of 100 ml_/ft2 of liquid collected for optimum spore
sample recovery.
Phase II: Commercially-Available Wet Vacuum Cleaner Evaluation
The Phase II study evaluated commercially available wet vacuum cleaners for surface
spore sampling efficacy. The wet vacuums were selected by the project team considering the
information from the Consumer Reports1. The selection criteria were ease of use, separate
cleaning and recovery of (dirty) tanks, suction power, portability, cost and a heated cleaning
option.
1 Consumer Report. 2015. Carpet Cleaners of 2015. The reader will require subscription to
Consumer Reports, https://www.consumerreports.org/products/carpet-cleaner/ratinqs-overview/
-------
Phase II evaluated three classes (residential, commercial, portable) of commercially-
available wet vacuum cleaners as well as a wet/dry vacuum (Shop-Vac) for their effectiveness
in sampling Bacillus spores. Three material types were used for this evaluation: carpet,
concrete, and vinyl flooring, inoculated with either B. atrophaeus var. globigii (Bg), or B.
thuringiensis subsp. kurstaki (Btk), both surrogates of B. anthracis, at a spore concentration of
107-10s CFU/ft2.
Each vacuum cleaner was evaluated based on cleaning patterns and time listed in the
ASTM F-1284-09 (ASTM. 2009) standard method for evaluating dirt removal effectiveness of
residential vacuum cleaners from carpet surfaces. The operational aspects and ratings were
reviewed and the following three commercially available wet vacuum cleaners and one dry
vacuum were selected for testing:
• Bissell Little Green portable wet-vacuum cleaner,
• Rug Doctor ProX3 commercial wet-vacuum cleaner,
• Hoover Dual Steam wet-vacuum cleaner,
• Shop-Vac wet/dry vacuum cleaner.
The sampling efficiencies of the selected wet vacuums were assessed by comparing
their recoveries to the recoveries obtained by currently available surface sampling methods.
Sampling efficiencies for porous surfaces (carpet and concrete) were determined by comparing
the recoveries obtained by the four vacuum cleaners to the recoveries obtained by the existing
sampling methods such as vacuum sock (carpet) and 37 mm cassette (concrete) (Calfee et al.,
2013), respectively. For the nonporous surface (vinyl flooring), the wet vacuum sampling
approach was compared to the Polyester Rayon Blend (PRB) wipe sampling method.
The overall results show that sampling via wet vacuum is a viable alternative to these
traditional sampling methods. All wet vacuum cleaner spore recoveries were within the order of
magnitude of the material-specific EPA-accepted sampling methods (PRB wipe, vacuum sock,
and 37-mm cassette).
A two-way analysis of variance (ANOVA) was performed to examine the two different
categorically independent variables (material type/sampling method) on the sampling efficacy of
the wet vacuum cleaners for spores. The results of the analysis (72 samples: four vacuum
cleaners, three materials, two surrogates, sampled in triplicate), demonstrated that the effect of
the material type on the mean recoveries is not statistically significant for all the types of
vacuum cleaners (F-value = 0.446, p-value = 0.642), while the effect of sampling methods on
the mean recovery is statistically significant (F-value = 3.03, p-value = 0.036). The interaction of
the two factors showed no significant difference in the mean recovery (F-value = 2.06, p-value =
0.07) at the 0.05 level.
The overall spore recovery efficiencies for the wet vacuum cleaners, independent of
material and spore types, varied between 32 + 20% for the portable, 25 + 26% for Shop-Vac, 33
+ 17% for the residential, and 55 + 52% for the commercial wet vacuum cleaner. In terms of
both usability and repeatability, the operators chose the residential wet vacuum cleaner as a
better sampling option over other tested cleaners for a wide area sampling of Bacillus spores.
v
-------
This is based on the operators' assessment for its lowest RSD (51%), weight, application speed,
and less prone to cross-contamination.
-------
Contents
Disclaimer
Acknowledgments
Executive Summary i
Figures
List of Tables x
Acronyms and Abbreviations xi
1. Introduction 1
2 Study Approach 1
2.1 Phase I: Evaluation of Wet Vacuum Cleaner Operational Parameters 1
2.1.1 Liquid for Sampling 1
2.1.2 Elapsed Time 1
2.1.3 Liquid Volume 1
2.1.4 Material and Equipment 1
2.1.4.1 Wet Vacuum Sampling Device 1
2.1.4.2 Liquid Agent Spray 2
2.1.4.3 Electric Backpack Sprayer 2
2.1.5 Coupon Preparation 2
2.1.5.1 Carpet Coupons 2
2.1.5.2 Concrete Coupons 2
2.1.5.3 Laminated Wood Coupons 2
2.1.5.4 Stainless-Steel Coupons 2
2.2 Phase II - Commercially-Available Wet Vacuum Cleaner Evaluation 2
2.2.1 Wet Vacuum Cleaners 2
2.2.1.1 Residential Cleaners 2
2.2.1.2 Commercial Cleaners 2
2.2.1.3 Portable Cleaners 2
2.2.1.4 Wet/Dry Vacuum Cleaners 2
2.2.2 Phase II Coupon Preparation 2
2.2.2.1 Phase II Carpet and Vinyl Coupons 2
2.2.2.2 Phase II Concrete Coupons 3
2.2.3 Test Matrix 3
2.3 Testing and Sampling Approaches 3
2.3.1 Material and Equipment Sterilization 3
2.3.2 Spore Preparation 3
vii
-------
2.3.2.1 Bacillus atrophaeus var. globigii 33
2.3.2.2 B. thuringiensis subsp. Kurstaki (Btk) 34
2.3.2.3 MDI Preparation 34
2.3.2.4 MDI Spore Concentration Validation and Spatial Distribution 34
2.2.3 Inoculation of Coupons 38
3 Sampling Procedures 40
3.1.1 Wet Vacuum Sampling Kit Preparation 40
3.1.2 Spraying Sequence 41
3.1.3 Sampling Sequence 42
3.2 Phase II: Commercially Available Wet Vacuum Cleaner 43
3.3 Wipe Sampling 44
3.4 Vacuum Sock Sampling 45
3.5 Cassette Sampling 45
3.6 Swab Sampling 45
3.7 Liquid Collection 45
4 Analytical Procedures 47
4.1 Sample Extraction 47
4.1.1 Wipes 47
4.1.2 Vacuum Socks 47
4.1.3 Small vacuum 37-mm cassettes 47
4.2 Spiral Plating and Filter Plating 47
4.3 Wet Vacuum Sample Processing 49
5 Results and Discussion 50
5.1 Phase I: Evaluation of Wet Vacuum Cleaner Operational Parameters 50
5.1.1 Selection of Sampling Liquid 50
5.1.1.1 Liquid Volume Recovery 50
5.1.1.2 Spore Recovery as a Function of Liquid Type 51
5.1.2 Elapsed Time 53
5.1.2.1 Sample Liquid Recovery Volume 53
5.1.2.2 Spore Recovery as a Function of Elapsed Time 55
5.1.3 Liquid Volume 57
5.1.3.1 Sample Liquid Recovery Volume 57
5.1.3.2 Spore Recovery as a Function of Liquid Volume Sprayed 59
5.1.4 Phase I: Summary 60
5.2 Phase II - Commercially-Available Wet Vacuum Cleaner Evaluation 60
5.2.1 Control Sample Recoveries 61
viii
-------
5.2.2 Wet- Vacuum Cleaner Evaluation 61
5.2.2.1 Carpet 61
5.2.2.2 Vinyl Flooring 63
5.2.2.3 Concrete 65
5.2.3 Phase II: Summary. 66
6 Quality Assurance and Quality Control 68
6.1 Criteria for Critical Measurements/Parameters 68
6.2 Integrity of Samples and Supplies 69
6.3 NHSRC BioLab Quality Checks 69
7 References 71
APPENDIX A: PHASE II: DATA REPORT 73
C1: Bg Spores 73
a) Carpet Results 73
b) Concrete Results 74
c) Vinyl Results 75
C2: Btk Spores 76
a) Carpet Results 76
b) Concrete Results 77
c) Vinyl Results 78
ix
-------
Figures
Figure 2-1. Wet Vacuum Sampling Assembly 19
Figure 2-2. Spray Box with Lid in Open Position 20
Figure 2-3: 16-in. x 16-in. Stainless-Steel Spray Template 21
Figure 2-4. Electric Backpack Sprayer 21
Figure 2-5. Carpet Coupon 23
Figure 2-6. Concrete Coupons 23
Figure 2-7. Laminated Wood Coupon 24
Figure 2-8. Stainless Steel Coupon 24
Figure 2-9. Wet Vacuuming with Residential Vacuum Cleaner (Hoover) 25
Figure 2-10. Wet Vacuuming with Commercial Vacuum Cleaner (Rug Doctor) 26
Figure 2-11 Wet Vacuuming with Portable Vacuum Cleaner (Bissell ProHeat) 27
Figure 2-12. Wet Vacuuming with Wet/Dry Vacuum Cleaner (Rigid ProPack) 27
Figure 2-13 Phase II Test Coupon Schematic 28
Figure 2-14. Carpet Coupon 29
Figure 2-15. Vinyl Coupon 29
Figure 2-16 Concrete Coupons 30
Figure 2-17. Sampling Approach 32
Figure 2-18. MDI Actuator Adapter for Small, 18 mm Coupons (A), Catalent MDI
Canister (B), Actuator Adapter (C) 35
Figure 2-19. RMCs Placed on 14 in. x 14 in. Stainless Steel Coupon 36
Figure 2-20.ADA Used on 14 in. x 14 in. Stainless Steel Coupon 36
Figure 2-21. MDI Content Mixing and Purging Prior to Inoculation 37
Figure 2-22. Color Coded MDI Spore Distribution Heat Map over 40 RMCs after
Actuation of Bg Canister #1 (CFUs per RMC coupon, 2 in2) 37
Figure 2-23. Phase I Carpet Coupon with Skirt and ADA 38
Figure 2-24. Inoculation of Phase II Test Coupons 39
x
-------
Figure 3-1. Wet Vacuum Adapter 40
Figure 3-2. Uncapped Wet Vacuum Sampling Kit 41
Figure 3-3. Wet Vacuum Sampling 43
Figure 3-4. Wet Vacuuming Technique 44
Figure 4-1. Bg (panel A) and Btk (panel B) Bacterial Colonies (CFU) on a Spiral-plated
Agar Plate 48
Figure 4-2. Bg (A) and Btk (B) Bacterial Colonies (CFU) on a Filter Plate 48
Figure 5-1. Spore Recovery for Material/Liquid Collection Combination 52
Figure 5-2. Spore Recovery at Different Elapsed Times 56
Figure 5-3. Spore Recovery for Different Number of Partitions on the Concrete Coupon 57
Figure 5-4. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Carpet 62
Figure 5-5. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Vinyl Flooring 64
Figure 5-6. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Concrete 66
Figure 5-7. Sampling Efficacy of the Various Sampling Methods for Bacillus Spores
Independent of Type of Material 67
xi
-------
List of Tables
Table 2-1. Sampling Liquid Test Matrix (Task 1) 17
Table 2-2. Elapsed Time Test Matrix (Task 2) 18
Table 2-3. Liquid Volume Test Matrix (Task 3) 19
Table 2-4. Description of Test Coupon Materials and Material Preparation 22
Table 2-5. Phase II Material Description 28
Table 2-6. Phase II Test Matrix 31
Table 5-1. Volume of Liquid Recovered from Carpet Coupons 50
Table 5-2. Volume of Liquid Recovered from Concrete Coupons 51
Table 5-3. Volume of Liquid Recovered from Laminated Wood Coupons 51
Table 5-4. Spore Recovery as a Function of Liquid Collection Type 52
Table 5-5. Volume of Liquid Recovered from Carpet at Various Elapsed times 54
Table 5-6. Volume of Liquid Recovered from Different Concrete Coupon Surface
Partitions 54
Table 5-7. Volume of Liquid Recovered from Laminated Wood at Different Elapsed
Times 55
Table 5-8. Spore Recovery as a Function of Elapsed time for Carpet and Laminated
Wood Coupons 56
Table 5-9. Spore Recovery as a Function of Number of Partitions on the Concrete
Coupon 57
Table 5-10. Liquid Volume Recovery from Carpet Coupon 58
Table 5-11. Liquid Volume Recovery from Concrete Coupon 58
Table 5-12. Liquid Volume Recovery from Laminated Wood Coupon 59
Table 5-13. Spore Recovery from Carpet and Laminated Wood Coupons 59
Table 5-14. Spore Recovery from Concrete Coupon 60
Table 5-15. Spore Recovery from Concrete Coupon 61
Table 5-16. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spore
Recoveries from Carpet 62
xii
-------
Table 5-17. Sampling Efficacy of the Various Sampling Methods for Spores (Btk and Bg
data pooled) Inoculated on Vinyl Flooring 63
Table 5-18. Tukey Pairwise Statistical Test Results for Various Sampling Methods on
Carpet 63
Table 5-19. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Recovery on Vinyl Flooring 64
Table 5-20. Sampling Efficacy of the Various Sampling Methods for Both Btk and Bg
Spore Recovery on Vinyl Material 65
Table 5-21. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spore
Recovery on Concrete Material 65
Table 5-22. Sampling Efficacy of the Various Sampling Methods for Both Btk and Bg
Spore Recovery on Vinyl Material 66
Table 5-23. Two-Way ANOVA on the Mean Sampling Efficacy of the Wet Vacuum
Cleaners 67
Table 5-24. Overall Sampling Efficacy of the Various Sampling Methods 67
Table 6-1. DQIs and Acceptance Criteria Validation for Critical Measurements 69
Table 6-2. Additional Quality Checks for Biological Measurements 70
xiii
-------
Acronyms and Abbreviations
ADA Aerosol Deposition Apparatus
ANOVA analysis of variance
B. Bacillus
Ba Bacillus anthracis
Bg Bacillus atrophaeus var. globigii
BioLab NHSRC RTP Microbiology Laboratory
Btk Bacillus thuringiensis subsp. kurstaki
cm centimeter(s)
CFU Colony Forming Unit(s)
COTS Commercial Off-The-Shelf
Dl Deionized
DTRL Decontamination Technologies Research Laboratory
DQI Data Quality Indicator
DQO Data Quality Objective
EPA U.S. Environmental Protection Agency
ft foot (feet)
g gravity
HEPA high-efficiency particulate air
HSRP Homeland Security Research Program
H2O2 Hydrogen Peroxide
in. inch(es)
L liter(s)
Lpm liter(s) per Minute
m meter(s)
MDI Metered Dose Inhaler
|jm micrometer(s)
mL milliliter(s)
mm millimeter(s)
min minute(s)_
MOP Miscellaneous Operating Procedure
NHSRC National Homeland Security Research Center
NIST National Institute of Standards and Technology
OP operating procedure
ORD Office of Research and Development
OSB oriented strand board
ppm part(s) per million
PPE Personal Protective Equipment
PRB Polyester Rayon Blend
psi pound(s) per square inch
QA quality Assurance
QC quality control
RMC Reference Material Coupon
-------
rpm
revolution(s) per minute
RSD
relative standard deviation
RTP
Research Triangle Park
sec
second(s)
TSA
tryptic soy agar
VHP®
Vaporous Hydrogen Peroxide
XV
-------
1. Introduction
The U.S. Environmental Protection Agency's (EPA's) Homeland Security Research
Program (HSRP) strives to provide expertise and products that can be widely used to prevent,
prepare for, and recover from public health and environmental emergencies arising from
terrorist threats and other contamination incidents. HSRP conducts research to provide
expertise and guidance on the selection and implementation of surface sampling methods that
may ultimately provide the scientific basis for a significant reduction in the time and cost of
sampling surfaces contaminated with Bacillus anthracis.
The currently-used surface sampling methods for biological agents include swabs,
wipes, and vacuums fitted with filter-type collection media. Individual methods are material-
dependent for application and limited in sampling area (1-4 square feet (ft2) per sample).
These methods may be physically demanding for application in a wide area incident. To
improve the sampling capability for responding to a wide area incident, this study evaluated
commercial wet vacuums as a sampling tool. The wet vacuums are applicable on both porous
and nonporous surfaces, widely available, and easy to operate for collection of biological
agents. In addition, the wet vacuum can sample more than 100 ft2 per sample and generate
liquid samples that may reduce the post-collection processing steps.
This study is composed of two phases: Phase I, Evaluation of Wet Vacuum Cleaner
Operational Parameters and Phase II, Evaluation of commercially Available Wet Vacuum
Cleaners. Phase I optimized the collection efficiency of a custom-made wet vacuum-based
surface sampling device on both porous and nonporous surface types through a set of
controlled operating parameters. The parameters tested for the wet vacuum sampling were
type of liquid (Task 1), temporal lapse between liquid application and suction (Task 2), and
liquid volume (Task 3). The vacuum-based sampling technique consisted of an optimized
sampling nozzle along with a liquid dispenser and a liquid collection sample vessel that would
eliminate post-collection processing since agents were to be captured directly into a liquid that
could be analyzed. As part of Phase II tests, commercial wet vacuums were evaluated to
determine their effectiveness for spore surface sampling on both porous and nonporous
surfaces, using the results of Phase I. A field-usable operating procedure (OP) was developed,
based on the results of the Phase II study.
16
-------
2 Study Approach
The Phase I experimental approach consisted of selecting a sampling liquid, optimizing
the temporal lapse between liquid application and suction, and evaluating the volume of liquid
collected as a function of surface type. The outcomes of Phase I were assessed to determine
the parameters resulting in an optimal sampling approach before moving on to Phase II for
evaluating commercially-available wet vacuums. The experimental approach for each phase is
presented in the next section.
2.1 Phase I: Evaluation of Wet Vacuum Cleaner Operational Parameters
A custom-made wet vacuum cleaner was designed to collect liquid samples from wet
surfaces contaminated with Bacillus atrophaeus var. globigii (Bg), a surrogate for the spore-
forming bacterial agent Bacillus anthracis. Three material types were investigated: carpet,
concrete, and laminated wood. The parameters tested for the wet vacuum sampling were type
of liquid, temporal lapse between liquid application and suction, and liquid volume. Spore
recoveries from the tested vacuum samplers were compared to recoveries from Polyester
Rayon Blend (PRB) wipe samples collected from stainless steel coupons.
Preliminary work under this project demonstrated that a typical wet vacuum cleaner
collects between 1.5 and 2.5 liters per minute (Lpm) of liquid during sampling. Between 20%
and 50% of the liquid applied was recoverable from carpet. Typical cleaning operation using a
wet vacuum cleaner is performed at a rate of approximately 2.5 seconds per stroke. (ASTM.
2009) The test coupons were sampled at a linear speed of three to five seconds per square foot
throughout the entire coupon. A 45-millimeter (mm) vacuum nozzle attached to a wet vacuum
adapter was used to sample the coupons. Coupons were vacuumed by traversing the coupon
using overlapping strokes in one direction, then again in a second direction at 90-degrees to the
first.
2.1.1 Liquid for Sampling
Three types of liquid (deionized (Dl) water, phosphate-buffered saline with Tween® 20
(PBST), and Dl water with Tween® 20 at 0.05% concentration) were tested for sample collection
on three material types (laminated wood, carpet, concrete) inoculated with a target spore
surface concentration of 2 x 106 colony forming units (CFU)/square foot (ft2). The test matrix for
this task is shown in Figure 2-1.
Table 2-1. Sam
pling Liquid Test Matrix (Task 1)
Test ID
Material
Variable: Liquid types (ID code)
1A
Carpet
Dl water (W), PBST (P), Dl water with Tween® (T)
1B
Concrete
Dl water (W), PBST (P), Dl water with Tween® (T)
1C
Laminated Wood
Dl water (W), PBST (P), Dl water with Tween® (T)
17
-------
2.1.2 Elapsed Time
Three different elapsed times for the liquid application and vacuuming were assessed.
Based on the results from Task 1, this optimization test was conducted using only the most
efficient extraction liquid, Dl water with 0.05% Tween® 20 (Dl-Tween). The total volume of liquid
applied for sampling was 250 ml_ for carpet coupons, 400-500 ml_ for concrete coupons, and
100 ml_ for laminated wood coupons. Different liquid volumes per surface type were determined
based on the collected liquid volume, targeting 50-100 ml_ for analysis. For carpet and
laminated wood, the elapsed time between liquid spraying and collection was tested at 1, 30,
100, and 300 seconds (sec). For the one-sec elapsed time, the liquid was suctioned
immediately following spraying.
For the concrete coupons, the surface liquid was retrieved immediately after the liquid
spraying was completed. Due to the high liquid-absorptive nature of the material, tests were
conducted in which the surface area was partitioned into smaller testing areas to increase the
liquid volume for collection. For the first test, the entire 500 milliliter (ml_) volume was sprayed
onto the concrete coupon and then vacuumed. For the second test, each half of the coupon was
sprayed and vacuumed consecutively using half of the volume (250 ml_) on each half. In the
third test, the coupon was sprayed and vacuumed consecutively in quarters, using 125 ml_ on
each quarter. For each concrete test, each coupon's sample was cumulative (comprised of 1, 2
or 4 sub-samples, respectively) and the same liquid collection system was used. The test matrix
is shown in Table 2-2.
Table 2-2.
Elapsed Time Test Matrix (Task 2)
Test ID
Material
Volume Sprayed (mL)
Variable: Elapsed Times (sec)
2A
Carpet
250
1b, 30, 100, 300
2C
Laminated Wood
100
1, 30, 100, 300
2B1
1ax 500
1
2B2
Concrete
2x250
1
2B3
4x125
1
indicates the number of concrete surface subdivisions to be vacuumed.
b1 sec elapsed time represents sampling that started immediately after spraying.
2.1.3 Liquid Volume
The objective of this test was to determine the effect of changing the volume of liquid
applied for sampling on the spore recovery at a constant elapsed time. For each test, the
volume of Dl-Tween liquid was sprayed onto the carpet and the laminated wood coupons and
remained for the prescribed elapsed time before vacuuming. For the concrete coupons, with a
1-sec elapsed time, each coupon surface was divided into quarters (2B3 from Table 2-2). The
test consisted of four consecutive spraying and vacuuming combinations of each quarter, with
the sample being cumulative and collected with the same custom-made vacuum device with a
liquid collection system. The test matrix for this test is shown in Table 2-3.
18
-------
Table 2-3. Liquid Voiume Test Matrix (Task 3)
Test ID
Material
Variable: Volume Sprayed (mL)
Elapsed Times (sec)
3A
Carpet
100, 50
30
3C
Laminated Wood
50, 10
30
3B
Concrete
100 (4a x 25), 40 (4a x 10)
1
a Indicates the number of concrete surface subdivisions to be vacuumed.
2.1.4 Material and Equipment
2.1.4.1 Wet Vacuum Sampling Device
The wet vacuum sampling device used in this evaluation was a prototype apparatus
made from off-the-shelf components, including a wet vacuum adapter comprised of 1 meter (m)
of latex tubing (Fisher Cat #: 14-178-2BB); a home-made 45-millimeter (mm) m. 1-mm cross-
sectional area-enhanced vacuum nozzle made of impact-resistant polycarbonate material
(McMaster Cat. No. 1749K399) with tube fitting; and a Cord-Grip fitting connected to a 1-liter (L)
Nalgene bottle (Fisher Cat. No. 02-543-03) with cap (see Figure 2-1). The wet vacuum sampling
assembly was connected to a self-contained service vacuum pump (Omega Plus HERA
Vacuum pump, Atrix International Inc., Burnsviile, MN) to provide the suction at 628 watts and a
filtration efficiency of 99.9% at 0.3 micrometers (jam).
Figure 2-1. Wet Vacuum Sampling Assembly
Sampling liquid was sprayed onto a horizontally placed coupon and then vacuumed
through the collection apparatus with the nozzle. The vacuumed liquid was collected in a clean,
sterile Nalgene bottle. New, sterile nozzles and inlet tubing were used for each sample and
coupon.
19
-------
2.1.4.2 Liquid Agent Spray
A spray box apparatus (Figure 2-2) was designed to facilitate repeatable spraying of the
28 inches (in.)- x 28 in.-coupons and was equipped with a hinged cover with an 18 in.- x 18 in.-
opening and a liquid collection vessel. The spray box lid opening was used as a template to
ensure consistent sprays between coupons. The hinged cover was opened following the wetting
of the coupons, to allow for vacuum sampling of a larger area.
Figure 2-2. Spray Box with Lid in Open Position
Three liquids were evaluated in Phase I: Dl water, PBST, and Dl-Tween. A High purity
Dl water system (Dracor Water System, Durham, NC) was used to obtain Dl water for each test
in this study. The system features 1- to 0.2-|jm x 10-in. pre-filters, acid-washed activated
carbon, and two mixed-bed deionizers. The PBST solution was prepared by dissolving one
packet of PBS with Tween® 20 (Cat. No. P-3563, Sigma Aldrich Corporation, St. Louis, MO) in 1
L of Dl water. The final sampling solution, Dl-Tween, was prepared by adding 500 |jL of Tween®
20 (Cat. No. BP337-100, Fisher Scientific, Hampton, NH) to 1 L of Dl water. All three solutions
were sterilized and placed into two sterile 500-mL reagent Nalgene bottles using a 500-mL
bottle top filter with a 33-mm neck and a 0.22-|jm cellulose acetate filter (Part No. 431118,
Corning Inc, Corning, NY).
Each sterile sampling solution was aseptically transferred to a clean backpack sprayer
vaporous hydrogen peroxide (VHP®)-sterilized). The sampling liquid (-100 ml_) was applied
onto the center of each test coupon marked by an area of 16 in. x 16 in. This area is
demarcated by a stainless-steel template (Figure 2-3). The entire coupon was then vacuumed
using the custom-made device shown in Figure 2-1, performing five replicates for each test. The
liquid collected was analyzed for spores, and the analytical results were compared to the
number of spores recovered from the stainless-steel control coupons.
20
-------
Figure 2-3: 16-in. x 16-in. Stainless-Steel Spray Template.
2.1.4.3 Electric Backpack Sprayer
The material coupons were wetted with the target wetting agent (Dl water, PBST, or Dl-
Tween) using a rechargeable backpack sprayer (SRS-600 ProPack, SHURflo®, Cypress, CA)
(Figure 2-4). The sprayer was maintained at a pressure of 35 pounds per square inch (psi) and
a flow rate of approximately 1 L/minute (min) for most of the tests. The chemical-resistant
backpack sprayer comes with a four-gallon tank and can spray up to 120 gallons on a single
battery charge. Four pump speeds allow for adjustable spray patterns. For Phase I tests, the
backpack sprayer pump was set at 35 psi and a target flow rate of approximately 1 L/min. The
backpack sprayer was decontaminated by soaking the tank with pH-adjusted bleach for 10 min,
then rinsing three times with the sterilized wetting agent. Separate backpack sprayers were
used for each wetting agent.
Figure 2-4. Electric Backpack Sprayer
21
-------
2.1.5 Coupon Preparation
Three material types (carpet, concrete, and laminated wood (28-in. x 28-in.) coupons)
were investigated in this study. These materials were prepared following standard procedures
for representativeness and uniformity so that tests could be reliably reproduced. Control
coupons for inoculation of surrogate organism checks were made of stainless steel.
Specifications for all test coupon materials and material preparation instructions are detailed in
Table 2-4.
Table 2-4. Description of Test Coupon Materials and Mai
erial Preparation
Material
Description
Manufacturer/
Supplier Name,
Location
Coupon Surface
Size
LxWxH (in)
Material Preparation
Carpet
100% Nylon
Multiplicity Tile
#54594
Shaw Industries,
Dalton, GA
24x24x0.25
Remove wood particles using
soft-bristle brush.
Sterilize (VHP®)*
Concrete
Quikrete Type I & II
Portland Cement;
Quikrete All
Purpose Sand
Lowe's
Companies, Inc.,
Mooresville, NC
28 x 28 x 1
Remove particles by power
washing.
After power washing, allow to air
dry in climate-controlled
environment for at least five
days.
Sterilize (autoclave).
Laminated
Wood
Winchester Oak
Smooth Laminate
Wood Planks
Lowe's
Companies, Inc.,
Mooresville, NC
28x28x0.276
Remove particles and dust by
wiping clean with water and
wipe dry.
Sterilize (VHP®)
Stainless
Steel
Multipurpose
Stainless Steel (48
in x 48 in), type
304, #2B mill
(unpolished), 0.036
in thick
McMaster-Carr,
Atlanta, GA
14x14x0.036
Remove any lubricant/grease
from shearing with acetone and
wipe dry.
Remove particles and dust by
wiping clean with water and
wipe dry.
Sterilize (autoclave).
"VHP1" = Vaporized hydrogen peroxide.
2.1.5.1 Carpet Coupons
The carpet coupons (Figure 2-5) were prefabricated 24-in. x 24-in. (0.61 m by 0.61 m) 100%
nylon tile, affixed in the center of a 28- by 28- by 7/16-in Oriented Strand Board (OSB) (Norbord
Technology, Ville St. Laurent, Quebec, Canada) using an adhesive caulk (Model LN-601 CP,
Liquid Nails® Adhesive, Strongsville, OH, USA). Coupons were clamped together and allowed to
dry overnight before use.
22
-------
Figure 2-5. Carpet Coupon
2.1.5.2 Concrete Coupons
The concrete coupons were prepared on-site using QUIKRETE® sand/topping mix that
consists of a uniformly blended mixture of Portland cement, commercial-grade sands, and other
approved ingredients. The concrete coupons were made using custom 28 in.x 28 in. x 1 -in.
deep forms. The concrete was prepared according to the package instructions, using a trough
and a garden hose for the water supply. Following preparation of the concrete, coupons were
covered with plastic and allowed to cure for no less than five days before use (Figure 2-6).
Figure 2-6. Concrete Coupons
2.1.5.3 Laminated Wood Coupons
The laminated wood coupons (Figure 2-7) were cut to 28 in. *: 28 in. from Project
Source 7.6-in. x 4.23-ft Winchester oak smooth laminated wood planks. The coupons were
glued to a 7/16-in.-thick OSB using an adhesive caulk (Model No. LN-601 CP, Liquid Nails®
Adhesive, Strongsville, OH). Coupons were clamped together and allowed to dry overnight
before use.
23
-------
Figure 2-7. Laminated Wood Coupon
2.1.5.4 Stainless-Steel Coupons
The stainless-steel coupons (Figure 2-8) were cut to 14 in. x 14 in. from 48 in. x 48-in
Multipurpose Type 304 stainless steel sheets using heavy duty power hydraulic shear (National
Sheet Metal In., Smartt, TN). Disposable Manila paper templates, 12 in. x 12-in. (30.5
centimeters (cm) x 30.5 cm) (Part# 225-2416, SKC Eighty-Four, PA), were used to identify the
inoculated area.
Figure 2-8. Stainless Steel Coupon
2.2 Phase II - Commercially-Available Wet Vacuum Cleaner Evaluation
Phase II evaluated three types (residential, commercial, portable) of commercially-
available wet vacuum cleaners as well as a Shop-Vac cleaner for spore sampling. Each vacuum
cleaner type was evaluated based on cleaning patterns and time listed in the ASTM F-1284-09
(ASTM. 2009) standard method for evaluating dirt removal effectiveness of residential vacuum
cleaners from carpet surfaces.
2.2.1 Wet Vacuum Cleaners
The Phase II study evaluated commercially-available wet vacuum cleaners for surface
spore sampling efficacy. The wet vacuums were selected by the project team considering the
information from the Consumer Reports. The operational aspects and ratings were reviewed,
and three commercially available wet vacuums and one wet/dry vacuum were selected. The wet
vacuums were selected based on reviews regarding ease of use, separate clean tank (to
24
-------
contain liquid before dispensed) and recovery tank (to contain dirty liquid from surface), suction
power, portability, cost and a heated cleaning option. The down-selected vacuum cleaners are
discussed in the following sections.
2.2.1.1 Residential Cleaners
Based on the Consumer Reports reviews of residential vacuum cleaners, the Hoover
Dual V Steam Vac All Terrain with Spinscrub (Model No. F7452900, The Hoover Company,
North Canton, OH), shown in Figure 2-9, was chosen for this study. This vacuum cleaner can be
used on a variety of surfaces (hardwood flooring, laminated flooring and upholstery) as well as
in challenging sampling situations (cold environments) by applying heat directly to the floor.
The Hoover F7452900 has brushes and two nozzles to deliver equal suction power
across the width of the nozzle. The cleaning nozzle is approximately 13 in. wide. The wet
vacuum cleaner has separate solution (clean) and recovery (dirty) liquid tanks (one-gallon
capacity) as well as hand tools for cleaning hard-to-reach areas. The vacuum cleaner was set to
"Wash Auto Rinse" mode (a unit function for spraying liquid automatically while vacuuming)
during vacuum sampling.
Figure 2-9. Wet Vacuuming with Residential Vacuum Cleaner (Hoover)
25
-------
2.2.1.2 Commercial Cleaners
Like the Hoover Steam Vac, the Rug Doctor Pro X3 (Rug Doctor, Inc., Piano, TX)
commercial vacuum cleaner, Figure 2-10, can be used on a variety of surfaces such as
hardwood and vinyl flooring. The Rug Doctor ProX3 comes pre-assembled and thus requires
little preparation prior to sampling. The Pro X3 cleaning nozzle is approximately 10 in. wide with
a vibrating brush for increased extraction power. It is equipped with separate solution (clean)
and recovery (dirty) liquid tanks with a large capacity (four gallons), allowing for sampling larger
areas in one sampling event. The vacuum cleaner was chosen for this study due to its wide
availability and ease of use.
Figure 2-10. Wet Vacuuming with Commercial Vacuum Cleaner (Rug Doctor)
2.2.1.3 Portable Cleaners
Portable vacuum cleaners are more suitable for harder-to-reach areas. While they are
not meant for cleaning entire rooms, portable cleaners are ideal for spot cleaning. The Bissell
Little Green ProHeat machine (Model No. 14259, Bissell Corp, Grand Rapids, Ml) (Figure 2-11)
was chosen for this study. As with the other cleaners used in this study, the ProHeat also has
separate clean and dirty tanks (48-ounce capacity) and comes with a built-in water heater for
heated cleaning in cold environments.
26
-------
Figure 2-11 Wet Vacuuming with Portable Vacuum Cleaner (Bissell ProHeat)
2.2.1.4 Wet/Dry Vacuum Cleaners
Wet-dry vacuum cleaners are available in a variety of sizes, ranging from 2.5 gallons to
14 gallons. To aid with portability and ease of use in this study, the mini wet-dry Rigid Pro Pack
vacuum (Model No. WD4550, RIGID Tool Company, Elyria, OH) (Figure 2-12) was selected.
The ProPack has a 2.5-gallon tank capacity but does not come with a liquid dispenser.
Figure 2-12. Wet Vacuuming with Wet/Dry Vacuum Cleaner (Rigid ProPack)
27
-------
2.2.2 Phase II Coupon Preparation
Coupons of three materials (vinyl flooring, carpet, and concrete) were used in Phase II. The test
coupon sizes were based on carpet test area sizes listed in the ASTM F-1284-09 (ASTM. 2009),
standard method for evaluating the effectiveness of residential vacuum cleaners in removing dirt
from carpet surfaces. A typical test coupon for Phase II, with the inoculated areas, is illustrated
in Figure 2-13.
92"
Figure 2-13 Phase II Test Coupon Schematic
Table 2-5 lists the characteristics of the test coupons used in this study.
Table 2-5. Phase II Material Description
Material
Manufacturer/Supplier Name
(Location)
Test Coupon Size
LxW
(in.)
Control Coupon
Size LxW
(in.)
Concrete
16-in. x 16-in. pewter concrete step stone (Cat.
No. 204659, Home Depot, Atlanta, GA)
96x48
16x 16
Carpet
Beaulieu Solutions Laredo Sagebrush loop carpet
(Cat. No. 409921, Home Depot, Atlanta, GA)
92x50
14x 14
Vinyl
12-ft.-wide River Park staggered slate brown
multi-vinyl sheet (Cat. No. 732233, Home Depot,
Atlanta, GA)
92x50
14x14
Stainless steel
16-gauge type 304 mill-finished stainless steel
(Dillon Supply Company, Raleigh, NC)
-
14x 14
The test coupons were prepared following standard procedures for representativeness
and uniformity so that tests could be reliably reproduced. Control coupons for inoculation of
surrogate organism checks were made of stainless steel. Preparation of the test coupons for
Phase II is summarized in the next sections.
28
-------
2.2.2.1 Phase II Carpet and Vinyl Coupons
All carpet (Figure 2-14) and vinyl flooring (Figure 2-15) test coupons had a surface area
of 32 ft2 (92 in. x 50 in.) and were cut to size from larger (18 ft x 12 ft) sheets and glued onto
plywood sheathing (Plytanium 15/32 CAT PS1-09 pine plywood sheathing, Home Depot,
Atlanta, GA) using, respectively, TEC Skill Set carpet flooring adhesive (Model No.
7047485021, Lowe's, Durham, NC) and TEC Multi-Floor flooring adhesive (Model No.
7074255021, Lowe's, Durham, NC). The positive and negative control coupons for the same
materials were 14 in. x 14 in. and fabricated in the same way as the test coupons but scaled
down to size.
Figure 2-14. Carpet Coupon
Figure 2-15. Vinyl Coupon
29
-------
2.2.2.2 Phase II Concrete Coupons
Due to their size and difficulty in transporting, concrete test coupons (Figure 2-16) were
not fabricated in-house. instead, 24 concrete pavers sized 16 in. x 16 in. (Cat. No. 204659,
Pewter Concrete Step Stone, Home Depot, Atlanta, GA) were aligned to yield a larger test
coupon, similar in size (96 in. x 48 in.) to the carpet coupons. The seams between the paver
stones were covered with 2 in. duct tape to prevent the sampling liquid from seeping through.
Concrete positive and negative control coupons consisted of a single concrete paver that was 16
in. x 16 in.
All positive control and inoculation control coupons were tested in triplicate (i.e., three
replicates per test). A negative control that was not inoculated but was subjected to the same
sampling techniques as its inoculated counterpart was also included.
2.2,3 Test Matrix
The Phase II test matrix was developed as the tests progressed based on the results
and ease of testing (sampling recovery, sterilization methods, and inoculation methods used).
Test conditions were scaled up from Phase I laboratory results. The overall test matrix for
commercial wet vacuum cleaner testing is shown in Table 2-6 for each surrogate contaminant
(.Bg and B. thuringiensis subsp. kurstaki (Btk)).
Figure 2-16 Concrete Coupons
30
-------
Table 2-6. Phase II Test Matrix
Target Spore (Bg and Btk)
Concentration
Test ID
Material
Vacuum Cleaner Type
Vacuum Cleaner Model
1
Portable
Bissell Little Green
2
Carpet
Commercial
Rug Doctor ProX3
3
Residential
Hoover Dual Steam Vac
4
Wet/Dry
Shop-Vac
5
Portable
Bissell Little Green
107 CFU/ft2
6
Concrete
Commercial
Rug Doctor ProX3
7
Residential
Hoover Dual Steam Vac
8
Wet/Dry
Shop-Vac
9
Portable
Bissell Little Green
10
Vinyl
Commercial
Rug Doctor ProX3
11
Residential
Hoover Dual Steam Vac
12
Wet/Dry
Shop-Vac
2.3 Testing and Sampling Approaches
The general testing approach, for both Phase I and Phase II testing, consisted of
inoculating coupons with Bacillus spores, using an aerosol deposition method that delivered
approximately 106-107 spores of a surrogate organism for B. anthracis on a material surface.
The inoculated material surfaces underwent the vacuuming process, and recovery of spores
from treated surfaces (test samples) was compared to recovery from stainless steel surfaces
that were inoculated but not treated (inoculation control samples). Surface sampling efficacy
was calculated as the difference between the average inoculation control recoveries and the
post-sampling recovery on each treated surface (test sample).
The approach for all tests followed the layout shown in Figure 2-17. Coupons for a
selected test material (concrete, carpet, or laminated wood) were inoculated and sampled using
the wet vacuuming (Wet Vac) method, yielding liquid samples for microbiological analysis. Wet
Vac samples were also collected from blank (un-inoculated) material coupons, and these are
referred to as Test Blanks.
31
-------
Inoculated Blanks
Wet Vacuum Test Samples-4
test coupons were vacuumed
and the liquid collected counted
as one sample)
Positive Controls - coupons were
sampled with an alternative
sampling method as a check on
sampling procedures.
One
12- by 12-in
Material
Coupon
One
12-by 12-in
Material
Coupon
One
12-by 12-in
Material
Coupon
One
12-by 12-in
Material
Coupon
Inoculum Controls- a separate
coupon set is inoculated to check
MDI functionality. Samples are
collected with PRB wipes.
One
14-by 14-in
Stainless
Steel Coupon
V
f -N
One
14-by 14-in
Stainless
Steel Coupon
V
( \
One
14- by 14-in
Stainless
Steel Coupon
f
One
14-by 14-in
Stainless
Steel Coupon
V
Figure 2-17. Sampling Approach
Positive control coupons that consisted of the selected test material were also
inoculated. These samples were collected using an alternative (non-wet vacuum) sampling
method, which served as a check on sampling procedures.
Each test included inoculum controls designed to check metered dose inhaler (MDI)
performance consistency. All inoculum controls consisted of stainless steel coupons sampled
with PRB wipes, which were processed for microbiological analysis. An un-inoculated stainless-
steel coupon was also wipe-sampled for sterility, and it is referred to as the Inoculum Control
Blank. Blanks were included for each sampling method to check for cross-contamination.
The general experimental approach used to meet the project objectives is described
below:
• Preparation of representative coupons of test materials: Tests used coupons
made of carpet, laminated wood or vinyl flooring, and concrete. Coupons for Phase I
had a surface area of 784 in2 (28- by 28-in), described in Section 2.1.5, while Phase
II used larger coupons with a surface area of 4600 in2 (92- x 50-in.), described in
Section 2.2.2.
• Sterilization of the coupon materials: Prior to use, the coupons were wrapped in
Tyvek bags and sterilized using a 4-hour Vaporous Hydrogen Peroxide (VHP®)
sterilization cycle at 250 ppm (Section 2.3.1).
• Inoculation of coupons: Test coupons were inoculated using an aerosol deposition
method, as described in Section 2.2.3. Briefly, a known quantity of the surrogate
32
-------
organism (105 - 107Bg or Btk spores) was deposited onto a coupon using an MDI.
The inoculation occurred a minimum of 18 hours prior to testing.
• Wet vacuum cleaner sampling process: The wet vacuum sampling approach for
Phase I described in Section 3.1 consisted of applying a sampling liquid to the test
coupon via a backpack sprayer and using a custom-made vacuum device to recover
the liquid and the spores. For Phase II, the vacuum sampling approach consisted of
using commercial off-the-shelf (COTS) wet vacuum cleaners, following the sampling
procedure described in Section 3.2. Inoculation control coupons underwent sampling
techniques involving PRB wipes, 37 mm cassettes and vacuum socks.
• Sampling Evaluation: The spore recovery efficiencies relative to the number of
spores deposited was estimated for each of the four wet vacuum cleaners, and for
each material-specific traditional surface sampling method. To account for differing
inoculation levels achieved across numerous test days and MDIs, recovery was
compared to the inoculum control coupon inoculated using the same MDI and
collected on the same test day as the test samples.
Spore Recovery Efficiency = (Test Wet Vacuum Recovery/Inoculum Check
Recovery) x 100
• Sample Sterility Evaluation: Swab samples collected from materials prior to testing
were analyzed (growth/no growth) to demonstrate sterility of the test materials.
2.3.1 Material and Equipment Sterilization
Test coupons were placed in Tyvek® bags (steam component bags, General Econopak,
Philadelphia, PA) prior to sterilization. Batches of carpet and laminated wood coupons were
exposed to 250 ppm hydrogen peroxide (H2O2) vapor for four hours using a STERIS VHP®
ED1000 generator (STERIS Corporation, Mentor, OH). Stainless-steel coupons were sterilized
for 30 min in an autoclave cycle; the concrete coupons were used as is. After sterilization, the
coupons treated with VHP® sat for up to two weeks at room temperature to force off-gassing of
H2O2 and prevent any biocidal effects. Sterility was evaluated by swab sampling one coupon
from each sterilization batch. Any dust or debris on each concrete coupon was removed using a
clean RIGID Pro Pack vacuum cleaner (Model No. WD4550, RIGID Tool Company, Elyria, OH).
2.3.2 Spore Preparation
The test organisms for this work were powdered spore preparations of Bacillus
atrophaeus var. globigii (Bg), and Bacillus thuringiensis subsp. kurstaki (Btk), obtained from the
U.S. Army Dugway Proving Ground Life Sciences Division (Dugway, UT). The powdered spore
preparation for this procedure is described elsewhere (Brown et al.. 2007).
2.3.2.1 Bacillus atrophaeus var. globigii
Bg, a surrogate for the spore-forming bacterial agent Bacillus anthracis, was used for
this project. Like B. anthracis, Bg is a soil dwelling, Gram-positive, spore forming, aerobic
microorganism, but unlike B. anthracis, it is non-pathogenic. Bg forms an orange pigment when
grown on nutrient agar, a desirable characteristic when there is a need to detect viable spores in
33
-------
environmental samples. Bg has a long history of use in the biodefense community as a simulant
for anthrax-associated biowarfare and bioterrorism events (Gibbons, et al.. 2011).
2.3.2.2 B. thuringiensis subsp. Kurstaki (Btk)
Btk, another surrogate for B. anthracis, was also used for this project. Like Bg, Bacillus
thuringiensis strains are not considered human pathogens, but unlike Bg, Btk produces no
orange pigment when grown on nutrient agar such as tryptic soy agar (TSA). Btk colonies are
whitish, round to irregular in shape, and have a matte or opaque texture. Btk is also known for
parasporal crystal formation, which makes it useful as a biopesticide. Multiple Btk strains are
registered with the USEPA as biopesticides (USEPA. 2017), and strain HD-1 is found in
commercial products, like Foray, used to control gypsy moths (Valaderes et al.. 2001).
Strain HD-1 has also been used to develop genetically tagged strains, whereby unique
and stable genetic signatures or "barcodes" have been integrated into non-protein coding
regions of the chromosome (Emanuel et al. 2012; Buckley et al. 2012). This barcoding system
facilitates detection of Btk by real-time PCR assay, even in the presence of non-tagged
background Btk or other Bacillus species. This project utilized Btk with barcode T1B2 (Buckley
et al. 2012).
2.3.2.3 MDI Preparation
MDIs were used to inoculate material surfaces with spore preparations of either Bg or Btk. Dry
spores received from Dugway Proving Ground were resuspended in 100% ethanol, then
combined into each MDI canister with 1,1,1,2-tetrafluoroethane (HFA-134a), a non-ozone
depleting propellant. Each MDI was charged with a volume of spore preparation plus propellant
sufficient to deliver 200 discharges of 50 |jL per discharge. The number of discharges per MDI
was tracked to ensure that the use did not exceed this value.
2.3.2.4 MDI Spore Concentration Validation and Spatial Distribution
Following the manufacturing process, MDIs were tested for concentration and spore quality.
MDIs were actuated to deposit either Bg or Btk spores on five 18-mm aluminum coupons. The
MDI canister (Figure 2-18, panel A) was situated inside an actuator (Figure 2-18, panel B) and
fitted into an adapter (Figure 2-18, panel C) that securely held the coupon so that each time the
actuator was depressed, a repeatable number of spores was deposited on the coupon (Lee et
al.. 2011).
34
-------
Figure 2-18. MDI Actuator Adapter for Small, 18 mm Coupons (A), Catalent MDI Canister
(B), Actuator Adapter (C)
After inoculation, coupons were placed into sterile 50-mL conical tubes, then extracted
with 10 mL of sterile phosphate buffered saline with Tween® 20 (PBST) by sonicating for 10 min
and vortexing continuously for two min. Following extraction, a 5-mL volume was transferred
into a fresh 50-mL conical tube and heat-treated in an 80 °C water bath for 10 minutes. Aliquots
from both non-heat-treated and heat-treated tubes were then spiral-plated (Autoplate 5000,
Advanced Instruments, Norwood, MA) in triplicate on TSA plates. Plates were enumerated with
the QCount (Advanced Instruments, Norwood, MA).
Mean results (CFU/mL) for each of the five 18-mrn coupons were averaged and then
multiplied by the total volume (10 mL) to determine the MDI concentration per actuation. The
percent relative standard deviation (% RSD) was calculated by dividing the standard deviation
by the average mean, then multiplying by 100. Spore quality was estimated by comparing heat-
treated results to non-heat-treated results for each canister. MDIs with a % RSD less than
approximately 50% and a spore quality score of greater than approximately 85% were put into
the inventory for testing.
Selected MDIs (Bg canister #1 and Btk canister #1) were also actuated to determine the
spray distribution over the center section (approximately 12 in. x 12 in.) of a 14-in. x 14 in. test
coupon. Forty sterile reference material coupons (RMCs, 1" x 2") were manually arranged in a
grid (5x8 RMCs) on a sterile 14 in. x 14 in. stainless-steel coupon as shown in Figure 2-19.
After RMC placement, the coupon was covered with an Aerosol Deposition Apparatus (ADA)
(Calfee et al., 2013a) like the ADA shown in Figure 2-20.
35
-------
Figure 2-19. RMCs Placed on 14 in. x 14 in. Stainless Steel Coupon
Figure 2-20.ADA Used on 14 in. x 14 in. Stainless Steel Coupon
-------
To inoculate the RMCs, each MDI canister was placed in a 50-mL conical tube adapter
that was attached to a Vortex Genie (Part #EF3030A, Daigger Scientific, Vernon Hills, IL), then
vortexed (see Figure 2-21, panel A) vertically (stem up) for two min at top speed. The MDI was
rotated in the adapter (stern down) and vortexed for another two min, then purged three times
with a 10-second side vortex between each purge (see Figure 2-21, panel B).
Figure 2-21. MDI Content Mixing and Purging Prior to Inoculation
After purging, the MDI was placed on the ADA and discharged once. Spores were
allowed to settle overnight, and the RMCs were collected into individual 50-mL conical tubes.
After addition of 20 ml_ sterile PBST, RMC tubes were sonicated for 10 min, then vortexed
continuously for two min. Aliquots were spiral-plated on TSA plates and incubated at 35 C ± 2
C for 18 to 20 hours (Bg) or 28 C ± 2 C for 18 to 22 hours (Btk). Plates were enumerated by
QCount. Results were used to develop heat maps, like the Bg MDI example shown in Figure 2-
22.
MDI Spore Distribution
O.OOE+00-5.00E+04 5.00E+04 1.00E+05 ¦ l.OOE+OS-l.SOE+OS
Figure 2-22. Color Coded MDI Spore Distribution Heat Map over 40 RMCs after Actuation
of Bg Canister #1 (CFUs per RMC coupon, 2 in2)
37
-------
While the highest spore counts were observed on RMCs placed near the middle of the
test coupon, spores were also distributed along the outer edge of the test area. Cumulative
recovery across all RMCs was 16 x 10s CFU, which was approximately 27.6% of the average
5.79 x 106 CFU/actuation observed during the 18-mm aluminum coupon test for Bg canister#!
The higher recovery for the 18-mm coupon test, compared to the cumulative RMC recovery,
was accounted for as follows: 1) variations in recovery efficiencies for 18-mm coupons and
RMCs (i.e., losses to the pyramid surfaces), and 2) losses due to the spaces between the
RMCs, which were inoculated but not recovered during the RMC test. Similar results were
obtained for the selected Btk MDI (canister #1). Cumulative recovery across all Btk Canister #1
RMCs was 3.68 x 106CFU, which was approximately 26.9% of the 1.37 x 107 CFU/actuation
observed during the 18-mm coupon test.
2.2.3 Inoculation of Coupons
Test coupons for Phase I were inoculated with approximately 106-107 aerosolized spores
on the same day using an aerosol deposition method (Figure 2-23). A sterile stainless-steel skirt
with the same dimensions as the test coupon, except for a 14-in. x 14-in. area cut open in its
center, was placed on each coupon to maintain sterility of the coupon surface that was not
inoculated. A single ADA was placed over the open 14-in. x 14-in. area in the center of the skirt
of each test coupon.
Figure 2-23. Phase I Carpet Coupon with Skirt and ADA.
Each ADA was designed to cover a 14-in. x 14-in. area of any coupon of any thickness.
Just prior to dosing, each ADA lid was opened, and an actuator with the MDI was placed in the
opening and depressed to release the spores (Figure 2-24). The ADA lid was closed after
inoculation
The same approach was used for the inoculation of Phase II test coupons. A sterile
stainless -steel skirt with the same dimensions as the test coupon, except for a 56-in. x 14-in.
area cut open in its center, was placed on each coupon to maintain sterility of parts of the
coupon surface that were not inoculated. Four ADAs were placed over the open 56-in. x 14-in.
area in the center of the skirt of each test coupon. Skirts were used only for the tests that
involved carpet coupons. Vinyl flooring and concrete coupons were covered using only four
ADAs per test coupon. The carpet test coupon layout inoculation is shown in Figure 2-24. For
38
-------
the paver stones, spores were inoculated on the taped surface. The estimated taped surface area
is approximately 10% of the entire inoculated paver area.
Figure 2-24. Inoculation of Phase II Test Coupons
Stainless steel coupons (14 in. x 14 in.), which served as quality control (QC) checks for
the MDI and each inoculation event, were also inoculated as the first, middle, and last coupons
during an inoculation event. Coupons were inoculated a maximum of 48 hours and a minimum
of 18 hours prior to the sampling event
Positive control coupons were sampled using a non-wet vacuum, an alternative
sampling method to check on sampling procedures. Carpet samples were collected using
vacuum socks, and concrete samples were collected using 37-mm cassette filters.
39
-------
3 Sampling Procedures
Prior to the sampling event, all materials needed for sampling were prepared using
aseptic A two-person sampling team, using aseptic techniques whenever possible, was
designated to perform the sampling procedure following a strict sampling protocol that listed
each person's role during the sample test preparation, sampling, and sample handling. The
sampling procedure consisted of the operations described below and were performed in
sequence,
3.1.1 Wet Vacuum Sampling Kit Preparation
The wet vacuum sampling kit was prepared prior to the start of each testing sequence. The
following procedures were followed by a two-person sampling team designated as sampler and
support person:
1. The support person wears new gloves, opens the sampling supply bin and removes the
sample kit from the bin.
2. The support person opens the outer bag and allows the sampler to remove the nozzle
and tubing from the kit.
3. The support person removes the sterile Nalgene bottle from the overpack bag and
allows the sampler to remove the cap and place it back in the overpack bag.
4. The support person holds the Wet Vacuum Adapter so the sampler can install the latex
tubing from the bottom of the adapter through the Cord-Grip fitting (Figure 3-1). The
sampler leaves approximately 3-4 in. of tubing below the cap and connects the tubing to
the nozzle fitting once threaded (Figure 3-2).
Figure 3-1. Wet Vacuum Adapter
40
-------
Figure 3-2. Uncapped Wet Vacuum Sampling Kit
5. The support person places the wet vacuum adapter onto the sterile Nalgene bottle,
taking care not to touch the inside of the bottle, the latex tubing, or the Cord-Grip fitting
used to connect the latex tubing to the apparatus.
6. The support person connects the Atrix Vacuum Nozzle to the Wet Vacuum Adapter
fitting. The wet vacuum sampler is then ready for sampling.
The wetting and sampling operations were performed by the same team that prepared the
wet sampler kit, with the addition of an assistant on an as-needed-basis. The sampling team
followed a strict protocol for sampling (Calfee et al., 2013b) as summarized below.
3.1.2 Spraying Sequence
1. The three-person sampling team members each don new Personal Protective
Equipment (PPE) before beginning a sampling set. A single set of clean lab-coat and
gloves is worn unless a change is required due to:
• a material type change (e.g., stainless steel to carpet),
• going to a lower contamination level (positive controls to test samples),
• any possible contamination of the current PPE at the sampler's discretion, or
• when the contamination level is unknown.
2. The support person removes the sterilized test coupon from the Tyvek® bag and places
it on the spray box (Figure 2-2) with the lid in the open position, then removes the ADA.
The support person then closes the spray box lid.
3. The sampler and support person ensure that the correct sample coupon has been
selected, referencing the coupon code on the sampling bag.
4. The support person or assistant records the coupon code (when required) on the
sampling log sheet next to the corresponding bag number that was just recorded.
41
-------
5. With the vacuum nozzle in one hand, the sampler uses their free hand to spray the 18-
in. x 18-in. center area of the coupon in an evenly dispersed pattern until the required
volume of wetting agent has been dispensed.
a. The spray wand nozzle is held at a 90° angle and 3 in. above the spray
box lid opening. The spray box lid has an 18-in. x 18-in. opening that
served as a template to ensure consistent sprays between coupons.
Once the spraying is completed, the sprayer wand is holstered in a
manner that keeps the nozzle from contacting any surfaces.
b. Each pass of the nozzle overlaps 50% with the previous pass.
c. Each coupon is sampled at a rate of approximately three min per
coupon.
3.1.3 Sampling Sequence
1. At the end of the spraying sequence, the support person then fully opens the lid of the
spray box, exposing the coupon for sampling.
2. The support person or assistant is prepared to record the duration of sampling. Prompts
are given to the sampler so that the sample duration is as close to the value indicated in
the test plan as possible.
3. The sampler:
a. Checks the fitting to the nozzle and adjusts, if necessary.
b. Turns on the vacuum.
c. Vacuums "horizontally" from one side of the coupon to the other, starting
from the lower right corner of the coupon (Figure 3-4). The sampling
covers the center 20- by 20-in. area (or 26- by 26-in. area, depending on
the coupon) of the material surface, and the nozzle is kept at a 45-
degree angle. The nozzle width is used to estimate a 2-in. border around
the coupon. The nozzle is pushed forward so that the coupon is sampled
in the direction of the larger angle (135 degrees). Each pass of the
coupon with the vacuum overlaps 50% with the previous pass.
42
-------
Figure 3-3. Wet Vacuum Sampling
d. The same area is vacuumed "vertically" using the same technique.
e. After sampling, the sampler lifts the nozzle so that the nozzle is one
meter directly over the Nalgene bottle, with the tubing taut, so that the
tubing is clear of liquid.
3.2 Phase II: Commercially Available Wet Vacuum Cleaner
All test vacuum cleaners, regardless of nozzle width, were moved back and forth in a
specified pattern as shown in Figure 3-4. Briefly, each test coupon was divided into N strips,
with N being half the width of a vacuum cleaner nozzle. The wet vacuum cleaner was placed on
the sampling area so that the front edge of the vacuum cleaner nozzle lip coincided with the line
defining the beginning of the sampling area. The first vacuuming stroke was a backward stroke,
while Dl-Tween was simultaneously sprayed onto the coupon surface. The second stroke was a
forward vacuum stroke to vacuum any residual liquid not covered by the first stroke. Next, the
vacuum nozzle was moved horizontally by half the size of the width of the nozzle, so that the
previous vacuum path was overlapped. This process was repeated until the entire surface of the
test coupon was covered.
The wet vacuuming sampling protocol is summarized below:
1. Don a fresh pair of sterile boot covers before stepping back into the sweep area.
2. Charge each vacuum cleaner with approximately 1 L of 0.05% Tween® 20. Record all
volumes in the laboratory notebook.
3. Place the vacuum cleaner nozzle so that the front edge of the vacuum cleaner nozzle lip
coincides with one edge of the test coupon, as shown in Figure 3-4.
a. Stroke 1: Backward stroke until reaching the end of the opposite edge of the test
coupon.
b. Stroke 2: Forward stroke until the initial start location is reached.
c. Move the vacuum cleaner horizontally by half of the nozzle width area.
43
-------
d.
Repeat the process (steps a-c) for the next sampling strip until the entire coupon
is reached.
e. Immediately post-sampling, aseptically transfer the liquid from the wet vacuum
"dirty" tank to a pre-labeled clean Nalgene bottle. Record the volume collected in
the laboratory notebook.
f. Sterilize the outside of the Nalgene bottle using Dispatch® bleach wipes (Caltech
Industries, Inc., Midland, Ml) and place the bottle in a secondary containment unit
such as a Twirl'Em® Sterile Sampling Bags (Labplas, QC Canada). Sterilize the
outside of the secondary containment unit using bleach wipes.
g. Place the whole sample into the collection bin.
width location Vacuuming
Figure 3-4. Wet Vacuuming Technique
3.3 Wipe Sampling
For Phase II of this study, stainless steel inoculum control coupons and vinyl flooring
positive control coupons were wipe-sampled. A moistened sterile noncotton sponge (2 in. x 2
in., 4-ply; Curity all-purpose sponge, Cat. No. 8042, Covidien PLC, Dublin, Ireland) was used to
wipe a specified area to recover bacteria, viruses, and biological toxins. Sampling was
conducted on one coupon at a time. All coupons were placed horizontally for sampling.
The sponges were prepared in the BioLab by aseptically removing them from their
packing and placing them into an unlabeled sterile 50-mL conical tube (Cat. No. 14-959-49A,
Fisher Scientific, Waltham, MA) using sterile forceps (Part No. 7190Busse Hospital
44
-------
Disposables, Hauppauge, NY). Each transferred sponge was then moistened by adding 2.5 mL
of sterile PBST and capped. The surface area for all samples was 1 ft2. A template was used to
cover the exterior (1 in.) of each coupon, leaving a square (12 in. x 12 in.) exposed for sampling
for all coupons. The outer 1 in. of each coupon was not sampled to avoid edge effects.
3.4 Vacuum Sock Sampling
Recovery efficiencies for carpet material sampling were determined by comparing the
recoveries of commercially-available vacuum cleaners to the recovery obtained using a vacuum
sock sampling method. During vacuum sock sampling, a 14-in. x 14-in. sterile positive control
coupon and a sterile sock/nozzle attachment were used to collect the sample. Holding the
nozzle at a 45-degree angle over the sample area, samples were taken using horizontal and
vertical S-strokes. This method is a modified version of the method detailed in the study by
Brown (Brown et al.. 2007)
3.5 Cassette Sampling
Concrete surfaces were sampled using a 37-mm filter cassette. A vacuum pump at the
back end of the filter pulled 20 L/min of air through the filter. A 3-cm section of Tygon® tubing
was cut to an angle of 45° on one end, the non-angled terminus was attached to the cassette,
and the angled end was used as a nozzle. The nozzle and filter were moved along the coupon
at approximately 4-in./sec in both directions (i.e., horizontally and vertically). A single coupon
per sample was used for these methods. The nozzle was extracted separately, the nozzle
extract was then combined with the filter extraction vessel, and filter extraction commenced.
3.6 Swab Sampling
The general approach for swab sampling was to use a swab (BactiSwab® Collection
and Transport System, Remel, Thermo Fisher Scientific, Waltham, MA) to wipe a specified area
to recover bacterial spores. The liquid in the swab is listed as "Modified Stuart medium" and is
part of the BactiSwab® package. Swab samples were collected from all decontamination
procedure equipment before use to serve as sterility checks.
3.7 Liquid Collection
Liquids vacuumed from coupons during Phase I tests were collected directly in Nalgene
bottles (Cat. No. 02-923-90, Fisher Scientific, Waltham, MA) fitted to the custom-sampling
device. Sampling liquids from the vacuum cleaners were aseptically transferred to Nalgene
bottles or specimen cups for Phase II tests.
For each test, the total mass of sampling liquid collected was recorded for comparison of
the collection vessel final weight to the initial weight value. If a large volume of liquid, i.e., more
than 1 L, was collected in the dirty tanks of the vacuum cleaners in Phase II tests, a
representative aliquot of 100 mL was obtained for bioanalysis. Each 100-mL aliquot was taken
via aseptic technique using a new 100-mL sterile serological pipette and sterile specimen cup
(Cat. No. B1202-1O-OR, Starplex Scientific, Cleveland, TN). The liquid in the dirty tanks was
homogenized by gently stirring the dirty tank before obtaining an aliquot. The liquid samples
45
-------
were then double-contained in sterile bags (Cat. No. 01-002-53, Fisher Scientific, Waltham, MA)
and transported to the BioLab for analysis.
-------
4 Analytical Procedures
The NHSRC Research Triangle Park (RTP) BioLab analyzed samples either qualitatively
for spore presence (quality control, swab samples) or quantitatively for the number of viable
spores recovered per sample. Results were reported in colony forming units (CFU) per unit
volume. Details of the analytical procedures are provided below. A laboratory notebook was
used to document the details of each sampling event (or test).
4.1 Sample Extraction
4.1.1 Wipes
Wipe samples were received in 50-mL conical tubes. Spores were extracted from the
wipes by adding 20 ml_ PBST to each sample, then agitating the tubes using a vortex mixer (set
to maximum rotation) for two minutes in 10-second intervals. Aliquots were then removed for
plating. (Brown et al.. 2007)
4.1.2 Vacuum Socks
Spores were extracted by first cutting the vacuum sock into small pieces with sterile
scissors. The vacuum sock pieces and any residual debris were deposited into a specimen cup
preloaded with 20 ml_ sterile PBST. Specimen cups were agitated on the rotary shaker at 300
rpm for 30 min, then aliquots were removed for plating (Brown et al., 2007 and Calfee et al.,
2013b).
4.1.3 Small vacuum 37-mm cassettes
Spores were extracted from the 37-mm cassette filter with a total of 10-mL sterile PBST.
PBST was pipetted from the 10-mL stock into the cassette to wet the inside contents thoroughly,
then the cassettes were opened. The wetted filter and debris were placed in a 120-mL
specimen cup. Additional stock volumes of PBST were used to further rinse the inside of the
cassette, and all liquid was pipetted into the specimen cup.
The nozzles were extracted separately using 50 ml_ sterile PBST. After agitation of the
PBST over the nozzle, PBST was removed to a sterile 50-mL conical tube and centrifuged for
15 min at 3500 x g (gravity) to pellet the spores. Approximately 46 to 47 mL of the PBST was
removed and discarded. The remaining liquid was vortexed to re-suspend the spore pellet, then
all contents were added to the specimen cup. The total extraction volume was then recorded.
4.2 Spiral Plating and Filter Plating
Sample extracts that required dilution were plated in triplicate using a spiral plater
(Autoplate 5000, Advanced Instruments Inc., Norwood, MA). The automated spiral plater
deposits the sample in exponentially decreasing amounts across a rotating agar plate in
concentric lines to achieve three 10-fold serial dilutions on each plate. Plates with Bg samples
were incubated at 35 ± 2 °C for 18 to 20 hours. Plates with Btk samples were incubated at 28 ±
2 °C for 18 to 22 hours. During incubation, the colonies develop along the lines where the
47
-------
sample was deposited (see Figure 4-1). Colonies on each plate were enumerated using a
QCount® colony counter (Advanced Instruments Inc., Norwood, MA).
Figure 4-1. Bg (panel A) and Btk (panel B) Bacterial Colonies (CFU) on a Spiral-plated
Agar Plate
Positive control samples were diluted 100-fold (10~2) in PBST before spiral plating;
samples of unknown concentration were plated with no dilution and with a 100-fold dilution.
Samples with known low concentrations were plated with no dilution. The QCount® colony
counter was used to automatically calculate the CFU/mL in a sample based on the dilution
plated and the number of colonies that develop on the plate. The QCount® recorded the data in
an Microsoft Excel spreadsheet.
Only plates that met 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® result. Filter plating was performed using 100-mL capacity Micro-
Funnel™ unit with 0.45 pm GN-6 Metricel membranes (Pall Corporation, Port Washington, NY)
and a vacuum manifold (Pall Corporation, Port Washington, NY). The filters were placed onto
TSA plates and incubated at 35 ± 2 °C for 20 to 24 hours before manual enumeration. Figure 4-
2 shows filter plates with colonies of Bg (panel A) and Btk (panel B).
48
Figure 4-2. Bg (A) and Btk (B) Bacterial Colonies (CFU) on a Filter Plate
-------
4.3 Wet Vacuum Sample Processing
Samples with large amounts of debris were homogenized by manually shaking the
sample by hand followed by spiral plating as described above. Although the debris caused
minor discoloration of the TSA plates, the samples still produced reliable and reproducible
colony counts and required no further analysis. Spiral-plated colony counts are unreliable for
samples with a small number of spores. While the existing protocol was to filter-plate the
samples with low colony counts, the debris clogged the filters and prevented any colonies from
growing on the plates. A new process was developed to process large sample volumes with
large amounts of debris and background contamination while maintaining a high sensitivity to
the low concentration of spores.
Sample containers with small numbers of spores and large amounts of debris were
continuously shaken by hand for 2 min for thorough mixing. A homogenized aliquot was
collected from the bulk volume (aliquot volume determined based on anticipated sample
concentration) from each sample and was heat treated at 80 °C for 20 min while being vortexed
every 5 min. Following this heat shock, the samples were centrifuged at 5500 x g for 15 min at 4
°C. The supernatant was filter-plated, and the pellet was resuspended in PBST and manually
spread-plated on TSA plates. The plates were incubated at 35 ± 2 °C for 18 to 24 hours prior to
manual enumeration.
49
-------
5 Results and Discussion
This section discusses the results of the Phase I that consisted of determining the
operational parameters that would allow the highest Bacillus spore recovery, and Phase II,
which consisted of evaluating four types of commercially-available wet vacuum cleaners
(portable, residential, commercial and wet/dry cleaners) for spore sampling efficiency, ease of
use, and overall reliability.
5.1 Phase I: Evaluation of Wet Vacuum Cleaner Operational Parameters
The Phase I experimental approach consisted of assessment of a sampling liquid, the
temporal lapse between liquid application and suction, and volume of liquid collected as a
function of surface type.
5.1.1 Selection of Sampling Liquid
5.1.1.1 Liquid Volume Recovery
Three types of liquids were tested for sample collection to determine the proper
sampling liquid type. Tables 5-1 through 5-3 show the volumes of liquid applied and volumes
recovered for each type of material/liquid type combination. The volumes collected were
dependent on material type. The highest volume recovery was from laminated wood (72-80%),
followed by carpet (39-49%), and with the lowest recovered collection liquid from concrete (16-
19%).
Table 5-1. Volume of
iquid Recovered from Carpet Coupons
Liquid
Coupon
Spray
Time
Elapsed
Time
Volume
Applied
Volume
Recovered
Liquid
Recovery
Average
Recovery
Collection
Time
sec
sec
mL
mL
%
%
sec
1
19
160
317
107
34
300
2
19
160
317
108
34
240
3
19
160
317
157
50
39
240
-t—¦
1
4
19
120
317
121
38
180
Q
5
19
150
317
117
37
240
1
20
160
333
139
42
180
2
20
160
333
144
43
240
3
20
160
333
120
36
39
180
PBST
4
20
160
333
130
39
180
5
20
160
333
121
36
150
1
18
160
300
132
44
180
-i—1
2
17
160
283
143
50
180
3
18
160
300
153
51
49
180
1 ®
I %
Q F
4
18
160
300
165
55
180
5
17
160
283
122
43
150
50
-------
Table 5-2. Volume of Liquid Recovered from Concrete Coupons
Liquid
Coupon
Spray
Time
Elapsed
time
Volume
Applied
Volume
Recovered
Liquid
Recovery
Average
Recovery
Collection
Time
sec
Sec
mL
mL
%
%
sec
1
30
60
500
96
19
150
2
24
60
400
76
19
60
o
> C\J
2
25
60
417
78
19
60
3 ®
§ 5
> ^
3
25
60
417
39
9
16
112
4
25
60
417
75
18
60
Q ^
5
25
60
417
89
21
60
Table 5-3
. Volume of Liquid Recovered from Laminated Wood Coupons
Liquid
Coupon
Spray
Time
Elapsed
time
Volume
Applied
Volume
Recovered
Liquid
Recovery
Average
Recovery
Collection
Time
sec
sec
mL
mL
%
%
sec
1
6
60
100
78
78
150
2
6
100
100
66
66
149
Dl Water
3
6
100
100
70
70
72
159
4
6
100
100
75
75
150
5
6
100
100
73
73
150
1
7
104
117
73
62
149
2
7
90
117
77
66
151
3
7
105
117
89
76
72
143
PBST
4
7
97
117
89
76
138
5
7
70
117
90
77
143
1
6
120
100
72
72
140
Dl Water with
Tween® 20
2
6
105
100
86
86
140
3
6
76
100
78
78
80
142
4
6
76
100
84
84
114
5
6
74
100
78
78
144
5.1.1.2 Spore Recovery as a Function of Liquid Type
The recoveries for each type of material/liquid collection type combination are presented
in Table 5-4 and illustrated in Figure 5-1. The stainless-steel recovery results were collected by
PRB wipe sampling in Figure 5-1. The percent recoveries for each type of material/liquid
collection were calculated as a percent of the total recovery on the stainless-steel control
coupons. Dl-Tween liquid solution was the highest performing liquid among the three
liquid/material combinations tested with an average recovery of 53% with laminated wood
51
-------
coupons, while Dl water had the lowest average recovery of 9 % when used with concrete
coupons. Overall, all three solutions performed within one order of magnitude of the stainless-
steel coupon average recovery.
Table 5-4. Spore Recovery as a Function of Liquid Collection Type
Spore Recovery (CFU) Summary for Different Liquid Types
Test
ID
Liquid Used
Material
Positive Control
Recoveries
Wet Vacuum Recoveries
Average
Stdev
Average
Stdev
%
1A
Dl Water
Carpet
8.19E+05
2.49E+04
2.03E+05
8.83E+04
25
Concrete
6.31 E+04
2.67E+04
9
Laminated Wood
2.02E+05
1.20E+05
27
"IB
PBST
Carpet
6.80E+05
2.79E+04
1.47E+05
2.34E+04
18
Concrete
9.62E+04
3.93E+04
14
Laminated Wood
3.65E+05
1.08E+05
49
1C
Dl Water with
Tween® 20
Carpet
7.50E+05
1.16E+05
3.02E+05
1.43E+05
37
Concrete
9.96E+04
7.26E+04
15
Laminated Wood
3.95E+05
1.87E+05
53
1(f-
o
&
CD
5
(D
a:
CD
8.
CO
.O
105-
104-
103-.
Di Water
PBST
Tween20
Stainless Steel
Carpet Concrete
Material Type
Laminate
Figure 5-1. Spore Recovery for Material/Liquid Collection Combination
52
-------
5.1.2 Elapsed Time
5.1.2.1 Sample Liquid Recovery Volume
This test was designed to determine the effect of delay between liquid deposition and
retrieval on sampling efficiency. The tests were conducted using the Dl water with 0.05% Tween®
20. The elapsed time between the liquid spray application and vacuuming of the target material
was evaluated. Concrete paver coupons are highly water-absorptive, so it is difficult to retrieve
enough liquid volume for analysis.
In this test, the liquid retrieval from the concrete surfaces was conducted right after the
surface spraying. For concrete coupons, the overall contact time that was comprised of the spray
time, elapsed time, and sampling time was used as an alternative parameter for the optimization
process. The overall contact time was varied by partitioning the surface area of each coupon to
enable better collection of the sample liquid.
The results for the total liquid volume applied and total volume recovered for each type
of material/elapsed time combination are presented in Table 5-5 for carpet material, Table 5-6
for the concrete material/surface partition combination, and Table 5-7 for laminated wood
material. The average liquid agent volumetric ratio was calculated using the actual volume
collected over the target volume applied determined by the number of sprays from the backpack
sprayer. The volume of liquid sample collected was found to be independent of elapsed time,
but dependent on material type.
Recoveries for carpet and laminated wood, averaged over all elapsed times tested, were
52 ± 5.8% and 123 ± 8.8%, respectively. Partitioning the concrete coupon surface area into two
halves or four quadrants increased the liquid sample collection from an average of 15 ± 4.3% for
an average contact time of 96 sees (see Table 5-2), 29 ± 9.2 % for a contact time of 45 sees, to
an average of 33 ± 5.3% for a contact time of 21 sec.
53
-------
Table 5-5. Volume of Liquid Recovered from Carpet at Various Elapsed times
Elapsed
Time
Coupon
Spray
Time
Volume
Applied
Volume
Recovered
Collection
Time
Liquid Recovery
per Test
Average Liquid
Recovery
Sec
ID
Sec
mL
mL
Sec
%
%
1
15
250
132
53
2
15
250
133
53
1
3
15
250
145
58
54
4
15
250
123
49
5
15
250
140
56
1
15
250
107
43
2
15
250
152
61
30
3
15
250
129
52
52
4
15
250
142
160
57
5
15
250
118
47
1
15
250
131
52
100
2
15
250
146
58
54
3
15
250
126
50
1
15
250
97.0
39
2
15
250
125
50
300
3
15
250
110
44
49
4
15
250
138
55
5
15
250
137
55
Table 5-6. Volume of Liquid Recovered from Different Concrete Coupon Surface Partitions
Spray Time
Liquid
Coupon
Surface
Partition
Coupon
Time
1
Time
2
Time
3
Time
4
Applied
Recovered
Recovered
per Test
Average
Contact
Time
ID
sec
sec
sec
sec
mL
mL
%
%
sec
1
25
X
X
X
417
56
13
60
2
25
X
X
X
417
78
19
60
Full
3
25
X
X
X
417
39
9
16
112
4
25
X
X
X
417
75
18
60
5
25
X
X
X
417
89
21
60
1
15
15
X
X
500
119
24
Halves
(0.5)
2
15
15
X
X
500
105
21
3
15
15
X
X
500
176
35
29
45
4
15
15
X
X
500
116
23
5
15
15
X
X
500
211
42
1
6
6
6
6
400
136
34
Quarters
(0.25)
2
6
6
6
6
400
136
34
3
6
6
6
6
400
136
34
33
21
4
6
6
6
6
400
101
25
5
6
6
6
6
400
160
40
54
-------
Table 5-7. Volume of Liquid
Recovered from Laminated \N
ood at Different Elapsed Times
Elapsed
Time
Coupon
Spray
Time
Volume
Applied
Volume
Recovered
Collection
Time
Liquid Recovery
per Test
Average Liquid
Recovery
sec
ID
sec
mL
mL
sec
%
%
1
6
100
137
137
2
6
100
118
118
1
3
6
100
116
116
122
4
6
100
127
127
5
6
100
114
114
1
6
100
117
117
2
6
100
128
128
30
3
6
100
116
116
120
4
6
100
123
160
123
5
6
100
118
118
1
6
100
130
130
100
2
6
100
116
116
124
3
6
100
125
125
1
6
100
118
118
2
6
100
148
148
300
3
6
100
122
122
125
4
6
100
116
116
5
6
100
121
121
5.1.2.2 Spore Recovery as a Function of Elapsed Time
The spore recoveries as a function of elapsed time for both carpet and laminated wood
are presented in Table 5-8 and illustrated in Figure 5-2. The results show that for these two
materials, the elapsed time (1-300 sec) between liquid application and suction and the type of
material had little or no effect on average recovery. However, for concrete, partitioning the
coupon area into halves or quarters had a marked effect on the total spores recovered, as
shown in Table 5-9 and illustrated in Figure 5-3.
The spore recovery increased from an average of 16.0 % (see Table 5-2) for one
spraying/sampling combination sequence covering the whole coupon surface (96 sec contact
time) to 28.7% for two consecutive sequences covering two halves of the coupon (45-sec
contact time), and 59.3% for four consecutive sequences covering four quarters of the total area
of the coupon (21-sec contact time).
55
-------
Table 5-8. Spore Recovery as a Function of Elapsed time for Carpet and Laminated Wood
Coupons
Spore Recovery (CFU) Summary for Various Elapsed times
Test
ID
Material
Elapsed
Time
Positive Control
Recovery
Wet Vacuum Recovery
Sec
Average
(CFU)
Standard
Deviation
Average
(CFU)
Standard
Deviation
%
2A
Carpet
1
6.27 x 105
5.35 x 104
1.61 x 105
2.86 x104
26
30
1.95 x 105
6.48 x104
31
100
2.35 x 105
5.41 x 104
38
300
1.91 x 105
4.97 x104
31
2C
Laminated
Wood
1
5.09 x 105
8.79 x 104
1.55 x 105
1.02 x105
31
30
1.46 x 105
4.37 x104
29
100
1.10 x 105
4.02 x104
22
300
1.99 x 105
3.93 x104
39
Stdev = Standard deviation
106-r
Residence Time (sec)
Figure 5-2. Spore Recovery at Different Elapsed Times
56
-------
Table 5-9. Spore Recovery as a Function of Number of Partitions on the Concrete Coupon
Spore Recovery (CFU) Summary for Various Elapsed Times - Divided Area
Test
ID
Material
Surface
(partitions)
Contact Time
(sec)
Positive Control
Recoveries
Average CFU Recovered
sec
Average
(CFU)
Standard
Deviation
Average
(CFU)
Standard
Deviation
%
2B
Concrete
2
45
5.96 x 105
2.26 x 105
1.71 x 105
1.09 x 105
29
4
21
3.54 x105
3.21 x 105
59
l l Stainless Stee
i I Concrete
112 4
Number of Partitions on the Coupon
Figure 5-3. Spore Recovery for Different Number of Partitions on the Concrete Coupon
5.1.3 Liquid Volume
5.1.3.1 Sample Liquid Recovery Volume
This test was designed to determine the impact of applied liquid volume on the sampling
efficacy, using the most efficient extraction liquid (Dl water with 0.05% Tween® 20). The results
from these initial tasks determined that the elapsed time had no effect on spore recovery for
carpet and laminated wood; therefore, elapsed time was set at a constant value of 30 seconds.
For concrete surfaces, a coupon partitioned into quadrants was used for the liquid volume test
based on the results in the previous section.
The results for the total volume recovered for each type of material and volume applied
combination are presented in Table 5-10 for carpet material, Table 5-11 for the concrete
/surface partition combination, and Table 5-12 for laminate wood. The liquid volume recovery
57
-------
fraction on carpet coupon decreased with less liquid volume sprayed: an average recovery of
5.4% for an average of 107 mL of initial spray, 8.5% for 161 mL, and 49% for 293 mL (Table 5-
1). These results suggest that the spraying liquid volume on carpet requires at least 161 mL per
ft2 to generate the target liquid volume (10 mL) for analysis.
rable 5-1C
). Liquid Volume Recovery from Carpet Cou
pon
Volume
Applied
Coupon
Spray
Time
Elapsed
Time
Collection
Time
Liquid Recovery
per Test
Average
Liquid
Recovery
Fraction
mL
ID
sec
sec
sec
mL
%
%
107
1
6
24
100
5.8
5.4
5.4
2
6
30
99
7.7
7.2
3
10
30
107
12
6.7
4
6
30
131
4.7
4.4
5
6
30
117
3.3
3.1
161
1
9
30
112
7.6
4.7
8.5
2
9
30
144
9.7
6.0
3
9
30
142
17.9
11.1
4
9
30
137
27.6
17.2
5
9
30
168
5.8
3.6
For concrete, decreasing the combined volume applied during the four-consecutive
spraying/sampling combination sequences did not show any real trend. The average recovery
was 36% collected for a net volume sprayed of 160 mL per coupon, 42% for a volume of 220
mL, and 33% for 500 mL (from Task 2). Contact seems to be the main driver for liquid sample
recovery from this type of material.
Table 5-11. Liquid Volume Recovery from Concrete Coupon
Volume
Applied
Coupo
n
Spray Time (sec)
Collection Time
(sec)
Liquid
Recovery per
Test
Average
Liquid
Recovery
Contact
Time
mL
ID
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
mL
%
%
Sec
1
3
2
2
2
29
34
40
34
50
31
160
2
2
2
2
2
33
38
33
37
57
40
352
35
3
2
2
2
2
36
46
42
41
67
47
4
2
2
2
2
29
19
-
31
30
NA1
1
3
3
4
3
42
38
44
34
105
45
220
2
3
3
3
3
31
32
33
29
97
45
42
39
3
3
3
3
3
33
38
44
38
78
36
4
3
3
3
3
38
33
34
36
90
42
1 Volume recovered by the NHSRC BioLab; the collection time for Q3 was not recorded.
2The average liquid recovery was calculated using the recorded collected times.
58
-------
The calculated volumetric fraction (%) of the collection liquid recovered from the
laminated wood coupons was found to be dependent on the total volume sprayed with an
average recovery of 57% for 18 ml_ of initial spray, 71% for 54 ml_, and 80% for 100 ml_ (from
Task 1, at an equivalent elapsed time of 30 sec). The low recoveries observed from this
nonporous material at the low volume (18 ml_) may be the result of the relative losses in the
nozzle and tubing of the wet vacuum sampling device.
Table 5-12. Liquid Volume Recovery from Laminated Wood Coupon
Target Volume
Applied
Coupon
Spray
Time
Elapsed
Time
Collection
Time
Liquid Recovery per
Test
Average Liquid
Recovery
mL
ID
sec
sec
sec
mL
%
%
1
1
30
164
10.5
58
2
1
30
199
12.9
72
18
3
1
30
167
12.7
71
57
4
1
31
183
NA
21
5
1
30
171
11.3
63
1
3
30
183
37.5
69
2
3
30
194
40.5
75
54
3
3
30
164
33
61
71
4
3
30
157
32
59
5
3
30
234
49.8
92
NA = Not available; data corrupted
5.1.3.2 Spore Recovery as a Function of Liquid Volume Sprayed
The spore recoveries as a function of total sample liquid volume sprayed onto both
carpet and laminated wood materials are presented in Table 5-13. The results show that for
carpet, decreasing the volume of liquid spray onto the coupons had a negative effect on the
overall spore recovery. The spore recovery for a 100 and 150 mL volume spray resulted in
average spore recoveries between 3.4 and 4.8%, while the recovery for a 250-mL spray was
31% (at an equivalent elapsed time of 30 sec). For the laminated wood material, the effect on
the spore recovery of the volume sprayed was found to be insignificant, with an average
recovery varying between 29 to 38% for volume of liquid agent sprayed between 20 and 100 mL
Table 5-13. Spore Recovery from Carpet and Laminated Wood Coupons
Spore Recovery (CFU) Summary for Various Sprayed Volumes
Test ID
Material
Volume
Sprayed
Spray
Duration
Positive Control
Recovery
Average CFU Recovered
mL
sec
Average
(CFU)
Stdev
(CFU)
Average
(CFU)
Stdev
(CFU)
%
3A
Carpet
100
6
1.56 x
10®
1.65 x
105
7.49 x104
4.01 x 104
4.8
150
9
5.37 x104
4.50 x 104
3.4
3C
Laminate
20
1
1.47 x
10®
1.58 x
105
4.51 x 105
1.60 x 105
31
50
3
5.60 x 105
2.15 x 105
38
59
-------
For the concrete coupons, the spore recovery was not affected by the liquid volume
sprayed as seen in Table 5-14. The spore recovery varied from a low average of 29% (500 ml_,
contact time 45 sec) as seen in Table 5-6 to a high recovery of 36% (220 ml_, 39 sec) found in
Table 5-14.
Table 5-14. Spore Recovery from Concrete Coupon
Spore Recovery (CFU) Summary for Various Sprayed Volumes
Test
ID
Material
Surface
Partition
Spray
Duration
Volume
Sprayed
Positive Control
Recoveries
Average CFU Recovered
Per
Coupon
sec
mL
Average
(CFU)
Standard
Deviation
(CFU)
Average
(CFU)
Standard
Deviation
(CFU)
%
3B
Concrete
4
3
220
1.33 x 10®
3.74 x 104
4.77 x 104
1.61 x 104
36
2
160
4.24 x 104
2.27 x 104
32
5.1.4 Phase I: Summary
The Phase I test results showed the following suggested sampling conditions:
• Dl water-Tween® liquid achieved the highest recovery among the tested liquid types.
• The spore recoveries were found to be dependent on the recovered volumetric fraction
of liquid sprayed onto the coupons for carpet and concrete.
• Elapsed time between the liquid application and vacuuming of the target material had
little or no effect on the wet vacuum sampling spore recovery for laminated wood and
carpet materials. Concrete surface test results showed a marked effect of elapsed time
on the spore recovery effectiveness. Concrete surfaces need to be vacuumed
immediately after the liquid is applied.
• The total volume of liquid sprayed onto the carpet material seems to affect the spore
recovery. Carpet surface requires enough volume of liquid to be applied to the surface
for higher spore and liquid volume recovery. For laminated wood, the effect of the
volume sprayed on the spore recovery seems to be insignificant. For concrete, the liquid
volume seems to have a negligible effect within the tested conditions.
5.2 Phase II - Commercially-Available Wet Vacuum Cleaner Evaluation
The spore recovery efficiencies were estimated for four types of wet vacuum cleaners.
Sampling efficiency was estimated by comparing the spore recovery from test coupons to the
recovery from inoculum control coupons. Since each test coupon was inoculated four times
using the same MDI, while control coupons were inoculated once, the resulting spores
recovered from control coupons were multiplied by four when calculating comparative
recoveries. The spore recoveries for all control coupons (stainless steel, and material-specific),
then the commercial vacuum cleaner recoveries, and finally proposed sampling and extraction
procedures are discussed in the next sections.
60
-------
5.2.1 Control Sample Recoveries
Stainless steel inoculum control coupons were used to verify the magnitude and
repeatability of spore loadings for every inoculation event. A total of 36 stainless steel coupons
were inoculated with Bg spores, and 38 stainless steel coupons were inoculated with Btk
spores. Both test organisms were sampled using the wipe sampling approach during all the
reported testing events.
The data obtained for the inoculum controls are shown in Table 5-15 and illustrated in
Figure 5-4. The repeatability of the inoculation control checks was within the data criteria set at
50% RSD for these levels of inoculation controls.
Table 5-15. Spore Recovery from Concrete Coupon
Target Organism
Replicates
Spore Recovery (CFU)
Average
Standard
Deviation
RSD (%)
Bg
36
1.85E+08
9.31 E+07
50.3
Btk
38
8.71 E+07
4.15E+07
47.6
5.2.2 Wet-Vacuum Cleaner Evaluation
The sampling efficiency of the commercially-available cleaners is a measure of the
spores recovered from a contaminated material surface by the vacuums as compared to the
spores recovered from inoculum control coupons, sampled using wipes. The spore recovery
efficiencies relative to the number of spores deposited was estimated for each of the four
vacuum cleaners (portable, commercial, residential, and Shop Vac), for all three materials
tested (carpet, vinyl flooring, and concrete) inoculated with Bg and Btk spores. The efficacy of
each vacuum cleaner was also compared to the alternative sampling methods. Detailed data for
Phase II are shown in Appendix A.
To determine if there were significant differences in the mean spore recovery among the
vacuum cleaners, material types, and among the vacuum cleaners and the traditional sampling
methods, a Tukey Multiple Comparison Test (Tukev, 1949) was used to compare all possible
differences in spore recoveries between each pair of sampling method means (example:
portable vacuum cleaner and vacuum sock for carpet material). This method is based on the
one-way statistical analysis of variance or ANOVA. The Null Hypothesis used for the
comparison testing is that the means at all levels are equal, and the alternative hypothesis is
that the means of one or more levels are different. The confidence interval level was set at 95%.
5.2.2.1 Carpet
The sampling efficacies for each vacuum cleaner and the vacuum sock for carpet are
presented in Table 5-16 and illustrated in Figure 5-4. The one-way ANOVA returns Btk spore
recoveries, suggest that at the 0.05 significance level, the means of the wet vacuum cleaners
are not significantly different (F-value = 3.30, p-value = 0.078) but are significantly different from
the established vacuum sock sampling method (F-value = 6.49, p-value = 0.0076). However, for
Bg spores, the results suggest that one commercial vacuum cleaner type outperformed all the
other vacuum cleaners as well as the inoculum controls and the vacuum sock positive controls.
61
-------
Table 5-16. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spore
Recoveries from Carpet
Btk Spore Recovery (%)
Sampler Type
Replicates
Mean
Standard Deviation
Standard Error of Mean
Portable
3
24
5
19
Commercial
3
29
7
24
Residential
3
8
4
44
Shop Vac
3
17
14
83
Vacuum Sock
3
48
15
41
Bg Spore Recovery (%)
Portable
3
20
6
28
Commercial
3
140
43
31
Residential
3
46
21
46
Shop Vac
3
10
5
44
Vacuum Sock
3
27
13
48
100
200-
Sampling Method
Sampling Method
Figure 5-4. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Carpet
The one-way ANOVA performed for all vinyl results, independent of the type of
surrogates, is summarized in Table 5-17. The three vacuum cleaners (portable, residential, and
shop vac) showed nearly the same recoveries as the vacuum sock sampling method. The
commercial wet vacuum cleaners, however, seem to be biased with the results obtained from
the Bg spore test.
62
-------
Table 5-17. Sampling Efficacy of the Various Sampling Methods for Spores (Btfcand Bg
data pooled) Inoculated on Vinyl Flooring
BtklBg Spore Recovery (Pooled Data)
Sampler Type
Replicates
Mean
Standard
Deviation
RSD
Portable
3+3
22
5
23
Commercial
3+3
84
66
79
Residential
3+3
27
25
92
Shop Vac
3+3
14
10
73
Vacuum Sock
3+3
39
16
41
Tukey statistical tests, listed in Table 5-18, were performed with and without the results
obtained with the commercial vacuum cleaners for Bg spores. In the absence of the commercial
vacuum cleaner results, the one-way ANOVA shows that there is no statistically significant
difference in the means between all four vacuum cleaner recoveries and the vacuum sock
sampling technique. The one test with the commercial wet vacuum biased the comparison when
it was compared to other sampling techniques.
Table 5-18. Tukey Pairwise Statistical Test Results for Various Sampling Methods on
Pairwise Comparisons
With Commercial Wet
Vacuum/Bg Test
Without the Commercial
Wet Vacuum/Bg Test
p-Value
Sig1
p-Value
Sig
Commercial/Portable
0.02337
1
0.96476
0
Residential /Portable
0.99839
0
0.97037
0
Residential/ Commercial
0.04466
1
0.99989
0
Shop Vac /Portable
0.99238
0
0.88367
0
Shop Vac /Commercial
0.00844
1
0.63004
0
Shop Vac/Residential
0.95018
0
0.54536
0
Vacuum Sock /Portable
0.88867
0
0.31414
0
Vacuum Sock/Commercial
0.15905
0
0.87182
0
Vacuum Sock/Residential
0.9694
0
0.66603
0
Vacuum Sock/ Shop Vac
0.66667
0
0.05836
0
1Sig equal to 1 indicates that the difference in the means is significant; when equal to 0, indicates
that the difference is not significant at the 0.05 level.
5.2.2.2 Vinyl Flooring
The sampling efficacies for each vacuum cleaner and the PRB wipe method for the vinyl
flooring are presented in Table 5-19 and illustrated in Figure 5-5. The one-way ANOVA returns
suggest that at the 0.05 significance level, the means of the wet vacuum cleaners are not
significantly different (Btk\ F-value = 1.33, p-value = 0.317; Bg\ F-value = 1.39, p-value = 0.314).
However, the PRB wipe sampling method out-performed the wet vacuum cleaners. Confirming
the results obtained for Bg recovery on carpet, the commercial vacuum cleaner had a better
spore recovery than the other wet vacuum cleaners.
63
-------
Table 5-19. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Recovery on Vinyl Flooring
Btk Spore Recovery (%)
Sampler Type
Replicates
Mean
Standard Deviation
RSD (%)
Portable
3
31
38
123
Commercial
3
18
11
62
Residential
3
25
4
14
Shop Vac
3
28
24
86
PRB Wipes
3
106
39
37
Bg Spore Recovery (%)
Portable
3
48
9
19
Commercial
3
87
59
68
Residential
3
41
7
17
Shop Vac
3
68
7
11
PRB Wipes
3
95
9
10
160-
140-
~ 120-
20-
Btk Spores
y
Sampling Method
J?
Ł
CL
c/>
Sampling Methods
Figure 5-5. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Vinyl Flooring
The one-way ANOVA returns for all vinyl flooring results, independent of the type of
surrogates, are summarized in Table 5-20. The returns, suggest that at the 0.05 significance
level, the means of the wet vacuum cleaners are not significantly different (F-value = 0.397, p-
value = 0.756). As expected, the PRB wipe sampling method, in general, out-performed the wet
vacuum cleaner sampling approach.
64
-------
Table 5-20. Sampling Efficacy of the Various Sampling Methods for Both Btk and Bg Spore
Recovery on Vinyl Material
Btk /Bg Spore Recovery
Sampler Type
Replicates
Mean
Standard Deviation
RSD (%)
Portable
3+3
40
26
10
Commercial
3+3
53
54
22
Residential
3+3
33
10
4
Shop Vac
3+3
48
27
11
PRB Wipes
3+3
100
27
9
5.2.2.3 Concrete
The sampling efficacies for each vacuum cleaner and the established 37-mm cassette
method for concrete are presented in Table 5-21 and illustrated in Figure 5-6. The one-way
ANOVA results suggest that at the 0.05 significance level, the means of the wet vacuum
cleaners are not significantly different for Btk spores (F-value = 1.33, p-value = 0.317).
However, for Bg spores, the Shop-Vac and the 37-mm cassette did not perform as well as the
other wet vacuum cleaners.
Table 5-21. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spore
Recovery on Concrete Material
Btk Spore Recovery
Sampler Type
Replicates
Mean
Standard Deviation
RSD (%)
Portable
3
20
13
65
Commercial
3
35
10
28
Residential
3
29
9
31
Shop Vac
3
47
12
26
37-mm
Cassette
3
44
26
60
Bg Spore Recovery
Portable
3
50
19
39
Commercial
3
20
2
7.5
Residential
3
49
8
15
Shop Vac
3
6
3
51
37-mm
Cassette
3
13
8
58
65
-------
100
d)
>
o
o
<1)
OH
CL
CO
Sampling Method
Figure 5-6. Sampling Efficacy of the Various Sampling Methods for Btk and Bg Spores
Inoculated on Concrete
The one-way ANOVA results for all concrete test results, independent of the type of
surrogate, are summarized in Table 5-22. The results suggest that at the 0.05 significance level,
the means of the wet vacuum cleaners and the 37-mm cassette method are not significantly
different (F-value = 0.474, p-value = 0.754).
Table 5-22. Sampling Efficacy of the Various Sampling Methods for Both Btk and Bg Spore
Recovery on Vinyl Material
Btk/Bg Spore Recovery
Sampler Type
Replicates
Mean
Standard
Deviation
RSD
(%)
Portable
3+3
37
22
59
Commercial
3+3
28
11
38
Residential
3+3
39
13
34
Shop Vac
3+3
26
24
92
PRB Wipes
3+3
28
24
86
5.2.3 Phase II: Summary
All the wet vacuum cleaner spore recoveries were comparable to the spore recoveries of
the alternate sampling methods. A two-way ANOVA (Material type/Sampling method) performed
with the wet vacuum cleaners (72 samples) is shown in Table 5-23. The results demonstrated
that the effect of material type on mean recoveries was not statistically significant for all types of
vacuum cleaners (F-value = 0.446, p-value = 0.642), while the effect of sampling methods on
the mean recovery was statistically significant (F-value = 3.03). The interaction of the two
factors showed no significant difference in the mean recovery (F-value = 2.06, p-value = 0.07) at
the 0.05 level.
66
-------
Table 5-23. Two-Way ANOVA on the Mean Sampling Efficacy of the Wet Vacuum Cleaners
Source of Variation
SS
df
MS
F
p-Value
F critical
Material Type
846.0533
2
423.0267
0.446071
0.642244
3.150411
Sampling method
8628.085
3
2876.028
3.032698
0.036072
2.758078
Interaction (Sampling method/Material type)
11736.86
6
1956.143
2.062702
0.071075
2.254053
The overall results for the analysis of the wet vacuum cleaners (independent of material
type and surrogate type) are presented in Table 5-24 and in Figure 5-7. In terms of both
usability and repeatability, the residential wet vacuum cleaner was found to be the better
universal wet vacuum cleaner for wide area sampling of Bacillus spores. The residential wet
vacuum cleaner was found to be more precise (RSD = 50%), more user-friendly (lighter than the
commercial wet vacuum cleaner, and less cumbersome to use than the portable vacuum
cleaner (the sampler needs to bend to sample flooring), and less prone to cross-contamination
than the Shop-Vac, which needs a second device for wetting the contaminated surface.
Table 5-24. Overall Sampling Efficacy of the Various Sampling Met
Btk/Bg Spore Recovery
Sampler Type
Replicates
Mean
(%)
Standard
Deviation
RSD
(%)
Portable
18
32
20
63
Commercial
18
55
52
96
Residential
18
33
17
51
Shop Vac
18
25
26
100
140
120
S? 100-
>
O
o
or
O
Q_
W
iods
c" «-
Wet Vacuum Sampling Method
Figure 5-7. Sampling Efficacy of the Various Sampling Methods for Bacillus Spores
Independent of Type of Material
67
-------
6 Quality Assurance and Quality Control
All test activities were documented in laboratory notebooks and digital photographs. The
documentation included, but was not limited to, a record for each decontamination procedure,
any deviations from the quality assurance project plan, and physical impact on materials. All
tests were conducted in accordance with established EPA Decontamination Technologies
Research Laboratory (DTRL) and NHSRC RTP Microbiology Laboratory procedures to ensure
repeatability and adherence to the data quality validation criteria set for this project.
The following sections discuss the criteria for the critical measurements and parameters,
Data Quality Indicators (DQIs), and the quality assurance (QA) and quality control (QC) checks
for the project
6.1 Criteria for Critical Measurements/Parameters
The Data Quality Objectives (DQOs) are used to determine the critical measurements
needed to address the stated objectives and specify tolerable levels of potential error
associated with simulating the prescribed decontamination environments. The following
measurements were deemed to be critical to accomplish part or all project objectives:
• Volume or mass of sampling liquid,
• Spray time,
• Run time,
• Incubation temperature,
• Plated volume,
• CFU counts.
The DQIs for the critical measurements are listed in Table 6-1. DQIs were used to
determine if the collected data met the quality assurance objectives. Decisions to accept or
reject test results were based on engineering judgment used to assess the likely impact of the
failed criterion on the conclusions drawn from the data. The acceptance criteria were set at the
most stringent levels that can be achieved routinely. The integrity of the 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 project used established and approved operating procedures for the maintenance
and calibration of all laboratory equipment. All laboratory measuring devices such as scales and
pipettors used in this project were certified as having been recently calibrated or were calibrated
at the on-site EPA Metrology Laboratory at the time of use. Deficiencies, if any, were noted and
the instrument replaced to meet calibration tolerances. All DQIs were within the target
acceptance criteria set for this project as shown in Table 6-1.
68
-------
Table 6-1. DQIs and Acceptance Criteria Validation for Critical Measurements
Measurement
Parameter
Analysis Method
Acceptance Criteria
Pass or
Fail Test
Mass of sampling liquid
Scale
Accuracy: 0.1 g
Pass
Volume of sampling liquid
Serological pipette - certified as
calibrated
Subdivision: 0.5 mL
Pass
Time
National Institute of Standards and
Technology (NIST)-calibrated
stopwatch
± 1 minute
Pass
Counts of CFU per plate
Q-count
1.82 x 104
-------
unanticipated contamination. Replicate coupons were included for each set of test conditions to
characterize the variability of the test procedures.
Additional QC samples were collected and analyzed to check the ability of the BioLab to
culture the test organism, as well as to demonstrate that materials used in this effort did not
contain spores. The checks included the following:
• Procedural blank coupons: Material coupons sampled in the same fashion as test
coupons but not contaminated with surrogate organism prior to sampling.
• Stainless-steel and carpet positive control coupons: Coupons inoculated in tandem
with the test coupons and meant to demonstrate the highest level of contamination
recoverable from an inoculation event.
Additional QC checks for BioLab procedures are listed in Table 6-2. These QC checks
provide assurances against cross-contamination and other biases in microbiological samples.
Table 6-2. Additional Quality Checks for Biological Measurements
Sample Type
Frequency
Acceptance Criteria
Information Provided
Corrective Action
Positive control coupon -
sample from material
coupon contaminated
with biological agent and
sampled using the
existing sampling
methods
Minimum of
three per test
105 - 107 for B. globigii
50% RSD between coupons
in each test set
Used to determine the
extent of inoculation on
the target coupon type
If outside range, Identify
and remove source of
variability if possible
Inoculum Control coupon
- stainless steel coupon
contaminated with
biological agent and
sampled using PRB
wipes.
Minimum of
three per test
105 - 107 for B. globigii
50% RSD between coupons
in each test set
Used to determine the
extent of inoculation on
the target coupon type
If outside range, Identify
and remove source of
variability if possible
Procedural blank
coupon without biological
agent that underwent the
sampling procedure
One per test
Non-detect
Controls for sterility of
materials and methods
used in the procedure
Analyze extracts from
procedural blank without
dilution; identify and
remove source of
contamination if possible
Blank tryptic soy agar
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
%. Reportable CFU are
between 30 and 300 per
plate
Used to determine the
precision of the replicate
plating
Re-plate sample
Most of the wet vacuum control blank (negative control), the EPA accepted sampling
procedure blank (procedural blank), and the inoculum control blank (stainless-steel control blank)
were non-detectable (> 93%). Some control blanks were found to be contaminated, but they had
little or no effect on the final results. The source of this contamination is unknown. For negative
controls, the contamination may have occurred by incomplete inactivation of spores from the
materials during VHP® cycle. Procedural blanks may have become contaminated due to their
presence in an area with inoculated coupons.
70
-------
7 References
ASTM F1284-09. 2009. Standard Test Method for Evaluating Carpet Embedded Dirt Removal
Effectiveness of Residential Central Vacuum Cleaning Systems.
Brown, G.S., Betty, R.G., Brockmann, J.E., Lucero, D.A., Souza, C.A., Walsh, K.S., Boucher,
R.M., Tezak, M.S., and Wilson, M.C. Evaluation of Vacuum Filter Sock Surface Sample
Collection Method for Bacillus Spores from Porous and Non-Porous Surfaces. Journal of
Environmental Monitoring, 2007. 9(7), 666-671.
Buckley, P., Rivers, B., Katoski, S., Kim, M.H., Kragl, F.J., Broomall, S., Krepps, M., Skowronski,
E.W., Rosenzweig, C.N., Paikoff, S., Emanuel, P. and Gibbons, H.S. 2012. Genetic Barcodes for
Improved Environmental Tracking of an Anthrax Simulant. Applied and Environmental
Microbiology, 78(23), 8272-8280.
Calfee, M.W., Lee, S.D., and Ryan, S. A Rapid and Repeatable Method to Deposit Bioaerosols
on Material Surfaces. 2013a. Journal of Microbiological Methods, 92(3), 375-380.
Calfee, M.W., Rose, L.J., Morse, S., Mattorano, D., Clayton, M. Touati, A, Griffin-Gatchalian, N.,
Slone, C., and McSweeney, N. Comparative Evaluation of Vacuum-based Surface Sampling
Methods for Collection of Bacillus spores. Journal of Microbiological Methods. 2013b. 95(3),389-
96.
Emanuel, Peter A., Buckley, P.E., Sutton, T.A., Edmonds, J.M., Bailey, A.M., Rivers, B.A., Kim,
M.H., Ginley, W.J., Keiser, C.C., Doherty, R.W., Kragl, F.J., Narayanan, F.E., Katoski, S.E.,
Paikoff, S., Leppert, S.P., Strawbridge, J.B., VanReenen, D.R., Biberos, S.S., Moore, D., Phillips,
D.W., Mingioni, L.R., Melles, O., Ondercin, D.G., Hirsh, B., Bieschke, K.M., Harris, C.L., Omberg,
K.M., Rastogi, V.K., Van Cuyk, S. and Gibbons, H.S. Detection and Tracking of a Novel
Genetically Tagged Biological Simulant in the Environment. 2012. Applied and Environmental
Microbiology, 78(23), 8281-8288.
Gibbons, H.S., Broomall, S.M., McNew, L.A., Daligault, H., Chapman, C., Bruce, D., Karavis, M.,
Krepps, M., McGregor, P.A., Hong, C., Park, K.H., Akmal, A., Feldman, A., Lin, J.S., Chang, W.E.,
Higgs, B.W., Ddmirev, P., Lindquist, J. Liem, A., Fochler, E., Read, T.D., Tapia, R., Johnson, S.,
Bishop-Lilly, K.A., Detter, C., Han, C., Sozhamannan, S., Rosenzweig, C.N. and Skowronski,
E.W. Genomic Signatures of Strain Selection and Enhancement in Bacillus atrophaeus var.
globigii, a Historical Biowarfare Simulant. 2011. PLoS ONE 6(3): e17836.
doi: 10.1371/journal. pone.0017836.
Lee, S.D., Ryan, S.P. and Snyder, E.G. Development of an Aerosol Surface Inoculation Method
for Bacillus Spores. 2011. 77(5), 1638-1645.
Lee, S.D., Calfee, M.W., Mickelsen, L., Griffin, J., Wolfe, S., Clayton, M., Griffin-Gatchalian, N.,
and Touati, A. Evaluation of Surface Sampling for Bacillus Spores Using Commercially Available
Cleaning Robots. Environmental Science & Technology. 2013. 47(6), 2593-2601.
71
-------
Tukey, J.W. Comparing Individual Means in the Analysis of Variance. Biometrics. 1949. 5 (2), 99-
114.
USEPA, 2017. Biopesticide Active Ingredients, Biochemical and Microbial, as of September 30,
2017. (https://www.epa.gov/ingredients-used-pesticide-products/biopesticide-active-ingredients)
last accessed on 3/13/2018.
Valadares de Amorim, G., Whittome, B., Shore, B. and Levin, D.B. Identification of Bacillus
thuringiensis subsp. kurstaki Strain HD1-like Bacteria from Environmental and Human Samples
after Aerial Spraying of Victoria, British Columbia, Canada, with Foray 48B. 2001. Applied and
Environmental Microbiology, 67(3), 1035-1043.
72
-------
APPENDIX A: PHASE II: DATA REPORT
C1: Bg Spores
a) Carpet Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery
(%)
Stdev Recovery
(%)
Carpet
Portable - Bissell
Pro-Heat*
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
2.42E+00
0%
0.00%
Positive Control
Vacuum Sock
3.35E+07
20%
32%
15%
8.32E+07
49%
4.74E+07
28%
Inoculum Control
Gauze Wipe
1.21E+08
71%
100%
27%
2.10E+08
123%
1.80E+08
106%
Test Coupon
Wet Vacuum
3.12E+07
18%
20%
5%
2.52E+07
15%
4.34E+07
26%
Commercial -
RugDoctor
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
1.06E+02
0.00005%
0.00005%
Positive Control
Vacuum Sock
1.15E+08
59%
42%
21%
3.59E+07
18%
9.68E+07
50%
Inoculum Control
Gauze Wipe
1.68E+08
86%
100%
15%
2.26E+08
116%
1.92E+08
98%
Test Coupon
Wet Vacuum
4.84E+08
248%
187%
62%
3.64E+08
187%
2.44E+08
125%
Home Use - Hoover
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
Positive Control
Vacuum Sock
1.61E+07
16%
16%
6%
2.13E+07
21%
1.02E+07
10%
Inoculum Control
Gauze Wipe
7.89E+07
78%
100%
19%
1.16E+08
115%
1.07E+08
106%
Test Coupon
Wet Vacuum
2.46E+07
24%
46%
21%
4.74E+07
47%
6.74E+07
67%
ShoVac
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
5.60E+01
0%
Positive Control
Vacuum Sock
4.70E+07
25%
17%
8%
2.88E+07
15%
1.84E+07
10%
Inoculum Control
Gauze Wipe
2.02E+08
106%
100%
6%
1.81E+08
95%
1.87E+08
98%
Test Coupon
Wet Vacuum
1.14E+07
6%
10%
4%
1.82E+07
10%
2.83E+07
15%
73
-------
b) Concrete Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery {%)
Stdev Recovery {%)
Concrete
Portable - Bissell
Pro-Heat*
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Cassette
Wipe
1.58E+01
0%
Positive Control
37 mm cassette
1.52E+07
11%
11%
4%
1.01E+07
7%
2.19E+07
16%
Inoculum Control
Gauze Wipe
1.15E+08
83%
100%
34%
1.93E+08
139%
1.08E+08
78%
Test Coupon
Wet Vacuum
9.25E+07
67%
50%
19%
7.57E+07
55%
3.98E+07
29%
Commercial -
RugDoctor
Liquid
Procedural Blank
ND
ND
Blank inoculum Control
Wipe
ND
ND
Positive Control
37 mm cassette
9.90E+06
6%
5%
3%
2.19E+06
1%
9.90E+06
6%
Inoculum Control
Gauze Wipe
1.32E+08
82%
100%
23%
2.03E+08
126%
1.50E+08
93%
Test Coupon
Wet Vacuum
3.20E+07
20%
20%
1%
2.97E+07
18%
3.45E+07
21%
Home Use - Hoover
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
6.71E+02
ND
Blank Inoculum Control
Wipe
9.12E+01
0.00005%
Positive Control
37 mm cassette
1.17E+07
7%
13%
7%
3.62E+07
21%
1.94E+07
11%
Inoculum Control
Gauze Wipe
1.71E+08
100%
100%
5%
1.79E+08
105%
1.62E+08
95%
Test Coupon
Wet Vacuum
8.57E+07
50%
49%
8%
7.02E+07
41%
9.57E+07
56%
ShoVac
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
1.01E+02
Positive Control
37 mm cassette
4.41E+07
27%
23%
12%
5.15E+07
32%
1.45E+07
9%
Inoculum Control
Gauze Wipe
1.46E+08
90%
100%
10%
1.78E+08
110%
1.62E+08
100%
Test Coupon
Wet Vacuum
7.46E+06
5%
5%
3%
1.39E+07
9%
5.26E+06
3%
74
-------
c) Vinyl Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery {%)
Stdev Recovery {%)
Liquid
Procedural Blank
ND
ND
Negative Control
Gauze Wipe
5.60E+02
0.0001%
0.0001%
4.28E+08
94%
Positive Control
Gauze Wipe
4.38E+08
96%
96%
2%
4.44E+08
98%
Portable - Bissell
Pro-Heat*
4.66E+08
103%
Inoculum Control
Gauze Wipe
4.66E+08
103%
100%
5%
4.30E+08
95%
2.53E+08
56%
Test Coupon
Wet Vacuum
1.71E+08
38%
48%
9%
2.27E+08
50%
Liquid
Procedural Blank
ND
ND
Blank inoculum Control
Wipe
ND
ND
Negative Control
Gauze Wipe
1.15E+02
0.00008%
0.00008%
1.87E+08
131%
Positive Control
Gauze Wipe
1.23E+08
86%
106%
23%
Commercial -
1.46E+08
102%
RugDoctor
2.05E+08
143%
Inoculum Control
Gauze Wipe
1.15E+08
81%
100%
38%
1.09E+08
76%
8.59E+07
60%
Test Coupon
Wet Vacuum
6.57E+07
46%
87%
59%
Vinyl
2.21E+08
155%
Liquid
Procedural Blank
4.32E+04
0.02%
Negative Control
Gauze Wipe
ND
ND
Blank Inoculum Control
Wipe
ND
ND
1.50E+08
79%
Positive Control
Gauze Wipe
1.58E+08
83%
84%
6%
Home Use - Hoover
1.72E+08
90%
2.94E+08
154%
Inoculum Control
Gauze Wipe
1.40E+08
73%
100%
47%
1.40E+08
73%
9.48E+07
50%
Test Coupon
Wet Vacuum
7.22E+07
38%
41%
7%
7.02E+07
37%
Liquid
Procedural Blank
6.45E+04
0%
Negative Control
Gauze Wipe
2.66E+02
0%
Blank Inoculum Control
Wipe
ND
ND
1.82E+08
129%
ShoVac
Positive Control
Gauze Wipe
8.80E+07
62%
95%
33%
1.33E+08
94%
1.83E+08
130%
Inoculum Control
Gauze Wipe
8.08E+07
57%
100%
38%
1.59E+08
113%
1.06E+08
75%
Test Coupon
Wet Vacuum
9.62E+07
68%
68%
7%
8.57E+07
61%
75
-------
C2: Btk Spores
a) Carpet Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery {%)
Stdev Recovery {%)
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
4.40E+07
47%
Positive Control
37 mm cassette
4.88E+07
52%
62%
22%
Portable - Bissell
Pro-Heat*
8.24E+07
88%
9.68E+07
103%
Inoculum Control
Gauze Wipe
9.68E+07
103%
100%
6%
8.72E+07
93%
1.92E+07
21%
Test Coupon
Wet Vacuum
2.74E+07
29%
24%
5%
2.10E+07
22%
Liquid
Procedural Blank
7.58E+05
ND
Blank Inoculum Control
Wipe
ND
ND
Negative Control
Vacuum Sock
ND
ND
4.09E+07
39%
Positive Control
Vacuum Sock
2.70E+07
26%
32%
9%
Commercial -
No Data
RugDoctor
1.14E+08
109%
Inoculum Control
Gauze Wipe
7.55E+07
72%
100%
25%
1.25E+08
119%
3.43E+07
33%
Test Coupon
Wet Vacuum
3.46E+07
33%
29%
7%
Carpet
2.18E+07
21%
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
3.05E+07
34%
Positive Control
Vacuum Sock
4.90E+07
55%
50%
13%
Home Use - Hoover
5.26E+07
60%
1.17E+08
132%
Inoculum Control
Gauze Wipe
9.12E+07
103%
100%
34%
5.72E+07
65%
7.90E+06
9%
Test Coupon
Wet Vacuum
3.97E+06
4%
8%
4%
1.04E+07
12%
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
4.58E+07
45%
ShoVac
Positive Control
Vacuum Sock
3.87E+07
38%
27%
24%
3.51E+00
0%
7.27E+07
71%
Inoculum Control
Gauze Wipe
1.18E+08
115%
100%
25%
1.18E+08
115%
2.12E+06
2%
Test Coupon
Wet Vacuum
3.12E+07
30%
17%
14%
1.95E+07
19%
76
-------
b) Concrete Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery {%)
Stdev Recovery {%)
Concrete
Portable - Bissell
Pro-Heat*
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
Positive Control
37 mm cassette
1.20E+07
16%
18%
8%
2.10E+07
27%
8.80E+06
11%
Inoculum Control
Gauze Wipe
5.80E+07
75%
100%
23%
7.92E+07
103%
9.36E+07
122%
Test Coupon
Wet Vacuum
1.02E+07
13%
20%
13%
2.74E+07
36%
9.35E+06
12%
Commercial -
RugDoctor
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
Positive Control
37 mm cassette
1.48E+08
86%
59%
38%
2.53E+07
15%
1.27E+08
74%
Inoculum Control
Gauze Wipe
1.59E+08
93%
100%
24%
2.16E+08
126%
1.38E+08
80%
Test Coupon
Wet Vacuum
7.96E+07
47%
35%
10%
5.28E+07
31%
4.80E+07
28%
Home Use - Hoover
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
Positive Control (Can 1)
37 mm cassette
7.15E+07
123%
73%
43%
2.77E+07
48%
Positive Control (Can 3)
2.83E+07
49%
3.87E+07
67%
3.18E+07
55%
Inoculum Control (Can 1)
Gauze Wipe
8.48E+07
146%
100%
38%
6.85E+07
118%
Inoculum Control (Can 3)
2.78E+07
48%
6.34E+07
109%
4.60E+07
79%
Test Coupon
Wet Vacuum
1.57E+07
27%
29%
9%
1.26E+07
22%
2.27E+07
39%
ShoVac
Liquid
Procedural Blank
ND
ND
Negative Control
Vacuum Sock
ND
ND
Blank Inoculum Control
Wipe
ND
ND
Positive Control
37 mm cassette
1.00E+07
27%
26%
3%
1.00E+07
27%
8.20E+06
22%
Inoculum Control
Gauze Wipe
2.66E+07
72%
100%
36%
3.22E+07
88%
5.14E+07
140%
Test Coupon
Wet Vacuum
2.16E+07
59%
47%
12%
1.76E+07
48%
1.27E+07
35%
77
-------
c) Vinyl Results
Material
Vacuum Type
Sample Identification
Sample Method
Recovery (CFU)
Recovery {%) Compared to
Average Inoculation Control
Average Recovery {%)
Stdev Recovery {%)
Liquid
Procedural Blank
ND
ND
Blank inoculum Control
Wipe
ND
ND
Negative Control
Gauze Wipe
ND
ND
9.92E+07
129%
Positive Control
Gauze Wipe
8.40E+07
110%
105%
27%
Portable - Bissell
5.85E+07
76%
Pro-Heat*
8.16E+07
107%
Inoculum Control
Gauze Wipe
8.24E+07
108%
100%
12%
6.58E+07
86%
5.73E+07
75%
Test Coupon
Wet Vacuum
3.49E+06
5%
31%
38%
1.07E+07
14%
Liquid
Procedural Blank
ND
ND
Blank inoculum Control
Wipe
ND
ND
Negative Control
Gauze Wipe
ND
ND
8.32E+07
165%
Positive Control
Gauze Wipe
7.70E+07
153%
160%
6%
Commercial -
8.16E+07
162%
RugDoctor
6.24E+07
124%
Inoculum Control
Gauze Wipe
3.52E+07
70%
100%
28%
5.34E+07
106%
1.36E+07
27%
Vinyl
Test Coupon
Wet Vacuum
1.07E+07
21%
18%
11%
2.78E+06
6%
Liquid
Procedural Blank
ND
ND
Negative Control
Gauze Wipe
ND
ND
Blank Inoculum Control
Wipe
ND
ND
5.34E+07
82%
Positive Control
Gauze Wipe
5.41E+07
83%
86%
5%
Home Use - Hoover
5.90E+07
91%
4.51E+07
70%
Inoculum Control
Gauze Wipe
6.15E+07
95%
100%
33%
8.80E+07
136%
1.72E+07
26%
Test Coupon
Wet Vacuum
1.83E+07
28%
25%
3%
1.40E+07
22%
Liquid
Procedural Blank
ND
ND
Negative Control
Gauze Wipe
ND
ND
Blank Inoculum Control
Wipe
ND
ND
7.48E+07
53%
ShoVac
Positive Control
Gauze Wipe
8.32E+07
59%
72%
27%
1.43E+08
102%
1.04E+08
74%
Inoculum Control
Gauze Wipe
1.47E+08
105%
100%
24%
1.70E+08
121%
1.83E+06
1%
Test Coupon
Wet Vacuum
1.33E+07
9%
5%
4%
5.51E+06
4%
78
-------
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
------- |