p
CBRN CM AT
Methyl Bromide
Field Operation Guidance
(MB FOG) Report
April 13, 2015
Based on an Operational Decontamination Assessment of Methyl Bromide
Fumigant on Bacillus anthracis Sterne
-------
Questions concerning this document or its application should be addressed to:
Leroy Mickelsen, EPA
Co-Program Manager
mickelsen.leroy@epa.gov
Shannon D. Serre, EPA
Co-Program Manager
serre.shannon@epa.gov
2
-------
Disclaimer
This report has been peer and administratively reviewed and has been approved for publication
as an EPA document. It does not necessarily reflect the views of the EPA. No official endorsement
should be inferred. The EPA does not endorse the purchase or sale of any commercial products
or services. This report includes photographs of commercially available products. The
photographs are included for purposes of illustration only and are not intended to imply that the
EPA approves or endorses the product or its manufacturer.
3
-------
Acknowledgements
This study required the collaboration of Federal, academia, and contractor personnel for planning
and successful execution. The project could not have been successfully accomplished without
the collective commitment and contributions of all involved.
The following are acknowledged for their project planning and execution leadership:
Leroy Mickelsen, CBRN Consequence Management Division, EPA
Shannon Serre, National Homeland Security Research Center, EPA
The following are acknowledged for scientific planning, coordination, and execution:
Larry Kaelin, CBRN Consequence Management Division, EPA
Leroy Mickelsen, CBRN Consequence Management Division, EPA
Joe Schaefer, Environmental Response Team, EPA
Tim Boe, National Homeland Security Research Center, EPA
Worth Calfee, National Homeland Security Research Center, EPA
Marshall Gray, National Homeland Security Research Center, EPA
Shannon Serre, National Homeland Security Research Center, EPA
Joe Wood, National Homeland Security Research Center, EPA
John Archer, NERL, ORD, EPA
Rob Fox, OEM, OSWER, EPA
Stephen Ball, Region 4, EPA
Ben Franco, Region 4, EPA
Matt Huyser, Region 4, EPA
Jeremy Arling, Stratospheric Protection Division, OAR, EPA
Matt Clayton, Arcadis US
Nicole Griffin, Arcadis US
Tim McArthur, Arcadis US
Rene Borja, Cardinal Professional Products
Jeff Edwards, Dead Bug Edwards
Anne Busher, Dynamac Corp
Neil Daniell, Dynamac Corp
Ray Cardenas, Hammerhead Termite Control
Mark Weinberg, Hammerhead Termite Control
Bill Kern, University of Florida
Renny Perez, University of Florida (Fumigation School Director)
Rudi Scheffrahn, University of Florida
4
-------
The following are acknowledged for their role in the primary data analysis and authorship of this
report:
Leroy Mickelsen, CBRN Consequence Management Division, EPA
Tim Boe, National Homeland Security Research Center, EPA
Worth Calfee, National Homeland Security Research Center, EPA
Marshall Gray, National Homeland Security Research Center, EPA
Shannon Serre, National Homeland Security Research Center, EPA
Joe Wood, National Homeland Security Research Center, EPA
Tim McArthur, Arcadis US
Anne Busher, Dynamac Corp
Neil Daniell, Dynamac Corp
Bill Kern, University of Florida
Rudi Scheffrahn, University of Florida
The following are acknowledged for their role as the primary editor:
Katrina McConkey, Booz Allen Hamilton
5
-------
Table of Contents
Acronyms and Abbreviations 8
1. Introduction 12
1.1 Background 12
1.2 MB Usage and Properties 15
1.3 Health and Safety 17
1.4 Study Objectives 18
1.4.1 Objective 1 19
1.4.2 Objective 2 19
1.4.3 Objective 3 19
1.4.4 Objective 4 19
1.4.5 Objective 5 19
2 Materials and Methods 20
2.1 Facility 20
2.2 Sealing the House 23
2.3 Circulation Fans, Heaters and Humidifiers 30
2.4 Temperature and Relative Humidity Monitoring 32
2.5 Coupon Preparation 33
2.6 Analysis of Test Coupons 35
2.7 Spatial Assessment of Efficacy (Qualitative Test) 36
2.8 Temporal Assessment of Efficacy (Quantitative Test) 38
2.9 Pre and Post Sponge Stick Sampling 39
2.10 Activated Carbon Scrubber 39
2.10.1 Air Flow Rate at Inlet to First Carbon Vessel 40
2.10.2 Scrubber Temperature and Relative Humidity 41
2.10.3 Scrubber MB Concentration 41
2.10.4 MB Mass Balance Calculations for Activated Carbon Scrubber 42
2.11 Ambient Air Monitoring 42
2.12 Leak Detection 45
2.13 MB Fumigation Process 46
3 Results and Discussion 48
3.1 Results from Release and Monitoring of the MB 48
6
-------
3.2 House Temperature and Humidity Results 51
3.3 Leak Monitoring Around the Perimeter of the House Results 53
3.4 Biological Indicator (Bl) Results 55
3.4.1 Pre- and Post-Test Bl Population Comparison 55
3.4.2 Spatial Assessment of Efficacy (Qualitative Test) Results 55
3.4.3 Temporal Assessment of Efficacy - Quantitative Test (Time-Series Test) Results.. 57
3.5 Surface Sample (Sponge Wipe Samples) Results 58
3.6 Ambient Air Monitoring Results 59
3.7 Activated Carbon Scrubber Results 61
3.7.1 Flow Rate at Inlet to First Carbon Bed 61
3.7.2 Temperature and RH at the Carbon Scrubber during Scrubbing 62
3.7.3 Temperature and RH between the Carbon Beds and in the Scrubber Stack 63
3.7.4 MB Levels during Aeration 65
3.7.5 MB Mass Balance for Activated Carbon Bed System and for Entire Fumigation ... 67
3.8 Dispersion Modeling and Results 70
3.8.1 Scenario 1 71
3.8.2 Scenario 2 72
3.8.3 Modeling Discussion 73
3.9 House Entries 74
4 Conclusions and Recommendations: 74
4.1 Objective 1, Conclusion 74
4.2 Objective 2, Conclusion 75
4.3 Objective 3, Conclusion 75
4.4 Objective 4, Conclusion 75
4.5 Objective 5, Conclusion 75
4.6 Recommendations 76
5 References 78
Appendix A. Overall Operation of the Project: Lessons Learned 81
Appendix B. Ambient Air Monitoring Figures 86
Attachment 1: MB Fumigation Health and Safety Plan
Attachment 2: MB Fumigation Ambient Air Monitoring Plan
Attachment 3: MB Field Operational Guidance to New York City
7
-------
Acronyms and Abbreviations
AAMP Ambient Air Monitoring Plan
ACGIH American Conference of Governmental Industrial Hygienists
B Bacillus
Ba Bacillus anthracis
Bl biological indicator
CAA Clean Air Act
CBRN Chemical, Biological, Radiological, and Nuclear
cfm cubic feet per minute
CFU colony forming unit(s)
CIO2 chlorine dioxide gas
CMAD CBRN Consequence Management and Advisory Division
CRZ Contaminant Reduction Zone ("warm zone")
CT concentration times time (dose)
CUE critical use exemption
°C degrees Celsius
DHS Department of Homeland Security
EPA Environmental Protection Agency
ERT Environmental Response Team
EtO ethylene oxide
EVOH ethylene vinyl alcohol
EZ Exclusion Zone ("hot zone")
FAWN Florida Automated Weather Network
FID Flame ionization detector
ft feet
ft3 cubic feet
GAO Government Accountability Office
HASP Health and Safety Plan
HAZWOPER Hazardous Waste Operations and Emergency Response
HBr hydrogen bromide
hr hour
ID inner diameter
lb pound
m2 square meter
m3 cubic meter
MB methyl bromide, bromomethane, CHsBr
MET Metrological
mg/L milligrams per liter
mg/L-hr milligrams per liter times hour(s)
MOP method of procedure
NHSRC National Homeland Security Research Center
NIOSH National Institute for Occupational Safety and Health
OEL occupational exposure limits
8
-------
OSHA
Occupational Safety and Health Administration
PBST
phosphate buffered saline with Tween20
PEL
permissible exposure limit
PID
Photoionization detector
PPE
personal protective equipment
PPm
parts per million
PVC
polyvinyl chloride
QAPP
Quality Assurance Project Plan
QPS
quarantine and pre shipment
QUIC
Quick Urban & Industrial Complex Dispersion Modeling System
RAP
Remediation Action Plan
REL
recommended exposure limit
RH
relative humidity
RTP
Research Triangle Park
SAP
Sampling and Analysis Plan
SCBA
self-contained breathing apparatus
SERAS
Scientific, Engineering, Response and Analytical Services
SPM
single point monitor
SZ
Support Zone ("cold zone")
T
temperature
tarp
tarpaulin
TLV
threshold limit value
TSA
trypticase soy agar
TWA
time-weighted average
UF
University of Florida
U.S.
United States
USDA
United States Department of Agriculture
USGS
United States Geological Service
VHP
vaporized hydrogen peroxide
VOC
Volatile organic chemicals
9
-------
Executive Summary
To better understand the use of methyl bromide (MB) in an operational environment, specifically
its ability to inactivate Bacillus anthracis (Ba) contamination in structures, a University of Florida
(UF) 1,444 cubic meters (51,000 cubic feet [ft3]) house was fumigated with MB at 212 milligrams
per liter (mg/l) (212 ounces per 1000 ft3) on December 9-11, 2013. The fumigant, MB, was
selected because it has shown to be efficacious in the inactivation of Ba spores during laboratory
testing; MB is less corrosive than most alternative fumigants; and MB can be captured on
activated carbon, mitigating the potential ozone depleting effects. The study was conducted by:
the United States Environmental Protection Agency (EPA) Chemical, Biological, Radiological, and
Nuclear Consequence Management and Advisory Division; the UF; EPA's National Homeland
Security Research Center (NHSRC); EPA's Environmental Response Team; three EPA Region 4 On-
Scene Coordinators; and several contractors.
Three 24 x 30 meters (80 x 100 feet) sections of MB resistant tarpaulins (tarps), made from
ethylene vinyl alcohol, were hoisted onto the roof and arranged to cover the entire house by the
tenting contractor. The sections were joined by overlapping and rolling adjacent edges together
and binding them with clamps, while the tarp skirt and apron were weighted to the ground
directly below the roof line with 18 kilograms (kg) (40 pound [lb]) sand "snakes". To add strength
and protect the first tarp in the event of strong winds; a second tarp was placed over the first
tarp and secured in the same manner. Interior preparation of the house included placing seven
85 cubic meters per minute (3,000 cubic feet per minute) fans, four 1,500-watt radiant heaters,
and 16 warm-steam vaporizers to help maintain temperature, and relative humidity (RH)
equilibrium throughout the house.
Spores of Ba Sterne 34F2, the vaccine strain, were used as surrogates in lieu of using virulent Ba
spores and placed on coupon materials. Two coupon material types, glass and wood, were
chosen for preparation of customized biological indicators (Bis) as these materials were found in
laboratory studies to be most resistant to Ba-spore inactivation using MB. Test coupons (87 glass
slides and 87 wood discs) were inoculated with approximately 1 x 106 colony forming units of Ba
Sterne per coupon and were placed at 22 separate locations throughout the house. All coupons
were analyzed by the EPA NHSRC Research Triangle Park Microbiology Lab.
An activated carbon scrubber system was rented and used for the study to mitigate the release
of MB into the environment. The system consisted of two scrubber vessels each containing
approximately 2,495 kg (5,500 lb) of activated carbon, a blower, duct, and fittings. The inlet to
the scrubber system was installed in the office-room window located on the northeastern side
of the house.
Ambient air monitoring was achieved by placing photoionization monitors at six stationary
locations around the house. In addition, hand-held monitors with the same technology were
used to leak test the tenting materials and to provide monitoring at those locations not covered
by the six stationary monitors. Monitors detected small leaks near the tented house enabling
leak-reduction measures to be deployed as needed. Ultimately, the monitors proved effective
10
-------
and provided a successful health protection measure for the site workers, as well as offsite
people.
Liquid MB was released gravimetrically from commercial cylinders, gasified using a propane-
fueled heat exchanger, and then introduced into the house at approximately 66 degrees Celsius
(°C). Temperature, RH, and concentration of MB were monitored inside the house and were
maintained above the set points of 27 °C, 75%, and 212 mg/l, respectively, throughout the
fumigation. The activated carbon scrubber was effectively deployed and used to reduce the
concentration of MB inside the house from approximately 55,000 parts per million (ppm) to
below 150 ppm in 4 hours. Of the 243 kg (536 lb) of MB entering the scrubber at the end of the
48-hour fumigation, a total of 241 kg (532 lb) of MB were captured by the carbon beds, 99%
efficient. After the fumigation, all test coupons were removed from the house and incubated for
growth potential. All of the 174 coupons were negative for growth.
Based upon the lessons learned during this operational study, and after the test was completed,
the Health and Safety Plan and the Ambient Air Monitoring Plan for this operational fumigation
test were revised and are available (as attachments) for the response community to use and
adapt to site-specific requirements. Based on this field study, it is recommended that the
temperature, RH, and concentration of MB be maintained above the set points of 27 °C, 75%,
and 212 mg/l, respectively, for 36 hours when fumigating a structure for Ba. Additionally, a
guidance document was written to review the tactical use of MB as a fumigant for inactivation of
Ba. Completeting this operational study and updating the operational documents provides EPA
with a greater resiliency and capacity to respond to and recover from a Ba release or other
biological incident.
11
-------
1. Introduction
The United States Environmental Protection Agency (U.S. EPA) in partnership with the University
of Florida (UF) conducted an operational test to further develop results supporting methyl
bromide (MB) fumigation as a Bacillus anthracis (Ba) decontaminant. The test was conducted in
an effort to gain large-scale information on the use of MB as a structural fumigant for
decontamination of Ba spores and to develop site-specific plans and guidance that could be
modified and employed in a real-world incident.
The operational fumigation took place on December 9-11, 2013, at a house located on the UF
Campus in Davie, FL. Project planning, coordination, and execution involved members from:
EPA's Chemical, Biological, Radiological, and Nuclear (CBRN), Consequence Management and
Advisory Division (CMAD); EPA's National Homeland Security Research Center (NHSRC); EPA's
Environmental Response Team (ERT); EPA's Region 4; the UF faculty and staff; and several
contractors.
While this report discusses the results of the operational fumigation, it also provides guidance
documents that could be used during a response at a later date. The plans used to govern the
fumigation for this test site were revised based on the lessons learned (listed in Appendix A) at
this site. Additionally, a field operational guidance document, detailing the use of MB for the
fumigation of buildings, rooms, and sensitive items, was written for New York City. The revised
plans and guidance are included in this report as attachments, and can be modified and used at
other sites requiring MB fumigation. They include:
Attachment 1: Health and Safety Plan (HASP)
Attachment 2: Ambient Air Monitoring Plan (AAMP)
Attachment3: MB Field Operational Guidance to New York City
1.1 Background
In 2001, a series of letters containing Ba spores were mailed to various locations throughout the
U.S. It was determined that initial and residual contamination from Ba spores was difficult to
detect, identify, and decontaminate in an efficient and expedient manner. Additionally,
significant costs were incurred during decontamination and clean-up efforts of buildings and
equipment that were suspected of being contaminated. Comments from the Government
Accountability Office (GAO) reports and congressional inquiries pointed out that sampling and
decontamination methods were not standardized and/or validated; and that deficiencies were
observed when attempts were made to locate and characterize Ba contamination (GAO Report -
06-756T, 2006). The GAO recommended standardizing and validating procedures that could be
used to characterize biological agent contamination and increasing capacity to effectively
decontaminate buildings and associated areas. The research covered by this report is focused
on efficient decontamination using fumigation; specifically, operational aspects of MB fumigation
to increase remediation capacity in preparation for a response to Ba incidents.
12
-------
Sandia and Lawrence Livermore National Laboratories used the U.S. Department of Homeland
Security (DHS), national planning scenario number two as a basis to produce a systems analysis
report for a Ba wide-area release (IBRD, 2008). The resulting area of > 100 spores per square
meter (m2) deposition was 6 square miles, and the resulting area of 10-100 spores/m2 deposition
was 160 square miles in total. Based on a number of considerations as well as the current state-
of-the-science, EPA and CDC (CDC-EPA, 2012) recommend that "no detection of viable spores" is
the best practicable clearance goal. With this as the clearance goal, the entire area, 166 square
miles would require remediation. This IBRD report estimated that it would take 15 to 18 years
to complete the remediation using the current remediation capacity.
The authors of the IBRD report considered using vaporized hydrogen peroxide (VHP) and chlorine
dioxide gas (CIO2) to achieve building and surface decontamination. Bleach solutions or other
liquid oxidants were also considered for the decontamination efforts in portions of buildings
where the primary source of contamination was determined to be tracked in by humans or
animals. The IBRD authors did not include MB as one of the fumigants even though the report
highlighted several important gaps; one being the "limited resources for indoor fumigation."
Corrosion and discoloration of materials are associated with the use of the majority of current
Ba remediation technologies. Even if the capacity of the current technologies is increased, the
collateral damage caused during fumigation or liquid application could generate a significant
volume of waste, thus increasing remediation time and cost. In the case of sensitive or historic
infrastructure, corrosive (methods relying on oxidation) remediation techniques are not an
option even if capacity is achieved. Several studies have been conducted to look at fumigant
efficacy against B. anthracis and the corrosion caused by fumigants. The studies listed below
highlight findings for MB.
• The U.S. EPA has conducted several studies looking at decontamination agent's material
compatibility with electronics and items of historical value.
o An unpublished U.S. EPA study1 on historical materials examined the impact of
CIO2, VHP, ethylene oxide (EtO), and MB on several types of materials. This study
provided insight into the risk for damage from a decontamination scenario using
different fumigants. Based on this work, VHP, EtO, and MB can be considered the
most compatible (of those fumigants and materials tested) with museum quality
objects. MB would be a viable alternative for a whole-building decontamination
scenario when materials such as books, documents and photographs are present.
o In another study (U.S. EPA, 2012), personal computers were exposed to MB
fumigation under the same conditions necessary to inactivate spore forming
bacteria. The fumigant included 2% chloropicrin mixed in with the MB. The
chloropicrin appeared to oxidize some components in the computer system.
1 Point of Contact: Dr. Shannon Serre, ORD, EPA
13
-------
• Laboratory studies by U.S. EPA (U.S. EPA, 2011) on seven different building materials
found that MB fumigation was efficacious for the decontamination of Ba Ames (a virulent
strain of Ba spores) on a broad range of indoor building materials tested.
• Corsi et al. (2007) concluded that MB does not engage appreciably in sorptive
interactions with indoor materials, although some diffusion can occur into porous
materials. Desorption of adsorbed MB from indoor materials appears to be rapid. It
also appears that exposure of some building materials to elevated concentrations of MB
leads to an increase in the off-gassing rate of carbonyls and several methylated aliphatic
compounds. However, the absolute increases appear to be small and are likely not a
major concern for either disinfection workers or those who reoccupy a building after a
disinfection event.
• Juergensmeyer et al. (2007) established that a MB minimum effective dose of 80 mg/L
was lethal to a concentration of 107 spores of Ba, specifically, nine different strains
(including Ames and Sterne) on glass slides after a 48-hr exposure at 37 degrees Celsius
(°C). In addition, at the same exposure conditions, 10 strains of Bacillus anthracis (ATCC
10, ATCC 937, ATCC 4728, ATCC 9660, ATCC 11966, ATCC 14187, AMES-1- RIID, AMES-
RIID, ANR-1, and STERNE) were equally susceptible to MB and were not dependent upon
virulence factors. The study showed that B. atrophaeus and B. thuringiensis were more
resistant than Ba to MB when tested at similar conditions. All B. thuringiensis and B.
atrophaeus spores tested showed a dose-dependent reduction in spore numbers, but
they were not reduced below detection level by any MB concentrations tested. The
authors concluded that MB has several advantages as a fumigant: First, because MB is a
registered structural fumigant, personnel trained in its use are available nationally.
Additional training in decontamination procedures would be minimal for these
professionals. Second, decontamination is rapid, occurring within 48 hours. Extensive
preparation of the contaminated item is not required, and all furnishings or other internal
structures or items may remain in place. Third, MB leaves no residue, and is a
noncorrosive alkylating agent that does not damage commodities (e.g., food supplies),
furnishings, documents, or even sensitive electronic equipment.
• Weinberg et al. (2004b) conducted a MB field trial within a 30,000 cubic foot structure.
Filter paper coupons containing 106 spores of one of three species, Geobacillus
stearothermophilus, B. atrophaeus and B. thuringiensis, and stainless steel coupons with
106 spores of B. atrophaeus were placed in 50 locations within the structure. Fumigation
was conducted using 312 mg/L of MB for 48 hr at 35.5 °C with overall mean RH of 76%.
The results of the field trial found that only one location, a sealed refrigerator, contained
viable spores of B. atrophaeus on a single coupon. It was noted that the performance of
sensitive electronics and electronic media placed in the structure were unaffected by the
MB fumigation.
• The Bio-response Operational Testing and Evaluation (BOTE) Project (U.S. EPA 2013) was
a multi-agency effort designed to operationally test and evaluate, at the scale of a
moderately sized building, a response to a Ba release from initial public health and law
enforcement investigation through environmental remediation. The BOTE Project was
14
-------
divided into two phases: a field-level decontamination assessment and an operational
exercise. Phase 1 tested three decontamination methods on inactivating a Ba simulant,
fumigation with hydrogen peroxide vapor, surface application of pH-adjusted bleach, and
fumigation with chlorine dioxide gas. It was proposed that one of these three methods
would be used during the Phase 2 exercise; however, because all three had already been
evaluated during Phase 1, in an effort to expand our knowledge of potential response
tools, a forth alternative, fumigation with methyl bromide gas, was selected for use during
Phase 2. The fumigation process was successful; however, there were some technical
difficulties that affected the outcome of the spore inactivation and the subsequent
aeration process.
1.2 MB Usage and Properties
MB, also known as bromomethane, is a colorless, odorless (at low concentrations), and
nonflammable gas and is classified as an alkyl bromide. MB is containerized as a liquid under
modest pressure, approximately 2 atm. MB is used primarily as a pesticide to control insects,
nematodes, weeds, pathogens, and rodents. MB was originally registered by EPA for various
applications including: soil fumigation (injected into the soil before a crop was planted to
effectively sterilize the soil); commodity treatment (used for post-harvest pest control);
structural pest control (used to fumigate buildings for termites; and warehouses and food
processing facilities for insects and rodents); and quarantine uses (used to treat exported and
imported commodities such as logs, fresh fruits and vegetables).
MB fumigant concentrations and contact times vary with the commodity or structure being
treated, the target pest, temperature, and RH. MB is an effective pesticide because it acts as a
methylating agent that disrupts an organism's internal enzymatic protein chemistry. However,
the production of MB was reduced (2005) under an international treaty called the Montreal
Protocol, and by EPA under the Clean Air Act (CAA) (http://www.epa.gov/ozone/mbr/) due to its
stratospheric ozone-depleting potential. Use now requires an exemption by the EPA under
appropriate provisions in the CAA. MB is currently used in the U.S. only under these exemptions
and is manufactured in the U.S. by Chemtura Corp. with label provisions developed by Great
Lakes Corp. Allowable exemptions include: the Quarantine and Pre Shipment (QPS) exemption,
to eliminate quarantine pests; and the Critical Use Exemption (CUE), designed for agricultural
users with no technically- or economically-feasible alternatives. Under those exemptions there
are approximately seven-million pounds of MB used annually in the U.S. In addition, there is a
third allowable exemption "the Emergency Exemption" that is not well defined at this time.
Due to the need to find an effective fumigant or method to inactivate Ba spores, the EPA
continues to research decontamination technologies, including MB at relatively low
temperatures and RH levels (U.S. EPA, 2014).
Before phase-out began in 2005 as part of the Montreal Protocol, as an ozone-depleting
substance, MB fumigation was widely used for 60 years against soil and structural pests.
Currently, MB is still used for quarantine fumigations against pests that harbor in perishable
commodities. Most major U.S. seaports, and some airports, have United States Department of
15
-------
Agriculture (USDA) regulated facilities for MB fumigations of imported fruits and vegetables.
These facilities have crews trained in MB fumigation using much of the same equipment and
methods as used in structural fumigations. While the crews have the technical expertise to
conduct lawful fumigations, only a small percentage of fumigation crews currently working in the
industry meet the requirement to enter a biological agent remediation site. Requirements would
include Occupational Safety and Health Administration (OSHA) Hazardous Waste Operations and
Emergency Response (HAZWOPER) certification, medical clearance to wear respiratory
protection, and the annual respiratory protection training (medical clearance, self-contained
breathing apparatus [SCBA] use, and respirator training are already existing requirements for
licensed fumigators). In addition, fumigation workers would need site-specific training with a
focus on the hazards of Ba and on conducting their fumigation tasks while in level-C, most likely
including power air purifying respirators. Initial HAZWOPER technician training is a one-time 24-
hour event with subsequent 8-hour refresher training required annually. To overcome this
deficiency, fumigation industry workers without the required HAZWOPER training could be
prepared with minimal training to meet these requirements as needed for emergency response
remediation work.
The structural fumigation industry (mostly non-MB usage) is located in Florida, the Gulf Coast,
the Southwest, and Hawaii. The quarantine fumigation industry (MB usage) in mainly located at
large sea ports and airports where international cargo is imported. In a national emergency
involving the release of Ba, this industry could potentially be used to increase our remediation
capacity; especially when the building material and building contents are deemed incompatible
with other remediation technologies.
MB penetrates quickly and deeply into sorptive materials at normal atmospheric pressure. Also,
at the end of a fumigation treatment, its vapors dissipate rapidly from those materials (Corsi et
al., 2007). Another important property of MB is the fact that freshly harvested produce has been
shown to be tolerant to this gas in insecticidal treatments, offering to potential outdoor
applications. MB is nonflammable and non-explosive under ordinary circumstances and may be
used without special precautions against fire.
In the absence of oxygen, liquid-phase MB reacts with aluminum to form methyl aluminum
bromide. Methyl aluminum bromide ignites spontaneously in the presence of oxygen. Liquid
MB should never be stored in cylinders containing any appreciable amount of the metal
aluminum and aluminum tubing should not be used for application of the liquid phase of the
fumigant.
The chemical properties of MB are summarized in Table 1 below.
Table 1. MB Chemical Properties
Chemical fnnmiki
CH3Br
Boiling point
3.6 °C
Freezing point
-93 °C
Molecular weight
94.95
Specific gravity gas (air= 1)
3.27 at 0 °C
16
-------
Liquid (water iit 4 °C= 1)
1.732 at 0 °C
Viipor Pressure
1400 mmHg at 20°C
Latent hent or Viipori/iition
61.52 calories per gram (cal/g]
FUimimibility limits in ciir
Flammable between 10-15% (some say 20%] in air
Solubility in water
1.34 g/100 ml at25°C
Odor
Odorless at low concentrations; strong musty or sickly sweet odor
at high concentrations (greater than 1,000 ppm]
Pertinent chemiail
properties (Liquid phcise
only)
Powerful solvent of organic materials, especially natural rubber.
Reacts with aluminum and its alloys to form methylated
aluminum compounds that are spontaneously flammable in air
(see text below]. Reacts with zinc, magnesium, tin, and iron
surfaces in the presence of impurities such as water or alcohol.
Avoid the presence of acetylenic compounds, ammonia,
dimethylsulfoxide, ethylene oxide, oxidizers, and hot metal
surfaces. Attachment 1 provides further details regarding MB.
1.3 Health and Safety
With all fumigants, human exposure is a concern that should be addressed and managed because
of theirtoxic nature and inhalation hazard. MB is no exception. MB is a toxic chemical. Because
MB dissipates so rapidly to the atmosphere, it is most dangerous at the actual fumigation site
itself. Human exposure to high concentrations of MB can result in central nervous system and
respiratory system failure, as well as specific and severe deleterious reactions affecting the lungs,
eyes, skin, kidneys, and liver.
The compound has a history of industrial use, and it is fairly well characterized in terms of human
toxicity, including recommended and regulatory occupational exposure limits (OEL). For the
purposes of this study, a detailed HASP was developed that integrated personnel and area
monitoring, emergency response, medical monitoring, personal protective equipment (PPE)
requirements, clearance thresholds, and more.
Although this study was a research project, the fumigation site was managed as if it were an
emergency response site with the designation of an Exclusion Zone (EZ) or "hot zone", a
Contaminant Reduction Zone (CRZ) or "warm zone", and a Support Zone (SZ) or "cold zone. The
three zones were delineated based upon the most conservative airborne OEL provided by the
OSHA, the National Institute for Occupational Safety and Health (NIOSH), and the American
Conference of Governmental Industrial Hygienists (ACGIH). The ACGIH threshold limit value (TLV)
is 1 ppm as an 8-hour time-weighted average (TWA) and the OSHA permissible exposure limit
(PEL) is 20 ppm as a ceiling value that cannot be exceeded in any part of the workday. The NIOSH
immediately dangerous to life or health (IDLH) value is listed as 250 ppm. It should be noted that
there is no NIOSH recommended exposure limit (REL), as NIOSH considers MB a potential
occupational carcinogen. Other organizations, such as the International Agency for Research on
Cancer (1986), the National Toxicology Program (1992), and the EPA (1988) do not classify MB as
a potential human carcinogen.
17
-------
In addition to inhalation exposure limits, the OELs annotate a skin notation, which suggests
potential adverse effects to the skin, and/or absorption through the skin. The reports of Jordi
(1953) and Hezemans-Boer (1988) suggest that sweating increases vulnerability to skin
absorption in humans. Yamomoto (2000) studied cutaneous exposure of rats to MB, but it is not
clear whether the exposure was to liquid or vapor. They found an immediate rise in plasma
bromide ion, with a plasma clearance half-life of 5.0 - 6.5 days.
For purposes of this MB fumigation study, zones were established as follows: EZ > 0.5 ppm; CRZ
> non-detect and < 0.5 ppm; and SZ = non-detect. Wind directional flags were used throughout
the fumigation, and the SZ was maintained upwind from the fumigation. Personal protective
equipment including SCBAs and foot and hand protection were prescribed based on worktask(s).
SCBAs were required for entry into an area with airborne concentrations consistently exceeding
the action level (a level of MB concentration that requires mitigative actions), 0.5 ppm. Two
Certified Industrial Hygienists (American Board of Industrial Hygiene) served as the site safety
officers (SOs) and provided 24-hour oversight of the project during all fumigation activities.
Personal breathing zone samples were collected on EPA and contract personnel by the SOs during
tasks identified as having potential for MB exposure. These tasks included coupon extraction and
carbon scrubber operations conducted during the aeration of the test house.
The HASP restricted entry into the test house from the time fumigation began until the
fumigation was complete and airborne concentrations were measured to be below 5 parts per
million (ppm). Workers in SCBAs could enter the house when the MB concentration was below
5 ppm and could enter without SCBAs when the MB concentration fell below the action level, 0.5
ppm. When dispensing MB from cylinders, workers wore loose fitting clothing, as required by
the MB labeling, to reduce the risk of trapping liquid MB under clothing next to the skin.
Engineering controls, work practices, and required PPE were all detailed in the site-specific HASP.
The risk of exposure to MB, without sufficient warning, is significant because MB is a colorless
and odorless gas (odorless at working concentrations). To address this significant risk, a detailed
AAMP for MB monitoring and HASP outlining measures to protect workers and adjacent building
occupants was followed.
After completion of the test and review of the lessons learned, a revised HASP (Attachment 1)
was created to serve as an example HASP that can be modified and used at other sites requiring
MB fumigation.
1.4 Study Objectives
The overall goal of this test was to conduct and evaluate the operational aspects and the efficacy
of MB fumigation for the inactivation of a nonpathogenic Ba surrogate for pathogenic Ba spores
in a single-story ranch-style house. The following five objectives were developed in order to
reach this overarching stated goal:
18
-------
1.4.1 Objective 1
To develop a Quality Assurance Project Plan (QAPP) for the fumigation of the University
of Florida, Hurricane House in Davie, FL using MB for the inactivation of the chosen non-
pathogenic surrogate spores. This study included the development of a Remediation
Action Plan (RAP); a site-specific HASP; Sampling and Analysis Plan (SAP), as part of the
QAPP; and an AAMP, to govern this site-specific MB fumigation. With minor changes,
these site-specific plans (i.e., RAP, HASP, SAP, and AAMP) could be easily modified and
used at other sites for MB fumigation of Ba.
1.4.2 Objective 2
To conduct the fumigation process safely, economically, and effectively. To monitor and
maintain MB concentrations, temperature and relative humidity (RH) during the testing
to assure dose requirements were reached inside the house during fumigation; > 212
mg/l, > 27 °C, and > 75%, respectively, during a 48-hour period. Furthermore, MB
concentration, temperature, and RH will be monitored from outside the house before,
during, and after the same 48-hour period.
1.4.3 Objective 3
To evaluate the efficacy of the fumigation by measuring the post-fumigation viability of
surrogate spores. This was accomplished by inoculating Ba Sterne onto coupons (wood
and glass) and placing them in 22 locations throughout the house prior to fumigation,
followed by analysis of viability.
1.4.4 Objective 4
To operationalize and evaluate the effectiveness of activated carbon for the capture of
the MB fumigant during the aeration portion of the fumigation cycle. In addition, during
aeration of the house, to monitor MB breakthrough status of the activated carbon and
provide an estimate of house re-entry time.
1.4.5 Objective 5
To monitor the effectiveness of MB containment and provide for the health and safety of
workers during the entire fumigation process.
The HASP, RAP, and AAMP provided detailed procedures for air monitoring and for handling
elevated levels (>0.5 ppm) of MB in the ambient air during all aspects of the fumigation.
Achieving these objectives will result in greater resiliency and capacity to respond to and recover
from a Ba release or other biological incident.
19
-------
2 Materials and Methods
2.1 Facility
The study building was the University of Florida "Hurricane Resistant Model Home" (house)
(Figures 1, 2). The house encompassed 1,444 cubic meters (m3) (51,000 cubic feet [ft3]), including
the exterior volume contained below the edge of the eaves. Constructed in 2005, the ranch-style
house is used for teaching and meeting functions and contains two large open meeting areas, an
office, a kitchen (with refrigerator and oven), two restrooms, two utility closets, an HVAC room,
and a storage/computer room. The house also contains two desktop computers, a computer
router, two LED monitors, and an LCD projector (Figure 3). The house is on the campus of the
Fort Lauderdale Research and Education Center, institute of Food and Agricultural Sciences,
University of Florida, 3205 College Avenue, Davie, FL 33314 (26.08343, -80.24115, two meter
elev.). Adjacent to the study house were two buildings that served as student housing. As part
of the health and safety protocols, these two buildings were evacuated during the fumigation. A
diagram of the campus grounds is shown in Figure 4. As an additional safety precaution, a nine
meter (30 ft) radius away from the perimeter of the house was cordoned off to non-authorized
personnel.
Figure 1. The House used for this Study on the Campus of the Fort Lauderdale Research and Education Center,
University of Florida, Davie, FL
20
-------
9 high
9'high 15'peak
45'
B
_J9!_peak _
9' high
/
9' high
¦ 74'
9' high
12'peak
9 .high.
/
16'
Figure 2. External Dimensions of House used in this Study.
A) Back Porch, B) Main Structure, C) Front Porch
21
-------
Figure 3. The Electronics that Remained in the House during Fumigation
-------
Yards
©'SpfenSirfetMap (and) contributors, CC-BY-SA
Hurricane House | Parking
^ Command Post Road
| Greenhouse Fence
^ Storage
J Student Housing
Figure 4. Area Around the House
2.2 Sealing the House
In preparation for sealing the house, the turfgrass surrounding the study house was mowed and
edged with a string trimmer to a height of 3 cm. Fine sand was then applied in areas around the
perimeter where foundation drop-offs or uneven or porous surfaces were found. Next, a 1.8-
23
-------
meter-wide tarpaulin (tarp) apron made of 6-mil high diffusion-resistant polyethylene vinyl
alcohol (EVOH) with polyester scrim reinforcement (GeoCHEM Inc., Renton, WA) was taped to
the house at ground level around the entire perimeter of the house using 8-cm-wide resilient
tape (Shurtape, Cardinal Pro. Prod., Anaheim, CA). A narrow bead of silicone sealer was also
applied overthe edge of the tape where it adhered to the house foundation (Figure 5). The EVOH
tarp has a white surface on one side and a black surface on the other side. For the apron
installation, the tarp was used white side down. Three 80 x 100 ft sections of EVOH tarps were
hoisted onto the roof and arranged to cover the entire house, black side out, by the tenting
contractor (Dead Bug Edwards, Fort Lauderdale, FL). The sections were joined by overlapping
and rolling adjacent edges together and binding them with plastic-tipped, metal spring
fumigation clamps. As the house was covered, horizontal "skirts" were dropped from the roof
onto the apron (Figure 6). The skirt and apron were held down on the ground directly below the
roofline with overlapping 40-lb sand "snakes" bags. The edges of the skirt and apron were then
tightly rolled and secured with clamps (Figure 7). To add physical strength and protect the first
tarp in the case of strong winds, a second commercial fumigation tarp, vinyl-coated nylon fabric,
12 ounces per square yard, white in color (Figure 8), was placed over the first tarp in the same
manner as the first and was secured to the ground atop the skirt and apron with additional
overlapping sand snakes (Figure 9).
Figure 5. Attachment of Tarp Apron to Foundation of House with Tape and Caulk
24
-------
Figure 6. View of the Skirt of EVOH Tarp Dropped to the Apron on the Ground
Figure 7. To Seal, the Skirt and Tarp Covering the House Were Rolled and
Clamped at Edges and Weighed Down with Sand Snakes
25
-------
:y -.-W-
Figure 8, Second Tarp (White) Positioned Over Black EVOH Tarp
Figure 9. In Preparation to Fumigate, the House was Sealed with Two Tarps
26
-------
The tarp sea! around the house was opened at two skirt seams to accommodate the 24-inch
diameter, fresh-air inlet port (Figure 10) and 24-inch diameter, exhaust port (Figure 11) for the
aeration procedure. Both the inlet port and the exhaust port were sealed with metal lids during
the fumigation. The ground seal was also opened between the skirt and apron of the first tarp
and the second tarp to allow the insertion of 4-inch diameter, polyvinyl chloride (PVC) pipes that
were used as conduits for two MB introduction or "shooting" hoses (3/4" braided chemical
resistant, high temperature [149 °C rating], and high pressure [> 200 pounds per square inch
rating]), monitoring lines (6.4 millimeter outer diameter, nylon), and a test coupon slide designed
to extract test coupons during intermediate phases of the fumigation (Figure 12). The two MB
shooting hoses were extended into the tested house with a similar inside diameter polyethylene
tubing connected together with compression fittings. These hoses were clamped and taped into
the bottom of a 5 gallon plastic bucket placed in the entry way of the tested house (Figure 13).
The bucket also contained a concrete block as weight for stability. The bucket was used as the
release point for the fumigation inside the tested, house and to protect the house floor. As MB
gas flows out of the shooting hoses, the bucket collects oils, rust, or other non-volatile
contaminants that might be present in the cylinders. After the shooting and monitoring lines
were routed through two pipes, any voids in the PVC pipe and pipe chases were filled and sealed
with expanding polyurethane foam (Figure 14). The test coupon slide (PVC-pipe construction)
was sealed at its exterior terminus with a threaded PVC cap.
»F1
Figure 10. Fresh-Air Inlet Port in Open Position (Closed during Fumigation)
27
-------
Figure 11. Flange of Carbon Scrubber Exhaust Port Joined to Tarp at a Seam
Figure 12. Monitoring Lines, Shooting Lines and Coupon Slide Inserted
Between the Skirt and Apron (View from Inside House)
28
-------
Figure 13. Bucket Release Point of Fumigant Inside House with Mixing Fan
Figure 14. PVC Pipe Voids and Pipe Chases Filled with Expanding Polyurethane Foam.
Mote: Threaded Cap on Coupon Slide {White PVC Pipe) in Center of Photograph
(Outside House View)
29
-------
2.3 Circulation Fans, Heaters and Humidifiers
Interior preparation of the house included the placement of seven 3,000 cubic feet per minute
(cfm) fans (Figure 13), sixteen 1-gallon capacity warm steam vaporizers (Figures 15 & 16)
(Walgreens Brand Model 21413ktc, Springfield, IL), and four 1,500 watt radiant heaters
(Delonghi, Model EW7707CM, Woodridge, N.J.) (Figures 15 &16). The fans ran continuously
during the fumigation (including during scrubbing and aeration) to maintain temperature and RH
equilibrium throughout the house and to help disperse the MB gas when it was introduced into
the house. The power supply for the radiant heaters and steam vaporizers were extended with
two extension cords each and routed outside so they could be powered on or off as needed to
maintain temperature and RH, 27 °C and 75%, respectively. The day before fumigating the house,
the air conditioning system was turned off and the 16 steam vaporizers were turned on to
increase the humidity of the materials within the house. The steam vaporizers were refilled prior
to sealing the house and just before fumigation.
It is important to note that the contents within a volume to be fumigated should be factored into
the fumigation decision process. When determining a decontamination approach, consideration
must be given to the contents (e.g., paper, foam, water, fabrics, concrete, galvanized metal, etc.)
as they may adversely impact the efficacy of the fumigation. Specific contents, when found in
significant quantity, may act as sinks for fumigants, water vapor (humidity), and/or heat.
Fumigant adsorption may be followed by latent desorption (off-gassing) for extended periods of
time following the initial fumigation. Large amounts of paper, for example, may need to be
removed or may need to be pre-humidified before fumigation; and large amounts of foam, may
act as a sink for fumigant, requiring the foam to be removed or additional fumigant used to
overcome the loss of fumigant to the foam. The interaction of the contents with the fumigant
and fumigation parameters will dictate what actions may be needed; however, interactions are
not always known in advance and fumigation parameters must be monitored during the
fumigation to assure the parameters necessary for an efficacious decontamination are met.
30
-------
Figure 15. Radiant Heater and Two Steam Vaporizers in the House
(Fans not Shown)
All cabinets, appliances, interior doors, and two attic access panels were opened to aid reaching
concentration, temperature and RH equilibrium. Exterior doors and windows were open in the
large classroom side of the house but closed on the office side of the house where the scrubber
duct was attached. This arrangement was used to direct airflow during aeration of the house:
fresh air coming into one side of the house while exiting out the other side to the scrubber. All
food was removed from the house, but everything else was left in place (furniture, two bed
mattresses, telephones, computers, printers, televisions, tables, chairs, upholstered chairs and
couch, carpet, flooring, lamps, HVAC system, books, paper, window treatments). The house was
completely sealed on December 7 and 8, 2013.
31
-------
Wipe Sample Pre-Fumigation
• Wipe Sample Post-Fumigation
• MeBr Sample Location
• Heater Location
• Humidifier Location
Figure 16. Schematic of House Showing Locations of Heater, Humidifiers, and MB Sampling Lines
2.4 Temperature and Relative Humidity Monitoring
Temperature and RH inside the house, were monitored during the fumigation, including the
aeration phase, using a HOBO temperature and RH monitoring system (Model ZW-03, Onset
Computer Corporation, Bourne, MA 02532). The system included 4 wireless sensor nodes spaced
throughout the house and a router, which was placed on the front entryway under the tarp. The
wireless system transmitted real-time data to the receiving station that was located
approximately 100 feet from the house. Real-time temperature and RH data were collected and
displayed on a laptop computer using HOBOware Pro software (Onset Computer Corporation,
Bourne, MA 02532). These data were used to determine if heat and/or moisture needed to be
added. In addition to the 4 wireless sensors, HOBO temperature and RH loggers (Model U10,
Onset Computer Corporation, Bourne, MA 02532) were placed adjacent to 21 of the 22 coupon
locations inside the house.
32
-------
2.5 Coupon Preparation
A Ba surrogate for this study was selected based on a series of laboratory tests using several
spore candidates; non-virulent strains Geobacillus stearothermophilus, Ba NNR1A1, and Ba
Sterne. Spores of Ba Sterne 34F2, the vaccine strain (strain obtained from Colorado Serum Co.,
Denver, CO), were selected as surrogates for fully-virulent Ba spores. Spore production
procedures were conducted at Yakibou Labs, Inc. (Apex, NC), according to proprietary methods.
Coupon material type was selected based on a series of laboratory tests (U.S. EPA, 2014)
completed prior to this study using several material type candidates including glass, ceiling tile,
carpet, wallboard paper, wood, and concrete. Two coupon material types (Figure 17, B and C)
were selected for preparation of customized biological indicators (Bis); glass (premium pre-
cleaned microscope slides, VWR International, Cat# 48300-047, Radnor, PA) and wood (Maple
discs, 1.43 cm dia., 0.32 cm thickness, part# DIS-050, American Woodcrafters Supply Co.,
Riceville, IA). Glass and wood were chosen because studies have shown these materials were
most resistant to spore deactivation with MB. Glass was cut into coupons (approximately 15 mm
by 18 mm), washed in alkaline detergent to remove grime and grease, rinsed until no residue
remained, dried at 125 - 150 °C, and sterilized by Yakibou, Inc. (Apex, NC) by steam autoclave (1
hour, 121 °C, 103 kPa, method of procedure [MOP] 6570). Wooden coupons were sterilized using
ethylene oxide.
After sterilization, test and positive control coupons were inoculated using a liquid inoculation
proprietary protocol (Yakibou, Inc.) with a target final spore inoculum of 2.0 to 5.0 x 106 spores
(as determined by enumeration of colony forming units (CFU) per volume of inoculum) per
coupon. Negative control coupons and field blank coupons, although not guaranteed to be sterile
following packaging, remained un-inoculated. After inoculation, coupons were allowed to dry at
room temperature on a bench top, and subsequently packaged into custom-sized Tyvek®
pouches (Figure 17, A). The pouches were heat-sealed to prevent infiltration or exfiltration of
spores or particulate contaminants, thereby preventing escape of the spores and maintaining the
integrity of the Bis from the surrounding environment. Tyvek® pouches were pre-labeled with
an identifier unique to each product type.
Pre-test and post-test determinations of Bl population densities were performed at EPA, NHSRC,
Research Triangle Park (RTP) Microbiology Laboratory (Table 2), according to MOPs 6535a, 6565,
and 6566. These tests were conducted to determine the spore population on coupons prior to
testing and then after fumigation (non-exposed coupons). Spores were extracted from the
coupons, ten-fold serially-diluted, and then plated onto tryptic soy agar (TSA) plates. Following
incubation at 35°C for 18-24 hours, the resulting CFUs were enumerated. The CFU abundance
was used to estimate the total spore abundance on the coupons. Triplicate samples of each
material type were analyzed for population density before and after the field test. In addition,
ten replicate of stainless steel coupons (Figure 17, B), inoculated by Yakibou, Inc. at the same
time as the glass and wood coupons, were analyzed for population densities, before and after
the field test. These stainless steel coupons were expected to yield more accurate and repeatable
33
-------
estimates of pre- and post-test viable spore population densities than glass or wood (Calfee,
2011), as recovery of spores from stainless steel surfaces is highly efficient.
Table 2. Pre-Test and Post-Test Bl Population Densities Samples
Figure 17. (A) Tyvek® Bl Envelope, "BX3143W", Label Indicating Inoculated Wooden Coupons are Inside (B) Glass
and Stainless Steel Bl Coupons (C) Wooden Coupons
Sample Type
Location
Purpose
Frequency
Quantity
Analysis
Coupon
enumeration for
pre- and post-
fumigation QC
EPA
Microbiology
Laboratory
To determine
spore population
densities on
coupons pre- and
post-test
one set each
material
before test
and one set
after test
32 total:
10 stainless
steel
3 glass
3 wood
Enumeration
34
-------
2.6 Analysis of Test Coupons
Two types of test coupons were utilized during the test to evaluate the efficacy of the MB
fumigation (Table 3). These included: (1) Bis deployed throughout the house to qualitatively
assess fumigant efficacy spatially; and (2) collocated coupons positioned inside the house (at the
extraction port) and collected at specified time intervals to quantitatively assess fumigant
efficacy temporally (Figure 12 & 14).
Table 3. Test Coupon Samples used to Evaluate the Efficacy of MB Fumigation
Sum pic
Type
Locution
Purpose
Frequency
Quantity
Analysis
Test Bis
22 locations
inside the
fumigated
house
To determine
the presence of
viable spores
after fumigation
Once per test
174 total:
87 wood
87 glass
Qualitative
(growth, no
growth)
Temporal
Progression
Coupons
Inside the
fumigated
house at
extraction port
To determine
fumigation
efficacy as a
function of time
One set of samples
at 16, 24, 32, and
40 hrs. (four sets
total)
48 total:
6 wood/set
6 glass/set
Enumeration
Four types of control Bis were utilized during the tests; procedural blank Bis, positive control
Bis, negative control Bis, and lab-sterilized negative control Bis (Table 4). Procedural blank Bis
were not inoculated but were collocated with test Bis during the fumigation and were used to
determine the extent of cross-contamination from sample to sample during collection. Positive
control Bis were inoculated in the same manner as test Bis, but were not exposed to MB.
Positive control Bis traveled to the test venue, but remained in the sample shipment cooler for
the duration of the test. Negative control Bis were not inoculated, but were packaged in the
same manner as test Bis, traveled to the testing venue, remained in the sample shipment
cooler, and were not exposed to MB. Since procedures required for packaging Bis into
envelopes are not strictly aseptic, these Bis were not guaranteed to be sterile. Accordingly,
positive growth results from these controls should not be interpreted to indicate a compromise
in sample integrity through contamination. Lastly, lab-sterilized negative control Bis were Bis
received from Yakibou, Inc. and autoclaved (1 hour gravity cycle) upon arrival at the NHSRC RTP
Microbiology Lab to sterilize. These Bis were used to assess the handling technique of the
laboratory personnel during culturing procedures. Growth from these Bis would indicate a
compromise of sample integrity through contamination within the laboratory.
35
-------
Table 4. Control Bl Samples Utilized During the Test
Sample
Typo
Location
Purpose
Frequency
Quantity
Analysis
Procedural
Blank Bis for
Test Bis
Same 22
locations as
Test Bis
To go through
fumigation and
determine extent of
cross-contamination
Once per test
44 total:
1 wood/
1 glass/
location
Qualitative
(growth, no
growth)
Procedural
Blank Bis for
Temporal
Progression
Bis
Same location
as Temporal
Progression
Bis
To go through
fumigation and
determine extent of
cross-contamination
One set of
samples at
16, 24, 32,
and 40 hrs.
(four sets
total)
16 Total:
2 wood/
2 glass/
set
Enumeration
Positive
Control Bis
Bis went to
the site but
remained in
coolers, did
not get
fumigated
To determine the
presence or non-
presence of viable
spores on non-
fumigated Bis
Once per test
48 total:
24 wood
24 glass
Qualitative
(growth, no
growth)
Negative
Control Bis
Bis went to
the site but
remained in
coolers, did
not get
fumigated
To determine the
presence or non-
presence of viable
spores on non-
fumigated Bis
Once per test
48 total:
24 wood
24 glass
Qualitative
(growth, no
growth)
Lab
Sterilized
(Negative
Control) Bis
Lab negative
control (EPA
Microbiology
Lab only)
To demonstrate
sterility of Bis and
extraction
materials/methods.
Per analysis
of test
samples
6 total:
3 wood
3 glass
Qualitative
(growth, no
growth)
2.7 Spatial Assessment of Efficacy (Qualitative Test)
Four duplicate Bis of each type, wood and glass, along with one procedural blank of each coupon
type (non-inoculated wood and glass Bis) were positioned at 22 locations (Figure 18) throughout
the house prior to fumigation. Coupon locations included placement inside a desk drawer
(Location A), inside file cabinets (C, L), in a kitchen cabinet (E), in an oven (F), inside the HVAC
return duct (G), inside the metal hurricane shelter (I), in the attic (J), under insulation in the attic
(K), on porches (P,Q,V), near a drain in a sink (R), in restrooms (S,T), and in a utility closet (U).
These Bis remained within the house during fumigation, and were retrieved after MB air
concentrations within the house had subsided. Once removed from the house, the Bis were then
aired out to allow MB to escape, cold packed, and transported to the NHSRC RTP Microbiology
Lab where they were analyzed qualitatively for surviving spores (MOP 6566). Briefly, in a
biological safety cabinet, Bl coupons were carefully and aseptically removed from Tyvek®
envelopes and placed into bacterial growth media (10 ml of TSA). Culture tubes (18 mm x 150
36
-------
mm sterile borosilicate glass tubes for glass coupons or 25 mm by 150 mm sterile Pyrex* tubes
for wood coupons) containing broth and Bl coupon were then incubated at 35°C for 7 days.
Periodically (on days 1, 3, and 7), the turbidity of the tubes was observed and the results
recorded. Turbid media indicated the presence of bacterial growth, and hence incomplete
decontamination. Representative turbid and lucid culture tubes are depicted in Figure 19.
•V i
• Q P.
• Coupon Holder Location
• Extraction Port for Time Study
Figure 18. Test House, Location of Test Bis (22 red dots),
And Temporal Progression Coupons (blue dot)
37
-------
Figure 19. Photograph of representative culture tubes:
Left: lucid = growth negative
Right: turbid = growth positive
2.8 Temporal Assessment of Efficacy (Quantitative Test)
in order to assess fumigation efficacy as a function of time, six replicates of each coupon type
(wood and glass) along with two procedural blanks of each coupon type (non-inoculated wood
and glass coupons) were retrieved from the tented enclosure at the 16th, 24th, 32nd, and 40th hour
into the fumigation. All coupons for a particular time-point were collocated at the coupon slide
on a stainless steel spring, and retrieved from the house by pulling on a metal wire attached to
the spring (Figures 12 & 14). Samples were allowed to off gas MB then they were packaged (with
cold packs) and transported to RTP NHSRC EPA Microbiology Laboratory for extraction and
analysis. Using aseptic technique in the laboratory, the coupons were placed into 18 mm x 150
mm sterile borosilicate glass tubes (glass coupons) or 25 mm by 150 mm sterile Pyrex" tubes
(wood coupons) each containing 10 ml phosphate buffered saline with Tween20 (PBST). Each
vial was then sonicated for 10 minutes at 42 kilohertz and 135 Watts (MOP 6566). Then tubes
were vortexed two-continuous minutes to further dislodge spores from the glass, wood or metal
coupon. Immediately before dilution or plating, each vial was briefly re-vortexed to homogenize
the sample. The resulting extracts were subjected to five sequential 10-fold serial dilutions (MOP
6535a), and 0.1 ml of each dilution was inoculated onto TSA plates, spread with sterile beads
(MOP 6555), and incubated at 35 °± 2 °C for 18-24 hours. Following incubation, CFUs were
enumerated manually. A photograph depicting representative dilution plates is shown in Figure
20.
38
-------
Figure 20. Representative Dilution Plates Containing Ba Sterne Colonies Recovered from Biological Indicators
2.9 Pre and Post Sponge Stick Sampling
The surrogate spores remained on the test coupons and inside Tyvek® envelops throughout the
study (i.e., during transportation to the site; distribution, fumigation, and collection processes in
the house; and transportation back to the lab). However, sponge wipe samples were collected
on surfaces in the test house before test Bis were deployed to gain an understanding of
background contamination within the house and after test coupons were retrieved at the end of
the fumigation aeration cycle to determine if the test organism escaped the Bl Tyvek® envelopes,
or if other contamination was present on surfaces following the fumigation. Wipe sampling was
conducted according to MOP 3144 and based on CDC protocols (CDC 2012,
http://www.cdc.gov/niosh/topics/emres/surface-sampling-bacillus-anthracis.html). A total of
eight wipe samples were taken, four before and four after the fumigation (see Figure 16 for
sample locations).
2.10 Activated Carbon Scrubber
The activated carbon system was leased from TIGG Corporation (Oakdale, PA) and arrived via
commercial carrier on the morning of Thursday, December 5, 2013. The system was unloaded
from the truck and staged for subsequent placement and installation. The scrubber system
consisted of two scrubber vessels (N5000 PDB, TIGG, Oakdale, PA), each containing
approximately 5,500 pounds of activated carbon (TIGG 5CC 0408); one centrifugal blower with
damper (Model 40-2800, Northern Blower, Manitoba, CA); 75-feet of 20-inch inner-diameter (ID)
flexible rubber ducting with spring steel reinforced helix; 26-kilowatt Generator (Model DB-
05011, Whisperwatt, Los Angeles, CA); exhaust stack (7-inch x 20-inch ID); various galvanized
metal joint fittings; and a galvanized slide gate valve.
The inlet to the scrubber system was connected to the office-room window (Figure 11) located
on the northeastern side of the house. The window was removed and a 4-feet x 24-inch ID
galvanized duct extension was used to penetrate inside the house and connect to the gate valve,
39
-------
with the balance of the window space blocked with cardboard held in place with duct tape. Ten
feet of the flex duct were connected on the outlet side of the gate valve to an 8-foot section of
20-inch ID straight galvanized duct. This straight section was used to perform flow measurements
as described in Section 2.10.1. Following the straight section, ten feet of the flex duct were used
to connect to the inlet of the first activated carbon vessel. Twenty-five feet of the rubber flex
duct was used to connect from the outlet of the 1st vessel to the inlet of the blower that was
positioned ten feet from the vessel. Ten feet of flex duct were used to connect the outlet of the
blower to the inlet of the second carbon vessel. After traveling through both vessels, the
scrubbed gas would be exhausted to the atmosphere through a 7-foot stack located on top of
the second carbon vessel. The generator was positioned approximately 25 feet from the blower.
The entire system took three people approximately 16 hours to install (Figure 21).
Figure 21. Activated Carbon Scrubber Installed at the House
Three activated carbon samples were placed downstream of the slide gate valve. Each of the
three activated carbon samples was contained in a nylon mesh sock that allowed the MB, and all
other potential contaminants, to adsorb on the carbon. One of these carbon samples was
subsequently analyzed for disposal acceptance criteria as described at the end of this Section
3.7.5. The activated carbon used in the samples as well as in the scrubbers was 4x10 coconut
shell activated carbon (TIGG 5CC 0408).
2.10.1 Air Flow Rate at Inlet to First Carbon Vessel
Measurements of air velocity within the duct leading from the house to the first carbon vessel
were taken on 12/8/13 (prior to the fumigation), and after the fumigation on 12/12/13 at around
1200 hours (near the end of the carbon scrubber operation). Both sets of measurements were
made with the carbon scrubber blower on and with the blower's damper set about halfway open.
An opening in the tent near the back porch allowed for intake of make-up air when the blower
was on, (Figure 10). Air velocity measurements were made using a pitot tube connected to an
electronic micro-manometer (Shortridge Instruments, Inc. Scottsdale, AZ, AirData Multimeter
40
-------
ADM 860). Pitot tube traverses across the duct in both the horizontal and vertical directions
were conducted per U.S. EPA Method 1 (http://www.epa.gov/ttnemc01/promgate/m-01.pdf).
The micro-manometer was calibrated by the manufacturer prior to the field test. As required by
EPA Method 1, to minimize bias due to turbulent flow, gas velocity measurements must be made
at a location at least 8-duct diameters downstream and two diameters upstream from any flow
disturbance. Since the inlet duct was made of flexible wire and rubber, a 12-foot long rigid piece
of duct (20-inch inside diameter galvanized metal) was placed between the outlet of the house
and the inlet to the first carbon bed to facilitate undisturbed velocity measurements. However,
the rigid duct had to be shortened to 8 feet due to air leakage at the connections with the flexible
duct, resulting in a length of only 4.8-duct diameters. Thus the minimum duct-length criterion of
the method was not met. The implications of this are discussed in the Results Section 3.7.5.
2.10.2 Scrubber Temperature and Relative Humidity
Temperature and RH of the gas stream were measured at three locations within the carbon
scrubber: in the rigid duct at the inlet of the first carbon vessel, between the two carbon vessels
(at the outlet of the blower), and at the stack (outlet of the second carbon vessel). Temperature
and RH were measured and logged at the inlet to the first bed using a HOBO model U10 (Onset
Computer Corporation, Bourne, MA 02532), which was placed directly inside the duct. For the
other two sample locations, temperature and RH were measured and the data logged using new,
factory-calibrated HOBO Model 023-002 RH and temperature data loggers (Onset Computer
Corporation, Bourne, MA 02532). For these two locations, the sensor tip was inserted into small
holes drilled into either the metal housing of the blower or the metal stack, and then fastened to
the sheet metal.
2.10.3 Scrubber MB Concentration
MB levels were measured within the carbon scrubber at the same sample locations used for the
temperature and RH measurements. A dual-channel VIG Industries (Anaheim, CA) Model 20/2
flame ionization detector (FID) was used to continuously and simultaneously monitor MB levels
at two of the three sample locations. Hydrogen gas (Airgas, Inc., Fort Lauderdale, FL) was
supplied to the instrument from a pressurized gas cylinder for the flame source. MB data were
collected, logged, and stored using a data acquisition system (lOtech Corporation, Cleveland,
OH).
The FID calibration was checked before and after the carbon bed operation portion of the
aeration using both a direct span and a bias span (calibration gases traveled through the sample
line prior to detection). Calibration gases were obtained from Custom Gas Solutions (Durham,
NC), and included 4.96 ppm MB in air, a 996 ppm MB in air, and a 5.28% MB in nitrogen. The FID
was zeroed using ultra high purity air (Airgas, Inc.). The channel 1 detector of the FID served as
the high level MB monitor, and was calibrated using the 996 ppm and 5.28% MB gases, while
channel 2 was the lower level detector and was calibrated using the 4.96 and 996 ppm gases.
Gas samples from the carbon scrubber were pulled via the FID sample pump through unheated
%" Teflon® tubing at a flow rate of four liters/min. Sample line length from each location to the
41
-------
instrument was estimated to be less than 100 ft, yielding a response time of less than eight
seconds.
2.10.4 MB Mass Balance Calculations for Activated Carbon Scrubber
The total mass of MB that exited the house and entered the carbon scrubber was calculated via
integration of the area under the concentration versus time curve (see Results Section 3.7.4).
That is, the mass flow of MB for each time increment was calculated and then summed for the
entire time period in which MB was detected. The MB mass for each time increment (0.5 min)
was calculated via the ideal gas law, using the gas volumetric flow rate, the MB gas concentration,
and the temperature of the gas at the inlet to the first carbon vessel. A similar integration
approach was used to calculate the total mass flow of MB between the carbon vessels and the
amount of MB emitted to the atmosphere via the stack.
2.11 Ambient Air Monitoring
The study team monitored ambient conditions using both wireless air monitoring units and
weather stations. Prior to the start of fumigation, personnel from EPA's Environmental Response
Team (ERT) and Scientific, Engineering, Response and Analytical Services (SERAS) contractor
deployed sixambient air monitoring units (Figure 22) strategically around the house (Figure 23). The units
were deployed around the house and skewed downwind based on local meteorological data. Each unit
contained a RAE Systems AreaRAE (RAE Systems, San Jose, CA) and a Honeywell Analytics
(Morristown, NJ) MDA single point monitor (SPM). The AreaRAE, utilized a 10.6 eV lamp and a
wireless radio frequency modem. Each unit was a five-sensor gas detector with a photo-
ionization detector (PID) installed. The PID was calibrated to be responsive to MB using a 1.7
conversion factor (RAE Systems, 2005: http://www.raesystems.com/products/multirae-family ).
The Honeywell Analytics MDA SPM employed a hydrogen bromide (HBr) Chemcassettes and
ChemKeys (a MB Chemcassette was not available). The Chemkey stores HBr setup information
and other functional information (i.e., flow rate, alarm levels, and compound concentration
times) needed for accurate detection of target gases. The HBr Chemcassette is a medium onto
which a known quantity of ambient air is concentrated. The unit has an internal sample pump
which draws air at a manufacturers predetermined constant flow rate through a chemically
treated paper tape. The tape darkens on exposure to the desired compound. At the end of each
sample period the concentration is converted into an analog output signal. This output is then
digitally stored on an attached data logger.
2 See RAE Systems TN-106 for the proper way to implement a conversion factor. For high concentration
initial doses, it may be desirable to use a dilution fitting. See RAE Systems Technical Note TN-167.
42
-------
Figure 22. Air Monitoring Unit Containing an SPM (left) and an AreaRAE
(right).
43
-------
Hurricane House
Source Esn DigitalGlo
Goimappir-vg ^rogfik
Community*'
20 40
SO
120
160
b Feel
Legend
AreaRAE Location
N
Figure 23. Ambient Air Monitoring Locations.
After deployment, the air monitoring units were calibrated at the study site. SERAS calibrated
the AreaRAE units using zero air and volatile organic chemical (VOC) standards, (isobutylene 100
ppm). Once the AreaRAE units were calibrated to the VOC standard, a bump test was conducted
with MB gas (5 ppm) to ensure that the units were reading MB in the 3-5 ppm range. If any drift
occurred, the units were re-calibrated a second time to insure accuracy.
44
-------
The SPMs were put through ari internal calibration. Once the units were calibrated, they were
bump tested against 5 ppm MB gas. Unfortunately, each of the SPM units failed to read MB and
were removed from the air monitoring scheme.
Weathering stations used to monitor ambient conditions included the Florida Automated
Weather Network (FAWN) and a mobile weather station (Figure 24). FAWN is a group of
permanent weather stations positioned around the State of Florida. A permanent station is
located on the Fort Lauderdale campus just north east of the house. A 600 Series mobile weather
station (Weatherhawk, Logan, UT) was brought by EPA, and positioned just southeast of the
house (Figure 25). Data from both units were read via wireless transmission.
Figure 24. FAWN Permanent Weather Station on the Figure 25. Portable Weatherhawk Weatherstation
University of Florida Deployed on Site
2.12 Leak Detection
In addition to the RAE Systems AreaRAE monitors being used at the six stationary positions, two
MultiRAEs (RAE Systems, San Jose, CA) were used as hand-held detectors for leak testing near
the tarps surrounding the house. A team of two or more walked the perimeter of the cordoned
off area around the house with a MultiRAE, and noted any non-zero readings. When the
perimeter was below the action level, then the team entered the cordoned off zone (30 feet
around the perimeter of the house) and approached the house, while noting any non-zero
readings. When readings were above the action level (0.5 ppm) at the breathing zone of any
45
-------
team member, the team exited the area, dorined SCBAs arid completed the leak survey with
appropriate PRE. Readings were taken all the way around the house, including immediately
adjacent to the tarp at multiple locations. Elevated readings were reported to the tenting and
fumigation contractor for potential leak mitigation.
2.13 MB Fumigation Process
The MB (100%, Meth-O-Gas 100®, Great Lakes Chemical Co., West Lafayette, IN) was contained
as a liquid in commercial 100-lb. metal cylinders (Figure 26). MB without chloropicrin was used
to avoid the potential corrosive damage caused by chloropicrin. Since MB has a boiling point of
3.6°C, heat was added during introduction to insure that only gaseous MB was released from the
end of the shooting hose. This was done by affixing the cylinder valve, by hose, to a 5-gallon-
capacity heat exchanger (Figure 27), The heat exchanger contained a coiled metal tube through
which the MB passed. The coil was surrounded by a water/radiator coolant mixture (60:40)
which was heated by a propane burner to 90°C. The gaseous MB exited the heat exchanger
through the shooting hose at about 70°C and then traveled as a gas through the shooting hose
and exited into the shooting bucket inside the house. The certified applicator (Hammerhead
Termite Control, Big Pine Key, FL) placed the MB cylinder on a balance and donned a full-face
shield before he opened the MB cylinder (Figure 26). All MB released was measured
gravimetrically.
Figure 26. MB Cylinder Being Opened
46
-------
Figure 27. Heat Exchanger Used to Convert Liquid MB to Gaseous MB Exiting the Blue Shooting Hose
The working concentrations of MB during the fumigation were monitored at seven locations with
two Fumiscope® thermal conductivity detectors (Figure 28) (Key Chemical Co., Clearwater, FL,
accuracy approximately ±1 gram per cubic meter MB). One monitor was calibrated with MB in
November, 2013 by Key Chemical Co., and the other Fumiscope was calibrated for Sulfuryl
Fluoride and a correction factor was added to obtain MB equivalence. Fumiscope monitoring
locations included the large classroom (southwest corner), the large classroom podium, the
women's restroom (south), the attic and the room leading to the attic, the HVAC room (north),
and inside the HVAC return duct. The Fumiscopes were fitted with air pumps that pull the interior
MB-laden air through a monitoring line into the instrument which then gave a near real-time
reading of MB concentration. During fumigation and aeration, the MB concentration was
monitored 24hrs/day by authorized personnel (licensed and monitored by the state).
47
-------
1
Figure 28. Fumiscope® Used to Monitor Working MB Concentrations
3 Results and Discussion
3.1 Results from Release and Monitoring of the MB
The time and amount of MB released into the house is provided in Table 5. Initially, 700 lbs of
MB was introduced into the house over nine hours. The introduction of MB was delayed by two
heat exchanger malfunctions (a pressure gauge fitting blew open in the first unit and inlet/outlet
ports were reversed in the second unit) during the introduction of MB from the second 100
pound cylinder. The heat exchanger was repaired and the introduction continued. The target
concentration (212 mg/l) was reached/exceeded at 2100 hours on December 9, 2013, starting
the fumigation clock (time zero ort Table 5), and was maintained for 48 hours. Two additional
50-lb increments of MB were added at 21 and 35 hours after the fumigation start time to
maintain the MB concentration. The concentration of MB in each of the seven locations were
comparable, which indicated adequate mixing within the house. Figures 29 and 30 are examples
from two of those monitoring locations. Additional figures are in Appendix B.
48
-------
Table 5. MB Release (lbs) and Concentration in House.
Time of MB Released
Date
Time (hr)
Elapsed Time (hr)
Inside Cone, (mg/l)
Lbs
12/9/2013
1200
-9.0
0
Initiate
12/9/2013
1224
-8.6
34
100
12/9/2013
1627
-4.5
102
201
12/9/2013
1712
-3.8
136
100
12/9/2013
1811
-2.8
170
100
12/9/2013
1943
-1.3
204
100
12/9/2013
2100
0.0
212
Start
12/9/2013
2122
0.7
238
100
12/10/2013
1800
21
225
50
12/11/2013
0800
35
230
50
House Concentration of MB (mg/l) over Time (hours)
Location: Backroom
300
250
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Figure 29. Concentration of MB (mg/l) over Fumigation Time (hr), Backroom Location
49
-------
House Concentration of MB (mg/l) over Time (hours)
Location: Attic
300
250
200
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Figure 30. Concentration of MB (mg/l) over Fumigation Time (hr), Attic Location
There was a reduction in the MB concentration in the house over time. When fumigant was not
being added to the house, the loss of MB can be observed (negative slope of the concentration
line shown in Figures 29 and 30). We can assume that the loss of MB from the house was the
sum of four contributions: leakage around the tarp material through penetrations (e.g., ducting
connections from the house to the scrubbers), permeation through the tarp material, sorption
into materials, and transformation or chemical reaction. However, MB is relatively stable so the
transformation or chemical reaction contribution can be assumed to be close to
zero. Additionally, according to a study conducted by Corsi et al. (2007), sorption and
chemisorption with MB are negligible. The mattresses and other foams and fabric materials did
not seem to increase the demand for the fumigant, nor did they seem to lengthen aeration time
for this study. Sorption from these materials most likely reached equilibrium over a short time,
bringing the sorption contribution to zero as the fumigation time increased. Note: these
materials were only a small fraction of the overall volume of the space fumigated in this study.
Specific studies to determine these affects should be conducted.
Assuming the loss of MB follows a first-order decay rate, the MB loss can be calculated from this
concentration data. From this analysis, an estimated 91 kg (200 lb) of MB was lost during the
entire fumigation, with leakage and permeation being the most likely contributors. Some leakage
around the perimeter of the house was indicated by monitoring and observed, after the study,
by the killing of the grass under the apron, see Figure 42 in Section 4.6. Sand was placed around
the base of the house in effort to reduce the leakage. As noted in Table 6, the rate of MB loss
decreased as the fumigation progressed in time. The decrease was most likely due to the
implementation of these early leakage mitigation measures. Furthermore, toward the end of the
fumigation (after the leakage mitigation was completed), 35 to 48 hours into the fumigation,
there was little to no change in the leakage rate. At this point, the rate of about 1 mg/l/hr was
observed.
50
-------
To normalize based on the area of tent, the area that gas could potentially leak from, this leakage
rate was divided by the total area of the tent (approximately 7200 ft2) and multiplied by the
volume inside the tent 1,284 m3 (45,348 ft3), resulting in a leak rate of 178 mg/hr per square foot
of tented area. If conditions and materials are similar, this leakage value may be used at other
sites to estimate fumigant loss/leakage. However, every site will have its own unique
containment issues that will affect fumigant loss. In this case, the house is well understood and
may represent a best-case seal as compared to typical structural fumigations.
In addition to the MB losses that can be calculated from analyzing the fumiscope data, there are
also losses caused by added pressure when MB was introduced into the house. Over the course
of the fumigation a total of 363 kg (800 lbs) MB, about 94 m3 (3336 ft3) of gas (28°C and one
atmosphere of pressure), equaling about 7.4% of the entire volume of the enclosure, was
released into the house. A 7.4% increase in volume produces a positive pressure inside the
house. As an estimate of the loss of MB due to this positive pressure it is assumed that 7.4% of
the total 363 kg or 27 kg (59 lbs) of MB leaked from the house, close in time to when the MB was
added to the house. The mass balance of MB for the whole fumigation process will be discussed
following the mass balance of MB in the scrubber system, Section 3.7.5.
Table 6. Loss of MB from the House in Milligrams per Liter per Hour
Time Range
Loss in mg/l/hr
From
fhr)
To
fhr)
At fhr)
Podium
Ducting
Attic
Backroom
Average
0
21
21
1.46
1.66
1.65
1.67
1.66
21
35
14
1.15
1.12
1.11
1.11
1.11
35
48
13
1.15
1.02
0.79
1.13
0.98
3.2 House Temperature and Humidity Results
The heaters and steam vaporizers were turned on and off as described in Table 7. The average
temperature and RH for each Bl location inside the house is shown in Table 8. The average
temperature inside the house during fumigation was 27.8 °C and the RH was 82.9%, these values
slightly exceeded the desired fumigation conditions of 27 °C and 75% RH. One location only, the
Mechanical Room did not meet or exceed the temperature set point of 27 °C.
51
-------
Table 7. Heater and Humidifier on and off Cycles during Fumigation
Date: 2013
Dec. (day)
Time (hr)
Humidifier
Bank 1*
Humidifier
Bank 2*
Heater Bank
1*
Heater Bank
2*
7
1400
On**
On**
Off
Off
9
0710
On
On
9
1130
On
On
9
1300
Off
10
0130
On
10
0335
Off
10
0445
Off
10
0620
Off
Off
10
1950
On
10
2100
Off
*A set of eight steam vaporizers was considered a "humidifier bank" and a set of two heaters was considered a
"heater bank".
**The steam vaporizers were refilled several times during the two days between 1400 hours 12/7/2013 and
1200 hours 12/9/2013, however, they were sometimes left empty, not generating, during this time.
52
-------
Table 8. Average Temperature (T) and RH during 2-Day Fumigation Inside House
Location ID
HOBO ID
Location
T(°C)
RH (%)
A
29
Entry room
27.2
81.7
B
17
Entry room
28.1
79.7
C
18
Office
26.6
84.3
D
10
Office
26.1
89.6
E
47
Kitchen
27.4
86.1
F
31
Kitchen
27.6
83.2
G
24
Mechanical room
28.4
80.0
H
34
Mechanical room
26
91.9
I
22
Hurricane Shelter
27.9
83.7
J
38
Attic
29.1
75.0
K
42
Attic
29.2
77.9
L
None
Storage room
N/A
N/A
M
57
Storage room
28.9
79.6
N
20
Classroom
29.3
77.2
0
44
Classroom
28.8
80.1
P
54
Back-porch
27.2
87.0
Q
11
Back-porch
27.6
83.8
R
55
Custodial
27.4
84.5
S
21
Restroom (Men)
27.9
82.8
T
30
Restroom (Womens)
27.6
84.3
U
43
Janitors Closet
28.2
79.6
V
58
Front porch
27.3
88.3
Average
27.8
82.9
3.3 Leak Monitoring Around the Perimeter of the House Results
Leak detection evaluations were conducted using hand-held RAE Systems MultiRAE, with photo-
lonization detector with 10.6 eV lamp, around the cordoned-off area and directly next to the
house, close to the tarps. The surveys were done by a variety of team members over the course
of the fumigation, especially early in the process so that any leaks could be detected and
addressed if possible. Leak information was given directly to the tenting and fumigation
contractors, however, the readings obtained were not always recorded. The lack of notable
findings, zero readings during monitoring at the caution perimeter may have resulted in a sense
that the findings did not need to be recorded. Figure 31 shows the results of one leak-test survey
taken when winds were unusually calm (less than one mile per hour winds reduced the dilution
53
-------
affects and provided a worst case scenario leak detection). The readings recorded at the caution
perimeter in Figure 31 (0.5-1.0 and 0.5 ppm) were the highest instantaneous readings; however,
most readings were zero even at these locations. There were no sustained high (>0.5 ppm)
readings recorded at anytime at the caution perimeter. The concentration values near the house
were obtained directly at the tarp fabric and do not reflect breathing zone concentrations.
Breathing zone concentrations for the crew taking the readings were at least an order of
magnitude lower than the concentrations obtained directly at the tarp using instrumentation
extended away from body. Even when MB readings were high (an instantaneous high of 27 ppm
in Figure 31) when touching the tarp there were no sustained high (>0.5 ppm) breathing zone
readings for the monitoring crews.
Caution perimeter
0.2
1.0-4.0
0.5
1.8-2.0
0.5
Hurricane House
56.760 ppm
3.0-6.0
2.0 •
1.0-2.0
5.0 27 4.0-5.0
1.0-1.5
Florida Automated Weather Network, FLREC
http: //f a wn. ifas. ufl. ed u /d a ta /
0.5-1.0
0.5
60cm Tavg 2m DewPtavg RelHumavg 10m Windavg 10m Windmin Wdir avg
(F) (F) 2m (pet)
68.31 70.71 98
Period
12/11/2013 4:00
{mphj
0.9
(mph) 10m {(leg) (#obs)
0.13 338 4
Figure 31. Leak Detection Results from December 11, 2013 at Approximately 4:30 a.m. Calm Wind Conditions.
Dots without Numbers Indicate Zero MB Detected and Numbers Reflect Highest Instantaneous MB
Concentrations (ppm).
54
-------
3.4 Biological Indicator (BI) Results
3.4.1 Pre- and Post-Test BI Population Comparison
The spore population densities were recovered from the test coupons pre- and post-test and
compared. No statistically significant differences were detected in the population densities of
spores on pre- and post-test control samples (Table 9).
Table 9. Spore Population Densities on Pre- and Post-Test Control (Non-Exposed) Bis
BICoupon
Type
Pre-Test
Population
Post-Test
Population
n
p-value (two tailed
Student's t-test)
Stainless Steel
2.2 x 106
2.5 x 106
10 pre-test,
10 post-test
0.1297
Glass
2.0 x 106
2.2 x 106
3 pre-test,
3 post-test
0.2499
Wood
9.6 x 105
4.6 x 105
3 pre-test,
3 post-test
0.0659
The abundance of viable spores on non-exposed BI coupons were similar before and after the
field test, indicating that time in storage did not significantly affect the spore titer on the Bis.
Recoveries from glass and stainless steel were within the targeted range (2.0 to 5.0 x 106):
recoveries from wood, however, were lower than the amount inoculated onto these coupons.
These results were expected, as recoveries from glass and steel are typically between 75 - 95%
of the inoculum, while recoveries from wood have historically been between 1 - 25% of the
inoculum. From Table 9, Spore Population Densities on Pre- and Post-Test Control Bis, it is
apparent that glass and wood demonstrated recovery efficiencies of 91% and 44% of the stainless
steel control carriers for pre-test evaluations, and 88% and 18% for post-test evaluations, each
respectively. All carriers were inoculated with a population density (as determined by the BI
supplier) of 4.2 x 106 CFU / carrier. Accordingly, mean recovery efficiencies from stainless steel,
glass, and wood were 52%, 47%, and 23%, respectively. It is therefore presumed that the lower
estimates of spore population density on wood coupons were due to lower recovery efficiencies
from the porous wood surfaces and not due to inactivation of spores on the surface.
3.4.2 Spatial Assessment of Efficacy (Qualitative Test) Results
Results from all 22 locations are shown in Table 10.
55
-------
Table 10. Bl Results from the Spatial Assessment of MB Fumigation Efficacy
Location
Location
ID
Test Bis
(growth-positive Bis / total Bis)
Procedural Blanks
(growth-positive Bis / total Bis)
Wood
Glass
Wood
Glass
1
A
0/4
0/4
0/1
0/1
2
B
0/3
0/4
0/1
0/1
3
C
0/4
0/4
0/1
0/1
4
D
0/4
0/4
0/1
0/1
5
E
0/4
0/4
0/1
0/1
6
F
0/4
0/4
0/1
0/1
7
G
0/4
0/4
0/1
0/1
8
H
0/4
0/4
0/1
0/1
9
I
0/4
0/4
0/1
0/1
10
I
0/4
0/4
0/1
0/1
11
K
0/4
0/4
0/1
0/1
12
L
0/4
0/4
0/1
0/1
13
M
0/4
0/4
0/1
0/1
14
N
0/4
0/4
0/1
0/1
15
0
0/4
0/4
0/1
0/1
16
P
0/4
0/4
0/1
0/1
17
Q
0/4
0/4
0/1
0/1
18
R
0/4
0/3
0/1
0/1
19
S
0/4
0/4
0/1
0/1
20
T
0/4
0/4
0/1
0/1
21
U
0/4
0/4
0/1
0/1
22
V
0/4
0/4
0/1
0/1
Total
0/87
0/87
0/22
0/22
Positive Controls
(growth-positive Bis / total Bis)
Negative Controls*
(growth-positive Bis / total Bis)
Wood
Glass
Wood
Glass
Not Exposed
24/24
24/24
2/24
0/24
+Note: Negative Control Bis were not sterilized prior to packaging and were not guaranteed sterile as received
from Yakibou Inc., therefore growth-positive negative controls were not unexpected. Results from lab-sterilized
(autoclaved) Bis (Table 11) indicate whether Bl culture procedures were performed aseptically.
None of the 87 wood or 87 glass test Bis had viable spores following fumigation. One wood test
Bl (Location 2) and one glass test Bl (Location 18) were not analyzed. At Location two only three
wood test Bis, instead of four, were deployed. At Location 18 the glass coupon was missing from
inside the Tyvek® envelope (apparently it was not placed into the envelope during laboratory
preparation). Similarly, the 22 wood or 22 glass procedural blanks (not inoculated, fumigated)
showed no turbid media (no growth) following attempted culture. These results verify that the
MB fumigation was effective throughout the entire house, as no spatial differences in Bl
inactivation were apparent. No growth on any of the procedural blank Bis suggests that
inadvertent contamination during field or lab procedures was not apparent. All 24 wood and 24
glass positive-control coupons (inoculated, not exposed) were indeed positive for growth upon
56
-------
analysis. Two of the 24 negative control (not inoculated, not exposed) wood coupon Bis were
positive upon analysis, none of the 24 negative control glass Bis were positive. The two growth
positive negative control wood Bis were not surprising, as the Bl producer did not guarantee
sterility of these coupons as provided. None of the microbiology lab negative controls were
positive for growth, suggesting that inadvertent contamination of samples during lab procedures
was not apparent (Table 11).
Table 11. Results from the Analysis of Microbiology Laboratory Control Samples
Microbiology Lab Controls
Results
(number growth-positive / total analyzed)
10ml TSA in 25mm Tubes
0/3
TSA Plates
0/3
10ml TSA in 18mm Tubes
0/3
Inoculum-spreading beads
0/1
0.9ml PBST Dilution tube
0/1
Cell spreaders
0/6
Lab-Sterilized Negative Control
Bis
0/3 steel, 0/3 glass, 0/3 wood
3.4.3 Temporal Assessment of Efficacy - Quantitative Test (Time-Series Test) Results
Results from the quantitative tests of spore survival during the temporal assessments of efficacy
are shown in Table 12.
Table 12. Bl Results from the Temporal Assessment of MB Fumigation Efficacy
Time
Point
(hours)
Sample
ID
Test Coupons
Total CFU Recovered
Procedure
Total CFU R
il Blanks
recovered
Wood (n=6)
Glass (n=6)
Wood (n=2)
Glass (n=2)
16
W
O1?
828 ± 2027+
0
0*
24
X
0
0
0
0
32
Y
0
0
0
0
40
Z
0
0
0
0
tViable spores recovered from 1 of 6 replicate Bis. 4967 CFU recovered from replicate #6 of 6.
Tor replicate #2 of 6), one filter-plate sample (1ml analyzed) yielded 1 CFU. When 7.5ml from the same sample
was analyzed, zero CFU were observed. No CFU were observed from the other 5 replicate wood Bis at the 16
hour time point.
"Contamination by non-target bacteria was observed on both glass procedural blanks at 16h time point. For
replicate 1, 2 CFU and 14 CFU were observed on the 1 ml and 7.9 ml filter-plates samples, respectively. For
replicate 2, 4 CFU and 50 CFU were observed on the 1 ml and 7.8 ml filter-plate samples, respectively.
Analysis of the time-series coupons showed viable spores (4967 CFU) were recovered from only
one of the six replicate glass coupons exposed for 16 hours, resulting in an average recovery of
828 spores across the six replicates. The remaining 5 of 6 replicates showed zero recovered viable
spores. Similarly, only one wood coupon exposed for 16 hours had any viable spores. This wood
57
-------
coupon had only one CFU detected during the 1 ml filter-plate analysis. Interestingly, filter-plate
analysis of the remaining 7.5 ml resulted in no growth. Log reductions for all wood Bis during the
quantitative temporal assessment portion were greater than or equal to 5.7. All glass, other than
the 16-hour exposure, were greater than or equal to 6.3 LR. The 16-hour exposure for glass
yielded a 3.41 LR.
Contamination by non-target bacteria was detected on both glass procedural blank coupons at
the 16 hour exposure point. Contamination on procedural blanks is not unexpected, as discussed
before. Overall, the MB treatment was efficacious, as 46 of 48 test coupons were completely
negative for growth of Ba Sterne at any of the time points tested. Exposures at the 16 hour time
point, where 10 of 12 coupons were completely inactivated, just missed the 6-log reduction
efficacy criteria. These temporal results indicate that the fumigation was efficacious (> 6 LR) early
(after 24 hours) in the process, at the temporal extraction location in the house.
3.5 Surface Sample (Sponge Wipe Samples) Results
Results from the surface sample wipes are shown in Table 13. Two blank surface samples
collected showed no growth upon microbiological analysis. Three of the four collected pre-
fumigation surface wipe samples showed the abundance of background organisms (non-Ba
Sterne). Similarly, three of the four post-fumigation surface wipe samples showed background
contamination after fumigation (also non-Ba Sterne organisms). Coupon sample collection and
other post-fumigation activities occurred before wipe sample collection. Both of those activities
may have contaminated the surfaces that were later wipe sampled. The wipe samples that follow
fumigation should be conducted in concert with all of the other post-fumigation activities in
mind. Recontamination of the house, even with organisms that do not have negative health
consequences, may interfere with post-fumigation sampling.
58
-------
Table 13. Recovery Results from the Surface Samples Collected within the House
Sample
Sample ID
Location
Pre- or Post-
Fumigation
Recovery (CFU)
Swabl
HHMB 07051
Kitchen wall
near oven
Pre
0
Swab 2
HHMB 07050
Server room desk
Pre
TNTC
fnon-Ba]
Swab 3
HHMB 07036
Floor by back
door ("west")
Pre
TNTC
("non-Ba")
Swab4
HHMB 07118
Reception desk
Pre
TNTC
("non-Ba")
Swab 5
HHMB 07049
("blank")
Blank
Pre
0
Swab 6
BW001
Reception desk
Post
TNTC
("non-Ba")
Swab 7
FW002
Kitchen cabinet
Post
TNTC
("non-Ba")
Swab 8
LW003
Server room
desk
Post
0
Swab9
W004
Floor by back
door ("west")
Post
TNTC
fnon-Ba]
Swab 10
Blank
Blank
Post
0
TNTC = Too Numerous to Count (background organisms and/or contamination of sample)
Non-Ba = Organisms found on the wipe samples were not classified as Ba
3.6 Ambient Air Monitoring Results
Ambient outdoor conditions were monitored throughout the fumigation process. Perimeter
monitoring was continuously conducted during fumigation using the wireless AreaRAEs. During
this process, SERAS calibrated each unit daily against the VOC standard, and bump tested with 5
ppm of MB. Whenever any drift occurred, due to outside factors, the units were re-calibrated.
AreaRAE readings were logged throughout the fumigation process (See Appendix B). The study
team utilized the readings to determine compliance with the 0.5 ppm MB action levels developed
for this site during fumigation operations. Any readings that were above the action levels were
further investigated using a MultiRAE handheld unit.
On occasion elevated readings seen on the AreaRAEs were investigated with a handheld unit.
The elevated readings were shown to be false positives for several different reasons. A
phenomenon known as "Hotbox syndrome" (which is caused by the sun heating the black
Pelican™ case housing the AreaRAEs) caused the units to exceed their operable temperatures.
The units were then cooled or replaced to improve accuracy.
Vehicle emissions contributed to at least one known false positive reading during fumigation (and
later during aeration). A truck parked close to Location 101 on December 10, 2013. The AreaRAE
59
-------
unit exceeded the action level of 0.5 ppm and was investigated. Upon examination with a
handheld, the elevated readings were shown to be caused by the idling truck.
Other contributing factors that led to investigations were moisture from high humidity. High
humidity was encountered during each day of fumigation phase, often from around 9 PM until
just after dawn. Data from area weather stations indicated that RH rose above 80% several times
for extended periods. The high moisture content in the air can create interference for the
AreaRAE's PID sensor. Virtually all units indicated elevated VOC levels (Appendix B) at one time
or another. Investigations with a handheld MultiRAE unit determined that sustained elevated
readings were false positives possibly due to moisture. SERAS recalibrated the AreaRAE units
during times of high humidity to help clean the sensor and improve their accuracy. There were
no substantiated, sustained elevated levels (> 0.5 ppm) of MB at any of the AreaRAE monitoring
sites at any time during the fumigation.
At approximately 1700 hrs, 12/11/2013, the air monitoring units were re-deployed/re-named to
prepare for aeration and to better surround the scrubber units (Figure 32). Location 101 was
removed from the MB release point and repositioned as Location 201 near the command post.
Location 102 was renamed Location 202. Location 103 was removed from near the storage
buildings and redeployed as Location 205 north of the scrubbers (between the greenhouses and
the house). Location 104 relabeled as Location 203. Location 105 was stayed in the same location
but was given the name Location 204. Lastly, Location 106 moved closer to the personnel air
monitoring the scrubber process, and renamed Location 206.
Throughout the aeration process, the AreaRAE readings were datalogged (See Appendix B). As
noted earlier, vehicle emissions were recorded on AreaRAEs at Location 201 and 202 as vehicles
parked at the command post or traversed the nearby roadway. The remainder of the AreaRAE
units did not detect any significant readings.
Outdoor wind speed, temperature and humidity are also plotted by day in Appendix B. Wind
speed can have a significant effect on ambient MB concentrations. This effect is further
addressed in the Section 3.8, Modeling.
60
-------
Hurricane House
ISourca^Eg
^tmaopir!|
gbnil^uStVi
Legend
^ AreaRAE Location
Figure 32. Re-Deployed Locations of Air Monitoring Units during the Aeration Process
3.7 Activated Carbon Scrubber Results
3.7.1 Flow Rate at Inlet to First Carbon Bed
The average velocity, in the duct preceding the first carbon vessel, measured prior to
commencement of the fumigation was determined to be 1185 ft/min, corresponding to a flow
rate of 2583 ft3/min. The post fumigation velocity and flow rate measurements were taken
toward the end of the carbon bed operation phase (at time 0000 hrs on 12/12/13), and were
determined to be 1280 ft/min and 2790 ft3/min, respectively. While these pre- and post-
fumigation flow results are correlated fairly well with each other (within 8%), the minor
61
-------
difference is likely due to having a larger diameter opening in the tarp near the large porch at the
back of the house (during the latter measurement) that allowed less resistance for make-up air
to enter the house while operating the carbon bed system blower. Refer back to Section 2.2 and
Figure 10 of this report for further details and discussion of the custom made tarp opening
created for aeration.
3.7.2 Temperature and RH at the Carbon Scrubber during Scrubbing
The results for the temperature and RH measurements of the gas at the inlet to the first carbon
bed are shown in Figure 33. Both temperature and RH immediately elevated once the aeration
began (at approximately 2100 hours on 12/11/13). Temperature climbed from an ambient level
of approximately 21 to 27 °C, consistent with the house fumigation target temperature of 27 °C.
During the aeration process, the temperature at the inlet gradually decreased a few degrees until
the blower was shut off.
Once aeration began, the RH at the inlet spiked from a level of 88% (a level consistent with
ambient RH) to over 98%. This initial spike in RH may have been due to excess moisture
originating from the fumigation inside the house that had condensed within the duct near the
gate valve. Following this initial spike, the RH fell to 80% during the next 15 minutes, a level
consistent with the fumigation target RH. At this point, the RH level continued to decrease, but
at a lower rate, until it reached a minimum of 75% at 2224 hrs (1-hr, 24-min scrubber time). Then
the RH slowly began to increase again, in conjunction with a decrease in temperature. This is
consistent with having no change in absolute humidity of a gas, resulting in an RH increase with
a decrease in temperature.
62
-------
X
cr
CO
-------
temperature was likely due to MB breakthrough from the first carbon bed carrying over to the
second bed, with the accompanying heat of adsorption in the second bed.
blowerT
blowerRH
stackT
stackRH
110-
100-
90-
80-
70-
~
50-
D5
"O
40-
30-
blower shut off 0036
20-
20:45 22:08 23:32 0:55
time
Figure 34. Temperature and RH Levels at the Blower and in the Stack, during Scrubbing
The blower RH level was initially at a level of ~ 54%, then spiked to nearly 70% when aeration
began. This initial spike in RH may have been due to a number of factors, such as the driving off
of moisture that had condensed and accumulated in the duct, similar to what may have
happened at the carbon bed inlet. Following this short term initial increase in RH, at the blower,
it trended downward to a minimum of 25%, by 2230 hrs (1-hr, 30-min scrubber time). The RH at
the blower then increased, until it reached a level of about 45% at the time the blower was shut
off. This fall in the blower location RH, followed by its subsequent increase, may be due to the
first carbon bed's adsorption of water vapor to the point of reaching its capacity. That is, we
suspect that once all active adsorption sites on the first carbon bed were utilized for capture of
water vapor and MB, no additional water vapor could be adsorbed onto the carbon. Additionally,
preferential adsorption for MB rather than water vapor may also have contributed to the release
of water vapor from the first carbon bed back into the gas stream. There is also the possibility
that the RH at the blower may have been influenced by gas temperature at the blower, more so
during the latter half of the carbon bed operation, when the temperature was decreasing.
The average stack RH level just prior to the start of aeration was 72%. Once aeration began, the
stack RH plummeted to a level consistent with the initial RH levels seen at the blower. From then
64
-------
on, the downward trend in RH levels in the stack gas followed blower RH trends, with a similar
time lag observed with the gas temperatures. However, unlike the blower RH, the stack RH level
never appeared to reach a minimum level until the blower was stopped. This apparent continued
removal of water vapor in the second carbon bed is presumably due to the availability of
adsorption capacity (for both water vapor and MB) of the second bed. Once the blower was
turned off, stack RH levels quickly trended upward, toward ambient levels.
3.7.4 MB Levels during Aeration
Initially, gas sample from the inlet to the first carbon bed was routed to channel 1 of the FID, gas
sample from between the two carbon beds was measured using channel 2, and no gas sample
was taken at the stack. As the carbon bed system scrubbing operation proceeded and MB
breakthrough occurred for the first carbon bed, the study team switched the sample lines and
channels, as needed, to ensure the appropriate data were obtained from all three sample
locations. The result was that for the latter portion of the carbon bed scrubbing, more emphasis
was placed on securing data for the stack location, resulting in intermittent time periods in which
no MB data were available for a particular location. Towards the end of the carbon bed system
operation, gas sampling occurred at the inlet to the carbon beds and at the stack. Monitoring of
the carbon bed system was stopped (the blower was turned off at 0036 hrs (3-hr, 36-min
scrubber time) on 12/12/13 when the stack MB concentration was nearly equivalent to the MB
concentration in the house.
The MB levels observed in the duct at the inlet to the carbon bed during carbon bed operation
are presented in Figure 35. Just prior to turning the blower on to start the aeration process (at
2100 hrs), the MB level in the duct was ~ 5,000 ppm. The presence of MB in the inlet duct prior
to aeration indicated some leakage from the gate valve in the duct between the house and the
first carbon bed, which was expected. Once the blower was turned on, the MB level peaked
immediately to 41,000 ppm, and then thereafter gradually decreased over time. There were
three periods of time when FID gas inlet sampling was temporarily suspended to allow for
sampling at the stack: 1) at 2157 hrs (57-min scrubber time); 2) at 2246 hrs (1-hr, 46-min scrubber
time); and 3) at 2327 hrs (2-hr, 27-min scrubber time). The carbon system blower was shut off
when the MB concentration at the scrubber inlet decreased to 137 ppm, ending active aeration
through the carbon scrubber.
The MB levels measured at the blower location (between the carbon beds) are shown in Figure
36. As with the sampling at the inlet and stack locations, there were intermittent periods when
gas sampling for the blower location was stopped to allow for sampling of the other two
locations; hence no data are available for those time periods. Breakthrough of MB from the first
carbon bed occurred around 2145 hrs (45-min scrubber time). MB emission levels from the first
carbon bed then continued to climb until sampling stopped for this location at 2339 hrs (2-hr, 39-
min scrubber time), when MB levels had reached 2843 ppm. From approximately 2220 hrs until
2312 hrs, MB levels increased relatively rapidly at a rate of about 36 ppm MB per minute.
However, during the last few minutes of sampling at this location, MB emissions levels began to
65
-------
stabilize, with the rate of increase in MB levels diminishing to approximately 5 ppm MB per
minute.
40,000 -
£ 30,000
Q.
Q.
0)
-g
£
0
JD
>
SZ
-t—'
01
20,000
10,000-
1 , level in duct just prior to
starting blower at 2108 hours
0-
"I 1 1
21:07 21:56:30
—I 1 1 1 1
22:46 23:36:30 0:26:30
Time
Figure 35. MB Concentration in the Duct at the Inlet to the First Carbon Bed during Scrubber Operation
3000
2500
E
& 2000
(D
T3
| 1500
-O
">»
I 1000
500-
oH
¦
¦
/
I 1 I 1 I 1 I 1 I ' I ' I 1 I
21:07 21:31 21:56:30 22:21:30 22:46 23:11:30 23:36:30 0:01
time
Figure 36. MB Concentration at the Blower (Between the Carbon Beds) during Scrubber Operation
66
-------
The MB levels measured at the stack location during carbon bed system operation are shown in
Figure 37. Stack sampling occurred intermittently from about 2133 hrs until 2248 hrs. During
that time period, the MB levels were reading negative on the FID. (We note that the instrument
was zeroed using ultra high purity air, but would read negative values when sampling ambient
air.) Sustained positive readings on the FID (i.e., breakthrough of MB from the second bed)
occurred at 2305 hrs (2-hr, 5-min scrubber time), with MB levels continuing to climb up to 156
ppm, when blower operation was terminated at 0036 hrs on 12/12/13 (3-hr, 36-min scrubber
time).
Next, the study team shut down the scrubber system, removed the inlet duct from the carbon
beds and sealed the beds. The duct was also removed from the valve gate, and a fan was placed
at the valve gate opening to blow fresh air into the house aiding the natural aeration process.
The following morning additional openings were made in the tarps to aid the natural aeration
process. One of the MB sampling line (bathroom sample line) was switched from a Fumiscope
to the FID (channel 2) detector. This was done to enable more accurate MB readings within the
house, since Fumiscopes are not sensitive enough at low levels, i.e., < 1 mg/L (Fumiscope Version
5.1 Manual pg. 2; personal communication with Rudolf Scheffrahn, UF Professor of Entomology;
email on 2/6/14). In addition, personnel wearing SCBAs entered the house at this time to obtain
the Ba (Sterne) inoculated coupons. Sampling of the bathroom air represented a space in the
house with limited air ventilation and most likely some of the highest remaining MB
concentrations. Sampling of the bathroom with the FID instrument commenced around 1040
hrs, and continued until 1311 hrs on 12/12/13.
The results for MB sampling within the house (bathroom) during a portion of the natural aeration
are shown in Figure 38. When FID sampling began at 1042 hrs, the MB level was 196 ppm, and
had dropped to 16 ppm at 1310 hrs. From the curve on Figure 38, the concentration decay rate
in this bathroom is estimated to be 0.38 mg/l/hour.
3.7.5 MB Mass Balance for Activated Carbon Bed System and for Entire Fumigation
Based on the MB levels measured at the inlet to the first carbon bed, as shown in Figure 35, the
total mass of MB entering the carbon system was calculated to be 243 kg (536 lb). From
integration of the MB levels in Figure 36, the total mass of MB estimated to have exited the first
carbon bed was 96 kg (211 lb), and integration of the data in Figure 37 resulted in an estimated
1.8 kg (4 lb) exiting the stack while the carbon system was operating. By difference, it is estimated
that 147 kg (325 lb) MB were collected on the first carbon bed, and 94 kg (207 lb) were captured
on the second bed. With 2495 kg (5,500 lb) of carbon in each bed, the adsorption of MB onto
Bed 1 is 2.9 kg (6.5 lb) MB per 45 kg (100 lb) carbon, and the second bed adsorption was 1.9 kg
(4.1 lb) MB per 45 kg (100 lb) carbon. The overall removal efficiency of MB by the carbon bed
system, therefore, is 99% for MB that enters the scrubber system. Please refer to Figure 39 for
a diagram of the mass balance for the carbon system.
67
-------
160 -
140 -
blower shut off
120 -
£ 100-
60-
40-
20-
21:30 22:18:30 23:08:30 23:58 0:48:30
time
Figure 37. MB Concentration at the Stack during Scrubber Operation
200-i
E
o
o
|m
-C
150-
CD
-Q
(A
"c
Q)
£ 100-
O
5
c
Q)
50-
U-
CQ
(1)
10:42
11:16
11:49
12:22
12:55
time (12/12/13)
Figure 38. MB Concentration fPPM) within a Bathroom during Natural Aeration
68
-------
The MB concentration in the house measured by the Fumiscope was 218 mg/L (0.22 oz/ft3), just
prior to starting the carbon scrubber, which is equivalent to 56,400 ppm at 27 °C. Using a house
volume of 1,284 m3 (45,348 ft3) (UF provided estimate of air volume, excluding solid objects,
within the house), the total mass of MB in the house just prior to operation of the carbon
scrubber system was calculated to be 280 kg (617 lbs). At the end of the carbon bed operation,
the Fumiscope was reading 1 mg/L (0.001 oz/ft3) which is equivalent to 1.3 kg (3 lbs) total MB left
in the house, 279 kg (614 lbs) of MB was pulled from the house.
The initial reading taken by the FID after starting up the scrubber was 41,000 ppm, 27% lower
than the Fumiscope reading of 56,400 ppm. The most likely reasons for the difference would
include low FID reading caused by ambient dilution air entering the duct (e.g., where the duct
was connected to the house) or ambient dilution air entering through tent leaks near where the
duct was connected to the tent. All ambient air entering the house near, or at, the scrubber duct
would contribute to dilution of the MB concentration as this air is mixed with the gas pulled from
the house toward the first carbon bed where the FID sampling port was located. Make-up air
entering the house near the back porch may also have mixed poorly with gases throughout the
house (though mixing fans were left on during this period of time), traveling through channels
directly to the exhaust duct and to the FID sampling point.
There are numerous possibilities for the discrepancy between what was removed from the house
as calculated using the FID carbon scrubber measurements 243 kg (536 lb), 13% lower than the
MB mass calculated using the house volume and Fumiscope readings 279 kg (612 lbs).
Differences could be attributed to inaccuracy in MB measurements (for either the Fumiscope-
measured levels in the house and/or the FID-measured levels in the exhaust gas), an inaccurate
house volume estimate, or an inaccurate blower gas-flow-rate measurement.
As stated earlier, accurate gas velocity readings in the duct are critical for obtaining an accurate
mass emission rate. We followed U.S. EPA stack gas flow measurement procedures, with the
exception of not having a sufficient length of straight duct to minimize turbulent flow. Turbulent
flow may have been present due to shorter straight-duct length where measurements were
made, resulting in biased flow measurements. In addition, EPA Method 2 requires that static
pressure be measured within the duct, and that the volumetric flow rate should be adjusted for
this relative to atmospheric pressure. Static pressure was not measured within the duct, and so
this adjustment could not be made.
The mass balance of MB for the entire fumigation includes the total mass of MB released in the
house, 363 kg (800 lbs); the MB leakage rate calculated for the combination of penetration
around the tarp material, permeation through the tarp material and sorption into other materials
within the house, 91 kg (200 lbs), as estimated in Section 3.1; the MB forced from the house by
displacement of fumigant as a result of the addition of 363 kg (800 lbs) of MB was 27 kg (59 lbs),
as estimated in Section 3.1; the MB taken to the scrubber, 243 kg (536 lbs); and the MB left in
the house at the end of scrubbing, estimated to be 1.3 kg (3 lbs). Remarkably, based on these
estimates, the entire fumigation mass balance of MB (363-91-27-243-1=1 kg) results in only 1 kg
(2.2 lbs) of MB unaccounted for.
69
-------
4lbs/V
House
617 lbs*
Blower
Straight duct to do flow measurements
*Calculated MB mass based on Fumiscope reading of
218 mg/L, house volume of 45,348 ft3, air flow rate of
2790ft3/min, and FID readings in scrubber system ducts
Figure 39. Mass Balance of MB for the Activated Carbon Scrubber
Once the fumigation was completed the scrubber system was disassembled and staged for
pickup by a commercial carrier. The activated carbon samples that were placed in the duct prior
to the scrubber operation were removed and placed into sample containers. One of the samples
was sent to a commercial lab for analysis for contaminants such as heavy metals that would affect
the carbon regeneration process. The other two samples were extra or backup samples. Once
the analysis was complete TIGG and the state of Pennsylvania reviewed the analytical results and
determined that the carbon could be regenerated. This analysis and acceptance process took
about six weeks and equipment was then picked up on January 30, 2014. The activated carbon
vessels were delivered to Siemens (Darlington, PA) to regenerate the activated carbon. The
activated carbon was removed from the vessels and placed into a rotary kiln reactivation furnace
where the carbon was heated to 927 °C. Any adsorbed MB was volatized and the off-gas from
the kiln passed through an afterburner to further destroy the organic portion and pyrolize any
remaining volatile bromines. Then the flue gas was sent through a scrubber to remove any
halogenated compounds.
3.8 Dispersion Modeling and Results
The Quick Urban & Industrial Complex (QUIC) Dispersion Modeling System is a fast response
urban dispersion model that runs on a laptop. QUIC will account for the effects of buildings in an
approximate way and provide more realism than non-building aware dispersion models. The
QUIC model was used with inputs of MB leakage rate of 1 mg/l/hr (Table 6) as the source
strength, which calculates to 357 mg/s, and an array of metrological (MET) conditions (Table 14,
Scenario 1) that closely matched the conditions seen during leak testing (Figure 31). As a result
of best-fit dimensions, model domain and wind field grid sizes were fixed at 500x500x75 m.
70
-------
Neighboring infrastructure was limited to two student dormitories, approximately 20 m
northwest of the fumigated house. House positioning and dimensions were modeled using Light
Detection and Ranging (LiDAR) data retrieved from the United States Geological Service (USGS).
Vegetative canopies and attenuation coefficients were not used due to sparse biomass. In light
of the dilute nature of MB leaving the house and the very slight variations in terrain surrounding
the house, elevation was not considered.
Table 14. QUIC Input Parameters
Scenario 1
Scenario 2
Wind speed: .44 m/s (calm)
Wind speed: 4 m/s
Wind angle: 328 0
Wind angle: 125 °
Release type: Continuous
Release type: Continuous
Source strength: 357 mg/s
Source strength: 357 mg/s
Source geometry: 22x14x4 (m)
Source geometry: 22x14x4 (m)
Sample height: 1.5 (m)
Sample height: 1.5 (m)
3.8.1 Scenario 1
MET conditions for Scenario 1 were retrieved from the Florida Automated Weather Network
(FLREC) on 12/11/13 at 0400 hrs. Though the observed MET conditions were abnormal for this
area, they allowed for a direct comparison of measurements (see Figure 31) taken at 0414 hrs on
December 11th 2013, a period of very calm winds. Scenario 1 showed a small quantity of MB
(ranging from 0.86 to 3.45 ppm) being emitted from the house. The horizontal dimension of the
plume was seen to increase around the house and quickly dilute further downwind. Model plots
show the gas diffusing over an area approximately 30 m southeast of the house. As a result of
calm wind conditions, dilution affects were minimal. Near worst case concentration levels of MB,
therefore, were predicted near the house. Model results in Figure 40 were correlated with air
samples shown in Figures 31. It should be noted that even under these near worst case
conditions the MB concentrations at the 30 feet standoff perimeter were less than 1 ppm, less
than the OEL.
71
-------
PPM
a. iMni : niS -VS»«™ 1W3 2011 UTMZon* 1TM
' *twm Vrt'cHof
C-9J1 on 0 4 9 16 24 32
(Meters
Figure 40. QUIC Model MB Concentration Results near the House during Fumigation
3.8.2 Scenario 2
MET conditions for Scenario 2 were derived by averaging weather observations retrieved from
the Fort Lauderdale-Hollywood International Airport (AWS ID; 747830) between 12-9-13 - 12-
11-13. The derived MET conditions were characteristic of those seen over the duration of the
study. Scenario 2 results (Figure 41) show a dilute concentration of MB extending approximately
50 m northeast of the house. Lateral spreading is more pronounced due to steady wind speed
of 4 m/s resulting in reduced MB concentrations. These measurements were correlated with
collected ambient air monitoring data. Model plots showed a significant amount of building-
induced turbulence near the student housing to the northwest. Although concentrations of MB
were below any recommended exposure limits, in an effort to keep any exposures to MB as low
as technically feasible, the decision to evacuate the inhabitants of these buildings was supported
by the QUIC analysis using the prevailing winds in that location. The higher wind speeds of
Scenario 2 increased dilution, resulting in lower MB concentrations in ambient air as compared
to Scenario 1.
72
-------
Bj cogs
Clbdor Per meter
Figure 41. QUIC Scenario 2 Results
3.8.3 Modeling Discussion
Overall. QUIC was effective in predicting the dispersion of MB during fumigation activities as a
function of gas emission levels from the house. Based on an emission rate of 360 mg/s, model
predictions were commensurate with observed leak and ambient air monitoring data (Figure 31
and Figures in Appendix B). The resulting plots indicate a continuous release MB is expected to
remain hazardous only very near the house if you were to stay there for prolonged periods of
time during the fumigation. The following should be noted:
• Results from Scenario 1 were correlated with MB leak-detection readings taken on 12-11-
13 at 0414 hrs (Figure 31).
• Results from Scenario 2 were correlated with ambient air monitoring results which were
typically below the action level of MB during the entire fumigation.
• Although concentrations of MB were below the action level for scenario 2, pre-fumigation
evacuation of inhabitants in nearby structures was a prudent precaution, eliminating all
inhabitant MB exposure.
• QUIC successfully accounted for the turbulence phenomena of nearby buildings.
• Both scenarios support the use of the 20 to 30 feet caution perimeter CRZ.
The purpose of this modeling was to better understand the atmospheric dispersion of leaking MB
as a result of fumigation activities. While no dispersion model can perfectly predict a real-world
73
-------
outcome, the behavior of gaseous agents, such as MB, can generally be estimated. Based on the
results of this modeling, QUIC can be considered a feasible planning tool for both small- and
large-scale fumigation activities and may prove to be most beneficial in a building-rich urban
setting.
3.9 House Entries
Although no entry was anticipated, two unexpected events occurred during the fumigation that
required entry into the test house when concentrations were above 5 ppm. First, the MB
fumigant shoot line burst early during the fumigation (fewer than 200 of the planned 700 lbs of
MB needed to achieve the target concentration was released before the line burst), requiring
two trained fumigation contractors to enter the house for 15 minutes while wearing SCBAs (EPA
personnel remained ready with SCBAs as backup/rescue) to repair the shoot line. A second 30-
minute entry into the test house occurred post fumigation, as MB levels were continuing to
decrease by natural ventilation following the activated-carbon scrubbing, in order to retrieve the
test coupons in a timely fashion (before tenting and fumigation crews entered the house).
EPA personnel wore personal breathing zone samplers while conducting certain tasks identified
as having potential for elevated MB exposure. Site SOs collected these samplers and sent them
for analysis. MB was not detected above the quantification limit on any personal samples, and
exposure was determined to be below applicable OELs during these tasks.
Following final indoor aeration to achieve below lppm of MB (12 hours after initial passive
aeration), the tarps, sand snakes, and clamps were removed from the house. All electronics and
appliances were found to be operating normally. According to UF personnel, a transient residual
odor common to MB fumigations lingered in the house for about four days.
4 Conclusions and Recommendations:
Fumigation with methyl bromide gas was effective for the inactivation of Bacillus spores when
used at the following conditions:
• MB concentration of 212 mg/l
• Temperature of 27 °C
• Relative humidity of 75%
From the laboratory data and the time-series results of this study, a fumigation time of 36 hours
is recommended. As a result of this study, we now have another tool in the Ba decontamination
toolbox that increases our capacity to respond to a Ba release, especially in the case where high-
value and/or corrosion-sensitive items are involved.
4.1 Objective 1, Conclusion
A QAPP was developed by a group of research and field professionals priorto the December 2013
field test. The QAPP included a detailed RAP, AAMP, and HASP. The SAP was incorporated into
74
-------
the QAPP and was not a stand-alone document because, for this test, most of the sampling was
covered by using Bis designed for this specific test. The documents were finalized and signed by
EPA management in October prior to the December operational test. The HASP and AAMP were
revised (Attachment 1 and 2, respectively), for future incident use, after the test to include key
insights and lessons learned from the test.
4.2 Objective 2, Conclusion
The operational fumigation was conducted safely during the second week in December 2013.
The activated carbon scrubber was set up December 6th-7th and connected to the house and tarps
on the December 8th as the house was being tented. Humidification, heating, analytical and
fumigation equipment were set up and biological indicator coupons were distributed throughout
the house on December 9th. Fumigation began that afternoon and MB concentration inside the
house reached the target at 2100 hours that day. MB concentration, temperature, and RH were
monitored and maintained inside the house throughout the 48-hour fumigation.
4.3 Objective 3, Conclusion
Test coupons, 87 glass slides and 87 wood discs, inoculated with approximately 1 x 106 CFU per
coupon non-pathogenic B. anthracis (Sterne) and placed in sterilized Tyvek® envelops were
placed in 22 separate locations throughout the house prior to fumigation. Negative procedural
blanks were included at each location. Positive and negative controls, 24 of each for both glass
and wood were also taken to the site but did not go through the fumigation process. The
evaluation of efficacy of the fumigation as measured by the deployment of coupons was
successful. All the test coupons that went through the fumigation process were non-detect for
spores.
4.4 Objective 4, Conclusion
An activated carbon scrubber system was connected to the house, used at the conclusion of the
48-hour fumigation, and monitored for breakthrough. The scrubber was effectively deployed
and used to reduce the concentration of MB inside the house from approximately 55,000 ppm
to below 150 ppm in 3.5 hours. Breakthrough was monitored and obtained for both carbon beds
set up in series. Results are available for future scrubber design(s).
4.5 Objective 5, Conclusion
Ambient air monitoring was achieved by placing photoionization monitors at six stationary
locations around the house. In addition, hand-held monitors with the same technology were
used to leak test the tenting materials and to provide monitoring for locations not covered by
the six stationary monitors. The monitors were effective for MB monitoring and provided a
successful health protection measure for not only the on-site workers, but for offsite persons, as
well. MB monitors detected small leaks near the tented house which directed leak reduction
measures to be deployed, as needed. During leak detection, the MB concentration was observed
75
-------
to dissipate quickly as the hand-held monitors were moved out and away from the tented house.
Leak-reduction techniques abated leaks quickly after they were detected by air monitors.
4.6 Recommendations
Based on the lessons learned (Appendix A) during this fumigation field test, there were several
recommendations that are listed here that have been incorporated into the revised HASP and
AMMP.
• Coordinate communications among all on-site personnel. The performance of this
fumigation test required multiple disciplines and several different teams (e.g., tarping,
fumigation, exposure monitoring, scrubbing, and health and safety). Each team had
different responsibilities, standard protocols and discipline-specific terminologies that
needed to be understood across the disciplines in order for the operation to run
successfully. Site management must construct the bridges between the multiple-
disciplinary teams to assure that communication among teams is clear and accurate.
• In order to obtain and maintain the high humidity required during fumigation, as was
required in this study, it may be important to: turn off the existing HVAC system in the
house and add humidity to precondition the contents, prior to fumigation. Also, if there
are large amounts of "dry" contents, some of the contents may need to be removed prior
to fumigation so that the efficacious humidity can be reached and maintained.
• There are times when purchasing or leasing the correct equipment for the job is more
advantageous than "making something work"; instead of using an extension to the end
of the shooting hose, try to acquire longer reinforced high pressure shooting hoses since
they can be made to custom lengths with no extension needed (e.g., A 40-foot hose would
have cost approximately $450.00; and would have been well worth the cost).
• MB may have leaked at a higher rate during and directly after gas was added to the house;
this would be caused by the increased pressure resulting from this volume of gas being
added to the system. When adding MB into the house, therefore, consider removing a
similar volume of air out of the house through the carbon scrubber (low flow rates should
be used to reduce stresses on tent seams). This should be helpful in reducing subsequent
leakage caused by added pressure inside the house.
• It is important that all duct work connections should both go together and seal easily. To
ensure this, require in the procurement specifications, that all duct-work be pre-fit from
the manufacturer/leasing company prior to shipment.
• To achieve better containment of the MB, use quality leak-resistant valves between the
house and the scrubber duct. The simple sheet-metal blast gate used during the
fumigation test allowed MB to leak out of the house into the scrubber duct. The use of a
sealing gate- or ball-valve would help contain fumigant.
• Large scrubber vessels require using a more-expensive, heavy-capacity fork-lift. Smaller
scrubber vessels placed in series could be used during future fumigation efforts, reducing
the fork-lift requirements.
76
-------
• There was a potential hazard to workers performing tasks near the tent material when
the scrubber blower was turned on (tent material is suddenly pulled in by the blowers),
First, open the make-up air gate, then open the blast gate to the scrubber while
simultaneously tuning on the scrubber blower (the gate for make-up air must be open
before or at the same time as the blower is powered on).
• Although industry standards recommend loose fitting clothing when working around
liquid MB, responders should be prepared for Level "A" entries with fully encapsulated
suits rated for protection against MB, as prescribed by the site Industrial Hygienist / SO
or Incident Commander (IC) in the event that an entry would be necessary when
fumigation was ongoing. Both an entry team, and a rescue team are required.
• As seen in Figure 42 below, grass kill was the result of constant MB diffusion through the
ground-seal apron. Some diffusion plume trails are seen as brown grass extending
beyond the ground seal apron. Similar diffusion must have also occurred through the
entire raised surface resulting in loss of some 200 lbs of MB. The new aluminum-layered
Insul-tarp® planned for a future fumigation should reduce diffusive loss greatly. As noted
in the report, the other important loss of MB was related to ducting connections from the
house to the scrubbers and increased pressure from adding gas to the enclosure.
Figure 42. Grass Kill as a Result of MB Diffusion Through the Ground-seal Apron.
• As with any activity involving potential exposure to hazardous agents, personnel who
enter the EZ or CRZ must be included in an occupational medical surveillance program.
Baseline and post exposure bromide biological monitoring should be considered as
advised by the employees' occupational health physician.
• Identify MB-specific monitors to utilize in addition to the non-specific PID. As an example:
Develop MB key for Single Point Monitors.
77
-------
• A method to plug monitoring lines was needed when they were not attached to the
fumiscope, as the test team discovered MB leaking back out of the lines. These
polyethylene lines could be plugged with a cap or similar objects.
• Planning for demobilization after performing a field study is as important as planning for
the test, itself; a person familiar with International Air Transport Association shipping
guidelines needs to assist with labeling, shipping, and coordinating the return of: unused
gases, scrubber vessels, etc.
• Pre-2005 MB was purchased for this test. MB purchased for use under current
exemptions is much less expensive. During a response to a national incident an
emergency exemption would be requested and, if granted, would allow the procurement
of MB at current market rates. To control collateral damage due to corrosion, MB without
chloropicrin should be used. In addition, research looking at the recovery of MB from
activated carbon for reuse is recommended.
5 References
American Conference of Governmental Industrial Hygienists, Threshold Limit Values and
Biological Exposure Indices for 2014. American Conference of Governmental Industrial
Hygienists. Cincinnati, OH.
Calfee, M.W., Choi, Y., Rogers, J., Kelly, T., Willenberg, Z. and Riggs, K. (2011) Lab-Scale
Assessment to Support Remediation of Outdoor Surfaces Contaminated with Bacillus anthracis
Spores. Journal of Bioterrorism and Biodefense 2, 1-8.
CDC 2012, "Surface sampling procedures for Bacillus anthracis spores from smooth,
non-porous surfaces", Revised April 26, 2012.
http://www.cdc.gov/niosh/topics/emres/surface-sampling-bacillus-anthracis.html).
CDC, EPA (2012), "Interim Clearance Strategy for Environments Contaminated with Bacillus
anthracis -DRAFT" 2012.
Corsi, R. L., Walker, M. B., Liljestrand, H. M., Hubbard, H. F., & Poppendieck, D. G. (2007).
Methyl bromide as a building disinfectant: interaction with indoor materials and resulting
byproduct formation. Journal of the Air & Waste Management Association, 57(5), 576-585.
DHHS, CDC, NIOSH, National Institute of Occupational Safety and Health, Pocket Guide to
Chemical Hazards 2012
GAO-06-756T (2006) "Anthrax: Federal Agencies Have Taken Some Steps to Validate Sampling
Methods and to Develop a Next-Generation Anthrax Vaccine", May 9, 2006.
Hezemans-Boer M, Toonstra J, Meulenbelt J, Zwaveling JH, Sangster B, van Vloten WA. (1988)
Skin lesions due to exposure to methyl bromide. Arch Dermatol. 1988 Jun;124 (6): 917-921
IBRD (2008) Task 1 Systems Analysis Report, A joint report from Sandia National Laboratory and
Lawrence Livermore National Laboratory. February 2008.
78
-------
International Agency for Research on Cancer (IARC). IARC Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals to Humans: Some Halogenated Hydrocarbons and Pesticide
Exposures. Volume 41. World Health Organization, Lyon. 1986.
Jordi AU (1953) Absorption of methyl bromide through the intact skin: A report of one fatal and
two non-fatal cases. J Aviation Med 24, 536-539
Juergensmeyer, M.A.; Gingras, B.A.; Scheffrahn, R.H.; Weinberg, M.J., Methyl Bromide
Fumigant Lethal to Bacillus anthracis Spores. J. Environ. Health 2007, 69, 24-26, 46, 50.
National Toxicology Program. Toxicology and Carcinogenesis Studies of Methyl Bromide (CAS
No. 74-83-9) in B6C3F1 Mice (Inhalation Studies). Technical Report No. TR-385. 1992.
Fumiscope 5.1 manual, Pg 2.
Dr. Rudolf Scheffrahn, UF Professor of Entomology, personal communication email on 2/6/14.
Dr. Shannon Serre, Office of Research and Development, EPA (Point of contact for an
unpublished U.S. EPA study that examined the impact of MB on historical materials).
U.S. Environmental Protection Agency (EPA) (1088) Health Effects Assessment for
Bromomethane. EPA/600/8-88/022. Environmental Criteria and Assessment Office, Office of
Health and Environmental Assessment, Office of Research and Development, Cincinnati, OH.
1988.
U.S. EPA, "Systematic Investigation of Liquid and Fumigant Decontamination Efficacy against
Biological Agents Deposited on Test Coupons of Common Indoor Materials", EPA 600-R-076,
August 2011.
U.S. EPA, "Compatibility of Material and Electronic Equipment with Methyl Bromide and
Chlorine Dioxide Fumigation", EPA 600-R-12-664, October 2012.
U.S. EPA, "Material Effects of Fumigants on Irreplaceable Objects, Short- and Long-term
Effects", EPA 600-R-13-216, September 2013.
U.S. EPA, "Bio-Response Operational Testing and Evaluation (BOTE) Project, Phase 1:
Decontamination Assessment" EPA-600-R-13-168, 2013.
U.S. EPA CAA, Clean Air Act http://www.epa.gov/ozone/mbr/.
U.S. EPA "Method 1 - Sample and Velocity Traverses for Stationary Sources",
http://www.epa.gov/ttnemc01/promgate/m-01.pdf.
U.S. EPA "Methyl Bromide Decontamination of Indoor and Outdoor Materials Contaminated
with Bacillus anthracis Spores." EPA/600/R-14/170, 2014.
U.S. EPA OPP, Report of Food Quality Protection Act (FQPA) Tolerance Reassessment and Risk
Management Decision (TRED) for Methyl Bromide, and Reregistration Eligibility Decision (RED)
for Methyl Bromide's Commodity Uses. Case No. 0355.
79
-------
USDOL OSHA 29 CFR, 1910.1000, Table Z, Permissible Exposure Limits, Occupational Safety and
Health Administration.
Weinberg, M.F., R.H. Scheffrahn, and M.A. Juergensmeyer, PART 1: Efficacy of Methyl Bromide
Gas against Bacillus anthracis and Allied Bacterial Spores in Final Report: Whole-Structure
Decontamination of Bacillus Spores by Methyl Bromide Fumigation, U.S. Environmental
Protection Agency, Small Business Innovation Research Phase II. 2004.
Weinberg, M.J. and R.H. Scheffrahn, PART 2: Whole-Structure Decontamination of Bacterial
Spores by Methyl Bromide Fumigation in Final Report: Whole-Structure Decontamination of
Bacillus Spores by Methyl Bromide Fumigation, U.S. Environmental Protection Agency, Small
Business Innovation Research Phase II. 2004b.
Yamomoto O, Hori H, Tanaka I, Asahi M, Koga M. Experimental exposure of rat skin to methyl
bromide: a toxicokinetic and histopathological study. Arch Toxicol. 2000 Feb; 73 (12): 641-8
80
-------
Appendix A. Overall Operation of the Project: Lessons Learned
At the conclusion of the project, the test team members met to capture all important lessons
learned during the project. The goal was to document the onsite observations and identify areas
of potential improvement. The following bullets capture those important findings:
• Coordinate communications among all on-site personnel. The performance of this
fumigation test required multiple disciplines and several different teams (e.g., tarping,
fumigation, exposure monitoring, scrubbing, and health and safety). Each team had
different responsibilities, standard protocols and discipline-specific terminologies that
needed to be understood across the disciplines in order for the operation to run
successfully. Site management must construct the bridges between the multiple-
disciplinary teams to assure that communication among teams is clear and accurate.
• Initial collection of temporal progression coupons should be initiated earlier during the
fumigation process. By the time we collected the first set of temporal coupons, 16 hours
into the fumigation, the efficacy of the fumigation process was almost complete.
Valuable data was missed by not taking temporal coupons earlier in the fumigation.
• Though it did not happen during this project, there was a concern that the humidifiers
might run out of water during the fumigation. In order to obtain and maintain the high
humidity required during fumigation, as was required in this study, it may be important
to: turn off the existing HVAC system in the house and add humidity to precondition the
contents, prior to fumigation. If there are large amounts of "dry" contents, some of the
contents may need to be removed prior to fumigation so that the efficacious humidity
can be reached and maintained. Additionally, a refill system needs to be devised for the
humidifiers or larger reservoirs of water could be developed for the humidifiers used.
Tenting and Shooting
• Pre-2005 MB was purchased for this test. MB purchased for use under current
exemptions is much less expensive. During a response to a national incident an
emergency exemption would be requested, and if granted, would allow the procurement
of MB at current market rates. To control collateral damage due to corrosion, MB without
chloropicrin should be used.
• Heat exchanger operation:
o During the initial injection of MB into the house it was determined that the heat
exchanger inlet/outlet ports were incorrectly plumbed. Ensure that the temp
gauge and inlet/outlet ports are installed correctly prior to initiation of MB
shooting.
o Test the heating system ahead of time to ensure proper operation.
o In order to reduce the time needed to get to the target MB concentration, multiple
heat exchangers could be used.
81
-------
o It was difficult to monitor the propane heater flame during operation. If propane
heat exchangers are used during future tests, placing a mirror underneath the unit
will allow one to more easily monitor the propane flame from a standing positon.
o Since liquid MB can damage flooring and other objects, plan for placing protective
material(s) under the shooting lines to reduce damage in the event that a
malfunction occurs.
o A shoot line extension was used to reach the release point inside the house. That
extension failed early in the fumigation process. There are times when purchasing
or leasing the correct equipment for the job is more advantageous than "making
something work"; instead of using an extension to the end of the shooting hose,
try to acquire longer reinforced high pressure shooting hoses since they can be
made to custom lengths with no extension needed (e.g., A 40-foot hose would
have cost approximately $450.00; and would have been well worth the cost).
• It appears that the humidity is affected by the injection of additional MB gas into the
house. This is due to the lack of water vapor in the MB gas which is displacing and
warming humidified air. When injecting MB into the house, it is important to closely
monitor the humidity and add humidity as needed.
• Most of the MB leaks seemed to occur directly after gas was added to the house; this
could be due to the increased pressure resulting from this volume of gas being added to
the system. When adding MB into the house, therefore, consider removing a similar
volume of air out of the house through the carbon scrubber to balance the pressure. This
should reduce subsequent leakage caused by added pressure inside the house.
• During tenting the placement of shoot lines and extraction points for the temporal
coupons was located and re-located. To avoid re-establishing tenting seals, establish
shooting locations ahead of tenting the house.
• If the truck that contains the cylinder scale cannot be moved adjacent to the house being
fumigated (these truck usually contains an onboard scale for measuring fumigant used),
a large heavy-duty digital floor scale can be used to measure loss of weight in MB cylinders
during fumigant introduction.
Scrubbing & Aeration
• Scrubber duct work connecting house, carbon beds, blower, and stack did not easily fit
together. It is important that all duct work connections should both go together and seal
easily. To ensure this, require in the procurement specifications, that all duct-work be
pre-fit from the manufacturer/leasing company prior to shipment.
• The simple sheet-metal blast gate used during the fumigation test allowed MB to leak out
of the house into the scrubber duct. To achieve better containment of the MB, use quality
leak-resistant valves between the house and the scrubber duct.
• Large scrubber vessels require using a more-expensive, heavy-capacity fork-lift; while
smaller scrubber vessels placed in series could be used during future fumigation efforts.
82
-------
• Scrubber system set up:
o Large-diameter flexible ducting was difficult to install because it was heavy. For
ease of setup, recommend using a smaller ducting and connections.
o Caulking was very effective at stopping leaks; however, required a lot of caulk that
did not dry for several days. Recommend using better fitting connections.
o Galvanized metal should be smooth to allow duct to slide over with little effort.
Welded joints caused problems with installing the rubber flexible ducting.
Recommend using easy to connect ducting with tapered joints to allow the duct
to slide on easily.
o The galvanized band clamps were difficult to use. Recommend using ratchet
clamps in place of the band clamps.
• To reduce air-flow losses and reduce stress at joint connections, recommend that the
blower and scrubber vessels be setup as close as possible in a right triangle so that the
inlet and outlet ports align. Use as much as possible: short distances, straight runs, and
smooth duct to connect the scrubber components.
• Due to limited resources, the scrubber exhaust stack was not monitored continuously for
breakthrough. Add resources to monitor the stack continuously or take grab samples
using a Suma canister and analyze these samples later.
• There were some leaks noted around tent penetration including the makeup air inlet port.
Need a better connection around the tent penetration. Molding clay could be used to
seal ports passing under the tarp and other potential leaks.
• There is a potential hazard to workers performing tasks near the tent material when the
scrubber blower is initially started. The tent material is suddenly pulled in as the scrubber
blower is activated. To reduce this hazard the scrubber start up sequence should be: first
open the makeup air-inlet port, then turn on the scrubber blower and at the same time
open the blast gate to the scrubber (the port for make-up air must be open before or at
the same time as the blower is powered on).
• To increase the effectiveness of the carbon scrubber, reduce the scrubber blower flow
rate, this allows MB to adsorb more effectively onto the carbon.
• EVOH tarp material is very soft and several tears were found where the clips pinched a
hole. Recommend using soft tip clips on all connections, including for the outer tarp
because soft tip clips were used for the inner tarp but hard tips for outer tarp may have
caused this damage.
• Scrubber vessels cannot be shipped back to the vendor immediately after project. Carbon
samples must first be analyzed to assure hazardous materials are not trapped on the
carbon. Add time to project schedule for carbon samples to be shipped, analyzed, and
the system authorized for return before the scrubber system can be picked up.
83
-------
Safety
• Need to plan for proper PPE when entering a house with MB present.
o Site safety plan noted no entries would occur in the house until ambient
concentrations were 5 ppm or less. On two occasions, it was necessary to enter
at higher concentrations: 1) distribution line break, 2) retrieval of coupons prior
to removing of tarp.
o Future plans should address unanticipated entry.
o The Tychem QC hooded suits on site have not been tested against MB. An
appropriate number of MB Level-A suits should be on site prior to fumigation.
o Although industry standards recommend loose fitting clothing when working
around MB, responders should be prepared for Level "A" entries with fully
encapsulated suits rated for protection against MB, as prescribed by the site
Industrial Hygienist / Safety Officer or Incident Commander (IC) in the event that
an entry would be necessary when fumigation was ongoing. Both an entry team,
and a rescue team are required.
• As with any activity involving potential exposure to hazardous agents, personnel who
enter the EZ or CRZ must be included in an occupational medical surveillance program.
Baseline and post exposure bromide biological monitoring should be considered as
advised by the employees' occupational health physician.
Perimeter Air Monitoring
• Area RAE monitoring stations were effective in detecting low level MB concentrations.
As leaks occurred, the AreaRAE units often detected small temporary increases that
correlated to a leak on that side of the tent.
• Handheld PID units were effective for pinpointing leaks along the tent. Concentrations
often increased rapidly several inches from the tent (versus a slow steady increase as
you got near the tent material).
• The RDA Fumiscopes and VIPER were not synchronized together, this should be done.
• The ambient air temperature and RH readings (or other MET data) with ambient air
monitoring station was not connected to the VIPER network. Connect MET data to
VIPER system for central collection of all data.
• The PIDs used at this site are not MB specific. Thus, it is not known if the PID response
is from MB or other VOCs. Identify MB-specific monitors to utilize in addition to the
non-specific PID. As an example: Develop MB key for Single Point Monitors.
• It was difficult to match up event notes and observations with VIPER data. Need a way
to add written observation, comments, and event markers to VIPER database file along
84
-------
with AAMP data - time stamped. Be systematic about capturing all observational data
for correlation with instrument response.
• Some leak detection and routine collection of hand-held monitoring data was not
recorded. Need to have a schedule for leak detection testing and standard form so
uniform results may be recorded by different teams.
• There were ambient air monitoring false positives do to "hot box" and high humidity.
Work on the monitoring system to reduce and illuminate false positives.
Interior Monitoring
• A method to plug/cap the monitoring lines was needed when they were not attached
to the fumiscope, as the test team discovered, MB leaking back out of the uncapped
lines. These polyethylene lines could be plugged with a cap or similar object
(painters tap is not an effective plug).
• Collect air samples inside the house to see if any breakdown chemicals from foams and
rubbers can be identified.
• Need a minimum of two portable monitoring devices and one back up unit.
• Setup a monitoring schedule, when samples should be taken and recorded.
• Monitoring station needs to have sufficient lighting and seating.
• Instruments need to be calibrated per manufacturer's recommendations and
documentation kept with the instruments.
Other
• Approximately one medium dumpster (5 yards) of waste was produced during the entire
project. Most of waste was used food and drink containers from on-site personnel. Other
waste included; plastic wrapping, cardboard tubing, and plastic monitoring lines.
• Pallet strapping tools are needed for shipping scrubber parts back to the rental company.
• Planning for demobilization after performing a field study is as important as planning for
the test itself; a person familiar with International Air Transport Association shipping
guidelines needs to assist with labeling, shipping, and coordinating the return of: unused
gases, scrubber vessels, and activated carbon samples from the scrubber.
• Compressed gas cylinder pick-up should be arranged in advance and may be difficult to
ship from remote locations via common carrier.
85
-------
Appendix B. Ambient Air Monitoring Figures
10
9
8
7
I
6 i
3
2 I
I
0|.
1
MB Gas Intermittently
Released
1
M
~ 1
1
1
1
1
1
I
A A A A A A /•
^ ^o> ^ ^o> ^ ^
^ ^
/i /i A A > ^o> <$> ^o> <$>,
>. ^&.\P. <9 ^o.^O. v.,v. v., V.
'1S> '^p '% '*& S? -15> '^y "^o '~*o '*V '% -v&
•VOC Sensor Readings PPM
Figure B-1. VOC Data from Location 101 LINC 78,12/9/13
10
9
8
0 -¦-
Additional MB Gas
Released
W
I
nmi
Mil'
.j-
pi
A A A A A A A A A A A A A A A A A A A A A A A A
•'a, ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ *0^ ^ ^ ^ ^ ^ ^ ^ ^ ^
^o> ^ /^> ^o> ^ ^<9 ^ ^ ^o> ^o> ^o>,
O. f. *?. d> v^V. 5L S. A cP. sP^ yj? ^ 'i) ^ ^ ^ ^ ^ ^i> ^>
•% •*> % <9 •* %%%'%%!%'%%%%%%%
VOC Sensor Readings PPM
Figure B- 2. VOC Data from Location 101 LINC 78,12/10/13
86
-------
¦
r
k
Additional MB Gas
Released
J
i
1
1
t
i
&
i
JI
1 i 1
I. 1
:t
i
i
1 1
1
¦¦ruinr
vi ujwa
I
[TjiPO^
''. IT1
1
1
0 *" almr^ MW WIP 1 Ml 1_
/; A A A A A A A A A A A A A A A A A A A A
%%%%%%%%%%%%%%%%%%%%%%%%
o O. J. J. o> « V. S s.n S. >. >. d> S SQ xi> <£> >£>
VOC Sensor Readings PPM
Figure B- 3. VOC Data from Location 101 LINC 78,12/11/13
10
9
8
7
6
5
4
3
2
1
0
MB Gas Intermittently
Released
£±
a a a /) a a a /) a /)
*o>. *V. ^ ^ ^7. ^>. 'V*. ^o>. ^o>.
vvvvv\ *%\ -^o%\%%%%%%%
¦VOC Sensor Readings PPM
Figure B- 4. VOC Data from Location 102 LINC 109,12/9/13
87
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
Additional MB Gas
Released
> /i /a A -A -A /i /i A A /i A A -A A A A A A A A A A
O. /j. ^j_ y^o yjp *?q ^ i ^>
°0 °J> °S O ^ ^O <9 <9 % &>\
»VOC Sensor Readings PPM
Figure B- 5. VOC Data from Location 102 LINC 109,12/10/13
1
1
1^
Additional MB Gas
Released
1
1
¦
1 l_
II
¦
tfa
i . ¦
. JLl
¦aiifc
L
TP
iiiiF-^7 m.in
ill
NL
~
y^> <£> A A A y^> y^> A y^> y^l A y^>
-V, V, -V, -V, -v, -V, -V, -V,
% % Hp % % '^> ' '% '3p *%
VOC Sensor Readings PPM
Figure B- 6. VOC Data from Location 102 LINC 109,12/11/13
88
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
U
MB Gas Intermittently
Released
£
1
^ ^ A A A A A A A A A A A /> />
S> / ^o> ^ ^ ^ ^ ^ ^
y&. vr. "->. *->. M>. M>. \9. \S>. *0. *0. v. V. V. V. V..V. S9.y '°3 '%> 'Ar "*S '*6- '*9 'Ar "%> '<» '<> -, '% -v& '<>• '"3r
V. \r.
^ ^
»VOC Sensor Readings PPM
Figure B- 7. VOC Data from Location 103 LINC 76,12/9/13
¦
k
1 N
Additional MB Gas
Released
1
_l
1
:
1 I. I
i •
i «.i i 11
-A 1?.. sf,. A dP. A A A A A A v^c\ >£> A A ^9 A i ^>
% ^ ^ ^ ^ ^ ^ v*>?> ^ ^ *0\ •<%> -%W •
¦VOC Sensor Readings PPM
Figure B- 8. VOC Data from Location 103 LINC 76,12/10/13
89
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
1^
Additional MB Gas
Released
1
Hi M L
I
t
I
J i—Li.
" "- ¦ ¦ ¦ ¦ ¦¦¦ ¦
"A "A A A A A A A A A A Ia A A A A A A A A A -A
^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^
*>, <>, *>, *>, *>, *>, *>, ~0, ~0, ~0, ~0, ~0, v<, v<, v<, v<, *>-, *>, v<, v<, v<, v<, V', •C-',
<>> <>> ^ ^ <>> £> ^Vj ^ *13 4>> 4V <^v> •£> - -j Oj *2 -Of
•VOC Sensor Readings PPM
Figure B- 9. VOC Data from Location 103 LINC 76,12/11/13
MB Gas Intermittently
Released
| I—
. I
M
1
~!
I
1
1
1
"
i > S~> •£> A A y^> /:
r. y._ X..
% '^>
^ A A A I/) A A A A A y^> A
; xP. M>. "VP. \9. *©. *©, V. V. V. V. V. V. x?.
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
S.
Additional MB Gas
Released
I iJ i I
i-i i
¦o •£> ¦£>
0^ Q' Q" <1" ^o> ^o> ^o> :V<>> I?6> :V<>> ^o> ^o> > ^ ^o> 3j> ,
Jo "^'n '/~* y> <5U ^-A ^'a *?vv ^-A A A ^ ^ ^> y£p vjp ^ ^ ^>
°o ¦* ^ ^ ^ ¦% ^ *> ^ ¦*> * VVVWVvVVWW
VOC Sensor Readings PPM
Figure B-11. VOC Data from Location 104 LINC 33,12/10/13
Additional MB Gas
Released
1
I
1
| |
1 1
| |
.ll. J
li i iii 1
L1
/) /> A /> /> /> /> /) A A A A A A A A A A A A A A A A
%%%%%%%%%%%%%%%%%%%%%%%%
O. O. A, k>. A, 5^ ?. 1^, 5L 51 -^,
-------
10
9
8
7
6
I
MB Gas Intermittently
Released
3t
2 I
1
0
1
1 1
1 ;
1 I
LlJ-mJ
LI . ' .
' o> «
^ !o> >
. o> <
V?o>
-d?p'%
i* "O* ^
> OO <
%*%
*_¦> y_5 vj v_>
,
?. v. V. V1. V. V1. ». ^ *s * *r W <%,% ^ % ¦Os%% %
^o> ^o> ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ % % %,
¦VOC Sensor Readings PPM
Figure B-14. VOC Data from Location 105 LINC 80,12/10/13
92
-------
10
Additional MB Gas
Released
1
nt Ili
H
%%%%%%%%%%%%%%%%%%%%%%%%
,^yo>^1yo>„yo> ,> ,<>> ,<>> ,<>> ,
O. O. J. x3. oV, ?. vJVj &• S. A. (iL >A A yja ^7 y^>
°o *o ^ '<£ ^ ^ ^ ^ "** ^ 3> ^ -V %%%%%%%%*
*
•VOC Sensor Reading PPM
Figure B-15. VOC Data from Location 105 LINC 80,12/11/13
10
4
3 I
2 I
U
MB Gas Intemiittently
Released
q | - »j» - * ¦¦ ¦¦
A y^> y^> A /) /> -A A •£> A A A A 7i *^> A v^> A A A A
*% ^ ^ ^ ^ ^ ^0> r^O> ^0> ^0> C^O> ^>> ^0>,
J&. ^ "<9. <9. ^O ^O. ^
'«£> "3> '*£> ^ '*$> '4? '% '%> "*> ,yV -v& *?P '*£>
¦VOC Sensor Readings PPM
Figure B-16. VOC Data from Location 106 LINC 42,12/9/13
93
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
K
Additional MB Gas
Released
fi I 1
H idllflililn* i ¦!
S "A A A "A A A A A A A A A A A A AIA A A A
<2- <2- <2- <2-- <2- <2- *<2- Q- <2- <2- <2- <2- <2- <2- <2- <2- <2- <2^ <2- *2- Q- <2- <2-
^o> ^6>
^y. /^ A, A, A, A. A A A
% ^ <%> 3p J? % <£,%%% •;&% % %>%%
¦VOC Sensor Readings PPM
Figure B-17. VOC Data from Location 106 LINC 42,12/10/13
* t k-
is
Additional MB Gas
Released
JlL
-L —A-»|
%%%%%%%%%%%
vfc <5 <
J-Jl
^y^y^y^y^y^
Oy. O. / A. v5-. c?.. 7. v£_ J. 6V. -^j, (P., A A A A A A A >A A />
¦VOC Sensor Readings PPM
Figure B-18. VOC Data from Location 106 LINC 42,12/11/13
94
-------
10
9
8
7
6
5
4
3
2
1 Aeration Begins 1
maPv
%%%%%%%%%%%%%%%%%%%%%
/ ? / 7 /> *7 / 7 <7 / 7 / 7 / 7 / 7 v / 7 x 7 / 7 -v^ ^
<9. <9.^<9. <9.^0.^0.^0. ^.. V^V^V^V. v.^v.,v.,V. V.A '*i> '*& '^> '%> '^r '*%>'t? *
VOC Sensor Readings PPM
Figure B-19. VOC Data from Location 201 LINC 78,12/11/13
^3>^3>^3>^3>^3>^3>
. %%%
VOC Sensor Readings
Figure B- 20. VOC Data from Location 201 LINC 78,12/12/13
95
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
\
I
,
Aeration Begins
| .
1 ¦> fl »
I
-A A A A A A A A A A A A A A A A
<9. ~ <9. <9*0*0*0.^0. *0.*S. *•?.*•?.. V.„V.^ V., V.^ V. x?.. ^o>., xf.. x?. '%> '^> 'Q? '<$> '°£> "*> '°f '<£ *o '*£>
VOC Sensor Readings PPM
Figure B- 21. VOC Data from Location 202 LINC 109,12/11/13
£> A A A A A A A A A A A A A A A A A
r^>
j ^ <*L *?•» . o> ? vJp %> \^> ^ "y^>
/^%%%%%%%
. .
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
Aeration Begins
*%%%%%%%%%%%%%%%%%%%%%
/> /> /> /> /> /> /> /> /> /> /> /> /? /> /> /> X > /> /> /> /> />
\9._"\9., \9. _"<9.^ V?.„ V?., V?., V?. V. V. V. V. V.^V1. V. V. V». V./o". x* ^o>. ^
3>.
¦^> '%> '*%"¦ *3? '*&> '""V '^> *£? '*£? '%* '^p ^ '*£> '*£? ^ ^
VOC Sensor Readings PPM
Figure B- 23. VOC Data from Location 203 LINC 33,12/11/13
A /) A A A A A /j A A A A A A A A A A
%%%%%%%%%%%%%%%%%%%%%%
^ ^ ^*a ^ A A A A A A A v£> A A A A A A A
•% ¦* • ^ %%%%\%%%%%%%%%^
VOC Sensor Readings PPM
*
Figure B- 24. VOC Data from Location 203 LINC 33,12/12/13
97
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
1 \
Aeration Begins
1
%%%%%%%%%%%%%%%%%%%%%%
<9. <9. <9. <9. •*& •*. x?. x?. \9. \9.
<> <*o %> %¦%%%%^%%%%%%%%%¦%%%%
VOC Sensor Readings PPM
Figure B- 25. VOC Data from Location 204 LINC 80,12/11/13
'%%%%%%%%%%%%%%%%%%%%&<&
o j s ^ o1 r1 c ^ ^ ^ /(j. ^ ^ ^ O) r
Sensor Readings PPM
Figure B- 26. VOC Data from Location 204 LINC 80,12/12/13
-VOC 5
98
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
1
1
1 \
Aeration Begins
1
1
1
1
1
1
1
A ¦<£) A A A A A A A A A A A A A A A A A
% % % % % % % % % % % % % % % % % % % % % %
% % % % % % % % % % % % % % % % % % % % % %
7 "O V ^ % 7 \f ^ ^ V ?(T ^ V v ^ 7
VOC Sensor Readings PPM
Figure B- 27. VOC Data from Location 205 LINC 76,12/11/13
%%%%%%%%%%%%%%%%%%%%%%%
J* J* J, J, J, <>'* <>', &, J, J, J, J, J, J, J, Sp v/jv s%> ^ c?, &0 J0
VOC Sensor Readings PPM
Figure B- 28. VOC Data from Location 205 LINC 76,12/12/13
99
-------
10
9
8
7
6
5
4
3
2
1
0
10
9
8
7
6
5
4
3
2
1
0
!\r
Aeration Begins
" - . .j . .
%%%%%%%%%%%%%%%%%%%%%
<9. <9.,<9. <9.^0.^0.^0.^*b..V. V., V.,V. V>,'V.>'V.,'V.
'<> -v3p '*& '^> '^o '%> '^> '¦<
O. O. O. A dP. cP.» A. A A A A A A A A A A A A A A A
% ^ ^ ^ -<9 %%%%'\%%%%%\%%%%
VOC Sensor Readings PPM
Figure B- 30. VOC Data from Location 206 LINC 42,12/12/13
100
-------
90
80
70
60
50
H 40
30
20
10
0
o l-»
o o
00000000
hJ hJ o
!V !*> o
NJ NJ
O l->
RelHum avg 2m (pet) VOC (ppm)
Figure B- 31. VOC Levels & Ambient Relative Humidity, Location 101 & 201 LINC 78,12/09/2013.
oooooooooooooo
RelHum avg 2m (pet) VOC (ppm)
Figure B- 32. VOC Levels & Ambient Relative Humidity, Location 101 & 201 LINC 78,12/10/2013.
101
-------
100
90
80
70
ai
u
1-
ai
a.
60
50
40
30
20
10
0
,Fr
If
I I
III! HIM 111
U "... ' T".l,
X
10
9
8
7
6
5
4
3
2
aiNJ(jj
oooooooooo^rrrirrrrrr.' jv r? ~ 77 '.r rr o
ooooooooooooooooooooooooo
oooooooooooooo
E
a.
a.
1
h1- 0
¦RelHum avg 2m (pet)
¦VOC(ppm)
Figure B- 33. VOC Levels & Ambient Relative Humidity, Location 101 & 201 LINC 78,12/11/2013
a>
u
1-
a>
a.
100
90
80
70
60
50
40
30
20
10
NOTE: All VOC readings were 0.0 during this
hi I l l l I l l l I l l l I l l l I l l l I l l l I l l l I l l l I
10
6
5
4
3
2
Ol-»NJlJJ.^lnCT>^IOOlDI-»l-»l-»l-»l-»l-»l-»l-»l-»l-»NJNJNJNJO
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
E
Q.
Q.
¦RelHum avg 2m (pet)
¦VOC (ppm)
Figure B- 34. VOC Levels & Ambient Relative Humidity, Location 101 & 201 LINC 78,12/12/2013
102
-------
90
80
70
60
50
30
20
Nju^uimviooiflp
l-» NJ NJ NJ NJ O
ID O l-» NJ (JJ
RelHum avg 2m (pet) VOC (ppm)
Figure B- 35. VOC Levels & Ambient Outdoor Relative Humidity, Location 102 & 202 LINC109,12/09/2013
a>
a.
100
90
80
70
60
50
40
30
20
10
0
FT
rm
Ol—^NJUJ-P^UnCTi^JOOtX)!—>1—>1—>1—>1—>1—>1—>1—>1—>1—>NJNJN)N>0
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
10
9
5
4
2
1
>¦ 0
E
a.
a.
•RelHum avg 2m (pet)
¦VOC (ppm)
Figure B- 36. VOC Levels & Ambient Relative Humidity, Location 102 & 202 LINC 109,12/10/2013
103
-------
NO
100
90
80
70
60
50
40
30
20
10
0 '
\ ,
^ ^
1
. 1
, .
HI L.
Ill .
T1
¦ 11r;vn
10
9
8
7
6
5
4
3
2
1
0
ooooooooooi-rfTirirrry. jv
ooooooooooooooooooooooooo
oooooooooooooo
•RelHum avg 2m (pet)
•VOC (ppm)
Figure B- 37. VOC Levels & Ambient Relative Humidity, Location 102 & 202 LINC109,12/11/2013
NO
100
90
80
70
60
50
40
30
20
10
%
I I I
10
9
8
7
6
5
4
3
Ol-»NJlJJ.^lnCT>^IOOlDI-»l-»l-»l-»l-»l-»l-»l-»l-»l-»NJNJNJNJO
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
¦RelHum avg 2m (pet)
•VOC (ppm)
Figure B- 38. VOC Levels & Ambient Relative Humidity, Location 102 & 202 LINC 109,12/12/2013
104
-------
90
80
70
_ 60
S?
~ 50
c
ai
a 40
ai
30
20
10
~p
T
' I " ' I'
Jnn
an cn
00 ID
NJ NJ NJ NJ
O P M W
Oi->NJ(jj^Ln
bbbbbbbbbbPh'!V^,^fy., rriY^o
ooooooooooooooooooooooooo
oooooooooooooo
•RelHum avg 2m (pet)
¦VOC (ppm)
10
9
8
7
6
5
4
3
2
1
0
E
Q.
Q.
Figure B- 39. VOC Levels & Ambient Relative Humidity, Location 103-205 LINC 76,12/09/2013
100
90
80
ai
a.
70
60
50
40
30
20
10
0
It
' I ' " I " ' I ' " I '
o
NJ
UJ
¦p*
U"l
CD
00
to
NJ
NJ
NJ
NJ
o
o
o
o
o
o
o
o
o
o
o
o
i—^
NJ
UJ
-p*
U"l
CD
00
to
O
NJ
UJ
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
10
7
6
5 I.
a.
4
3
2
1
¦RelHum avg 2m (pet)
•VOC (ppm)
Figure B- 40. VOC Levels & Ambient Relative Humidity, Location 103-205 LINC 76,12/10/2013
105
-------
NO
100
90
80
70
60
50
40
30
20
10
¦ i ¦¦ ¦ i ¦
¦ i ¦¦ ¦ i ¦
10
9
8
7
6
5
4
3
^ 0
Ol—^NJUJ-P^UnCTi^JOOtX)!—>1—>1—>1—>1—>1—>1—>1—>1—>1—>NJNJN)N)0
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
•RelHum avg 2m (pet)
¦VOC (ppm)
Figure B- 41. VOC Levels & Ambient Relative Humidity, Location 103-205 LINC 76,12/11/2013
100
90
80
70
E 60
1 50
U
40
30
20
10
NOTF.: All VOC readings were 0.0 during tlii
Ol—^NJUJ-P^UnCTi^JOOtX)!—>1—>1—>1—>1—>1—>1—>1—>1—>1—>NJNJN)N)0
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
10
9
8
7
6
5
4
3
2
E
Q.
Q.
•RelHum avg 2m (pet)
¦VOC (ppm)
Figure B- 42. VOC Levels & Ambient Relative Humidity, Location 103-205 LINC 76,12/12/2013
106
-------
90
80
70
60
— 50
+¦»
c
ai
^ 40
1—>1—>1—>1—>1—>1—>1—>1—>1—>NJNJN)N)0
bobbbbbbbbP^fV^^^^^PPHPP^fV^b
ooooooooooooooooooooooooo
OOOOOOOOO00000
E
Q.
Q.
Figure B- 44. VOC Levels & Ambient Relative Humidity, Location 104-203 LINC 33,12/10/2013
107
-------
100
90
80
70
a>
u
30
10
\ ^^
\ ^
111..
JU, J
i
oooooooooo^T'^7!V^,^^'., r: ^ ^ ^ r: :r ^ o
ooooooooooooooooooooooooo
OOOOOOOOO ooooo
•RelHum avg 2m (pet)
¦VOC (ppm)
10
9
8
7
I—*1—*1—*1—*1—^NJNJNJMO
Lncn--J00UDOh^NJUJ
E
Q.
Q.
Figure B- 45. VOC Levels & Ambient Relative Humidity, Location 104-203 LINC 33,12/11/2013
NO
100
90
80
70
60
50
40
30
20
10
0
10
9
8
7
6
5
4
3
2
1
mhtth • aii \rr\r readings were 0.0 during this q
Ol—^NJUJ-P^UnCTi^JOOtX)!—>1—>1—>1—>1—>1—>1—>1—>1—>1—>NJNJN)N)0
bbbbbbbbbo°!r'!V^^^,?^??^°!r'!V^b
ooooooooooooooooooooooooo
oooooooooooooo
E
Q.
Q.
¦RelHum avg 2m (pet)
¦VOC (ppm)
Figure B- 46. VOC Levels & Ambient Relative Humidity, Location 103-205 LINC 76,12/12/2013
108
-------
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
2m Avg. Windspeed (mph) 2m Max. Temperature (°F) 2m Relative Humidity (%)
Figure B- 47. Ambient Temperature, Relative Humidity, and Windspeed, 12/09/13
2m Avg. Windspeed (mph) 2m Max. Temperature (°F) 2m Relative Humidity (%)
Figure B- 48. Ambient Temperature, Relative Humidity, and Windspeed, 12/10/13
109
-------
90
80
70
60
50
40
30
20
10
0
.00
90
80
70
60
50
40
30
20
10
0
2m Avg. Windspeed (mph)
2m Max. Temperature (°F) 2m Relative Humidity (%)
Figure B- 49. Ambient Temperature, Relative Humidity, and Windspeed, 12/11/13
2m Avg. Windspeed (mph)
2m Max. Temperature (°F) 2m Relative Humidity (%)
Figure B- 50. Ambient Temperature, Relative Humidity, and Windspeed, 12/12/13
110
-------
Concentration of MB (mg/l) over Time (date)
Location: Classroom southwest
300
250
Classroom
southwest (green
200
150
2 per. Mov. Avg.
(Classroom
southwest (green))
100
50
NJ
NJ
NJ
NJ
NJ
NJ
NJ
Concentration of MB (mg/l) over Time (date)
Location: Womens Bath south
300
200
Womens Bath
south
100
NJ
NJ
NJ
NJ
NJ
NJ
NJ
Concentration of MB (mg/l) over Time (date)
Location: AC room north
300
250
200
AC room north
150
2 per. Mov. Avg.
(AC room north)
100
50
NJ
U3
NJ
NJ
NJ
NJ
NJ
NJ
Figure B- 51. Concentration of MB (mg/l) over Time (date), Three Manually Monitored Locations
111
-------
Concentration of MB (mg/l) over Time (date)
Location: Ducting
300
250
200
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Figure B- 52. Concentration of MB in mg/l over Time (hr), Return Ducting Location
Concentration of MB (mg/l) over Time (date)
Location: Podium
300
250
200
150
100
50
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
Figure B- 53. Concentration of MB (mg/l) over Time (hr), Classroom Podium Location.
112
-------
Attachment 1: MB Fumigation Health and Safety Plan
Attachment 2: MB Fumigation Ambient Air Monitoring Plan
Attachment 3: MB Field Operational Guidance to New York City
113
------- |