OSHAO
        United States
        Environmental Protsetion
        Agency
                 United States.
                 OccupMionBl Safety antd
                 HsaJtfe Adr*Mttwtion
EPA 560-fl-99-003
April 1999
        EPA/OSHA JOINT
        CHEMICAL
        ACCIDENT
        INVESTIGATION
        REPORT

        BPS, Inc.
        West Helena, Arkansas
EPAxnstQSHA

-------

-------
                  The EPA/OSHA Joint Accident Investigation Program

       EPA and OSHA work together under conditions detailed in a Memorandum of
Understanding (MOU) to investigate certain chemical accidents. The fundamental objective of
the Joint EPA/OSHA chemical accident investigation program is to determine and report to the
public the facts, conditions, circumstances, and causes or likely causes of any chemical accident
that results in a fatality, serious injury, substantial property damage, or serious off-site impact,
including a large scale evacuation of the general public.  The ultimate goal of the accident
investigation is to determine the root causes in order to reduce the likelihood of recurrence,
minimize the consequences associated with accidental releases, and to make chemical production,
processing, handling, and storage safer. This report is a result of a Joint EPA/OSHA investigation
to describe the accident, determine root causes and contributing factors, and identify findings and
recommendations.

       Under section 112(r)(l) of the Clean Air Act Amendments of 1990 (CAA) and under the
OSH Act of 1970, industry has a general duty to design and maintain a safe facility taking such
steps as are necessary to prevent releases, and to minimize the consequences of accidental releases
which do occur, and to provide a safe and healthy workplace for workers.  In addition, OSHA
has promulgated the Process  Safety Management Standard at 29 CFR 1910.119 for the
prevention of chemical accidents that impact workers. EPA, under section 112(r)(7) of the CAA,
has promulgated regulations for the preparation of risk management programs and plans for the
prevention of accidental chemical releases that harm the public and the environment. However,
compliance and enforcement  with these provisions are not the focus of this report but will be
addressed by EPA, OSHA or both as necessary in separate reports or actions.

       Prior to releasing an accident investigation report, OSHA and EPA must ensure that the
report contains no confidential business information. The Freedom of Information Act (FOIA),
the Trade Secrets Act, and Executive Order 12600 require federal  agencies to protect confidential
business information from public disclosure.  To meet these provisions, OSHA and EPA have
established a clearance process for accident investigation reports in which the companies who
have submitted potentially confidential information used in the report are provided a portion of
the draft report. This portion contains only the factual details related to the investigation (not the
findings, the conclusions nor  the recommendations).  Companies are asked to review this factual
portion to confirm that the draft report contains no confidential business information. As part of
this clearance process, companies often will provide to OSHA and EPA additional factual
information.  In preparing the final report, OSHA and EPA consider and evaluate any such
additional factual information for possible inclusion in the final report.

       Chemical accidents investigated by EPA Headquarters are conducted by the Chemical
Accident Investigation Team  (CAIT) located in the Chemical Emergency Preparedness and
Prevention Office (CEPPO) at 401 M Street SW, Washington, DC 20460,  202-260-8600.  More
information  about CEPPO and the CAIT  may be found at the CEPPO Homepage on the Internet
at http://www.epa.gov/ceppo. Copies of this report can be  obtained from the CEPPO Homepage

-------
or by calling the National Service Center for Environmental Publications (NSCEP) at 800-
490-9198. OSHA Headquarters are located in the US Department of Labor - OSHA, 200
Constitution Ave NW, Washington, DC, 20210, 202-219-8118. More information about OSHA
may be found at the OSHA Homepage on the Internet at http://www.osha.gov.

Chemical Safety and Hazard Investigation Board (CSB)

      In the 1990 Clean Air Act Amendments, Congress created the Chemical Safety and
Hazard Investigation Board (CSB). Modeled after the National Transportation Safety Board
(NTSB), the CSB was directed by Congress to conduct investigations and report to the public the
findings regarding the causes of chemical accidents. Congress authorized funding in November
1997 and the CSB began operations in January 1998. Several investigations by the CSB are
underway. More information about CSB may be found at their Homepage on the Internet at
http://www.chemsafety.gov or http://www.csb.gov.

      EPA and OSHA plan to complete their work and issue public reports on investigations
initiated prior to funding of the CSB.  Under their existing authorities, both EPA and OSHA will
continue to have roles and responsibilities in responding to, and investigating, chemical accidents.
 The CSB, EPA, and OSHA (as well as other agencies) will be coordinating their efforts to
determine the causes of accidents and to apply lessons learned to prevent future events.

-------
                                  Executive Summary

       On May 8, 1997, at approximately 1:15 p.m., Central Daylight Time , a massive explosion
and fire occurred at Unit Two of the Bartlo Packaging Incorporated (BPS) facility located in
West Helena, Arkansas.  As a result of the explosion and fire, three West Helena firefighters were
killed.  Seventeen firefighters required medical attention due to heat exhaustion and injuries
during the response. The Unit Two structure was completely destroyed.  Hundreds of residents
and patients at a local hospital were either evacuated or sheltered-in-place.  The Mississippi river
traffic and major roads were closed for approximately twelve hours due to the release of toxic
materials from the facility.

       Prior to the explosion, BPS employees observed smoke in the Unit Two warehouse.
Following established procedures, all employees evacuated the building.  The company placed an
emergency call to local emergency response groups.  Members of the West Helena Fire
Department (WHFD) responded to the scene within minutes. A reconnaissance team composed
of four firefighters was outside of the Unit Two warehouse when an explosion occurred inside the
building. Three firefighters were fatally injured when they were struck by materials blown out of
a falling cinder block wall.  The fourth firefighter was seriously injured.

       EPA and the OSHA conducted a joint investigation of the incident.  The Joint Chemical
Accident Investigation Team (JCAIT) determined that the incident was most likely caused by the
decomposition of a bulk sack containing the pesticide Azinphos methyl (AZM) SOW which had
been placed against or close to a hot compressor discharge pipe.  Under this scenario, the heat
from the discharge pipe would have caused the pesticide material to decompose and give off
flammable vapors which resulted in the fatal explosion.

        The investigation team could not eliminate the possibility that the AZM SOW arriving at
BPS the day of the accident was already decomposing.  This alternate scenario could either be an
initiating event by itself or a factor influencing the preferred scenario. In other words, a
decomposing bag of AZM SOW could have been placed closed to the compressor discharge pipe.

       The JCAIT identified the following root causes and contributing factors of the event:

$      MicroFlow Company (MFC) and BPS did not have a full understanding of the hazards
       associated with AZM.

$      BPS did not assess the potential hazards of a hot pipe in an area where hazardous
       chemicals were to be stored when the new warehouse addition was constructed.

$      BPS did not have standard operating procedures for material storage and handling.

$      On-site information provided to the WHFD was conflicting and incomplete.
                                           in

-------
       The following recommendations were developed by the JCAIT to address the root causes
and contributing factors and to prevent recurrence of similar incidents at other facilities:

$      Manufacturers should be proactive in testing potentially hazardous materials. Testing for
       actual conditions and elevated temperatures during storage should be conducted to
       determine safe storage conditions.  Screening tests, such as Differential Scanning
       Calorimetry (DSC), can be helpful in determining the need for additional testing.
       However, thermally unstable materials which are intended to be packed and shipped in
       large volume containers should be tested beyond screening levels.

$      Facilities which store, use, handle, manufacture or move hazardous materials should
       develop and implement a system to review potential hazards of modifications to facilities,
       equipment, chemicals, technology, or procedures.  The system should analyze potential
       impacts to safety, health, and the environment and take appropriate actions before the
       modifications are implemented. OSHA=s Process Safety Management (PSM), EPA=s Risk
       Management Program (RMP), and the Center for Chemical Process Safety (CCPS)
       guidelines can help facilities develop such system.

$      Facilities that store hazardous chemicals should  develop standard operating procedures for
       material storage and handling that address storage restrictions. Such facilities should
       adhere to applicable practices outlined by CCPS and the National Fire Protection
       Association (NFPA). Pesticide facilities are encouraged to also follow NFPA 43D (Code
       for the Storage of Pesticides), specifically the non-mandatory Appendix B.

$      Facilities storing hazardous chemicals should develop an inventory management system
       with information regarding composition, compatibility, storage, location, and quantity of
       incoming products.  This management system can help the facility comply with storage
       restrictions and provide  emergency responders useful information during a response
       action.

$      EPA and OSHA, in conjunction with interested parties, should facilitate a workshop to
       make recommendations on how to improve the quality of hazardous materials information
       available during response actions.  The workshop should review appropriate uses of
       Material Safety Data Sheets by local emergency response groups and how to provide
       these groups information describing the behavior of hazardous materials when they begin
       to react or decompose and what responders should look for during a chemical emergency.
                                           IV

-------
                                   Table of Contents


1.0    Background	1

       1.1    Introduction	1

       1.2    Facility Description	1

       1.3    Chemicals in the New Warehouse Addition	4

2.0    Description of the Incident	5

       2.1    Sequence of Events	5

       2.2    Emergency Response Actions	10

             BPS Emergency Preplanning	10

             Initial Response	10

             Response Actions Under the Incident Command System	11

       2.3    Public Health and Environmental Issues	12

3.0    Investigation and Analysis	13

       3.1    Investigation	13

       3.2    Analysis	14

             3. 2. A  Overview of Explosion Scenarios	14

                    Scenario 1:    Chemical in Supersack Decomposes when Placed Close to
                                 the Compressor Header Pipe	15

                    Scenario 2.    AZM decomposition begins before arriving to BPS	23

                    Scenario 3.    Incompatible Chemicals React	25

                    Scenario 4.    Malfunctioning  Compressor Overheats Supersacks Near the
                                 After-cooler Piping	28

-------
             3. 2. B Overview of Early Emergency Response	28




                    On-site information	29




                    Risk Perception/Risk Management	31




4.0   Root Causes and Recommendations	32




      4.1    Root Causes and Contributing Factors	32




      4.2    Recommendations	33
                                         VI

-------
                                   List of Figures
                                                                              Page

1.     Floor Plan-EPS, Inc. Unit Two	2

2.     New Warehouse Addition Compressor System	3

3.     After-cooler	4

4.     Compressor Discharge Pipe After the Incident	4

5.     Supersack set-up at BPS Unit One	17

6.     Aerial Photo BPS Unit Two	18

7.     Approximate Location of Chemicals in the New Warehouse Addition	19

8.     Ventilation Fan with Yellow Residue	20

                                    Appendices

                                                                             Page
$     Material Safety Data Sheets for Maneb 75DF, Azinphos methyl SOW, and Alliette
      Signature WDG	35

$     Summary of Laboratory Results	52

$     References	71
                                         vn

-------

-------
1.0    Background

       1.1    Introduction

              On May 8, 1997, an explosion and fire occurred at the Bartlo Packaging
Incorporated (EPS) facility located in West Helena, Arkansas.  As a result of the explosion and
fire three firefighters died and seventeen other firefighters required medical attention due to heat
exhaustion and minor injuries.  Hundreds of residents, including local hospital patients, were
evacuated or sheltered in place due to the threat of exposure to toxic chemicals released in the
blast. Major roads were closed and the Mississippi river traffic halted. Several emergency
response groups participated in the response action.  It took approximately two weeks to
extinguish the fire.

       The Environmental Protection Agency (EPA) and the Occupational Safety and Health
Administration (OSHA) conducted a joint investigation of this event in accordance with a
Memorandum of Understanding (MOU) signed in November 1996.  The agencies established a
joint chemical accident investigation team (JCAIT) made up of personnel from the EPA and
OSHA National Offices, OSHA=s Health Response Team, and Regional and contractor personnel
from both agencies.  This report contains a description of the incident and the results of the joint
investigation.
       1.2    Facility Description

       BPS is a corporation with facilities in Helena and West Helena, Arkansas.  The West
Helena facility is located in an industrial park three miles from the central business district of West
Helena, Arkansas.  The facility is located in a flat area used primarily for agricultural purposes.
The nearest residential area is located less than one mile northeast, and the Mississippi river is
located approximately three miles east.

       BPS is an agricultural chemical packaging facility.  No chemical manufacturing occurs at
the facility.  BPS receives bulk shipments of agricultural chemicals (pesticides, insecticides, etc.)
and repackages them in  smaller, water soluble, containers.  The operation is conducted for
chemical manufacturers using tolling contracts. Under a tolling arrangement a company contracts
with another company to perform  a specific operation.  In this case, chemical manufacturers
deliver agricultural chemicals in bulk containers, which BPS repackages according to the
manufacturers: specifications. BPS then ships the product back to the specified location.

       The West Helena facility employs approximately 130 workers.  At the time of the incident
65 employees were on duty. The facility consists of two production buildings (referred as units
One and Two), two satellite buildings, and a Agel@ building.  The production buildings are
constructed of corrugated metal with steel reinforcement. The Unit Two building (Figure 1),

-------
                                                  North Office Door
Packaging
  Room
      ROOM 6
   Satellite Two
Packaging
  Room
      ROOM 7
   Satellite One
Roll-Up
' Door
                                 Shop
Breezeway
                     Roll-Up
                     ' Door
Breezeway
                         Office
                                            Laundrj
        Breakroom
  Woman's
   Change
    Room
                      60 FT
                             Compressor
                                Room
                     Slitting
                     Room
ROOM
  8
                                        Shared Southern Wall
   Men's
  Change
   Room
                                                                        LOADING
                                                                        DOCK
 Non-Haz
Waste Area
                                                    ROOM
                                                       9
                                                                         ROOM
                                                                           10
                                                                                . Roll-Up
                                                                                  Door
                                        NEW WAREHOUSE ADDITION
                                                                 East Wall
                                                                   Door
                                                                  Cinder
                                                                /Block
                                                                   Wall
                                               100 FT
                       FIGURE 1: FLOOR PLAN - EPS, INC. UNIT #2

-------
                      18'
                                             22'
                                           70'
30'
                                                                                       11'
                                    ±19'
i After
oolef
      *V
50'



,.n|,| Compressor Room
!i 1 IS 2DHP 15 HP T




8'43"
	 a 	
Silting
Room

ROOM
Fork Lift Ha|way
Open
Area



NEW WAREHOUSE ADDFION
                                                               100'
                                                               128'10"
                      FIGURE 2  NEW WAREHOUSE ADDFION COMPRESSOR SYSTEM
                                                                                                           NOT TO SCALE

-------
 in which the incident occurred, had a 100' x 150' main area, a 16' x 34' loading dock, and two 50'
x 60' satellite buildings connected to the main area by breezeways.  In October 1995, a warehouse
 addition was added to the Unit Two building.  It shared the southern wall of the original building
(referred to as the new warehouse addition north wall). The addition was approximately 7800
square feet.  It was also constructed of corrugated metal with the exception of the outside eastern
wall. This particular area had two stories with an exterior (eastern) wall constructed of cinder
blocks.
       Repackaging operations in the Unit Two building required the use of two reciprocating air
compressors.  The compressors were located in the southern portion of the original building.  The
compressors: discharge pipes went through the new warehouse additions north wall into a
common header pipe (Figure 2). This header pipe was fifteen feet long and 5'11" from the
concrete pad floor. It ran parallel to the north wall to meet an after-cooler outside the new
additions west wall.  The output from the after-cooler was piped back along the same wall 3'7"
from the concrete floor carrying the cooled air back to the accumulator tanks under each
compressor (Figures 3 and 4).
           Figure 3. After-cooler
                           Figure 4. Compressor Discharge
                            Pipe After the Incident
1.3
Chemicals in the New Warehouse Addition
    The inventory information used by emergency responders during the response action was
based mostly on BPS= management recollection. The Agency of Toxic Substances and Disease
Registry (ATSDR) developed a table during the response action based on employee interviews
(After Action Report, EPS Pesticide Fire, ERSAB, ATSDR, August 4, 1997).  Several weeks
after the incident,  BPS provided to JCAIT information regarding the type and quantities of the
chemicals stored in the Unit Two building the day of the incident.  Based on the BPS inventory

-------
information and witness statements, the JCAIT determined that the following chemicals were
present in the Unit Two new warehouse addition at the time of the incident: Maneb 75DF,
Azinphos methyl (AZM) SOW, Alliette Signature WDG, Topsin WSB, Sevin 80 WSP, and
Penncozeb 75DF. Material Safety Data Sheets (MSDSs) for Maneb 75DF, Azinphos methyl
(AZM) SOW, and Alliette Signature WDG are included in Appendix A.

2.0    Description of the Incident

       2.1    Sequence of Events

       December 1995- May 7.1997

       In December 1995, BPS provided  MicroFlow Company (MFC) a quotation to repackage
bulk AZM SOW into 1 Ib. water soluble bags. BPS was to provide warehousing for a two-week
supply of materials being repackaged and  two weeks prior to and  following repackaging.  As part
of the contract arrangement, BPS requested MFC  to do a presentation to BPS workers and
managers on safety and health issues related to worker exposure and handling of the AZM SOW.
The request was based on AZM=s toxicity.  The presentation was to be delivered prior to the
repackaging operation.

       On January 29, 1996, BPS sent a letter to MFC expressing concern about the
reactivity/flammability of AZM SOW. Their concern originated through a conversation with a
representative of Bayer Agricultural Division. Bayer noted that it  had experienced a number of
incidents involving thermal decomposition and/or fires involving Guthion (Bayer=s AZM
formulation). The letter stated that many of Bayer=s fires were initiated in ribbon blenders and
transfer screws similar to those used at BPS. BPS noted in its letter that the Material Safety Data
Sheet (MSDS) provided by MFC did not have information to support a similar situation. BPS
questioned why the MSDS provided by MFC did not contain information similar to Bayer=s
MSDS on Guthion for flammability and reactivity. MFC=s MSDS  (of January 1995) had a
Hazardous Materials Incident System (HMIS) flammability and reactivity rating of 0 compared to
Bayer=s National Fire Protection Association (NFPA) rating of 2.  BPS requested MFC=s advice
since they Ahave little experience dealing with reactive materials and depend on our customers to
inform us of any problems inherent in their materials® (letter from BPS to MFC January 29,
1996).

       MFC and BPS personnel met on February  8, 1996, to discuss the suitability of the BPS
packaging equipment and the apparent inconsistency on the AZM  SOW fire and reactivity hazards.
As a result, BPS proposed to construct a water deluge system to accommodate a potential
smoldering of the product.  The parties agreed on a system that would run water lines to the
repackaging hopper, with valves located by the packaging room=s  door. In case of a Abad odor®
while running the equipment, the operator was supposed to flood the hopper with water.

       At BPS=s request, MFC made a safety presentation on February 12, 1996 to BPS workers

-------
and managers. The presentation included product background, toxicity, safe work practices, and
fire/reactivity issues.

       On February 13, 1996, MFC sent a follow-up memo by telefax to EPS.  It states that
A...AZM SOW will begin to smolder and smoke at approximately 170 degrees Fahrenheit.  This
temperature is consistent with the 167 degrees listed on our MSDS.@ In the same memo, MFC
stated that they were in the process of locating a sample of Guthion SOW to test and that they
would update BPS with any new findings. At the time of the incident MFC had not given any
additional information to BPS.

       The MSDS for AZM SOW provided by MFC to BPS did not reference any 167° F (75° C)
temperature.  MFC used a 90% pure AZM technical grade as the AZM SOW active ingredient.
The technical grade supplier has a 158° F ( 70° C) temperature in their MSDS Aconditions to
avoid® section.

       May 7, 1997: Tifton, Georgia, MicroFlow Warehouse

       MFC had made arrangements to ship two truckloads of AZM SOW to BPS from Tifton,
Georgia on May 7, 1997 via Milan Express.  Each truckload contained 26 bulk bags (supersacks)
with approximately 1600 pounds of AZM each. These supersacks are constructed of woven
polypropylene coated fabric and have a 45 cubic foot capacity. The supersacks on both trucks
had AZM SOW from batches produced from 10/96 to 4/97.

       Prior to his arrival  to Tifton, the first truck driver picked up the truck in South Bend,
Indiana. Then, he picked up plastic lawn mower parts in Elgin, Illinois and delivered them to
Macon, Georgia. At 2:45  p.m., MFC personnel started loading AZM SOW onto the first truck.
Upon completing the loading, truck driver one left the Tifton warehouse at 3:45 p.m.

       At 5:00 p.m. MFC personnel started loading the second AZM SOW truck. Truck driver
two had not hauled pesticides before. At 6:30 p.m., the second truck loaded with AZM SOW left
Tifton, Georgia bound for BPS.

       May 7 - 8.1997: Road

       Truck driver one pulled over and rested for two hours at Wyona, Missouri. He stated that
the AZM SOW odor was making him feel sick.  He transported AZM SOW a year earlier from the
MFC plant located in Macon,  Georgia. He stated that the AZM SOW smell was similar to the
previous truck load.  The smell had made him feel sick both times, but this particular time it
Areally got him.®

       Truck driver two stopped for an eight-hour rest in route to BPS. He stated that he could
smell the cargo from outside the truck. The AZM SOW smelled bad to him but did not make him
feel sick.

-------
       May 8.1997 : BPS. West Helena. Arkansas

       Before the 10:00 a.m. work break

       Truck driver one arrived at BPS, Unit One at 7:20 a.m. He was received by a BPS
employee who directed him to Unit Two.  Once in Unit Two, truck driver one broke the truck
seal at 8:00 a.m.  The truck was not unloaded immediately because the fork lift operators were
unloading Procure empty drums. The Procure truck was unloaded by 9:55 a.m.

       From 10:00 a.m. break to lunch break (11:55 a.m.)

       BPS fork lift drivers began unloading the first truck after the 10:00 a.m. break.  They had
to move other material in the new warehouse addition (empty cardboard and drums) to make
space for the incoming AZM. According to BPS forklift drivers the cargo was located along the
new warehouse=s north wall on a two row/double stack arrangement.  They also stated that AZM
pallets were spotted approximately six inches from the north wall.

       While unloading, fork lift drivers and nearby employees noticed and made comments
about the strong odor.  They reported that the AZM in the first truck smelled worse than the
AZM in the second truck and the AZM repackaged at BPS one year earlier.

       A fork lift driver reported a spill in the new warehouse addition right after the 10:00 a.m.
break.  Twenty to thirty pounds of Alliette Signature had leaked from the top pallet of a
previously patched supersack which had reopened. The spill reportedly occurred next to the new
warehouse addition north wall, near some empty drums on the west side.  The BPS waste monitor
began to clean the spill up around 11:30 a.m. He used a forklift to move the top pallet of Alliette
and took it to the stretch wrap area.  He then took the waste to room seven for disposal.

       The second AZM truck arrived at  11:30 a.m., when the first truck had only two pallets left
to unload. BPS employees finished unloading the first truck close to  lunch time.  The first truck
pulled away from the loading dock. Another truck, reportedly carrying cardboard, pulled in and
stayed at the loading dock for approximately ten minutes.  In the meantime, forklift drivers started
stacking two rows of Aother® product to the north wall of the new warehouse addition.  After the
cardboard truck pulled out, the second truck pulled into the loading dock. Truck driver two
broke the truck seal but one of the fork lift operators told him that the unloading would begin
after the lunch break.

       Lunch break (11:55 a.m.-12:25 p.m.)

       All work activities, with the exception of the spill cleanup, stopped during the lunch break.
 The BPS waste monitor completed the Alliette Signature spill clean-up around 12:20 p.m. He
called the shift supervisor to check on the  spill clean-up.

-------
       Truck driver one was dispatched to Grenada, Missouri and left the site before the
explosion.

       After lunch break (12:45 p.m.)

       The forklift supervisor returned from lunch, then went back to the slitting room to wrap a
pallet.  He was the first person to see the smoke. He described it as Aa yellow powder puffing
through the hole® around the compressor header pipes.  He reported that the smoke (or powder)
was coming from the new warehouse addition through the hole and forming in the air, not
dropping to the floor.  He also stated that the powder had the same smell as the AZM that had
been unloaded earlier.  He did not see fire but called Afire® on the radio at what he thought was
approximately 12:50p.m.  He grabbed a fire extinguisher and went to rooms eight, nine, and ten
to get people out. He then looked back to the compressor area and saw a large cloud of what
appeared to be powder. He tried to go into the warehouse area but the powder was too dense.
Another employee was in the warehouse with an extinguisher. Neither employee used his fire
extinguisher; they left the unused fire extinguishers in the warehouse and evacuated. In the
meantime, the shift  foreman called ACode Red® and the evacuation process continued.  Most
employees reported seeing yellowish smoke.  Others reported the smoke color to be lime green.
All the employees reported seeing the smoke coming from the new warehouse addition area
where the AZM had just been placed or through the wall holes around the compressor pipes into
the slitting room. Employees also reported a rotten egg/skunk odor.

       The production manager called 911.  The West Helena Fire Department (WHFD) received
first notification at  1:02 p.m. According to the 911 call transcription, EPS reported a small
smoldering fire with no flames.  The production manager stated: Ais where some product was set
next to a hot line off an air compressor. It=s  starting a little bit of a smother, but no fire. But it=s a
lot of smoke.® The caller also referred to a 1,500 pound supersack.  A second notification, to the
Helena Fire Department,  was received  at 1:09 p.m.

       Three maintenance employees went to Unit Two after the radio fire call. All of them
reported seeing smoke coming through the holes around the compressor header pipes. They
described it as light yellow close to the roof and thick grey/tan near the floor.  One  of the
employees turned the exhaust fans on.  Reportedly, this employee thought that Aone of the
supersacks of MicroFlow=s was leaning against the pipes.® The other employee went to the
electrical panels (the electrical panel was adjacent to the compressors room on the way to the
breeze way leading to satellite one) and turned the compressors off.

       During the evacuation of Unit Two, truck driver two observed yellow Astuff® coming out
of the back of the building. One of the fork lift operators told him that there was fire close to
some pipes.  Without having unloaded any product, he closed the doors to the truck and pulled
his rig away from the loading dock, taking it across the  street.

       The shift foreman took a roll  call and one employee was missing. A fire truck arrived at

-------
1:15 p.m., just after the first roll call. The firefighters stated they thought yellow product was
coming from the building.  The WH Fire Chief arrived shortly after the fire truck.  One of the
firefighters received an MSDS from a BPS employee. He checked the Department of
Transportation (DOT) Hazardous Materials Booklet and noted that one of the products on site
was water-reactive.  The production manager discussed the products: reactivity with the WH Fire
Chief.  He gave the Fire Chief a binder with the MSDS and a floor plan. The WHFD department
called volunteers, other emergency services, and the Helena Fire Department for backup. After
consulting with the WH Fire Chief, the maintenance manager closed the three roll-up doors to the
loading dock and satellites one and two.

       Several BPS employees went to satellite one to locate the missing employee. He was
located upstairs in the reclaim area and escorted out. A second roll call took place and all
employees were accounted for.

       The WH Fire Chief and the maintenance supervisor discussed the smoke location. The
Fire Chief observed that the Asmoke® looked more like powder or product and that it was seeping
instead of puffing.  The maintenance supervisor unlocked and opened a side door on the east side
of the new warehouse addition for a firefighter, but the yellow smoke was too thick for the
firefighter to enter.

        The firefighter reported back to the WH Fire Chief. The WH Fire Chief asked the
production manager to show him the building layout to check the location of the smoldering
supersack. The WH Fire Chief then asked about the danger of an explosion and the BPS
President said there was none.

       The four firefighters walked back toward Unit One to get a lifeline.  They returned to the
Unit Two building close to the room 9 exterior wall (east wall).  A bell started to ring inside the
building, and the maintenance supervisor explained to the firefighters that the sprinkler system
alarm had just gone off. The maintenance supervisor then observed water coming from the
sprinkler alarm on the east exterior wall indicating that the sprinkler system had in fact been
activated.  At 1:34 p.m. the alarm company received a fire notification.  (The on-site activation of
the  sprinkler system sends an electronic notification simultaneously to the alarm company.) The
maintenance supervisor asked the firefighters to wait for him to turn the power off before entering
the  building because the equipment was still energized.

       The maintenance supervisor went to the exterior office door by the north side of Unit One
to attempt to disconnect the power to the building.  The disconnect power box was located in an
interior hallway between the office and the maintenance shop. He entered the office and
proceeded to the door leading to the hallway.  Suspecting fire, he felt the door and found it hot to
the  touch.  He cracked the door and observed that the shop area was full of smoke. He
determined that he could not reach the disconnect box safely and retreated.  He notified the WH
Fire Chief that he was unsuccessful in disconnecting the power to the building.

-------
       An electrical company=s service man had an appointment with a nearby facility.  He saw
the police and firemen and went directly to EPS.  He tried to get in the building by the office
door, but felt heat on the walls and decided to turn the power off from the main power cutouts
outside near the transformer. He observed yellowish dust or smoke coming out of the vents. The
main power cut consisted of three individual legs.  The service man pulled the first leg.  As he was
getting ready to pull the second leg, an explosion occurred. A firefighter reported seeing a
mushroom cloud at the east side of the building. Another firefighter reported hearing a wuff
sound Alike throwing gasoline on a fire,® at the same time he saw a massive fireball coming from
the building. The explosion caused the cinder block wall to collapse.  The four firefighters
standing east of room 9 were struck by the collapsing wall.  Three of them were killed and the
remaining one was seriously injured.

       At the time of the explosion, the WHFD received a call from the New Jersey Bartlo
Packaging chemist.  The fireman reported that the chemist asked whether the sprinkler system had
activated and explained to the firefighter that two different types of chemicals were present at the
site.  According to the firefighter, the chemist said the chemicals would explode if water was put
on them.

       At 1:39 p.m. the alarm company was notified of the explosion.

       2.2    Emergency Response Actions

             BPS Emergency Preplanning

         BPS was an active member of the Local Emergency Planning Committee (LEPC).  BPS
had a written AEmergency Response and Contingency Plan® dated September 1995.  They had
made arrangements with the WHFD for emergency support and had provided copies of their
written plan and MSDSs. BPS had also invited the fire department to tour their facility and to
participate in their emergency drills.  A West Helena firefighter stated that fire department
personnel had toured the facility approximately one month before the incident.

       According to BPS employees, the facility had several fire extinguishers but they were to
be used only on non-chemical fires.  Employees were instructed not to fight chemical fires but to
immediately evacuate the building. The Unit Two building reportedly had a fire alarm system
which was backed up with radios and intercom. Safety meetings covered evacuation routes.  A
floor plan showing the evacuation routes was posted on the wall.

             Initial Response

       Upon being called to the site, the West Helena Fire Chief called in all volunteers and off
duty personnel.  He also called the Helena Fire Department, the emergency medical services, the
State Police, and  the Phillips County Office of Emergency  Services.  This office notified the State
Office of Emergency Services, schools and radio stations in accordance with the County and
                                           10

-------
LEPC Plan.

       At the time of the explosion, the Helena Fire department had just arrived. The priority
immediately after the explosion was to rescue the injured firefighters and control the fire.  Both
fire departments retreated from the fire after rescuing the only survivor from the reconnaissance
team. Police and emergency medical services also arrived on scene.  Several firefighters were
treated on-site because of minor injuries and heat exhaustion. The fire chiefs issued an initial
evacuation order downwind of the smoke plume, including the Helena Medical Center, and called
the West Memphis HazMat team.

       At 3:00 p.m. the West Memphis HazMat team arrived at the site to support fire fighting
efforts.  They provided the first air monitoring equipment. Due to the extreme toxicity of the
chemicals involved and changing wind conditions, the evacuation was extended to a three-mile
radius area. Most of the  Helena Medical  Center patients were taken to a community college and
others to a hospital in Clarksdale Mississippi. Residents of West Helena and nearby Helena were
sheltered in place.  A twenty-mile section of the Mississippi river was closed to river traffic due to
the prevailing winds at the time of the incident,.
              Response Actions Under the Incident Command System

       At 2:06 p.m. the National Response Center notified EPA Region 6 of the fire and
explosion at BPS Inc.  The initial notification had no information regarding fatalities, injuries or
evacuations. At 5:00 p.m., EPA received a second notification indicating that the incident was
out of control and requesting federal assistance. EPA Region 6 dispatched two On-Scene
Coordinators (OSC) and activated the Regional Response Team (RRT). Other federal groups
joined EPA in the response action.   DOD=s Pine Bluff Arsenal provided atropine and real time air
sampling equipment. The atropine was intended to be used as an antidote for AZM exposure of
responders and community members.

       The RRT contacted several chemical companies for scientific and technical support.
Among other companies, Mobay Chemical, DuPont, Bayer, Rhone Poulenc, and Elf Atochem
sent representatives to the site to voluntarily assist in the response action.  DuPont also deployed
its HazMat team to provide emergency response support.

       Response organizations continued air monitoring to determine if the plume contained
dangerous levels of toxins. Based on wind conditions and monitoring results the evacuation was
downgraded to stand-by  status.  Local authorities allowed evacuees in the two-mile radius return
to their homes.

       On May 9, 1997, the Incident Command System (ICS) was officially implemented.
Numerous Federal, State, and Local agencies and organizations provided support within the ICS,
including US EPA, US ARMY, Arkansas State Police, Office of Emergency Services, West
                                           11

-------
Helena/Helena/ West Memphis Fire Departments, OSHA, US Alcohol, Tobacco, and Fire Arms
(ATF), Center for Disease Control, ATSDR, NFPA, DuPont, EPS, and others.

       The EPA On-Scene Coordinator directed the response through the ICS operations.  As
more information regarding the quantity and nature of the chemicals involved in the fire became
available, the fire was allowed to burn with minimal active fire fighting efforts.  This decision was
made based on the potential water reactivity of the burning chemicals and the concern that
incomplete combustion products could be more harmful than those generated by complete
combustion.

       By May 14, 1997, Maneb was the primary chemical still burning at the facility. Maneb is
air reactive and water reactive. After several unsuccessful efforts to extinguish the fire,
emergency responders decided to spread the Maneb into thin layers and then to fog it with water.
This strategy was chosen based on information provided by Rhone Poulenc on a similar incident
in Brazil. All fire zones were extinguished and the site was downgraded from emergency
response. After inspection on May 15, 1997, the Arkansas State Police released the site from
crime scene  status.

       The EPA OSC opened the site for the JCAIT to take samples and document the scene
before the clean up activities could begin.  The JCAIT coordinated site documentation, sample
planning, and sample collection with all the on-scene investigative parties.  Once the JCAIT
completed sample collection, the EPA OSC released the site for cleanup.  The EPS contractor
began cleanup operations under EPA=s oversight on May 22,  1997.
       2.3    Public Health and Environmental Issues

       Several response organizations, including EPA, Arkansas Department of Health,
Mississippi Department of Environmental Quality, and the BPS contractor performed air
monitoring.  This information was used to determine whether the plume could present a threat to
public health or the environment.  Chemical companies provided technical assistance on
decomposition products and monitoring devices.

       The Arkansas Department of Health requested on-site assistance from ATSDR to address
the following public health issues: 1) acceptable exposure levels, 2) hospital reoccupation, 3)
decontamination of business and residences, and 4) consumption of exposed food products.

       On-site use of atropine was limited to one firefighter who exhibited exposure symptoms.
Reportedly, this firefighter was not wearing respiratory protection. Approximately 400 people
reported symptoms consistent with short term exposure to pesticides.  Thirteen of those cases
were referred for blood tests. These blood tests were reported as normal.

       The Arkansas Department of Pollution Control and Ecology collected point of entry water
                                           12

-------
samples from the community water system wells.  The impact was found to be minimal because
runoff from fire fighting efforts was contained on-site and no drinking water wells were in the
vicinity of the facility.

       ATSDR=s after action report concluded that no long-term public health effects were
expected from the fire and explosion at EPS.  This conclusion was based on the toxicology of the
chemicals involved and the maximum contaminant levels detected in and around the businesses
and residences.

3.0    Investigation and Analysis

       3.1     Investigation

       Members of the JCAIT interviewed BPS personnel and other individuals potentially
having knowledge about the incident. The JCAIT also requested documents from the facility,
documented the scene, and collected samples. Once the initial field activities were completed, the
JCAIT identified two distinct problem areas: the existence of a combustible atmosphere in the
new warehouse addition and the resulting three firefighter fatalities.

       The primary focus of the JCAIT investigation is on the events leading to the creation of
the combustible atmosphere.  Therefore, most of the initial investigation activities were conducted
to support the root cause analysis of this particular problem.  The JCAIT acknowledged that other
investigation groups, such as the NFPA and the U.S. Fire Administration (USFA), were
addressing the three fatalities. It is not the intention of this report to duplicate the work
performed by these groups. Instead, this report looks at general areas in the emergency response
system that could have  contributed to the firefighters: fatalities.

       The JCAIT did not attempt to analyze the explosion dynamics. Given the presence of a
combustible atmosphere, any source of ignition had the potential to initiate the explosion.
However, the most likely source of ignition was the arc(s) created in the facility equipment when
the electrical company service man began disconnecting the power to the facility.  The JCAIT did
attempt to identify the explosion origination point and the source of the combustible material in
the air as relevant to the immediate cause.  The investigation team used witness statements,
photo-documentation of the area, and laboratory analysis in this process. The process required
several iterations of analytical work.  Some of the laboratory  results are not discussed directly in
this report because they were either inconclusive (did not confirm or disprove a conjecture) or did
not include any detectable contaminants levels.  Summary reports on laboratory analyses are
included in Appendix B.
                                            13

-------
       3.2    Analysis

              3. 2. A  Overview of Explosion Scenarios

       BPS did not have standard operating procedures (SOPs) for material storage and handling.
The general practice at the facility was to store materials in the warehouse as space was made
available.  There were no established methods to ensure segregation of incompatible materials or
protection of stored materials from factors that could cause accidental releases, ignition or reaction
of ignitible or reactive materials.  According to the BPS Unit Two forklift supervisor, he was not
instructed to tell forklift operators where to spot materials in the warehouse. The fork lift
operators were supposed to find an Aempty spot® to locate incoming materials. There was no
attempt to determine the materials hazard classification and/or incompatibilities.

       BPS conducted a hazard review before agreeing to repackage any product. The written
procedure required going through a check list before beginning a repackaging  operation.  The
hazard review  did not address chemical handling and storage. There was no systematic review of
factors that could potentially affect warehousing of hazardous chemicals. For example, in October
1995, BPS added the warehouse area to the Unit Two building. The compressors: discharge pipe
was modified to pass through the new warehouse additions north wall and take a 90-degree elbow
turn to meet the outside  after-cooler. This modification resulted in a fifteen foot long discharge
header pipe running at a height of 6 feet inside the pesticide storage area.  No assessment of the
potential risks  associated with this change was performed.

       The incident occurred in early May, which is  a peak production month  for BPS as the
agricultural industry begins to prepare and place orders for various products for their growing
season. The morning of the event, forklift operators had to move materials around in the new
warehouse to make room for off-loading Procure  and AZM.  During the investigation, through
interviews of forklift operators and supervisors, investigators attempted to identify where materials
were spotted in the warehouse. The different accounts regarding what was located in the storage
area and where it was located indicated that there was no system in place to manage the  storage of
the various materials at the facility. This lack of an inventory management system, storage SOPs,
and a system to review potential hazards of changes in the facility could have led to a number of
warehouse incidents.

       All witnesses agree that the smoke originated near the warehouse additions north wall,
close to the compressor header pipe. Witnesses, including the fire fighters, also reported the
presence of Aproduct® or Apowder® in addition to smoke in this area.  This suggest the presence of
a hybrid dust/vapor mixture. The JCAIT found no visible crater for the explosion, which is
consistent with a dust/vapor explosion.  The explosion of an airborne flammable  vapor or dust
could occur at any location where a flammable concentration has accumulated. This could be at
some distance  from the source of the dust/vapor mixture. Presumably, the fan located on the
southwest side of the building could have drawn the hybrid mixture in that direction, affecting also
the direction of the blast. In any case, the explosion origination point is not necessarily the
                                            14

-------
location of the flammable material source. The source of the flammable material will be discussed
in the scenario analysis.

       Based on an event and causal factor diagram, analytical results, and professional judgment,
the JCAIT identified the following scenarios in the development of the combustible atmosphere
that led to the explosion:

$      Chemical inside supersack decomposes when placed close to the compressor header pipe
$      Decomposition of AZM SOW begins before arriving at BPS
$      Incompatible chemicals react
$      Malfunctioning compressor overheats a supersack

       The JCAIT concluded that a supersack placed close or against the compressor header pipe
was the most likely scenario.  Several of the chemicals stored in the new warehouse addition at the
time of the incident can decompose thermally while in contact with a surface within the
temperature range of the compressor header. However, the JCAIT concluded that AZM SOW had
a greater probability to initiate the event. It should be noted that most incidents are the result of
multiple factors rather than a single cause.  The JCAIT did not rule out the possibility that the
AZM SOW placed close to the compressor pipe was already decomposing before arriving at the
facility. Following an initial decomposition of the AZM SOW, the Maneb adjacent to it could have
also been involved in the subsequent explosion. The explosion cause scenarios are discussed
below.

Scenario 1:  Chemical in Supersack Decomposes when Placed Close to the Compressor
             Header Pipe

       Critical to the development of this scenario was the need to determine if a supersack was
actually placed against the pipe, which chemicals were most likely to have been placed in such
proximity, and whether the compressor pipe could reach temperatures high enough to cause the
chemical to decompose.

       Compressor Discharge Temperature

       At the time of the incident, BPS personnel stated that the surface temperature in the pipe
was approximately 145° F (63° C).  In order to confirm this statement and determine the potential
involvement of the compressor in the incident, the JCAIT  conducted a series of activities. First,
the team inspected and documented the compressor system conditions after the event.
Observations from this inspection were supplemented with interviews with BPS employees and
management and the compressor manufacturer.

       Second,  the JCAIT conducted a forensic analysis of the compressors and estimated the
anticipated temperatures in the discharge pipe=s system. The analysis is presented in a report
dated August 20, 1997 and referenced as DNV Project No. 232-8384, Insecticide Warehouse
                                          15

-------
Explosion Investigation. The summary report is included in Appendix B.

       Finally, the JCAIT participated in a simulation conducted by MFC at Tifton Industrial
Controls in Tifton, Georgia on May 15, 1998.  The simulation intended to measure a range of
temperatures in a compressor system re-constructed to simulate BPS operations at the time of the
event.

       Of particular interest was the surface temperature of the common header pipe at the
approximate point were the supersacks could have been  spotted. MFC, with the concurrence of
OSHA and EPA, developed a testing protocol to provide and connect two compressors to
simulate the BPS conditions. The compressors were connected under EPA and OSHA oversight.
 The JCAIT measured the piping surface temperature at several locations and under different
conditions (insulated vs non-insulated).

        A summary of the findings is presented below:

$      The piping configuration between the compressors and the after-cooler included
       approximately twenty feet of discharge piping and two short radius turns.

$      Discussions held with the manufacturer of the two compressors used at BPS indicated that
       there are a number of factors which can affect the compressor discharge temperature such
       as ambient temperature and discharge pressure.  However, under normal operating
       conditions the maximum discharge air temperature of the compressors at the cylinder head
       would be expected to be in the range of 300° to 350° F (149° - 177° C).

$      JCAIT estimated that the discharge temperature on a compressor system like the one used
       at BPS would be approximately 350° F (177° C). The associated external pipe
       temperature would be 280° F (138° C).  If the pipe is engulfed by an insulating type
       material, such as a supersack, the pipe would be expected to attain the same temperature
       as the discharge air.  See DNV Project No. 232-8384 in Appendix B.

$      During the MFC=s Tifton simulation, the team measured the non-insulated pipe
       temperature at the distance where the supersack could have been in contact with the
       header pipe.  Once equilibrium was reached, the  surface temperature at that point was
       approximately 255° F (124° C).  The group then wrapped a two foot section of the pipe
       with a fibrous glass insulation to roughly simulate the effect of a supersack against the
       pipe. The temperature increased from 255°  to 301° F (124° to 149° C) in less than 30
       minutes. The maximum insulated header temperature in the simulation was 336° F (169°
       C).

       The JCAIT also reviewed technical literature, including the compressor operators manual,
to determine how the BPS compressor system compares with industry practices:
                                           16

-------
 $     The Operators Manual for the Model 460 Compressor, Overheating Section, states that
       the piping to the after-cooler location should be as short as possible, preferably no more
       than three feet. For runs over three feet, the pipe size should be increased by one pipe size
       for each eight foot run.

$      The Compressed Air and Gas Handbook, published by the Compressed Air and Gas
       Institute, states that the discharge piping, i.e., the piping between the compressor and the
       after-cooler, the after-cooler separator, and the air receiver, should be as short and direct
       as possible and should use Along-radius® elbows where bends are necessary.

$      The American Society of Mechanical Engineers (ASME B19.1 - 1995) Safety Standards
      for Air Compressor Systems, Section 2.1.8. High Temperature states, flExternal surfaces
       subject to temperatures in excess of 175° F (80° C) which personnel may have contact,
       shall be guarded or insulated.®

       The JCAIT concluded that the common header pipe connected to the two-compressor
discharges was in fact substantially higher than the 145° F (63° C) estimated by BPS employees
and management. From the above results, the JCAIT estimates that the discharge header in the
warehouse could have been in excess of 300° F (149° C).

       Chemical Location

                    Shortly after the incident, BPS employees and management identified the
decomposing material as an AZM supersack placed against or close to the hot compressor
discharge pipe.  The JCAIT confirmed that supersacks of materials were being spotted in close
proximity or against walls at BPS. After the incident, JCAIT observed supersacks spotted along
the wall in the Unit One warehouse. In this case, the supersacks were stacked two-high.  The
edge of the bottom supersack was within inches of the wall. The top supersack was listing so that
it was in contact with the wall (Figure 5).
                                 L
                    Figure 5.  Supersack set-up at BPS Unit One
                                           17

-------
       Several of the chemicals stored in the new warehouse addition at the time of the incident
had the potential to decompose thermally while in contact with the hot compressor pipe.  The
JCAIT collected bulk samples of the combusted residue where the explosion occurred, attempting
to determine the exact location of each chemical pallet.  The analysis of these samples was of
limited use due to the total destruction of the area and combustion of the sampled material (Figure
6).  The collection and analysis of samples was supplemented with the analysis of other physical
evidence and witnesses: statements.
         Figure 6. Aerial Photo BPS Unit Two

       Based on witness interviews, the JCAIT identified the approximate location of the
chemicals in the warehouse area (Figure 7).  Even though witness accounts are somewhat
conflicting regarding the quantity and approximate location of the stored chemicals, most
statements agree that AZM supersacks had just been placed next to the compressor piping.
Witnesses also agree that the yellow smoke or powder was coming from this location. Forklift
operators recollect placing Maneb pallets by the compressor pipe in an attempt to make room for
the incoming AZM.  In addition to the witness statements the JCAIT:

$      Screened the bulk residue samples for various pesticides including AZM, Maneb, Topsin,
       and Sevin. Only  semi-quantitative values of AZM and Maneb were reported.

$      Secured and  analyzed the remains of the new warehouse addition ventilation fans which
       had visible yellow residue (Figure  8). AZM and its major decomposition products were
       confirmed.

       The JCAIT concluded that it was highly probable that pallets of both AZM and Maneb
were placed along the compressor pipe the day of the event.
                                           18

-------
               18'
30'
To After
     X
60'
                               22'
                                                            11'
                             70'
                        ±19'
e











Qife Compressor Room
, |S 20HP 1EH=> T
l:ztyz:'.:z:>,?l!*i:>,:i.
:;:|5:;:s:!^t:j:;:j:^:;:
Sitting
Room
;:;:S:;::5:;:S:;:S::::S:;:S:;:s::;:S:;:s
ROOM
8
:;:S:;:S:;:K:;::;:;:;:S::;:S::;:^:;:;;:;:;;:;::;:;::;:;::;::::;
O^>>OOO:H;d>O
a*fcNu3-<.g.gBia>
£ ^ S^Q-3 * M « £ tt
<_r :!
O "
flj
a
r"
)






^ Q 3
U]
a
_J> SPI.L

"0 O W > ;

3.


S
d
=

-------
                    Figure 8. Ventilation Fan with Yellow Residue

       AZM and Maneb Thermal Decomposition

       Both AZM and Maneb can decompose thermally if they are exposed to elevated
temperatures during a period of time.

       Maneb is classified by the Department of Transportation (DOT) for transportation
purposes as an ASpontaneously Combustible Material® unless it is stabilized. If it is stabilized,
Maneb is classified as a flDangerous When Wet Material.® This classification includes materials
that evolve flammable gas when in contact with water. Maneb presumably falls in this category
because of formation of carbon disulfide.  According to the MSDS, the Maneb at EPS was
stabilized.

       Data related to AZM decomposition temperature is rather conflicting.  MSDS do not
identify AZM as flammable  and most literature provides a decomposition temperature of 320° F
(160° C).

       The JCAIT requested representative  samples from the manufacturers of AZM SOW and
Maneb 75DF to conduct several thermal stability tests including decomposition temperature and
color changes associated with temperature.  The  JCAIT also conducted Differential Scanning
Calorimetry  (DSC) tests on other AZM formulations, including the 90% pure technical
formulation used as the active ingredient in AZM SOW. A summary of the tests findings and
literature search are presented below:
                                           20

-------
                                      AZM SOW

$      AZM SOW showed visible color change to dark tan at 217° F (103° C). The sample
       showed visible smoke at 340° F (171° C).

$      The DSC analysis showed the 90% AZM technical formulation decomposing exothermally
       (1100 J/g) at approximately 320° F (160° C). Other formulations, including AZM SOW,
       decomposed exothermically (600 J/g) at approximately 338° F (170° C). The smaller
       amount of heat released by the 50% formulations compared to the 90% pure technical
       grade is consistent with the addition of inert ingredients.

$      A basket test to determine safe storage temperatures for bulk AZM SOW showed
       decomposition of the sample beginning between 158°-176° F (70°-80° C). The
       decomposition temperature corresponds to an estimated safe storage temperature of 79° F
       (26° C), using a 10% safety factor, based on volume and surface area specifications for
       supersacks provided by MFC.  It should be noted that the test does not predict a safe time
       interval corresponding to this temperature.

$      MFC conducted a twelve month storage stability study in support of registration of its
       product. The procedure included the use of two 2.5 pound samples. For this test the
       product was stored at 68° F + 36° F (20° C + 2° C) for twelve months.

$      The EPA  Office of Pesticides-- Product Properties Test Guidelines (OPPTS 83 0.6317) for
       pesticide registration requires storage stability tests to be conducted under either of the
       following conditions: A) At 68 ° or 77° F(20° C or 25° C); B) Under warehouse conditions
       which reflect the expected storage conditions of the commercial product; C) The test
       parameters may be expanded to include accelerated conditions, such as elevated
       temperatures (104°-129° F) (or 40°-54° C) or cold temperature (-20°-0° C).

$      In a test to determine whether it would melt, decompose, or the vapor given off would
       ignite, the AZM sample turned yellowish brown, then black, gave off yellow smoke, and
       the vapors ignited. A second test confirmed these results.

       A study conducted by G. Bertoni and Co-workers; Lazioni Commercial! in Ambienti
Refrigerata,, Annali di Chimica,  1985, states that Aaccidental overheating of an AZM mixture may
occur during the mixing process and since the active principle melts at low temperatures
(m.p.l62° -165° F) (m.p. 72°-74° C) and decomposition begins at a temperature of about 212° F
(100° C), gases and vapors are set free.® The study concluded that:

$      The product begins decomposition around 100° C. As temperature increases an intense
       exothermic reaction occurs between 338° and 356° 1
       volatile products of about 40% of the initial weight.
exothermic reaction occurs between 338° and 356° F (170° and 180° C) with a loss of
                                           21

-------
$      AZM is a thermally unstable material; a slow process of degradation of the compound
       occurs below 122° F (50° C).

$      The spontaneous degradation of AZM is noticeably accelerated by any increase of
       temperature so that attention has to be paid to storage of this product and its commercial
       forms.
$      It is recommended to keep AZM away from any heating.  If the temperature rises above
       100° C decomposition is very fast and at 170°-180° C the product decomposes almost
       instantaneously.

                                    MANEB 75DF

•      The sample of Maneb 75DF showed a black spot beginning at 320° F. Visible smoke was
       observed at 340° F (171° C).

$      Under nitrogen atmosphere, Maneb 75DF did not release a significant amount of energy
       when heated during the DSC.  In the temperature range of 338°-410° F (170°-210° C) the
       samples heated in nitrogen showed an exothermic reaction followed by an endothermic
       reaction. The net result under these conditions was a slight absorption of heat with
       decomposition occurring at approximately 338° F (170° C).

$      Zi-Ru Liu,  et al. published a study in Thermochimica Acta 220 (1003) 229-235 entitled
       Heat Changes Associated with the Thermal Decomposition of Maneb andZineb.  This
       study focuses on the heat changes on thermal decomposition of Maneb and Zineb using
       DSC. It acknowledges that both endothermic and exothermic processes are present in
       their initial decomposition. The study concludes that the initial decomposition
       temperature in air is greatly  decreased compared with that in nitrogen. The study
       indicates that the thermal decomposition of Maneb is accelerated and is an exothermic
       process accelerated in air or oxygen gas.

$      The basket test results for Maneb 75DF showed an onset temperature between 221°-239°
       F (105°-115° C). Using similar procedures as described for AZM, an estimated maximum
       safe storage temperature of  181° F (83° C) was  calculated for a supersack of Maneb.  The
       test does not predict a safe time interval corresponding to this temperature.

•      When tested for melting, decomposition, and evolution of ignitible vapor, Maneb 75DF
       decomposed into a black material, white vapors evolved from the decomposing sample,
       the vapors ignited into a yellow/orange flame, and the vapor flame self-sustained several
       seconds after the removal of the ignition source. In a second test Maneb produced vapors
       that ignited as a yellow flame; at full  decomposition the sample produced white smoke.

       As mentioned before, both AZM SOW and Maneb 75DF could have been placed close or
                                          22

-------
against the compressor header pipe. The test on both substances indicate that decomposition
could have occurred at the temperature likely reached by the compressor exhaust pipe, but AZM
begins to decompose at a lower temperature than Maneb.  Statements provided by most witnesses
of the incident describe a yellow smoke or gas which is consistent with what was observed during
experimental tests.

       The JCAIT concluded that AZM SOW was the material responsible for the initial evolution
of the combustible atmosphere. If a supersack of AZM SOW was placed in contact with or in
close proximity to the hot compressor pipe, the heat could have initiated its thermal
decomposition.  The decomposing material would propagate away from the pipe in the direction
of the center of the supersack.  The contact of the decomposing material with the pipe in this
instance would not necessarily be prolonged.  The decomposition would be accompanied by the
evolution of gas and smoke (the products of decomposition).

       AZM=s volatile decomposition products, as all organic compounds, evolve flammable
constituents upon decomposition. In particular, a literature reference (Combustion Products from
Pesticides and Other Chemical Substances Determine by Use of DIN 53 436, L. Smith-Hansen
and K. Haahr-Jorgensen, Fire Safety Journal 23(1994),  51-66), lists six organic combustion
products from the decomposition of AZM. The article further states that generally, large numbers
of different organic species are formed during decomposition due to incomplete decomposition
and partial oxidation. As mentioned before, the flammable gases from decomposition would not
have been confined to the immediate area above the supersacks and could have ignited/exploded
at some distance from the origination point.

Scenario 2.   AZM decomposition begins before arriving to BPS:

       The JCAIT postulated as a possible scenario that a thermal decomposition was occurring
inside a supersack of AZM SOW before it arrived at BPS. This decomposition could have
generated the airborne flammable substances that exploded in the warehouse.  The scenario is
supported mostly by witness statements concerning the smell of the supersacks that were
unloaded the morning of the event.  The truck driver reported that the AZM SOW smell had  made
him feel sick and that he had to stop and rest for that reason. On separate interviews, BPS
employees stated that he had made the same remarks to them the morning of the incident.  Other
BPS employees reported the unusual smell as well.

       Chemical powders can undergo smoldering combustion. Hot temperature spots can
become entrapped in bulk containers (e.g. a supersack).  Smoldering can also occur as a result of
self heating when the temperature of a bulk material is raised to a level at which the rate of heat
production exceeds the rate of heat loss. In either case, the container and contents can thermally
insulate, allowing exothermic reactions to continue at a very  slow rate. When the container is
disturbed, the reaction can spread and the reaction rate can increase until the self heating reaction
reaches the surface. The hot material or its decomposition products may reach temperatures
sufficient to burst into flames, especially when disturbed.
                                           23

-------
       Less= Loss Prevention in the Process Industry, volume 2; 17.5 explains as follows: AA dust
deposit can undergo smoldering for a long period. It is not unknown for large piles to smoulder
for a matter of years. Both air access and heat loss are restricted so that combustion is very slow,
but is sustained.  Such smouldering may give no readily  detectable effects.  In particular, there
may be no smoke or smell from the burning.  This delay  between ignition and outbreak of flaming
can create hazards. Fire may break out unexpectedly in  a factory shut down overnight or at a
weekend, or the cargo of a ship may be discovered to be on fire when it is unloaded. Hazards of
dust fires include those of a dust explosion resulting from the formation and ignition of a dust
suspension, of the ignition of other flammable materials  and of the evolution of toxic combustion
products.®

       The AZM unloaded at BPS the day of the incident was contained in supersacks
approximately 1600 pounds of material each. The product had been stored in Tifton, Georgia.
NOAA reported a daily average temperature of 76° F (24° C)  for Tifton, GA during the month of
April.  The maximum temperature reported by NOAA for Helena, Arkansas on May 8, 1997 was
82° F (28° C).

       MFC files show eight minor incidents from 1987-1996.  These incidents were associated
with AZM SOW smoldering as a result of the material  coming in contact with hot surfaces (mostly
hot bearings) during production. In those instances, MFC flooded the smoldering product with
water.  A manufacturer of another AZM formulation reported twelve incidents in the 1960's, five
in the 1970's, and seven in the 1980's. All of them involved excessive heating during processing
or storage.  As stated in previous sections, MFC had discussed with BPS the product=s potential
for smoldering while in contact with hot bearings. Reportedly, MFC advised BPS to flood
production hoppers in the presence of a bad odor during the repackaging operation.

       In theory, one of the AZM SOW supersacks could have had a smoldering hot spot as a
result of the mixing operations.  A smoldering spot in a bulk container could have been in storage
without being detected. Sensors normally used in automatic fire protection systems cannot
usually detect this kind of condition.  This hot spot could have initiated a self heating reaction
which accelerated during the unloading at BPS.  The smoke or powder was discovered right after
the lunch break.  There is an approximate 15 minute time span from the time the waste monitor
left the new warehouse addition, and the smoke was discovered. After the discovery, the reaction
seemed to have continued at an increasingly accelerated  rate.  An accelerated reaction rate after
being disturbed is consistent with industry=s experience of smoldering spots insulated by the bulk
container.

       The scenario, however, is based solely on witness statements. These statements are not
consistent.  Thermally stable AZM has a very strong and persistent odor. The truck driver and
BPS employees were not familiar with AZM.  Their statements concerning whether or not this
load had a different odor from a previous one are at times contradictory.
                                           24

-------
       The JCAIT inspected the truck several days after the incident looking for evidence of
AZM SOW decomposition.  Prior to the incident, the truck had transported plastic lawn mower
parts. Therefore, there was no potential for the AZM SOW to react with a compound previously
transported in the truck.  Wipe samples conducted in the truck confirmed only the presence of
AZM.  Early AZM decomposition products would be in the form of volatile gases and vapors.
Since several days had passed since the incident, it was not reasonable to expect positive sampling
of volatile compounds.

       The JCAIT requested MFC split control samples from the production batches offloaded at
EPS.  Laboratory analysis showed no signs of thermal decomposition. The fact that the control
samples showed no signs of decomposition, however, does not rule out the possibility of a hot
spot entrapped in a supersack. In addition, these control samples  have been in a controlled
environment that could be substantially different from actual storage conditions. Similarly,  there
is no evidence that the AZM supersacks delivered at EPS were exposed to factors that could
induce its thermal  decomposition.

       The JCAIT concluded that this is a possible scenario but the available evidence is
uncertain and cannot substantiate it. However, the team acknowledges that a self heating process
could have either initiated the event or accelerated the thermal decomposition of a supersack
placed close to a heat source.

Scenario 3.  Incompatible Chemicals React

       The following pesticides were present in the Unit Two new warehouse addition at the time
of the incident: Azinphos methyl SOW, Maneb 75DF, Alliette Signature WDG, Topsin WSB,
Sevin 80 WSP, and Penncozeb 75DF.  In addition, a spill of twenty to thirty pounds of Alliette
Signature was reported next to the new warehouse addition north wall shortly before the incident.

       The team reviewed the chemical properties and reactivity of these pesticides to estimate
potential hazardous reactions that could have initiated the explosion and subsequent fire. This
review is discussed below and summarized in the table at the end of this section. The pesticides
present represent the following types of chemicals:

•      Carbamate - Topsin and Sevin;
•      Dithiocarbamate  - Maneb  and Penncozeb; and
•      Organophosphorus - Azinphos methyl and Alliette.

       The analysis showed that none of the pesticides would be expected to be highly reactive
with each other under normal conditions. Based on their chemical structures, there would be no
reason to expect any of these substances to react with each other if they were accidentally mixed
together. The form in which these substances were stored (i.e., solid formulations) and the
presence of inert ingredients would make reactions particularly unlikely.
                                           25

-------
       The carbamates and Dithiocarbamate are chemically similar; chemical reactions would not
be expected to take place between such similar chemicals.  Maneb and Penncozeb (Mancozeb), in
particular, are compounds of the same base chemical and are very similar; Maneb is the
manganese salt of dithiocarbamic acid, and Penncozeb is a compound of dithiocarbamic acid and
both manganese and zinc. Topsin (Thiphanate methyl), a carbamate, is combined in formulations
with both Maneb and Mancozeb (Farm Chemicals Handbook, 1994), indicating that o reaction
takes place when these substances are mixed. There appears to be no reason to expect a reaction
between the carbamates Topsin and Sevin (carbaryl), because of their chemical similarity, or
between Sevin (carbaryl) and Maneb  or Penncozeb, by analogy with Topsin.

       Alliette (Fosetyl-aluminum) and Azinphos methyl are Organophosphorus compounds, not
carbamates or Dithiocarbamate, but no reaction would be expected upon mixing with carbamates
or Dithiocarbamate, based on the chemical structures of these substances.  Fosetyl-aluminum is
combined in formulations with Mancozeb (Farm Chemicals Handbook., 1994), indicating that no
reaction would take place between these  substances. This type of formulation also provides
evidence that  Fosetyl-aluminum likely would not react with Maneb, because Maneb is very similar
to Mancozeb, and would be expected to react similarly.

       Several of the pesticides are reported to be incompatible with strong oxidizers, and it is
likely that all  of them would react with strong oxidizers under some conditions.  No oxidizers
were reported to be present, however. Based on this analysis, the JCAIT concluded that the
event was not initiated by the Alliette Signature spill or the reaction of incompatible chemicals
placed in proximity.
                                           26

-------
                              Reactivity and Flammability of Pesticides Present in New Warehouse Addition at EPS
    Name
 Common Name/Chemical Name/Formula
           of Active Ingredient
          Reactivity and Flammability Data
    Potential Reactions with Other
          Pesticides Present
 Azinphos
 methyl
Azinphos methyl
O,O-(Dimethyl S[(4-oxo-l,2,3-benzotriazin-
3 (4H)-yl)methyl] phosphorodithioate
C10H12N303PS2
Decomposes at elevated temperatures.
Hydrolyzed in alkaline and acidic media. Contact with
strong oxidizers may cause fires and explosions.
Combustible (conflicting data).
None expected.
 Topsin
Thiophanate-methyl
Dimethyl[(l ,2-phenylene)bis
(iminocarbonothionyl)]bis(carbamate)
C12H14N404S2
Compatible with other agricultural chemicals that are
neither highly alkaline nor contain copper.
No data on flammability (probably combustible).
None expected. May be combined in
formulations with Maneb and Mancozeb
(Penncozeb), indicating no reaction.
 Alliette
 Signature
Fosetyl-aluminum
Aluminum tris(O-ethyl phosphonate)
C6H18A109P3
Stable under storage conditions.
Incompatible with strong bases, mineral acids, strong
oxidizers, strong reducing agents.
Non-flammable.
None expected. May be combined in
formulations with Mancozeb (Penncozeb),
indicating no reaction.
  Sevin
Carbaryl
1 -Naphthyl Bmethyl carbamate
C12HnNO2
Stable under storage conditions.
Incompatible with alkalies and strong acids.
Combustible.
None expected.
 Maneb
Maneb
Manganese ethylenebis(dithiocarbamate)
(C4H6MnN2S4)x
Decomposes on prolonged exposure to air or water.
Incompatible with strong acids and strong oxidizers.
Classified by DOT as spontaneously combustible or
dangerous when wet.
None expected. May be combined in
formulations with Thiphanate-methyl
(Topsin) and Mancozeb (Penncozeb),
indicating no reaction.
 Penncozeb
Mancozeb
Manganese ethylenebis(dithiocarbamate)
complex with zinc ion
(C4H6MnN2S4)x(Zn)y
Stable under storage conditions.
Decomposed in acid and alkaline conditions, by heat, and
when exposed to moisture and air.  Incompatible with
strong acids and strong oxidizers.
Compatible with most common pesticides.
No data on flammability - probably similar to Maneb.
May be combined in formulations with
Thiphanate-methyl (Topsin) and Maneb,
and with Fosetyl-aluminum (Alliette)
indicating no reaction.
Sources:
Farm Chemicals Handbook >P4.Hazardous Substances Databank (HSDB), National Library of Medicine, for Azinphos methyl, Fosetyl-aluminum, Maneb, Mancozeb.
MSDS for Azinphos methyl SOW, Alliette Signature, Sevin, Maneb 75DF, Penncozeb 75DF.
TOMES for carbaryl. Worthing, ed., The Pesticides Manual (1987).]
                                                      27

-------
Scenario 4.   Malfunctioning Compressor Overheats Supersacks Near the After-cooler Piping

         Two multistage reciprocating air compressors were used in the Unit Two building. As stated
   previously, these compressors discharged into a common header pipe that was located on the
   warehouse side of the wall, approximately five feet above the deck. This header was 15 feet long and
   led to an after-cooler outside the building. The 15-hp unit suffered substantial damage during the
   incident. After the explosion, this unit was found lying on its side with no lubricating oil in the
   crankcase. The concrete foundation by the compressor had substantial heat damage and spalling in a
   configuration that suggested a liquid had burned on the surface.  The 20-hp unit had only moderate
   damage, remained in its upright position after the event, and had a substantial amount of oil in the
   crankcase.

         The JCAIT dismantled the 20-hp compressor.  The 20-hp compressor did not show any
   observable internal damage. The JCAIT also performed a forensic analysis of the 15-hp unit to
   determine whether or not the unit was working properly at the time of the event (DNV Project No.
   232-8384).  The forensic analysis conducted on the 15-hp compressor showed that:

   •      The aluminum bell housing for the electric motor and the aluminum header for the first stage had
         melted away.  The melted residue had been deposited on the engine and compressor mount
         platform immediately below the motor when it was still in the upright position. This indicates
         that the compressor was exposed to heat before falling on its side.

   •      The pulley side of the compressor had sustained direct flame impingement heat, but little was
         observed on the opposite side.  The damage areas indicate that an intense fire had been burning
         on the deck next to the pulley side while the compressor was still upright.

   •      The connecting rod and journal bearings had not been scored.  This indicates that the unit had
         sufficient lubrication when last run.

   •      A coke-like residue was  inside the crankcase.  This indicates that a lubricating oil fire had
         developed inside. Presumably, it was ignited by a liquid fire on the deck after the compressor
         fell over. It is also likely that the oil  leaked out through the pulley side bearing.

         Based on these findings, JCAIT concluded that the 15-hp compressor was not malfunctioning
   before the event.  Therefore, this scenario was discarded by the investigation team.

                3. 2. B Overview of Early Emergency Response

         There are many factors that could be root causes or  could have contributed to the three
   firefighter fatalities. A formal analysis requires a thorough review of operational parameters and human
   performance influencing factors including training, competency, pre-planning, policies and procedures,
   etc.  A critique of these factors and the local emergency response activities is outside the scope of this
   investigation. However, the JCAIT evaluated some general aspects of the emergency response system
   (related to the EPS explosion) which can foster unsafe situations. By doing this, the JCAIT attempts to
   promote efforts to provide local emergency response groups with information critical to their safety
   when responding to chemical incidents.

                                             28

-------
       BPS personnel believed and informed the WHFD that a smoldering Abag@ of AZM had initiated
the incident. This fact is indicated in witness accounts, the BPS 911 call transcript, and an early press
release from the facility management. As a repackaging tolling operator, BPS did not have in-house
expertise to test and identify the hazards associated with the chemicals they were handling. Instead,
BPS was relying on the chemical manufacturers information (in this case MFC) to address the chemical
hazards.  On the other hand, the WHFD relied on BPS to provide them with chemical hazard
information.

       On-site information

       BPS management told the WH Fire Chief that AZM would not explode. Neither the facility
personnel nor the documents handed to the fire department conveyed the danger of explosion. The
MSDS for AZM used by the BPS personnel and firefighters was provided to BPS by MFC. The
MSDS includes the following information concerning the thermal stability and reactivity and
flammability hazards of AZM SOW:

$      FDVIIS flammability rating of 0 (non-combustible).

$      HMIS reactivity rating of 0.

$      Stable under normal conditions.

$      High temperatures may cause hazardous vapors.

$      Store in cool, dry, well ventilated place.  Do not place near heat or open flame.

       There  is no data on the AZM SOW indicating the possibility of an explosion hazard.  It does not
include a safe storage temperature or a decomposition temperature.  It does, however, warn against
placing AZM near heat.  The JCAIT  reviewed several other MSDS for different AZM formulations, in
particular, the MSDS for Bayer=s Guthion, which BPS had discussed with MFC.  This MSDS includes
the following information:

$      NFPA flammability rating of 2. (JCAIT Note: An NFPA flammability rating of 2 applies to
       materials that must be moderately heated or exposed to relatively high ambient conditions before
       ignition can occur.  These materials would not under normal conditions form hazardous
       atmospheres with air, but under high ambient temperatures or under moderate heating might
       release vapor in sufficient quantities to produce hazardous atmospheres with air.)

$      NFPA reactivity rating of 2.  (JCAIT Note: An NFPA reactivity rating of 2 applies to materials
       that are normally unstable and readily undergo violent chemical change, but are not capable of
       detonation. It applies to materials that can undergo chemical change with  rapid release of
       energy at normal temperatures and pressures and materials that can undergo violent chemical
       changes at elevated temperatures and pressures.)

$      During routine handling of this material, there should be little risk of dust explosion.

                                           29

-------
$      Stable material.  Unstable in sustained temperature above 100 °F (38° C).

$      Storage temperature:  30-day average not to exceed 100 °F.

$      Store in cool, dry area away from heat source.

       It should be noted that NFPA 472 Standard on Professional Competency ofResponders to
Hazardous Material Incidents, 1997 edition, Non mandatory Appendix A-21.4 explains that: ASome
materials have products of combustion or decomposition that present a significant greater degree of
hazard that the inherent physical and toxic properties of the original material. The degree of hazard is
dependent on the conditions at the time of the incident®.

       In addition to the AZM SOW, the WHFD had the DOT=s Emergency Response Guidebook.  In
the 1993 edition, Guide Number 55 applies to AZM.  In the Fire and Explosion Section, Guide 55
indicates A Some of these materials may burn, but none of them ignites readily.  Container may explode
violently in heat of fire.®

       The WH Fire Chief reported during an interview that fire personnel received the following
HazMat training; two career firemen (both killed during the event) had 80 hours of technician level
training; all fire personnel had training through the awareness level; and other firemen were trained
through the technical and operational level.  As part of the emergency preparedness program, the
WHFD received MSDSs from BPS and had been invited to tour the facility and participate in their
emergency response drills.  As mentioned before, the fire department had toured the facility one month
before the incident.

        Training and pre-planning are critical to emergency response groups. Additionally, adequate
information is essential for incident-specific risk management.  Chemical emergency situations are
among the worst work environments for human performance.  It is in emergency situations where the
human information processing system is burdened with multiple and critical tasks. The information
provided to local emergency responders has to be structured and prioritized for this  specific use to
maximize human performance.

       MSDSs are developed to comply with OSHA=s Hazard Communication Standard to
communicate the hazards posed by chemicals to employees.  Additionally, they are extensively used by
emergency response groups during chemical releases. The JCAIT looked at the MSDSs present at BPS
at the time of the incident from a local emergency response standpoint.  The number of MSDSs at BPS
do not constitute a statistical representation  of the MSDSs developed by the chemical industry.
Evaluating the MSDSs present at BPS, the JCAIT found the following:

 $     MSDSs did not have a standard  format.  Information relevant or critical during an emergency
       response operation may not be readily available or may be presented in a confusing format.
                                           30

-------
$      Some MSDSs had a check-box format. In the case of the Maneb MSDS, information was
       incomplete or conflicting. For example, hazard information stated the chemical was Awater
       reactive.® However, the information on firefighting stated Ause water.® No further explanation
       on how the firefighting water will interact with the Maneb was provided (e.g. should the
       firefighters use fog vs a large stream).  Similarly, no information was provided related to Maneb
       being stored in the presence of a water-based sprinkler system.

$      Some terms were not clearly explained.  In the case of the AZM SOW MSDS,  information
       included in the section AUnusual Fire and Explosion Hazards® stated that the Avapors and fumes
       from fire are hazardous.® The term hazardous does not convey whether the vapors and fumes
       are toxic, combustible or both.

       The JCAIT concluded that the on-site hazard information was conflicting and incomplete.  It is
critical that fire departments collect as much hazard information as possible within the time, resources,
and training limitations. In addition to MSDS, NFPA 472 Standard on Professional Competency of
Responders to Hazardous Material Incidents, 1997 edition, Appendix A Explanatory Material,
identifies other sources of information for hazard identification such as the North American Emergency
Response Guidebook., hazardous material databases, technical information centers
(CHEMTREC/CANUTEC/SETIQ), shipper/manufacturer contacts, and monitoring equipment.

       Risk Perception/Risk Management

       The WHFD was reviewing MSDSs when the explosion occurred. The firemen that died were
close to the building getting ready to enter the building.  From witness interviews they were trying to
locate the smoldering Abag® in the warehouse.

       BPS employees were not aware of any explosion hazards.  The employees did  not show extreme
concerns to the WHFD. The facility personnel conveyed more the need of air-packs due to the toxicity
of the chemicals rather than any fire and explosion hazards. The production manager had entered the
building several times just before the explosion.  He actually closed the building doors (with the fire
chief approval) which in effect confined the combustible atmosphere.

       The lack of awareness of the potential explosion hazard played an important role in the tactics
used by the WHFD.  With a better understanding of the potential hazards, the WHFD would
presumably have been more cautious.  NFPA 1561, Standard on Fire Department Incident Command
System, Explanatory Appendix A-4-1.2, explains that the risk to fire department personnel is the most
important factor to be considered by the incident commander in determining the strategy to be
employed  in each situation. One of the factors involved in the management of risks levels is the
pessimistic evaluation of changing conditions.

       NFPA  1561, 4-1.2 states that AThe concept of risk management shall be utilized on the basis of
the following principles: (a) Activities that represent a significant risk to the safety of personnel shall be
limited to situations where there is a potential to save endangered lives; (b) Activities that are routinely
employed  to protect property shall be recognized as inherent risks to the safety of personnel, and
actions shall be taken to avoid these risks; (c) No risk to the safety of personnel shall be acceptable
where there is no possibility to save lives or property®.

                                           31

-------
       NFPA 1561, A-4-1.3, further explains: AThe acceptable level of risk is directly related to the
potential to save lives, the risk to fire department personnel must be evaluated in proportion to the
ability to save lives, the risk to fire department personnel must be evaluated in proportion to the ability
to save property of value. Where there is no ability to save lives or property, there is no justification to
expose fire personnel to any avoidable risk, and defensive fire suppression operations are the
appropriate strategy.®

       As stated in the previous section, the on-site information available to the WHFD was conflicting.
 The AZM SOW MSDS did not state the potential for an explosion hazard.  In addition, EPS
management may have given the WHFD a false sense of risk when asked about the danger of an
explosion. However,  chemical warehouses may present unique and unexpected hazards to emergency
responders because of unknown combustion products and chemical interactions. In the BPS incident,
the building had been evacuated and no lives were threatened. Factoring conflicting information and the
unexpected hazards presented in a chemical storing area into the risk management decision process
could have helped the emergency responders to develop a safer response strategy.

4.0     Root Causes and Recommendations

       4.1    Root Causes and Contributing Factors

       Root causes are the underlying prime reasons, such as failure of particular management systems,
that allow the faulty design, inadequate training, or deficiencies in maintenance to exist. These, in turn,
lead to unsafe acts  or conditions which can result in an accident. Contributing factors are reasons that,
by themselves, do not lead to the conditions that ultimately caused the event; however, these factors
facilitate the occurrence of the event or increase its severity. Although the JCAIT cannot precisely
determine the exact cause of this event, there is sufficient information to support several root  and
contributing causes. The root causes and contributing factors of this event have broad application to a
variety of situations and should be considered lessons for industries that conduct similar operations.
The JCAIT identified the following root causes and contributing factors of the event:

$      MFC and BPS did  not have a full understanding of the hazards associated with AZM.

       EPA=s Office of Pesticides requires manufacturers to conduct storage stability tests under one of
the following conditions: A) At 20° C or 25° C; B) Under warehouse conditions which reflect the
expected storage conditions of the commercial product; C) The test parameters may be expanded to
include accelerated conditions , such as elevated temperatures (or 40° C-54° C) or cold temperature (-
20° C-0° C).  MFC  conducted the study at 20°C + 2° C for twelve months and a two pound bag. In
order to comply with the Office  of Pesticides requirements, MFC should have tested for the actual
container size (1,600 pounds) and expected storage and transportation temperatures which can be
considerably higher than the 20°C used by MFC in their test.

       In addition, the Office of Pesticide Programs requires the use of DSC to test pesticides for
explosiveness. DSC is a screening test.  For thermally unstable materials, the DSC test does not provide
specific enough information to predict safe storage temperatures of large storage or shipping containers.
                                           32

-------
       MFC failed to provide BPS with adequate information on the hazards associated with the
chemical.  As MFC did not perform adequate testing, hazard information relative to the thermal stability
and explosiveness of AZM was not included in the MSDS.

$      BPS did not assess the potential hazards of a hot pipe in an area where hazardous
       chemicals were to be stored when the new warehouse addition was constructed.

       In October of 1995 BPS added the warehouse area to the Unit Two building. The compressors:
discharge pipe was modified to go through the area where hazardous chemicals were stored. A review
of the impact of the change should have identified the risks associated with this configuration to
workers and/or heat sensitive chemicals.

$      BPS did not have standard operating procedures for material storage and handling

       Standard operating procedures could have prevented BPS  from placing a thermally unstable
substance next to a heat source, in this case, the compressor header pipe.

$      On-site information provided to the WHFD was conflicting and incomplete.

       The AZM  SOW MSDS did not specifically identify an explosion hazard. Generally, chemical
hazard information on  MSDS is not structured and prioritized for local emergency response use.
MSDSs may not have  enough information to help emergency responders conduct safe operations and
should not be relied upon as the sole source of information during an emergency response. In fact,
DOT=s Emergency Response Guidebook on-site had a warning related to containers exploding violently
in the heat of fire.  Additional sources of information can help local responders to conduct safer
operations.

       4.2    Recommendations

       The following recommendations were developed by the JCAIT that address the root causes and
contributing factors to  prevent recurrence or similar incidents at other facilities:

$      Manufacturers  should be proactive in testing potentially hazardous materials. Testing for actual
       conditions and  elevated temperatures during storage should be conducted to determine  safe
       storage conditions.  Screening tests, such as DSC, can be helpful in determining the need for
       additional testing. However, thermally unstable materials which are intended to be packed and
       shipped in large volume containers should be tested beyond screening levels.

$      Facilities which store, use, handle, manufacture, or move hazardous materials should develop
       and implement a system to review potential hazards of modifications to facilities, equipment,
       chemicals, technology, or procedures. The system should analyze potential impacts to safety,
       health, and the  environment and take appropriate actions before the modifications are
       implemented. OSHA=s Process  Safety Management (PSM), EPA=s Risk Management Program
       (RMP), and the Center for Chemical Process Safety (CCPS) guidelines can help facilities
       develop such system.
                                           33

-------
$      Facilities that store hazardous chemicals should develop standard operating procedures for
       material storage and handling that address storage restrictions.  Such facilities should adhere to
       applicable practices outlined by CCPS and NFPA.  Pesticide facilities are encouraged to also
       follow NFPA 43D (Code for the Storage of Pesticides), specifically the non-mandatory
       Appendix B.

$      Facilities storing hazardous chemicals should develop an inventory management system with
       information regarding composition, compatibility, storage, location, and quantity of incoming
       products.  This management system can help the facility comply with storage restrictions and
       provide emergency responders useful information during a response action.

$      EPA and OSHA, in conjunction with interested parties, should facilitate a workshop to make
       recommendations on how to improve the quality of hazardous materials information available
       during response actions.  The workshop should review appropriate uses of MSDS by local
       emergency response groups and how to provide these groups  information describing the
       behavior of hazardous materials when they begin to react or decompose and what responders
       should look for during a chemical emergency.
                                           34

-------
                  Appendix A




AZM SOW, Maneb 75DF and Alliette Signature MSDSs
                      35

-------
                     KATERIAIi SAFETY DATA 5HB3T

               S3CTION 1 - MAKTTPACTtTRStt INFORMATION

MANUF/DIST:     M1CKO FI.O  CO.
                P. 0. 3ox  5948
                Lakeland,  Ft.  33237

SMERG2NCY  PHONE:   (300) 424-33CC-

     TRADE NAME /SYNONYMS    :   AZINPKOS METHYL SOW SOLUBLE
     CHEMICAL NAME/SYNONYMS :   See Hazardous Ingredients Below
     CHSMICAT, FAMILY        :   Or=ranouhos-sh.e™us Insecticide
     FORMULA        •        :   Cwi^pTjOjSS, ~"
     PRODUCT CCUS           :   Reg. # 51036-154

         HAZARSOUS MATS P. I RIi I DBMTZFI CATION SYSTSM  ''HMISi

                   *    KZAL.TH ..... 2          *
                   *    FLtAMMABILITY .  . 0          *
                   *    RSAC"TVITY  . .  .0          •
                   *    PROTECTION  . .  .2          *

                 SECTION 2  - HA.2A^SOC"3
THIS 5RCOUCT CONTAINS HAZARDOUS INGASDISNTS :   Yes

CHSKICA7.  MAMS _      CAS NT^KSS?.   %    P5I.-OSHA    TL.V-ACGIH
Azirpnos methyl            S5-5J-0      50.0  (skir.)     0.2mg/tn3
Iner1. ingredients
                      To«.al :

THIS PRODUCT CONTAINS  CARCINOGENS  CNTF,  1ARC. Or OSKA) :   NO

                   SECTION 3 - HSAIiTK HAZARD DATA
       SrFSCTS (Ac^ra and Chsx>n : o 5 :
 (LDSO Values  fcr azinphos snechyl technical)
Acute Oral    IjDSO (r=aL)        "   -  13-15
Acute Dermal  LD53 (ran.)            -  >  222  rag/Vcg
Acute Inhalation                   -  C.C3   mgy'l

MAY BE LETHAL IP INHALED,  SWALLCWEU OR  A3SOR3SD TKRCUGK SKIS

PRIt^ARY KOU'i'SS OF ENTRY:
Poisonous if swallowed,  inhaled, or absorbed Lhrouerh skin.  Rapidly
absorbed throuc.i akin "surf aces  & eyes.   Contaminated clothing must
be  removed  immediately.    Exposed  persons  must  receive  prompt
ir.udlcal ~
Physician Note:  This produce is a strong cholir.esterase inhibitor.

KEDICAi. CONDITIONS AGGRAVATED dV EXPOSURE:
Low cholinesueraee levels.

SIGNS & SYMPTOMS Of POISONING:
Headaches, nausea, vomiting, cramps, weakness, blurred vision/ pin-
                                 36

-------
n~-~t puoils, tightness in chest,  labored breaching, nervousness,
sweating", watering  of eyes, drcclir.g or frorhing of ir.outh &  nose,_.{;
muscle spasms, and  coma .

EMERGENCY FIRST AID PROCEDURES:
Call  a  doctor-  iphysiciar.) ,   clinic,   cr   hospital  iir.meG.a3.teiy •
Exp1 ain  that the  victim  has  been  exposed to  azinphos  methyl  &.
describe his condition.  If the doctor cannot come,  take the victim
to a hospital or- clinic at ones.
IF INXALEn-.                                                     .  .
Remove  victim to fresh air.   IF  not  breathing,  give artifi=ia_
resoiration,  preferably mouth tc  mouth,  St maintain until doctcr
sees victim.  If breathing is  diif icult , _give axygen.
If ZN 2VES OS. ON SKIN:
Iir.:r.ediately  flush with plenty of water for -5 mine . while removing
contaminated personal clothing i shoes to- avoid continued possible
exscsure .
  ~
Induce vomiting imtr.ediately by giving 2 glasses ot water & touching
back  of  throat  with  finger.    Do  Not  Induce  Vomiting  Or  Give
Anything  3y Mouth  To An Unconscious Person.   Have victim lie down
& x.eeo quiet .
r.'OTF.  TO  PHYSICIAN:    Antidote  - give  atropine  sulfate in  large
doses.  TWO to FCD3. tr.g. intraver.cusly or intramuscularly as soon as
cyar.csis  is  overcome.     Repeat  at  5-10  min .  intervals  until
atropinization signa appear. *2-?AK c'r.leride ir.ay be administered as
ai  a'djunct to,  but  not  substitute  for  atropir.e.   DO NOT  GIV3
MORPXIKE   OR   TRANQUIL IZSRS    BECAUSE   THSSB   KAY    ACCENTUATE
PHARMACOLOGIC  EffECT OF THIS PP-ODUCT .

                     SECTION 4  - CHSMrCAI.  DATA

BOIDIXG
5?3Cif J.C GRAVITY 7-i=ar.hir\s'  apparatus   
-------
orotactive clothing.

ENUS'OAL FIRS .B-NO EXPLOSION HAZARDS:
Vapcrs and futr.ea frcir. fire are harar^
avoid dispersing  powder.  Evacuate
fii-a.

INCOMPATISUITY  (Materials to avoid)
Alkaline materials.
phospho - -
tira  involving  this product.

W-LL  HAZARDOUS  POLYMERIZATION CCCCT :            .
Has not been  known to occur ur.der normal conditions.

I a THS PuOJJCT  STABLE:
Under normal  conditions.

CONDITIONS TO AVOID ?O.-< S7A3IL:TY:
High  tenperatures rosy rause hazardous vaccrs -   Avoid cor.tact witr.
strong cxidizers or &l!
-------
fMSHA/NIOSH approval  number prefix TC-21C)

    SCTIVIi UliOVSS :
OTHSS PROTECTIVE
Chemical -rasistant  fouLwear plus sccka
Prof- active eyewear  such as safety goggles  or a. face shield
Chemical -resistant  headgear for overhead exposure
CheniCcJl resistant  aoron when cleaning  equipment, mixinc, or
loading
Coveralls ever  short -sleevad shirt  and  sh-vrt pants

Discard  clothing and ether  absorbent  maUsrials chat  have  been
drenched cr haavilv contaminated with this product's  concentrate.
Do not reuse them. "Follow ir.anuf acturer'  s.-inst ructions for cleaning
and maintain ing PPE.   If  no  such instructions for washablea,  usa
detergent and hot water.  Xee? and  wash PPE separately  from other
laundry.

DSS?. SAFETY RSCOMSfc CATIONS
1. Wash har.da before eatir.g,  drinking,  chewing gurn, using tobacco
   or us J fig  the toilet
2. Reir.cve clothing  immediately if pesticide gets inside.  Th-t:i
   wash thoroughly  and put oii clean clothing
3. Rsracvs ??E irvnediately after handling this product.   Wash the •
   outside of gloves .before removing.   A= socn as possible,  wash
   thoroughly and change i::'_o -clean cl
REPEATED  EXPOSURES TO CKOLINSSTESASE INKI3ITORS SUCK AS
KE7IIYL MAY WITHOUT WARKING,  CA^S2 ?ROLOXG33 5LTSCt:PTI3ILITY TO VZRY
SMALL DCSSS OF AXV CKOLI2JSSTZXASE IKKI5ITCS .  Persons working with
this   product  should  have   frequent   bloud   tests   of   their
cholir_ess_erase  levels.   If it  falls below a  critical point,  no
further exposure  should  be allowed unt^l it has b«sss:i detsrjr.ir.ed by
ciood  tests  that  it  lias  returi;«ti  uo  a nor-n-.al  le-/el .   Keep 'all
unprotected persons & anix.als sway f rcir.  trs«±Lsd area or  where "there
is a danger or drift.  Uo Not rub'eyes cr mouth  wi-_h ha:-ds .   If you
feel sick In  any  way,  STO? wc.i-k and  get help right  away.

                  SBCTIOU 8 - SPEClar. PRECAUTIONS

PRECAUTIONS TO 3E TAXSN  IN j-ANOLING  AND STORAGE:
btore in a cool,  dry w&ll-ventilated  places.  Do  tio-^  store n*av heat
cr open f lame .  Xeea cut of reach  of children..   STORE IN ORIGINAL,
COHTAIKERS ONLY.   t)O JCCT  US3  CR STORE  IN OR AROUND HOME.   E:nu«_y
container retains product  residue.   Observe all labeled  safeguards
until  container  is disposed in  accordance  with state  and  local
l*w/s .

KAI.MTlia.ANCh; f^ECACJTIQNS :
wear full protective  clothing when working en  equipment that has
been used  to apply  cr  package  this product.   Residues  leCi..  in
equipment are extremely
Kespirators should  be cleaned  &  cartridg-es  replaced According to
                                  39

-------
instructions included wizh respirators.  Replace gloves frequently. *
Xeisr to product label  for  further prsvauuiona regarding reentry*
    worker warnings
ADDITIONAL
To ths best of cur knowledge , the Information contained herein is
accurate.   However, Miero-7lo does not assume any liability for the
accuracy or completeness of the information .   Final  detertnindticn
is the sole responsibility of the user.
SEGREGATE FROM FOOD,  FSEDSTUcFS. & CLOTHING (4SCFR 177.841)

                SECTION 9   TRANSPORT iy?OHMATION

D.O.T. Proper  Shipping Narr.e  (43C7:l  172.101):   Orgaiiuphoephcrus
PssLicides, Solid,  Toxic, N.O.s. (Contains  Azinphos  M«±Lhyl)
D.O.T. Hazards Class:   G.l
OtT/NA Number:   UN27S3
Packing Group:   ?G II
LsibeKd)  Required:   Primary - Poison
                  Subsidary - NA
Placard (s)  Required:   Primary - Poiacn
                    Subsidary - NA
Emergency Response Uuia« ;  #55

               SSCTION 10  R2g3ZATOR"g
CESCIA Repcrtable Quantity:  *l
RCHA Status:   Nu(_ Regulated
     7I7LS III:
          Section 302 Extreaely Hazardous Substance:  'YES
          Section 311 Hazard Categories:  Immediate
          Section 313 Toxic Chemicals:  NO
Revision-.   1/95
                                     40

-------
MANEB 75DF
elf atochem
EMERGENCY01*0'01116 ADDRESS.- /SSj©
TELEPHONE NUMBERS: ELF ATOCHEM NORTH AMERICA. INC.
4M215) 419-5054 (9 a.m.-5 p.m. EST) (PRIMARY) AGCHEM DIVISION FORM 4627
(800) 424-3300 (CHEMTHEC) 2000 MARKET ST.. 21 st FLOOR REV. 5/96
MATERIAL
SAFETY
DATA SHEET
PHILADELPHIA. PA 19103-3222
PRODUCT SHIPPING
IDENTIFICATION INFORMATION
PRODUCT PROPER SHIPPING.
NAMff DESCRIPTION
MANEB 75DF FUNGICIDE Sp^RATKDNS,
SR3F-"0- Sttggpo,,.
*Lf ATOCHEM NORTH AMERICA, INC. PLACARD
CODE NUMBER DANGEROUS WHEN WET
326 IMO: MARINE POLLUTANT
CHEMICAL NAME AND MANEB 75DF. when shipped
MOLECULAR FORMULA ""TSJS1)! 5 ElF1 J&SK££ahw*d
MANGANESE ETHYLENE BIS S}££RS£25& 3£S£K?V, T
(DITHIOCARBAMATE) (POLYMERIC) SOUD. NOT REGULATED BY O.O.T
(C.H.MnN,SJx (according to 49 CFR)

MANEB. TRIMANGOL PHYSICAL
CAS NUMBER PROPERTIES
OF ACTIVE INGREDIENT EEBEEfCEflEBBBBEEEH
12427-38-2

CHEMICAL FAMILY PHYSICAL STATE
DITHIOCARBAMATE BOILING POINT/HANAC

HAZARDOUS "*
INGREDIENTS MRTING POINT
B^BJB^BJBJBJEflBJEBJBJBJ DECOMPOSES BEFORE MELTING
MATERIALS FREEZING POINT
OH COMPONENTS % MrW NA

MANEB' 7S.O
ETU» trace MOLECULAR WEIGHT (CALCULATED)
(CAS9&45-7) (2853,
'This ingredient has been listed as toxic
under SARA 31 3. BULK DENSITY
•Etnytenemiourea (ETU), a trace 0 6S oArH at 20'C
contaminant and break-down product of "
Maneb 75DF, has been categorized as a _-._ __.
probable human carcinogenbylARC. VAPOR PRESSURE (mm Hg)
and as a group b carcinogen by NTP. NEGLIGIBLE AT
ETU has also caused birth defects in ROOM TEMPERATURE
laboratory animals.
+ WARNING: VAPOR DENSITY f AIR. «>
This product contains chemicals known NA
to the state of California to cause cancer.
(See ingredient 1 listed above.) SOLUBILITY IN H,O
This product contains chemicals known OISPERSIBLE
to the state of California to cause cancer
and birth defects or other reproductive _, „„. „„ „ _ v „„. ..,._
harm. (See ingredient 2 listed above.) % ™LATILES BY VOLUME
FIRE AND
EXPLOSION DATA
FLASHPOINT (TEST METHOD;
NE
FLAMMABLE LIMITS
NA
AUTOIGNITION TEMPERATURE/
FIRf POINT
NE
EXTINGUISHING MEDIA
J WATER SPRAY 2 DRY
CHEMICAL
!» WATER FOG G ALCOHOL
FOAM
G WATER STREAM 'g FOAM
2 CO. G EA"™ OR
SAND

SPECIAL FIRE FIGHTING
PROCEDURES
G DO NOT ENTER BUILDING
G ALLOW FIRE TO BURN
G WATER MAY CAUSE FROTHING
G CO NOT USE WATER
K OTHER:
WEAR SELF CONTAINED
BREATHING APPARATUS AND
PROTECTIVE CLOTHING.



UNUSUAL FIRE AND
EXPLOSION HAZARDS
G DUST EXPLOSION HAZARD
C SENSITIVE TO SHOCK
G CONTAMINATION
C TEMPERATURE
2 OTHER:
TOXIC AND FLAMMABLE FUMES.

REACTIVITY
DATA


STABILITY

Jf STABLE n UNSTABLE
REACTIVITY DATA,
CONTINUED
HAZARDOUS DECOMPOSITION
PRODUCTS, THERMAL AND
OTHER:
THERMAL - CARBON DISULFIOE
OTHER — ETHYLENE THKXJREA
CONDITIONS TO AVOID
HHEAT
g OPEN FLAMES
g SPARKS
[g IGNITION SOURCES
Si OTHER:
DAMPNESS OR WET
SPILL
OB LEAK
E^MEflBBHHEl

STEPS TO BE TAKEN IF
MATERIAL IS RELEASED
Off SPILLED
B FLUSH WITH WATER
D ABSORB WITH SAND
OR INERT MATERIAL
R NEUTRALIZE
B SWEEP OR SCOOP UP AND
REMOVE

G KEEP UPWIND. EVACUATE
ENCLOSED SPACES
C PREVENT SPREAD OR SPILL
G DISPOSE OF IMMEDIATELY
WASTE DISPOSAL METHOD:
Waste resulting from the use of this
product may be disposed of on site or at
an approved waste disposal facity.








SARATITLEIII

Hazards Classification (40 CFR 370):
  Immediate Health: YES
  Delayed Health: YES
  Fire: YES
  Sudden Pressure: NO
  Reactivity: YES
Section 313 (40 CFR 372): This product
contains the following chemicals subject
to SARA Section 313 reporting
requirements: Maneb.

HFPA RATINGS
FtammaWity: 1
Reactivity:!
Health: 2
EVAPORATION RATE
NA

APPEARANCE AND ODOR
GREENISH TO GRAYISH YELLOW
SOLID MICRO-GRANULE.
NO SPECIFIC ODOR.
CONDITIONS CONTRIBUTING
TO INSTABILITY
2 THERMAL DECOMPOSITION
G PHOTO DEGRADATION
1_ POLYMERIZATION
C CONTAMINATION

INCOMPATIBILITY — AVOID
CONTACT WITH
2 STRONG ACIDS
C STRONG ALKALIS
2; STRONG OXICMZERS
g OTHER:
   MATERIALS THAT REACT
   WITH WATER
                                                                HA  - NOT APPLICABLE
                                                                HE  = NOT ESTABLISHED
                                                                MM .DATA NOT AVAILABLE
                                                                 +  * SECTION REVISED
                                                                                                CONTINUED ON OTHER SIDE
                                                             41

-------
ELfATOCHeii MOUTH AMERICA, me.
- , '"'" MATED! At PRODUCT NAME:
„ MAItHIAL HAHEB 75DF FUNGICIOf.
ammr «««—
BEFORE USING PRODUCT, READ AND FOLLOW DIRECTIONS AND PRECAUTIONS ON PRODUCT LABEL AND BULLETINS.
TOXICITY
••••B^BM
ORAL lACVTmi
-D, (RAT) > 5.000 mg/kg
OtRUAL lACIfTf)
^DM > 2,000 mg/kg

" INHALATIOM (ACUTE)
1C, (RAT) > 2.0 mg/L (4 hr.)
CHRONIC. tUBCHKONIC, tTC.
flats lad 2 yam on a diet containing 250
3pm suffered no ill effects. In long term
(reding studies on rodents, some
iitfikxarbamates have induced
rardnooenicity and birtti detects at high
dietary intake levels.

HEALTH HAZARD
INFORMATION

tmcr* OF EXPOSURE
TO CONCENTRATE

K*M«*mur EXPOSURE LIMIT
(TtvmiA on CEIUHO ten
*CGIH TLV/TWA S rog/m
OSHA (QSmg/m1
IRRITATION
SEYE
_-.ERE [j SEVERE
MODERATE Q MODERATE
UMILO
(TRANSIENT)
comosiwrr
NA
SEHSITIZATIOH
NOT A SKIN SENSITIZER
INHALATION EFFECTS
NA
LUNG EFFECTS
COUGHING. IRRrTATION
UPON OVEREXPOSURE





















HEALTH HAZARD
INFORMATION,
CONTINUED

EHEROEHCY FIRST AID

INOCSTfON
fjg GET MEDICAL ATTENTION
g INDUCE VOMITING
Q DO NOT INDUCE VOMITING
jg GIVE TWO GLASSES OF WATER
ocmui.
SI GET MEDICAL ATTENTION IF
IRRITATION PERSISTS
ig FLUSH WITH SOAP AND WATER
2 CONTAMINATED CLOTHING —
REMOVE AND LAUNDER
O CONTAMINATED SHOES —
DESTROY

EYI CONTACT
D SET MEDICAL ATTENTION
50 FLUSH WITH PLENTY OF WATER
FOR AT LEAST 1 5 MINUTES
H OTHER:
GET MEDICAL ATTENTION IF
IRRITATION PERSISTS
INHALATION
;g GET MEDICAL ATTENTION
il REMOVE TO FRESH AIR
1 IF NOT BREATHING. GIVE
ARTIFICIAL RESPIRATION.
PREFERABLY MOUTH-TO-MOUTH
D GIVE OXYGEN






















OATK9/M

AOMCSSJ
ELF ATOCHEM NORTH AMERICA. INC.
AGCHEM DIVISION
2000 MARKET ST.. 21st FLOOR
PHILADELPHIA. PA 19103-3222
SPECIAL.
PROTECTION
INFORMATION
••••"•^••"•"•aY
VfHTILATION
JtB*Q(/MCMCMT$

ALWAYS MAINTAIN EXPOSURE
BELOW PERMISSIBLE
EXPOSURE LIMITS
[] CONSULT AN INDUSTRIAL
HYGIENIST OH ENVIRONMENTAL
HEALTH SPECIALIST
0 LOCAL EXHAUST
S3 USE WITH ADEQUATE
VENTIUVTION
C] CHECK FOR AIR
CONTAMINANT AND
OXYGEN DEFICIENCY
EYE
WEAR PROTECTIVE EYEWEAR WHEN
MIXING OR LOADING

HAND (GLOVE TYPf)
WEAR WATERPROOF GLOVES WHEN
APPLYING OR HANDLING
RCSPHUTOH TYPE —
NOT REQUIRED— USE NORMAL
SAFETY PRECAUTIONS
OTHER PROTECTIVE
EQUIPMEHT
Applicators and other handlers must wear
• coveralls over long-sleeved shirt and
long pants.
• waterproof gloves.
• shoes plus socks,
• protective eyewear when mixing or
loading,
• chemical- resistant apron when cleaning
equipment, mixing, or loading.



























SPECIAL
PRECAUTIONS
wmm*mmmm^^m
PRECAUTIONARY
LABELIHO
JX] WASH THOROUGHLY AFTER
HANDLING
(g DO NOT GET IN EYES, ON SKIN
OR CLOTHING
ffl 00 NOT BREATHE DUST. VAPOR
MIST. GAS
Ijrj KEEP CONTAINER CLOSED
jXl KEEP AWAY FROM HEAT
SPARKS AND OPEN FLAMES
C STORE IN TIGHTLY CLOSED
CONTAINERS
C DO NOT STORE NEAR
COMBUSTIBLES
n KEEP FROM CONTACT WITH
CLOTHING AND OTHER
COMBUSTIBLE MATERIALS
55 EMPTY CONTAINER MAY
CONTAIN HAZARDOUS
RESIDUES
GUSE EXPLOSION-PROOF
EQUIPMENT
OTHER HAHOL1HO AND
STORAGE CONDITION*
Store in the original container in a dry
area. If allowed to become wet the product
will deteriorate and represent a fire
hazard. Keep away from sources of
iginition (e.g. sparks and open flame).
Close bag when not in use. Do not store in
a manner where cross-contamination with
other pesticides, fertilizers, food or feed
could occur.








PLEASE
NOTE
EHAiochtm North Amgric*, Inc. betiwes tfi*t
ffw information and neonrnmnimliana
contauirthtnin (inducting dausnd
automata) an accurate « of tfw f«f and may no* b» voKd wfMv such
product a uawd in esomtoMflbo with any otfMr
mttfritis or in any praorss, Fwffwr, *oot Vw
conditions and /wtfwds tf UM of tfw product
•vitf of tto BiibuTjCun lytmtd to titnin 
beyond the aortml of Elf Atocfwn, Elf
Atochwnaxpn&ty diadems wy**ii*
liability 45 to *iy naiAa obtvnod or vising
*om try io» oS the product or roUmncm on sucfi
ffUbmution.
42

-------
CHIPCOCR) ALIETTE(R) SIGNATURE(TM) BRAND WDG FUNGICIDE
            |        i
Material Safety Data Sheet       Date Prepared:    2/05/96 Supersedes Date:   0/00/00
 /. CHEMICAL PRODUCE AND COMPANY DESCRIPTION
 RHONE-POULENC AG COMPANY
 AGRICULTURAL  CHEMICALS
 2 T.w. Alexander Drive
 Research Triangle Pk NO   27709

 Enargsnoy Phone  Nxuiborii
 FOR EMERGENCIES  INVOLVING A 3PIL1,  LEAK, FIRE,  EXPOSURE OR ACCIDENT
 CONTACT: CHEMTREC {800-424-9300}  OR RHONE-POULENC (800-334-7577).
           i         i
 For Product Information!
 (300) 334-&74S
                    i
 Product Qtatuai
 FIFRA regulated  uad  only.

• BPA FIFRA Ragiotratjion Kunbttri
 264-515
           i         !
 chamioal N»n« or Synonymi
 ETHYL HTOROOSN PHO^PHONATE,  FOSETYL-AL

 Molecular pontiula.1 :
 2. COMPOSITION/INFORMATION ON INGREDIENTS
 Ccaponwit
 FOSETTL-AL
 CRYSTALLIKS SILICA AS
 OTHER INOREClSNTS (TRADI'SSCRBTI
                                                      018%
                                        CU *«9 M)sab«r  Uitrd
                                         39U9-S4-B     r       80.0
                                         14808-60-7     Y     < 0.2S
                                         .....*.*.*.*    fj       BALANCf
               NT1F1CA1
3. HAZARDS IDENTIFICATION
 A. EXSROENCY OVBRV^BHi
     Page..„,.!..
                                        43

-------
 CHIPCOCR) AtlETTE(R) SIGNATURECTM) BRAND WDG FUNGICIDE

 Material Safety Data Sheet       Date Prepared;     2/05/96 Supersedes Data  0/00/00
y;^^^sAS3^.i
-------
 CHIPCO(R) ALIETtE(R) SIGNATUREfTM) BRAND WDO FUNGICIDE

 Material Safety Data Sheet        Date Prepared:     2/05/96 Supersedes Dates  0/00/00
                         fltMntsn JH^^
  4. FIRST AID MEASURES (Continued)
 victim is breathing. If breathing  is  difficult,  administer oxygen,  if
 available. If victim is not breathing, .administer CPR (cardio-pulmonary
 resuscitation) . Seek medical attention.

 Xngaotiom
 If victim ia conscious and alert,  give  2-3  glasses of water to drink and
 induce vomiting by touching back of throat  with  a finger.  Do not induce
 vomiting or give anything by mouth to an unconscious  parson.  Seek
 immediate medical attention. 3o not leave victim unattended.  Vomiting
 may occur spontaneously. To prevent aspiration of swallowed product,  lay
 victim on aide with head lower than waist.  If vomiting occurs and tha
 victim ie conscious, give water to further  dilute the chemical.

 MSDICAL CONDITIONS POSSIBLY AOORAVATSD  BY HXPOSURSi
 Inhalation of product may aggravate existing chronic  respiratory
 problems such as asthma, emphysema or bronchitis.  Skin contact may
 aggravate existing skin disease.
 NOTES to
 All treatments should bo based en observed  signs  and symptoms of
 distress in the patient. Consideration  should be  given to tha
 possibility that overexposure tc materials  other  than this product may
 have occurred.

 Treat symptomatically. No specific antidote available.
  5. FIRE FIGHTING MEASURES
 7XRB HAZARD CATAi

 Flaah Pointi
 Not Applicable

 Sxtinguiahlng Madia t
 Recommended  (small fires): dry chemical, carbon dioxide, Recommended
 (large fire): water spray, alcohol foam, polymer foam, ordinary foam.

 Special Fire Fighting Procedure*i
 Firefighters should wear NIOSH/MSKA approved self-contained breathing
 apparatus and full protective clothing. Dike area to prevent runoff and
End of Page	3.	.*	,„,.„.—CoxMttou«djoalifejct£BgfL
                                    45

-------
CHIFCO(R) AtrETta
-------
CHIPCO(R) ALlETTEdl) SIGNATUR£(TM) BRAND WDG FUNGICIDE

Material Safety Data Sheet        Date Prepared:    2/05/96 Supersedes Date:  0/00/00

          *^<«tWHi81!W5^
 7. HANDLING AND STORAGE ( Continued )
 dusta. Do not breathe vapors and mists. Do not  ingest.


 Storage i
 Store in an area that is away from foodstuffs or animal feed.
 8. EXPOSURE CONTROLS/PERSONAL PROTECTION
 Introductory
 These recommendations provide general guidance for handling this
 product. Bacauao specific work environments and material handling
 practices vary, safety procedures should be developed for each intended
 application. While developing safe handling procedures, do not overlook
 the nead to clean equipment and piping systems for maintenance and
 repairs. Waeto resulting from these procedures should be handled in
 accordance with Section 13: Disposal Considerations.

 Assistance with selection, use and maintenance cf worker protection
 equipment is generally available from equipment manufacturers .


 Zxpoiura Quid«lin«Bt
 Exposure limits represent regulated or recommended worker breathing
 zone concentrations measured by validated sampling and analytical
 methods, meeting OSHA requirements. The following limits  (ACGIH, OSHA
 and other) apply to this material, where, if indicated, S-skin and
 c-ceiling limit:


 CRYSTALLINE SILICA AS QUARTZ
                         Notoa               TWA                 STSL
 ACGIK                                  0.1 mg/cu m
 OSHA                                   0 . 1 mg/cu m
 Engineering Control •>
 Where engineering controls are indicated by usa conditions or a
 potential for axceaeive exposure exists, the following traditional
 exposure control techniques may be used to effectively minimize employee
 exposures .

 Raapiratory Protect ion i


End..of Page ... .,,,5 „..,. „,....„,-—— -------- jCtontta«fid.43tt,Nfixi.?age_,
                                        47

-------
           i

CHIPCOCR) AtlETTEtR) SIGNATURE(TM) BRAND WDG FUNGICIDE

Material Safety1 Data Sheet       Date Prepared:     2/05/96 Supersedes Date:   0/00/00

 5. EXPOSURE CONTROLS/PERSONAL PROTECTION (Continued)
When  respirators  are  required,  select NIOSH/MSHA approved equipment
based on actual or potential airborne concentrations and in accordance
with  the latest OSHA  atandard (29 CPR 1910.134)  and/or ANSI Z88.2
recommendations .

Under normal  conditions,  in the absence of other airborne contaminants,
the following devices should provide protection from this material up to
she conditions specified  by OSKA/ANSI: Air-purifying
 (half -iT.ask/full -face)  respirator with cartridgas/canister approved for
use against duets, mists  and fumes,  pesticides.
Under conditions  immediately dangerous to life or health,  or emergency
conditions .with unknown concentrations,  use a full-face positive
pressure air-supplied respirator equipped with an emergency escape air
supply uniu or use a  self-contained breathing apparatus unit.

By«/?«ce Protection i
3ye and face  protection requirements will vary dependent upon work
environment conditions and material handling practices .  Appropriate
ANSI  Z87 approved equipment should be selected for the particular USB
intended for  this material.

It is generally regarded  as good practice to wear a minimum of safety
glasses with  side shields when  working in industrial environments .

Skin  Protection i
Skin  contact  should be minimized through use of  gloves and suitable
long-sleeved  clothing (i.e.,  shirts and pants).  Consideration must  be
given both to durability  as well as permeation resistance.

Work  Practice Control* i
Personal hygiene  is an important work practice exposure  control measure
and the following general measures should be taken when working with or
handling this material:

 (1) Do not istora,  use, and/or consume foods,  beverages,  tobacco
    products,  or  cosmetics in areas where this material  is stored.
 (2) Waah hands and face carefully before eating,  drinking,  using
    tobacco,  applying cosmetics,  or using the toilet.
 (3) Wash exposed  skin promptly  to remove accidental splashes  of contact
    with this material.
                                       48

-------
CHIPCO(R) AilETTE(R) SIGNATURE(TM) BRAND WDG FUNGICIDE

Material Safety Data Sheet        Date Prepared:    2/05/96 Supersedes Date:   0/00/00
                ;:Bm«^^
 9. PHYSICAL AND CHEMICAL PROPERTIES
 Physical and Chemical properties here represent typical properties of
 this product. Contact the business area using the Product Information
 phone number in Section 1 for its exact specifications.


 Physical Appearancei
 green granules solid.

 Odors
 acidic odor.

 pHi
 3.5  to 4 at 5 wt/wt*.

 Specific Oravityi
 No Data Available

 Water Solubility!
 diaperaible

 Malting Point Kangei
 No Data Available

 Boiling Point Bangei
 No Data Available

 Vapor Pressure*
 No Data Available

 Vapor Deniiityi
 NO Data Available

 Molecular Weighti
 354.1
 JO. STABILITY AND REACTIVITY
Chemical Stability.
 This material is stable under normal handling and storage conditions
 described in Section 7.
                                        49

-------
CH1PCO(R) AL
[ETtE(R) SIGNATURE(TM) BRAND WDG FUNGICIDE
Material Safety Data Sheet         Date Prepared:      2/05/96 Supersedes Date:   0/00/00
 10. STABILITY AND REACTIVITY (Continued)
Conditions To Be Avoided:
 moisture

Materials/Chemlcali To Be Avoided:
 strong basep
 strong oxidizing aganea
 strong reducing agents
 mineral acids

Thi Following Haz^rdou* Decowfjosltlon Products Might Be Expected:
 Decomposition Type: thermal
   oxides of; phosphorus
   oxides of. carbon
   phosphine' gaa

Hazardous Polynjertatloh Will jfot Occur.
 Avoid The Following Td Inhibit Harardous Polymerization:
   not  applicable
 //. TOXrCOLQCICAL INFORMATION
Acute Eye Irritation:
 Toxicologlcal Informatlpa »nd interpretation
   eye - eyej irricatlon,  rabbit.
   Under tas.t conditions,  material caused mild irritation  effects
   in rabbit eyes.

Acute Skin Irrluaion:
 Toxicologlcal InfQrmaUpn cod Interpretation
   ekin -  akin irritation,  rabbit.
   Under toat conditions,  material caused slight irritation effects
   to rabbiti akin.

Acute Dermal Toxjclty:
 Toxicologlcal Information nod tnterpreUtlon
   LD50 -  lathal  dose 50% of teat species,  >  2000 mg/kg, rabbit.

Acute Respiratory  Irritation:
 No  test data found for  product.

Acute Inhalation toxlcity:


              A	^Continued, n
                                                50

-------
16. OTHER INFORMATION

National Fire Protection Association Hazard Ratings--NFPA(R):
2 Health Hazard Rating—Moderate
0 Flammability Rating—Minimal
1 Instability Rating-Slight

National Paint & Coating Hazardous Materials Identification
2 Health Hazard Rating—Moderate
0 Flammability Rating—Minimal
1 Reactivity Rating—Slight

Reason for Revisions:
Conversion to ANSI MSDS format.

Key Legend Information:
ACGIH - American Conference of Governmental Industrial Hygienists
OSHA -  Occupational Safety and Health Administration
TLV - Threshold Limit Value
 PEL - Permissable Exposure Limit
TWA - Time Weighted Average
STEL - Short Term Exposure Limit
NTP - National Toxicology Program
IARC - International Agency for Research on Cancer
ND - Not determined
RPI - Rhone-Poulenc Established Exposure Limits

Disclaimer:

The information herein is given in good faith but no warranty, expressed or implied, is made.
                                         51

-------
        Appendix B




Summary of Laboratory Results
            52

-------
                                U.S. DEPARTMENT OF LABOR
                   OCCUPATIONAL SAFETY AND HEALTH ADMINISTRATION
                                  SALT LAKE TECHNICAL CENTER
                          PHYSICAL MEASUREMENT AND ANALYSIS BRANCH
                                       1781 SOUTH 300 WEST
                                    SALT LAKE CITY, UTAH 84165-0200
                                                                               (801J-487-0073 ext 272
                                                                                  FAX (801J-487-1190
12 December 1997
Report of analysis of samples for EPS, INC.

For smoldering point analysis, a volume of approximately 2mL of material was introduced into a
modified Setaflash© flashpoint tester. The tester was modified by having removed the cup cover so that
the test could be carried out as an "open cup" procedure. The material was heated up from room
temperature until smoking was observed and then until a visible change occurred to the material.

Sample J56676 (Azinphos) showed a color change to dark tan at 217°F. A further change to a dark red-
brown color occurred as the temperature was increased.  The sample showed visible smoke at
approximately 340°F.

Sample R68650 (Maneb) showed a small black spot beginning at about 320°F. The sample changed to
black increasingly as the temperature increased. Visible smoke was observed at 415 °F.

The smoldering temperature results reported  are given for sample test results performed under laboratory
conditions, and may not be representative of smoldering temperature value(s) resulting under differing
conditions.
Report of analysis of Samples V30707, V30708, V30709 from inspection of explosion at EPS, Inc.

Three samples were submitted for materials analysis on 6/4/97. They were assigned sample numbers as
shown below:

V30707 - Fiberglass exhaust north (hanging from the compressor line)

V30708 - Fiberglass Enclosure (burned remnant of fiberglass paneling).

V30709 -New, unused "Supersack"

V30707 showed continuous fiber fiberglass of diameter approximately  12 micrometers embedded in
white, brittle ash, consistent with a partially burned structural fiberglass panel.  It was unclear as to
whether there was any latent supersack material left. The majority (80%) of the material remaining was
continuous fiber fiberglass

V30708 was continuous fiber fiberglass. The diameter of this fiberglass was nominally the same as for
V30707 at 12 micrometers. However, the index of refraction was different reflecting a different source
                                             53

-------
for the fiberglass by the manufacturer.  Little other material was present other than a small amount of
powder adherent to the fibers.

V30709 appeared to be a way, fibrous synthetic material and was confirmed to be polypropylene by a
telephone call to Dorothy Hullett, an employee of Rexam Mulox, the bag manufacturer. Ashing of the
sack material at 500 °C produced a black residue in an aluminum pan. No such residue was noted on
V30707.
                                             54

-------
                          U. S. DEPARTMENT OF LABOR
                 OCCUPATIONAL SAFETY & HEALTH ADMINISTRATION
                            Salt Lake Technical Center
                              1781 South 300 West
                                P.O. Box 65200
                           Salt Lake City, UT 84165-0200
Memo
                                                                     801-487-0680
                                                                     FAX 487-1190
                                                                  XIIHK
                                                               ACCREDITED
                                                               LABORATORY
DATE:                  November 4, 1997

MEMORANDUM FOR:    Al Heins
THRU:

FROM:

SUBJECT:


Analytical Screening
Lois Moncrief, Jerry Schultz, and Mike Shulsky

Wayne Potter, David Arm itage, Richard Lawrence, Pat Hearty,
and Joanna Shulsky
Organic Division Analytical Results for BPS, Inc. and
Microflow, Inspection # 6-009401
Samples J56742 through J56752 and J56755 through J56760 (field numbers 521-01
through 04, 521-06 through 11, Oil 1, 522-Fan SW, 522-Fan SE and 522-01 through 04)
were screened by HPLC for various pesticides including: azinphos-methyl, maneb, topsin
(thiophanate-methyl) and carbaryl (Sevin).   Only azinphos-methyl and  maneb were
reported,  and their  values should be  considered as  semi-quantitative  only.  Gas
Chromotagraphy-Mass Spectroscopy (GCMS) on J56742 (field number 521 -01) confirmed
the presence of azinphos-methyl as well as some of its decomposition products, the major
one being  0,0,S-trimethyl ester of phosphorodithioic acid. However, GCMS is unable to
determine  if the decomposition products were present before analysis or created during
the sample analysis.  J56742, as well as J56744, J56745 and J56758 (field numbers 521 -
01, 521-03,  521-04 and  522-02  respectively) were  also run  by GC-FPD.  Both
azinphos-methyl and the major decomposition product mentioned above were observed.
The samples reported as 
-------
1) Mass Spectroscopy

Azinphos-methyl was identified in sample J56755 (field number522-Fan SW) by HPLC.
Azinphos-methyl could not be confirmed by GCMS, but the major decomposition
product, 0,0,S-trimethyl ester of phosphorodithioic acid, was identified.  However,
azinphos-methyl was confirmed in sample J56755 by GC-FPD.  The material analyzed
was primarily the yellow stuff off  of the fan. The other fan  (sample J56756, field
number 522-Fan SE) had very little material on it, and was only screened for pesticides
(not including maneb).

2) Atomic Absorption

An initial screening for soluble aluminum compounds was  performed  by  atomic
absorption analysis on samples J56700 through J56721 (bulk field numbers 1A, 2A,
2B, 3A, 3B, 4A, 4B, 5A, 5B, 6A,  A1, AR, and also wipe field numbers BPS-1 through
BPS-10). The compound, fosetyl-aluminum (Aliette Signature) is a soluble aluminum.
Preliminary tests with a standard indicated that this procedure could be used to detect
the presence of fosetyl-aluminum.  Unfortunately the samples all contained high
background levels  of soluble aluminum and the results  of this procedure was
inconclusive. Fosetyl-aluminum is used as a premixwith mancozeb.

3) HPLC

   A) Bulks

   These same bulk samples (bulk field numbers 1A, 2A, 2B, 3A, 3B, 4A, 4B, 5A, 5B,
   and 6A) were given new lab numbers (J56637-46) and analyzed by HPLC-UVfor
   azinphos-methyl (0300), maneb (M177), thiophanate-methyl (D347) and carbaryl
   (sevin, 0525).  No detectable amounts of these compounds were found in any of
   these bulk samples except 2B and 4B. Maneb was found on 2B and 4B.

   B) Wipes

   Similarly, the wipe samples mentioned in the paragraph above (wipe field numbers
   BPS-1  through BPS-10) were also given new  lab numbers (J56649-58 ) and
   analyzed by  HPLC-UV for  azinphos-methyl (0300).  Detectable amounts of
   azinphos-methyl were found on samples J56649,  J56650 and J56652 (field
   numbers BPS-2, BPS-3  and BPS-1  respectively).  These samples were each
   confirmed by peak wavelength  ratioing.

   C) Air Samples

   J57349-53 (Field numbers A-1, A-2, A-4, B-1 and BLANK) were air samples that
   were analyzed by  HPLC-UV for azinphos-methyl and also  a  qualitative HPLC
   analysis. No detectable amounts of analytes were found on these air samples.

4) Fosetyl-Aluminum Samples

                                  56

-------
These samples (bulk field numbers 1A, 2A, 2B, 3A, 3B, 4A, 4B, 5A, 5B, 6A, A1, AR,
and also wipe field numbers BPS-1 through BPS-10) have not yet been analyzed for
fosety-aluminum.  Fosetyl-Aluminum  is  not  compatible with the  HPLC or  ion
chromagraphy (1C) analysis. A method in the literature analyzed fosetyl-aluminum by
GC-FPD using a process of methylation. Several attempts were made to methylate
fosetyl-aluminum but our attempts were unsuccessful. So currently our laboratory does
not have a method for the analysis of fosetyl-aluminum.
5) HPLC and GC

   A) Repirator Mask Samples

   A respirator used in the BPS  Exposion was examined for contamination of the
   organophosphorus insecticide Azinphos Methyl SOW, the fungicide Maneb and their
   decomposition products.  The respirator used during the evacuation by John
   Fernirola was analyzed for any decomposition products. It was reported that John
   Fernirola went into the smoke in effort to extinguish the source of the smoke. This
   respirator worn by John Fernirola (E62546) appeared to be coated with smoke and
   looked like a likely candidate to find the decomposition products. When Maneb is
   heated, the major decomposition products are carbon  disulfide and ethylene
   thiourea.  Carbon disulfide is collected with  charcoal and ethylene thiourea is
   collected with a glass fiber filter.  The respirator used by John Fernirola was a 3M-
   5300 using a 501 organic vapor cartridge. The cartridge has a glass fiber filter on
   the outside and charcoal on  the inside. The glass fiber filter was  analyzed for
   ethylene thiourea and the charcoal was analyzed for carbon disulfide. A portion of
   the charcoal was desorbed  with toluene and analyzed by gas chromatography
   using a flame photometric detector. Carbon  disulfide was not detected. A portion
   of the glass fiber filter was extracted with water and analyzed by HPLC using a UV
   detector. Ethylene thiourea was not detected.  The outside of the mask was wiped
   with several glass fiber filters  soaked with methanol,  toluene and water. These
   wipes were  analyzed by GCMS for analyte identification.  GCMS  identified  2-
   naphthalenol, phthalate ester,  and approximately C14-C18 acids.

   A portion of the glass fiber filter covering the charcoal on the organic vapor cartridge
   was extracted with acetonitrile  and analyzed by HPLC  using a  UV detector.
   Azinphos-methyl was not detected on the glass fiber filter or the charcoal portion of
   the cartridge. Mass spectroscopy also analyzed samples from the cartridge of the
   mask.   GCMS identified limonene,  ethanol,  isopropanol, methyl chloroform,
   ethalfluralin,  aliphatic   C9-C10  (approximately),   terpene,  C3-C4  benzenes,
   dichlorobenzene, and siloxane on the charcoal from the cartridge. GCMS identified
   naphthalenol, dursban, 0,S-dimethyltetrachlorothioterephthalate, an ethyl ester of
   a long chain acid, phthalate  esters, and a couple of unidentified  compounds,
   including an unknown amine, on the glass fiber filter portion of the cartridge.
                                   57

-------
Results
Air Samples

          Azinphos-methyl
J57349               ND
J57350               ND
J57351               ND
J57352               ND
J57353               BLANK
Qualitative HPLC for other Compounds
          ND
          ND
          ND
          ND
          BLANK
Bulk Samples
Field# Azinphos-methyl
J56637
J56638
J56639
J56640
J56641
J56642
J56643
J56644
J56645
J56646
Field
J56742
J56743
J56744
J56745
J56746
J56747
J56748
J56749
J56750
J56751
J56752
J56753
J56754
1A
2A
2B
3A
3B
4A
4B
5A
5B
6A
Maneb Topsin
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
.03%
ND
ND
ND
.6%
ND
ND
ND










# Azinphos-methyl Maneb
521-01
521-02
521-03
521-04
521-06
521-07
521-08
521-09
521-10
521-11
OIL1
FGLAS EXN
FGLAS END













9.0%
.06%
.3%
.2%
ND
.003% 2.0%
.002%ND
.01%
ND
.004%ND
NA
NA
NA
4.0%
.06%
.08%
.08%
ND


.1%
ND

NA
NA
NA
                                             Carbaryl
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                                             ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                         ND
                                58

-------
J56755
J56756
J56757
J56758
J56759
J56760
Field
J56700
J56701
J56702
J56703
J56704
J56705
J56706
J56707
J56708
J56709
J56710
J56711
Wipes
Field
J56649
J56650
J56651
J56652
J56653
J56654
J56655
J56656
J56657
J56658
Field
J56712
J56713
J56714
J56715
J56716
J56717
J56718
J56719
J56720
J56721
522-FAN
522-FAN
522-01
522-02
522-03
522-04
#
1A
2A
2B
3A
3B
4A
4B
5A
5B
6A
A1
AR

#
BPS-2
BPS-3
BPS-7
BPS-1
BPS-5
BPS-6
BPS-4
BPS-8
BPS-9
BPS-41
#
BPS-2
BPS-3
BPS-7
BPS-1
BPS-5
BPS-6
BPS-4
BPS-8
BPS-9
BPS-10
SW 1.3% ND
SE ND NA
.006%. 06%
.02% .001%
.01% .03%
.02% 5.0%
Aluminum (as Al), Soluble Salts
5.0%
9.0%
8.0%
7.0%
6.0%
9.0%
9.0%
6.0%
8.0%
9.0%
.2%
.8%

Azinphos-Methyl
46.0 //g
43.0 //g

-------
Respirators

      Field #      CS2   Ethylene Thiourea Azinphos-methyl   Maneb

E62546     EPA001     ND         ND               ND         NA
HPLC and GC analysis by Dave Armitage and Wayne Potter
AA analysis by Richard Lawrence
GC-Mass Spec, analysis by Pat Hearty and Joanna Shulsky
Compiled and written by Wayne Potter
                                    60

-------
 DET               (U£A.), INC






                                                                   HuuKS*. T»l» 77W4 USA
                   August 20, 1997

                   A. C.                PE
                   CWI, AK-5 JO,      Level HI

SUBJECT:
                   ONV       No.
Introduction
On            and Thursday,      1 1 in aid 12th, 1997. 1
EPA,                  asd Jay Fills,      from OSHA,  10 the            of 4e site snd
                             ia the           of tie BPS                     to
                 This     an        (hat                                        Tlte
              in this      WM to          the              I     air               to
the         and to        its            in the
            site            with            and                              follow.
      on the          of d»                     sad             I          that the
          of        was not the       of        The                   in the
       however,                         to           the
       the                         ia ibc           of            ps.  The
                   pe, if it          to be fed by the                    be          (o
                   »     or            of

The                      wttf      »t the
I,  Two                      air            wwe involved in the fiie,  Cfoe    t 15-bp
            had                                  the     on*     a 2i-hp unit,     had
                        The                          is       in      1 *

2,  The         on the                         the                                 in
         2, had          heat        tad        in t                       that »
   had       on tte       in          This    typified by  tfce             of        ia
   the joint in 'the
3,                                      the               by A
                                            61

-------
4   Ik? iv, o «imptewof discharges fed a cornoiosi header pips t.'ial was seated on '.he warehouse
    Side of Ike wall  approximately Jue fee; above the deck.  This header  led to mi aiieTtcyltT
    O'ii*iib the -juiliJinsi.  UK output frori  tlv ;ificrtoole: was             back  i-onp tlic same
    ftiill closer to Ilu* d«A;k carrying tin:  cookd JM bsck to she acxn-TiuIasar tank* utjdcr catb
    asrnprGbS-or.
f .   The- :5-rrp cyiTjpTessur, found OK its Sid* Csfter ihe fire, -.v;u rep«ted to  have to;i'-ai:i«-I na
    lubricating oil in fhcctKikc.*^ after ih-e  lire.
6,   The 20-lip compressor remained xiprighl ir= its operaiirig pDjiuen. snc suHicien: cil  was tbund
    The liluri ilium bttt housings fcr ihc elcciric motor and t?K aluminum inJec header for the firsi
    5*2 sic on ihe 15-hp cunijWissof  liad  rnticcd away.  The melted r&idue liad !K-«I deptMitc-d o;i
    the -eaginc and compr^sior tno'.ird plssSimi imiiieduidy hirlo* the siKiior when u wi:* still ::i
    lha upriaht iiG-sibon. 'fae ITW*CT snc cjimpressor crickcai.es are >Jiown so Fig5!res •• fti'd 4.
S.  The condiucn describe*.;! "m item ^ ksuicatei dial liic cc-rnp'trnsor was sxposec to heit (likeK
    radtaiu Ironi abovej beisre felJiag on. is side,
9.  The pulley sid-J of the ajnipri-^sor liad sasiamcci dirv^t flane impingement heat, but      *as
    Dbsei-ved or; tae opposite isdc.  Figures 5 &KI! 6 deptci ttis condition,  Tb*         areas
    i.n] of melted plastic when the compressor
    fell over on ils side.
11.1J}s'?t»  dismaiitling of  the compressor to examine the bearings-.  \' was obserwd  ihm ihe
    coaTjCtiiim rod ;ir,d iourtial  bearinp lud not been scoretl.  This iKdjcaed diat tic unit nad
    suliiciiT;' lubriccju^n wben last ran.
12. The        in itw cap half  of the cctrmect^g r«d  beafiiig        ihe pulley (Figure '?!  Had
    partially rndierl away,  arai shuwed M>F« firesh, bright scoring sn ihe rod  toll'. Hie  briahlness
    inditaies tjiat lilts scoring resulted s^bscijucrit lo "lie  f-re. 8'id  was likely caused by niatiu,;;
    Hjuilsui! v?i!,c jjullcy liui'jn.c inv«-su2»sticci.. Otbirwi-rM, it .iho^-o«3 flo darnasc.
13  The '•*•«) jounial "tj-eannts aie sho'*ii in Flexes SO and 13.  It is po-susblc thai joiriC of ihc
    crank case oil could have leaked uai af 'Llw bearing ;icxi io ths pullej- w'hile Use coiupres^or
    wa# t.»n tis s.K.b.
14. The first siage aluminum p:sc*in bad c»lwd awsy leaving i.inly part of ilid skin along tine
    cyiimler wall   |:ipre  [2 is  a            tskcrs after ilia      had      removed,  Nwle  tfe
    di€ wrist pin hat; ?kipp»J 10 ihe lei: itovurd the pulley side af the oonipresso?).  Sufficient
    huat h?.i? to btf present  itisice the craokca?*: sfter i"        ov*r to expsnd the aluminum alkn
    corinectiRg roc  suffiei5" itie crankcu>e.
    fiffxc  13  is :i v?e\v  looking Irons th-e oppijs;!e side of the pulky.
                                             62

-------
16. An aceuroalatioti  of coke-like  residue was found inside ihs ttarikca&e taexauiig  that a
              o»] fire hat! developed jnside. This h assumed to      b«».        by she liquid
   fire on the deck after die compressor fell over. It is     probable ctef -soaiii of the oil leaked
   Q-M        she pulley Aide journal        and vOfttributed to the plasiic tire
17. Disas.<.scil»Iy of the 20-hp vompressur issdscaied  no ofcservabk  ic'ental  dataap.* to  Uie
   bearings or other
                                        63

-------
      TECHNOLOGIES IMC
                                                                  !*
                                     Safety « Heollh • EnviiOfmar.t
                                      sjrrrf £.nvK)?:ffmt',t3i

                                    CON FIDENTIAL
August 31, 1998

To: Mr.                         !Cf, 9300 Lee Highway, Fairfax, V'A 22031-1267
    Ms.       Faenles, EPA, .Mail Code 5104.401 M      SW, Washington DC
                      f^"
Frew: Arthur Miller, William Fenion, SHE Technologies, 1050 Gravel R6j-673'»fl80. FAX: {7!6t-6?M34.1
Subject: SPS                                -  Final       of Test Results and
Corwl

INTRODUCriON

In aur earlier             of U27/98. >we recommended that On I worth TeeS»okhj% Inc.
perform "Basket Tests'* lo determine safe storage temperatures fo-r bulk rmatenals The
        of (he      was iwofoSd: to tubiain             stability datis !oi key materials
known to he in the BPS vtarehousc,    to prove a: disprove, if possible, any ol ibc ihrec
          root      decomposition          The ke)1 rtiatenaJs in the warehouse were
AZM (Azsnphos       SOW Soluble) and       75DF The scenarios were
              of AZM prior to            IB the          decomposition of AZM
       by exposure to the compressor       line, or decomposition of Maitea
by         to the                  I me,



1. AZM

         DSC    were perfanwed on AZM by the EPA's NBJC,. resusts reported on
12.''30.-^7 tndiaiieiJ an average  (exL'ap€iIsed)onstt temperatMns of apptoximate'y 16? C
The          performed by Qulworth Technology on AZM are .surninanzeil m their
report "Process Safety Tesl Results mi            for AZN-f          Methyl SOW
Swlwbte), Report                             June 30,1998)", a copy «r*hich has
         to 1C?       Imemation-al, lac. 'The test             is outlined m their
report     will not 'be         in this report.
The       of the 100 mm       test m the Chi I worth study showed ssn
                   7041) C;                                     would be
         to show even lov»er tei-nperatufes. For example, using the proei'siure o«l lined in
Chilwurth's report, artd the      of the       «f the         co-ntamec to the surface
area of the          an                      safe                  of 29.6 C was
   730 Hlghtower Way * «»twi!(H New vork      * to!; (716)
                                        64

-------
                         BUSINESS CONFIDENTIAL

        when      the volume and surface area specifications fo-r the AZK container,
as specified by Ms, Awiltia Fuentcs.      tfwl this               not
3 safe Juste interval corresponding lo this temperatwre. The extrapolation tor AZM not
      in Chi 1 worth's original report referenced        is summarized in attachment i
{plot of onset temperature vs.            This                  be         to
Chilworth's A2M

Pre-1 urinary      were also performed by Chilwonh on AZM lo determine if ihe material!
coe'd be safely          lo dust explosion hazard testing. These tests, consoling of an
ignition tube and flat      heating wsis, both       by an open flame, stowed
decomposition CM' the sample accompanied by ihs e volution oif ycl iowis'h srnofcc and
itamrrtihle vapors, hi an BPA draft summary (Pages 12 and i 3> prepared by Ms. Ku-cn-.es.
several BPS           » the same       of lowetm(«
i>f QKI&A teraperacun; relative lo DSC      as       by AZM. In sirttslw Fashion to
A7\1, an                                leroperiiiure o ( 83.1 C? w« obtained for
        The calculation*         the              speei.fkalions for       as used
fcr AZ.M.      tha? this           ss s.igmlietfttly higher than the 29.6 C          for
AZM, The            tor       not      in Chttwwth's report referenced above, s
$vmmafi/c<3 in sliifihrnent 2 (plot of onset           vs.       si:ze).  This atiachmcnt
      be          lo Chilworth's       report.

            and flat     heating     performed by Chilworth on Ma neb showed white
vapors were evolved fro-m the                      The  vapors evolved were
                                      65

-------
        CO
        o>

       -*-•  ^
        /IS  ^
        CD  v>

                              «
                           1?
                           2 b
                              §.
                             co

                             tfc.
3  8
                           eo
                           c

                          -i 2
                           » s
                           « E
                          03 2
                          •35
                          S '^
                           11
                          •g'3
                          la
                                                                          iq
                                                                          9

                                                                            O)
                                                                            o
                                                 •5

                                                 :>
                                                                         in
                                                                         CN
                                         66

-------
                            Differential Scanning Calorimetry Results
               Root Cause Investigation of Explosion at Bartlo Packaging South, Inc.
                                    NEIC Project SPOO30
INTRODUCTION

    On May 8, 1997, an explosion took place at the Bartlo Packaging South, Inc. facility in West Helena,
Arkansas. NEIC is assisting the joint EPA and OSHA investigation by providing information obtained by
differential scanning calorimetry (DSC) on the decomposition of two pesticides. The DSC results give
the amount of energy released by decomposition and the temperature at which decomposition occurs.

BACKGROUND

    The two pesticides analyzed by NF-IC are Azinphos-methyl (AZM) and Maneb. Eight samples of
AZM were received from the OSHA lab. Two samples of AZM were sent directly to NEIC by the
Agriculture Division of Bayer Corporation in Kansas City, Missouri. These are labeled Guthion, which is
Bayer's name for A.ZM. One is Guthion Technical, which is 93.6% AZM, and the other is Guthion 50%
Wettable Powder (WP), which is 50% AZM. The Guthion Technical is from Batch No. 7030105, and the
Guthion WP is from Batch No. 7030063, according to the information supplied by Bayer. A sample of
Maneb 75DF was received from the OSHA laboratory in Salt Lake City. It was packaged in 23 separate
bottles for shippino, purposes. Three of the bottles were analyzed by DSC.

    When a material undergoes a chemical change such as a decomposition, heat is either absorbed or
released. Often decompositions are initiated simply by raising the temperature of the materials.
Differential scanning calorimetry is a method for measuring the heat released or absorbed during a
decomposition or other reaction. The differential scanning calorimeter gradually increases the
temperature of a reference cell and a sample cell. If a reaction of the  material in the sample cell takes
place which either releases or absorbs heat, the DSC measures the amount of heat involved and the
temperature at which the reaction starts, called the onset
                                           2 of 10
                                             67

-------
 temperature.  The heat released or absorbed is measured in Joules per gram (J/g). In order to put
 the reported values in perspective, decomposition energies for a number of common explosives
 and other compounds are given below:

                     TABLE 1 Decomposition of Common Materials
, Material
Trinitrotoluene (TNT)
Ammonium nitrate
Dibenzoyl peroxide
Cellulose
Decomposition energy (J/g)
1690
1600
; 1400
330
 Reference: Theodor Grewer, 'The Influence of Chemical Structure on Exothermic
 Decomposition? Thermochimica Acta, 187 (1991) 133.

 RESULTS AND DISCUSSION
       The results of the DSC analyses are shown in Table 2. DSC scans for all four of the
 formulations are shown in the attached figures. The two types of A2M received from Bayer
 were each analyzed in triplicate. The agreement between the triplicates was good, as shown by
 the standard errors of the means in Table 2. Relatively pure AZM (93.6%) decomposes
 exothermically (1100 J/g) at 164 degrees Celsius. The Bayer Guthion 50% wettable powder
 decomposes exothermically (600 J/g) at 168 degrees Celsius. The smaller amount of heat
 released by the 50% powder compared to the technical grade AZM is consistent with dilution by
 an inert ingredient. The eight samples of AZM received from the OSHA lab were each scanned
 two times. Since no differences were observed between the different samples, the results were
 averaged.  The included figure for the AZM from OSHA shows two scans of the sample labeled
 J56673. One scan shows a well shaped Gaussian curve and the other shows a poorly shaped
 exothermic curve due to heat and mass transfer effects. Only well shaped curves were averaged
 together for the results in Table 2 (6 scans in all). All sixteen of the scans show a considerable
release of energy of about 600 Joules per gram.
                                      3  of 10
                                           68

-------
       In contrast to AZM, Maneb 75 DF does not release a significant amount of energy when it
 decomposes. In the temperature range of 170 to 210 °C two processes occur; one is exothermic
 and the other endothermic. This is evident in the three scans for Maneb which are shown in two
 figures. Treating these scans in the standard manner with;the instrument software gives the onset
 temperature and heat released shown in table 2. On average, there is a net absorption of heat, not
 a release.  As shown in the figures, the differential scanning calorimeter data for Maneb 75DF is
 not as reproducible as it is for AZM. Nevertheless, it is clear that little or no energy is released
 by the decomposition of Maneb 75DF, which starts slightly over 170 °C.
                , TABLE 2  Decomposition Temperatures and Heat Released
Sample
Bayer Guthion Technical
(93.6% AZM)
Bayer Guthion 50% Wettable
Powder
AZM from OSHA
Maneb 75 DF
Temperature (° Celsius)
(a)
163.6 (0.3)
168.0 (0.9)
166.6 (2)
183(5)
Heat Released (Joules/gram)
(a)
1096 (40)
596 (50)
519 (6)
-22 (30)
a: Standard error of the mean in parenthesis
EXPERIMENTAL
       A Perkin Elmer DSC 7 calorimeter was used for the measurements.  Samples were placed
in screw top stainless steel capsules sealed with a gold plated copper seal. Sample sizes ranged
from 1 to 6 mg, with most in the middle of that range. Samples were scanned from 50°C to
200°C at a scan rate of 10 degrees per minute.  The instrument was purged with nitrogen during
all measurements. The instrument was calibrated with indium, both for the onset temperature
and for the energy released. Results were plotted so that an exothermic reaction gave a
downward peak, and an endothermic reaction (such as the melting of indium) gave an upright
peak.  Instrument software was used to calculate the onset temperature and the heat released.
                                       4 of 10:
                                        69

-------
70

-------
Uncertainty in determining the baseline contributed to imprecision in the heat release data. The
DSC data for AZM and Maneb were not as reproducible or accurate as the data for the standard, which is
the melting of metallic indium. There are several reasons why this occurs. Organic substances do not conduct
heat as well as metals. Decompositions which produce gaseous or volatile products, such as those of AZM
and Maneb, show more variation.
                                            5 of 10
                                              71

-------
Appendix C




 References
    72

-------
                                         References

             ATSDR, ERSAB, After Action Report, EPS Pesticide Fire, August 4, 1997.

     American Institute of Chemical Engineers, Center for Chemical Process Safety; Guidelines for
               Preventing Human Error in Process Safety, ISBN 0-8169-0461-8, 1994

   American Institute of Chemical Engineers,  Center for Chemical Process Safety; Guidelines for Safe
              Storage and Handling of Reactive Materials, ISBN 0-8169-0629-9, 1995

   American Institute of Chemical Engineers,  Center for Chemical Process Safety; Guidelines for Safe
                      Warehousing of Chemicals, ISBN 0-8169-0659-9, 1998

   Bertoni and Co-workers, Annali di Chimica 76; Lazioni Commerciali inAmbienti Refrigerati, 1985

  Det Norske Veritas (USA) Inc, Memorandum Report; Insecticide Warehouse Explosion Investigation,
                          DNV Project No. 232-8384, August 20, 1997

EPA Office of Prevention, Pesticides, and Toxic Substances; Product Properties Test Guidelines OPPTS
                          830.1000, 830.6316, 830.6317, August 1996.

    Fire Safety Journal 23,1994; Smith-Hansen and K. Haahr-Jorgensen; Combustion Products from
             Pesticides and Other Chemical  Substances Determine by Use of DIN 53 436

  Less, Frank; Loss Prevention in the Process  Industry, Second Edition, Butterworth-Heinemann, 1996

  Mississippi Department of Environmental  Quality; A final Engineering Report for Ambient Air quality
                 Monitoring for the BPS Inc Fire,, Jackson, Mississippi, May 1997

                       NFPA 43D; Code for the Storage of Pesticides, 1994

                         NFPA 231; Standard for General Storage, 1995

NFPA 472; Standard on Professional Competency ofResponders to Hazardous Material Incidents, 1997

             NFPA 1561; Standard on Fire Department Incident Command System, 1995

  Zi-Ru Liu et al, Thermochimica Acta 220 (1003) 229-235; Heat Changes Associated with the Thermal
                             Decomposition ofManeb and Zineb.
                                             73

-------