United States Region 4
Environmental Protection 345 Courtland Street, NE
Agency Atlanta GA 30308
November 1980
904/9-80-062
&EPA Chemical Fire At
Drexel Chemical Company
Memphis, Tennessee
-------
PREFACE
This report reflects the activities of the Environ-
mental Emergency Branch (EEB) Region IV, EPA, as it was
concerned with the captioned event. The report outlines
the key actions of the Federal On-Scene Coordinator, the
Regional Response Team (RRT), and state, local, and Federal
members of the RRT. It reflects the complexity of respond-
ing to environmental emergencies and the need to coordinate
and plan in advance for a major incident such as this event.
This report is intended to satisfy the requirements of
the National Oil and Hazardous Substances Pollution contin-
gency Plan and to help others learn from our experiences.
Environmental Emergency Branch
library/region n
r-kunonMWlFNTAL
PROTECTION agency
-------
TABLE OF CONTENTS
SECTION PAGE
1. SUMMARY 1-1
2. THE SITUATION 2-1
3. PHOTOGRAPHS 3-1
4. SUPPORTING MATERIAL 4-1
-------
1. SUMMARY
At 9:15 a.m. on July 5, 1979, an explosion and fire occurred at
the Drexel Chemical Company, Memphis, Tennessee, which resulted in the
evacuation of numerous individuals along the path of the billowing
smoke. The presence of Class B poisons created the possibility of
significant public health and environmental impact. Local officials
evacuated the threatened areas while fire officials utilized minimum
amounts of water in an effort to contain the blaze to the production
plant. Despite these efforts, large amounts of contaminated water ran
from the plant site to a nearby drainage culvert, Mai lory Bayou, which
flowed to Nonconnah Creek and into McKellar Lake. City officials
acted quickly to dam the drainage at Horn Lake Road, thus containing
the contaminated water. Local contacts indicated that no air monitor-
ing took place during the initial fire.
When EPA arrived on-scene the air pollution problem was essen-
tially over, but several other situations had developed. Large
amounts of vapor had precipitated out of the dense smoke and left
large yellow droplets along its path. The plant itself was a complete
loss and presented a special cleanup problem because of the Class B
poisons. The contaminated water was contained at Horn Lake Road, but
was already in large quantities. The containment area was a flood-
plain for drainage from the South Memphis area, which made isolation
of the contaminated water virtually impossible.
1-1
-------
A Regional Response Team (RRT) meeting was held and potential
problem areas were discussed. Nearly every local agency was repre-
sented along with selected state and federal agencies. Task assign-
ments were made as follows:
FDA - commercial products
Health Department - residential areas
Poison Control Center - human exposure
EPA and State of Tennessee - plant site and contaminated water
Human exposure and chemical fallout were found to be minimal.
Plant site cleanup presented difficult solid waste questions concern-
ing neutralization and ultimate disposal. Tennessee Solid Waste
Control worked closely with the company to minimize disposal costs.
During the initial operation, the company seriously considered bank-
ruptcy versus committing to substantial cleanup costs that seemed
open-ended at the time. This was outweighed by the company's desire
to clean up their "mess" and continue in business.
Disposal of contaminated water presented a special problem. Sam-
ple analysis showed low levels of poisons but bioassay work indicated
the water was very toxic to aquatic life. The volume and concentra-
tion of the water made carbon filtration cost-prohibitive, but dis-
charge to McKellar Lake was equally undesirable because surrounding
communities used it for water sports and fishing. The solution was
limited treatment through the Sewage Treatment Plant (STP) and subse-
quent discharge into the Mississippi River. This proved to be an
acceptable method and resulted in the disposal of an estimated 14
million gallons of contaminated water.
Cleanup of the plant site lasted about three weeks using a com-
mercial contractor, and company officials seemed pleased with the
results. The company also entered into a cooperative effort with the
Poison Control Center to investigate degree and effects of fallout on
soil and vegetation. The work was funded by Drexel Chemical and
conducted by the Poison Control Center and a local university.
Chemical/Physical Characteristics
¦Drexel Chemical was predominantly a formulation plant for various
pesticide and herbicide mixtures used in agricultural applications
1-2
-------
with very little chemical reaction. The company handled three Class B
poisons: Methyl Parathion, EPN, and DNBP (Dinoseb). Other chemicals
in inventory presented a reduced concern when compared to the poisons.
Some surfactants, emulsifiers, etc., were present, which may have
added significantly to the toxic effects on aquatic organisms but were
not felt to be toxic in general.
Documentation of Significant Events
Initial concerns centered on the possible air pollution problems
resulting from the heavy smoke at the plant site. The smoke traveled
in a southwesterly direction, crossing the Mississippi River into
Arkansas over open fields and residential and industrial areas. Evac-
uation had minimized human exposure, but officials were faced with
determining a safe time for people to return home. Jim Littell of
EPA Air Monitoring checked the downwind area late in the evening of
July 5, and determined that any remaining air pollution threat would
exist only around the immediate area of the plant site.
Protection of human health became the first priority in this
incident. At the RRT meeting, an effort was made to determine all of
the potential health hazards resulting from the fire. Task assign-
ments were given to various groups and this information was distri-
buted to the news media by means of a press conference directly after
the RRT meeting. Agency names, functions, and phone numbers were
distributed to enable citizens to contact them directly. Information
about decontamination of household and outdoor items was also distri-
buted to the press after being discussed by EPA, State, Health Depart-
ment, and Poison Control Center officials.
Drexel Chemical officials looked to EPA and state personnel for
guidance on the necessary cleanup procedures. Initial speculation on
disposal costs caused the company to seriously consider whether it
should even begin cleanup operations if, in fact, they were going to
"break" the company. Lawyers and insurance agents advised company
officials against entering into an open-ended effort. The OSC worked
with the company and state officials for one and one-half days to
1-3
-------
devise a cleanup plan that was environmentally sound yet within reason
on cost. This effort allowed the company to clean up its problem
instead of laying the burden on the taxpayer.
Water decontamination posed one of the most serious problems.
The water was in a floodplain and could not be isolated. Until signi-
ficant cleanup occurred at the plant site, additional chemical runoff
and large volumes of water would collect in the holding basin. Acti-
vated carbon filtration was considered, but estimates indicated that
it could run as high as $500,000 to filter what appeared to be low-
concentration contaminated water. This did not seem reasonable if an
alternative could be found. Neutralization was tried using caustic
materials but it was not felt the water would be suitable for dis-
charge to McKellar Lake afterward. Discharge to the lake was undesir-
able because of its use for water sports and fishing, and also because
limited flow through the lake would not permit dilution and disper-
sion. These factors led to the decision to pump the waste through the
STP for limited treatment and discharge into the Mississippi River
where its toxic effects, if any, would be minimal.
The logistics of pumping all of the liquid to the STP proved to
be difficult because the plant was able to handle only a limited
amount (l%-5%) by volume without affecting the bacteria. At these
pump rates, it would have taken an estimated two weeks to empty the
pond. Any rainfall would significantly extend that period of time.
Finally, the decision was made by the 0SC and state to bypass the
treatment at the STP and pump as much contaminated water as possible.
This method posed a second problem: availability of large-volume pumps
because most were tied up irrigating crops. The city of Memphis
assisted by loaning a 10-inch pump from the STP. This proved invalu-
able in eliminating the contaminated water and allowing the drainage
canal to be reopened.
1-4
-------
2. THE SITUATION
Response and Observations: Chronological Events
July 5, 1979
An explosion and fire occurred at the Drexel Chemical Company,
Memphis, Tennessee, at 9:15 a.m. At 4:00 p.m., Jan Rogers, who had
been designated On-Scene Coordinator, arrived and was briefed on the
situation.
The initial fire caused significant air pollution concerns. SSW
winds blew the smoke across the Mississippi River and into Arkansas.
The river was closed to vessel traffic until 3:00 p.m. Civil Defense
had evacuated the area along the path of the smoke as a precaution.
The U.S. Coast Guard (USCG) contacted Ergon, Inc., and requested that
they put a boom across Nonconnah Creek to contain any floatable pollu-
tant runoff from the plant site.
The OSC, USCG, Civil Defense, and Ergon surveyed the plant site,
drainage ditches, and Nonconnah Creek to McKellar Lake. There was
still some fire in the building but the dense smoke had cleared.
Firemen were using minimal water to contain and extinguish. There
were large amounts of contaminated runoff from firefighting and broken
water lines. An earthen dam was containing contaminated water at Horn
Lake Road. The boom on Nonconnah Creek was not collecting any pollu-
tant .
2-1
-------
At 6:00 p.m., the OSC met with Drexel Chemical Company President
Robert Shockey and learned that the highly toxic chemicals involved
were Methyl Parathion, EPN, and DNBP. It had been determined that an
explosion in the "Hot Room," where the poisons were stored, had caused
the fire. The cause of the explosion was not known. Most of the
chemicals in the building were in liquid form. It had been recom-
mended to Drexel that contaminated runoff be treated with caustic to
neutralize the poisons.
Chemical representatives believed that in a hot fire 95% of the
chemicals would be destroyed. A dam had been installed by city per-
sonnel at Horn Lake Road to contain the contaminated runoff. Shelby
County Health Department officials were sampling Nonconnah Creek to
determine the extent of pollution below the containment dam. By 6:00
p.m., evacuated residents had been allowed to return to their homes.
The industrial area south-southwest of the plant was still evacuated.
Mr. Shockey agreed to provide a list of the inventory of the building.
Jim Littell and James Bricker of EPA Air Monitoring arrived at
6:30 p.m. and looked over the situation. EPA Emergency Response Team
(ERT) and Region VI personnel arrived on-scene and met with the OSC
and Mr. Littell and Mr. Bricker to discuss the incident. A Regional
Response Team (RRT) meeting was scheduled for July 6.
The OSC toured the plant, containment, and creek areas. The
company sampled behind the containment dam.
July 6, 1979
The Command Post was set up at the Drexel Chemical office. The
OSC contacted Jack Stonebraker, EPA in Atlanta, and briefed him on the
situation. The OSC requested that he contact EPA solid waste person-
nel and relay the potential disposal problem.
The OSC surveyed the drainage ditch and Nonconnah Creek. The dam
was holding all contaminated runoff in the drainage ditch. Nonconnah
Creek had very low flow and there was no visual indication of the
chemical reaching the mouth of McKellar Lake.
At the plant site, the fire department was hosing the building.
Sample results indicated low levels of contaminants. The OSC obtained
the inventory list from Mike Smith, the vice president of Drexel.
2-2
-------
The RRT meeting was held at 10:00 a.m. The OSC advised the group
of sample results and the status of the situation. The fire depart-
ment was preparing to bulldoze part of the building to permit fumes to
escape and limit the amount of water needed for smoldering areas. The
air pollution problem was discussed. Concern was expressed about
fuming in the plant. It was decided to advise that the trucking com-
pany keep personnel in the building and out of the section of the yard
downwind of the fire. It was felt that the air dispersion problem
downwind did not present a real danger.
Contamination of cars, gardens, homes, lawns, etc., was dis-
cussed. The Shelby County Health Department was given the assignment
to check on private facilities and FDA would check commercial facili-
ties. A statement was to be released that would advise on procedures
for decontaminating household and outdoor items.
Since some contaminated runoff may have reached McKellar Lake,
the Health Department preferred to continue the restriction on water-
contact use of McKellar Lake until further analysis could be done.
The fire department was advised on how to decontaminate their hoses
by washing with soda ash and rinsing with clean water. The Health
Department would do sample analysis on the commercial list if the need
arose. It was decided that each agency representative would discuss
his agency's area of expertise with the press. The meeting with the
press took place at 11:45 a.m.
The OSC visited the plant site at 12:30 p.m. Contaminated water
was still running from the plant site. The OSC instructed Mr. Shockey
to block all drainage from the plant site and attempt to shut off all
sources of water to the building. All fire was out. There was sub-
stantial flow in the drainage ditch. The company would develop a
plant sketch showing where material was stored.
The OSC contacted Ergon to remove the boom from Nonconnah Creek. '
Jack Stonebraker relayed the information that the EPA pesticides lab
in Bay St. Louis, Missouri, would do chemical analysis over the
weekend if needed.
An estimated 25-30,000 pounds of soda ash had been added to the
runoff at various points to neutralize the chemicals.
2-3
-------
Ben White of the Health Department called to report that his
people had not found any significant problems of chemical fallout or
fly ash in their visual inspection of residential areas. Reports from
the TVA power plant indicated that the cloud had traveled north and
west of the facility.
The OSC and John Gilbert, EPA ERT, visited the plant site and
found that the company was starting to move undamaged drums from the
plant and was spraying with caustic to decontaminate.
Mr. Cherry of TVA called and reported that he had taken samples
on July 5 and 6.
July 7, 1979
At the plant, the city would reinforce the dam at Horn Lake with
stone and asphalt at the state's request. The state estimated that
there were two million gallons of contaminated water at the Horn Lake
dam. The Horn Lake area is a floodplain and there was no way to con-
trol continued water collection. It was not considered feasible for
the treatment plan to stay ahead of water accumulation. The approach
to be used at the pond was to aerate to oxidize chemicals, neutralize
to pH 9, check for 1 ppm levels on Class B poisons, allow water to
build to dilution-level, and discharge several batches or one large
batch.
An RRT meeting was held at 10:15 a.m. Health Department person-
nel had found minor amounts of charred material or yellow spots on
vegetation. Fallout was considered minimal. There were no reports of
illness, but the contractor working around the levee reported that
some of his personnel were sick. FDA checked the trucking companies
and stated that sealed trucks or cars should not have been damaged.
The state presented results from samples taken July 5.
Samples were taken at the plant and at the Horn Lake Road dam and
sent to Bay St. Louis. Mike Higgs of Tennessee Water Quality Control
called with the results of bioassay work. He felt that toxicity tests
looked bad and that caution should be used regarding the release of
contaminated water.
2-4
-------
The OSC met with representatives of Drexel and OH Materials to
discuss how to approach treatment.
The OSC briefed A1 Smith, EPA Region IV Environmental Emergency
Branch Chief. The OSC also talked with Waynon Johnson, Department of
Interior, Fish and Wildlife Division, who felt that bioassay work
would be beneficial because it would show how the contaminated water
affected the ecosystem.
July 8, 1979
Pond pH 9.4-9.5, Bioassay 1 a.m.-8 a.m. 10% and 5% dead.
The OSC visited the plant and talked with OH Materials and the
company. The company had committed OH Materials to handling the
contaminated water problem. They were considering cleaning up the
plant themselves. The OSC informed the company that this was their
option within limits but that cleaning up contaminated rubble would
involve far more exertion and danger than normal work in the plant and
some employees would probably rather quit.
The OSC talked to Waynon Johnson, who said that toxicity was
probably due to surfactants, alcohols, emulsifiers, etc., since sample
analysis had shown only small quantities of chemicals.
At 12:50 p.m., George Sand of the Bay St. Louis lab reported a
rough analysis:
In the sample behind the dam:
Methyl parathion
Atrazine
EPN
DNBP
0.1 ppm
0.6 ppm
Non-detected
Non-detected
Samples from the drainage culvert at the plant site:
Methyl parathion
Atrazine
EPN
DNBP
Heavy Concentrations
2-5
-------
A meeting was held with the company. Terry Cothron of Tennessee
Water Quality Control outlined four alternatives for moving the water
as follows:
Put it in holding pond and wait. This would buy time but do
nothing.
Leave material where it was and explore alternatives.
Put material in sewer.
Do nothing.
At 5:45 p.m., George Sand reported the following:
Dam Sample:
Methyl parathion 0.07 ppm
Atrazine 0.61 ppm
EPN 0 ppm
DNBP (Dinoseb) 0 ppm
Ditch Sample:
Methyl parathion 1.0 ppm
Atrazine 101.5 ppm
EPN 0.41 ppm
DNBP 0 ppm
The bioassay unit arrived on-scene and was moved to the Nonconnah
pump station.
The 0SC met with company officials. They had talked with OH
Materials and were concerned with the costs. Lawyers and insurance
personnel advised the company against committing to an open-ended
cleanup; they would prefer to let the government clean up the site.
The company wanted to stay in business and was concerned with its
reputation. Company spokesmen said that they would like to clean up
the site if it were economically feasible. Representatives from OH
Materials stated that they would be ready to aerate the pond on July
9. The 0SC discussed with Drexel the approach to cleaning up the
plant site.
2-6
-------
July 9, 1979
The OSC met with:
Maynard Stiles - City of Memphis Public Works
Dr. John Smith - City of Memphis Public Works
Terry Cothron - Tennessee Water Quality Control
Wayne Max - Tennessee Water Quality Control
John Gilbert - EPA ERT
Alternatives for moving the water and other factors involved in
the cleanup were discussed.
At the Horn Lake Road dam:
OH Materials was moving pumps into place.
OH Materials was clearing dirt from the top of the dam, which
completely shut off the culvert and would prevent overflow in
case of a flood. City personnel requested that it be left that
way and they would break the dam if necessary.
More vegetation appeared to be dying in the pond and along the
edges. It was probably from the defoliants.
Reporters were on-scene and interviewed the OSC.
Harbor Patrol reported 30-40 fish dead at McKellar Lake.
The OSC told Bob Shockey to make a commitment to OH Materials to
start cleaning up the plant or he would take over and perform it as a
federal project. Mr. Shockey expressed some reservations about cost
but agreed to hire OH Materials.
The OSC contacted EPA Region VI and briefed Dick Peckham on the
possible use of the Sewage Treatment Plant (STP) to discharge contami-
nated water to the Mississippi River. Mr. Peckham requested that he
be notified before start-up.
The OSC received sample results from Bay St. Louis:
Pond (second look using acid extraction pH 2):
DNBP - 0.67 ppm
Ditch - 0 ppm
2-7
-------
At the dam site, OH Materials had two small pumps recirculating/
aerating water from the impoundment. A larger pump, 6 inches in
diameter, was in place and would be started shortly. Recirculation
was creating foaming, indicating the presence of surfactants, deter-
gents, etc. OH Materials had additional equipment en route.
The OSC talked with Bob Shockey and Mike Smith to discuss cleanup
operations. Later, the company, state, and EPA met and discussed the
following:
By pumping into the sewer at 1% dilution, about 500,000 gallons
would be pumped per day. Tests indicated this rate would not
hurt the STP.
Pumping into the sewer still looked to be the desirable method
of disposal. The pumping would go on all night.
The state would run a bioassay that night on circulated pond
water and would run a new respiration rate on bacteria in the
morning. Aerators were down at the STP.
The pond was visited and found to be circulating well. Mr.
Cothron discovered a small seepage under the dam but it was not posing
a threat.
OH Materials would pump all night. Two men would remain on-scene
at the pond. The state would collect a bioassay sample at 8:30 p.m.
and run it during the night. All three pumps were running. A line
would be laid up the levee to the manhole the next day. Pumping was
still causing foaming. OH Materials would begin cleanup of the plant
site the next morning.
At the plant, the company had diked off the plant along Pennsyl-
vania Street and would flush out the ditch on the west side of the
plant that night.
2-8
-------
July 10, 1979
At the dam site, water was being circulated. The OSC went to the
bioassay unit and talked with state personnel. Samples had been taken
the previous night at the dam, the middle of the pond, and at the
culvert entrance to the pond. The midstream sample appeared more
concentrated than the other two. Split samples went to the STP. The
city would also run for pesticides. The state would take additional
samples.
At the plant site, work had been completed along the railroad
track but there was still some discoloration in the ditch. There was
very little flow in the ditch.
The OSC talked to Terry Cothron:
The weather would be good until Friday with the possibility
of thundershowers.
Pumping into the sewer would begin at a 1% dilution rate and
bacteria would be monitored for effects. If none, it would be
possible to increase the rate.
Middle of the pond would be mixed with tailings due to a
difference in concentration and toxicity.
The STP would look at respiration rates every two hours to
observe effect.
The state, EPA, and the city gave approval for pumping into
the STP.
The state arranged to sample STP effluent while contaminated
water was being pumped.
Atrazine would be hydrolyzed by acid or base.
Dr. Nunez of the mid-South Poison Control Center called and said
that he was going to release a statement about the deaths of small
animals in the immediate area of Drexel. He also asked if soil
samples would be analyzed and was informed that there were no plans to
do so at that time.
LIBRARY/REGION IV
U.S. ENVIRONMENTAL PROTECTION AGENCY
345 COURTLAND STREET, N.E.
ATLAr!"'". ca
2-9
-------
At 10:30 a.m., John Gilbert told OH Materials to start pumping
into the sewer at 380 gpm. There was a press conference at Maynard
Stiles' office. The OSC, Terry Cothron, Mike Higgs, and Mr. Stiles
participated. The press was briefed on the situation and informed of
the decision to pump through the STP.
Mr. Sand reported that the samples taken July 7 contained no
chlorinated compounds.
The pumping rate was increased to 400 gpm at 12:50 p.m. The
state took an effluent sample at the STP.
The OSC spoke with Larry Ledford of Tennessee Solid Waste Control
and informed him that the company was ready to begin cleanup of the
building and needed to establish a plan for each item based on his
requirements for disposal.
OH Materials shut down for the day at 5:30 p.m., except to finish
the 6-inch pipe to the top of the levee. The company was going to
clear an area that night so the contractor could begin decontamination
operations in the morning.
July 11, 1979
The OSC checked the dam area. The level was down approximately
one inch. The OSC also checked the bioassay unit and found that 1%
and 5% tests looked OK. The effluent leaving the STP was being di-
luted 10 to 1 with TVA cooling water before entering the Mississippi
River.
At the plant site, the OSC, John Gilbert, Bob Shockey, and Larry
Ledford discussed the disposal technique to be used on materials in-
plant. OH Materials had started moving drums, tin, etc., but were
progressing slowly. The fire department arson squad requested that
the contractor work around the Hot Room while they inspected and
sampled the area.
Marlene Jones of the STP was reluctant to increase the flow,
although 1% and 5% tests looked OK at the 8:00 p.m. test. She was
requested to do tests up to 25% concentration to establish an upper
limit that the plant might tolerate. Mike Higgs would discuss in-
creasing the flow with Ms. Jones again.
2-10
-------
The state talked to the city about increasing the flow. They
were being cautious and indicated that if the STP had to be bypassed,
it would cut off the food supply to bacteria. Therefore, bypassing
would have to be scheduled for a period of time, and then switched
back to the STP for a few hours and continued in a cycle.
At the pond area, the OSC told OH Materials to dike the Delta
drainage ditch. Marlene Jones of the STP called and reported that
higher forms of bacteria made up 15% of the mass. The plant would
test every four hours starting at 6:00 p.m. when the 2% flow arrived
at the plant. The lake would remain closed. Mr. Cothron would set up
a schedule for possible STP bypass.
In the evening, the OSC had OH Materials set the pumps and hoses
to pump into the manhole on Horn Lake Road. The OSC also checked the
Delta Refinery drainage to make sure the dam would have no adverse
effects when the water backed up. It started to rain at 5:30 p.m.
The city approved the plan to increase pumping without bypassing.
Maximum volume, about 2,500 gpm, was being pumped into the sewer.
Mike Smith asked what to do with concentrated water accumulation
and was told to store it in tanks for the time being.
Terry Cothron stated that it was not feasible to pump uncontami-
nated water at the upper end of the pond over the levee.
July 12, 1979
The rain stopped at 4:00 a.m. The level of the pond was up one
foot. Nonconnah Creek water level was also up. The flow in the
drainage ditch was about the same.
State personnel arrived at 6:30 a.m. and stated that higher con-
centrations should be pumped at the pump intake versus the upper end
of the pond. The state would be running a bioassay on the water taken
at the back of the pond.
Bob Shockey told the contractor to move the Class B poisons. The
OSC told Mike Smith to put additional contaminated water into a hold-
ing tank.
2-11
-------
By 10:00 a.m., the pond level was approaching one inch below the
zero-inch mark. At the STP, aeration was on, bacteria was showing
minor effects, there was some color, no odor, some foaming. Samples
were taken and respiration rates had been checked every 30 minutes the
previous night.
At 3:00 p.m., the pump hoses were adjusted and the pumping volume
was 3,000 gpm.
At 4:50 p.m., there was a message from the STP that the bacteria
were dying. It was decided to pump maximum volume and bypass the STP.
At 6:00 p.m., the STP began bypass until midnight. The flow was cut
back to 1,000 gpm. Two 4-inch pumps were being used.
Although called bypassing, the water was going through contact
tanks, clarifiers, chlorine contact tanks, and discharge; no aeration
tanks.
Paul Christman of the STP suggested getting a 10-inch pump and
loading the sewer with maximum pumping for full bypass of the STP and
running in cycles until all the water was gone. Mr. Christman offered
the use of the STP1s 10-inch pump if none other was available.
July 13, 1979
The pond level continued to decrease.
Ben White called. The 0SC and Terry Cothron discussed opening
the lake. It was agreed that danger was remote at this time. Ben
White saw samples with very low ppb range in McKellar Lake. He de-
cided to open the lake with the suggestion that water-contact sports
be avoided. He could close the lake quickly if heavy rains washed out
the pond.
The maximum pumping/bypass cycle began at 12:15 with all hoses in
use. The schedule was 12 hours maximun pumping and 6 hours reduced to
1,000 gpm.
At 1:00 p.m., George Sand called with the final sample results:
Pond:
Methyl parathion
Atrazine
0.07 ppm
8.3 ppm
EPN
DNBP
PNP
8.9 ppm
1.7 ppm
0 ppm
2-12
-------
Ditch:
Methyl parathion
Atrazine
1.0 ppm
101.5 ppm
EPN
DNBP
PNP
0.41 ppm
25.3 ppm
7.1 ppm
At 3:15 p.m., the STP's 10-inch pump arrived at the pond site.
The OSC talked to Bob Shockey about solid waste disposal problem.
At 8:00 p.m., the 10-inch pump was started. The pumping rate
was estimated to be 5,000 gallons. At 11:00 p.m., the 10-inch pump
was shut down.
July 14, 1979
At the pond site, the contractor pumps were running. The 10-inch
pump started at 7:45 a.m.
Sewers were OK, there was no backing up or overflowing.
The pond was dropping rapidly. Mr. Shockey said that he thought
OH Materials was doing a good job.
It was decided to run the STP 10 hours (6:00 p.m. - 4:00 a.m.) to
build up bacteria, and then start pumping again between 4:00 a.m. and
4:00 p.m.
At 6:00 p.m. at the pond site, the 10-inch pump was reworked.
The flow was shut back to 1,000 gpm. The backhoe opened up drainage
from low areas of the pond as water returned to the confines of the
concrete culvert.
Plans were made with the state to pump down liquid on July 15 and
flush with dammed water and pump out. It was also planned to close
the floodgates at Nonconnah pump station on July 16, break the dam,
regrade the earth, and remove dirt from the ditch. The pump would be
returned to the STP, and Ben White would be consulted on sampling
McKellar Lake.
2-13
-------
July 15, 1979
Pumping at the pond started at 4:00 a.m., except for the 10-inch
pump because it wouldn't prime.
The pool would continue to be pumped down as much as possible. A
fire hydrant would be used to flush the culvert to the floodgates.
The dam would be broken and water allowed to flow to the floodgates.
A backhoe would be used to uncover an 18-inch storm drain.
At 4:30 p.m., the OSC contacted Paul Christman of the STP to
inform him that the flow had been cut back to 1,000 gpm at 4:00 p.m.
It was believed that another bypass would not be needed.
The OSC informed Mr. Shockey and Mr. Smith that the pond area
should be nearly drained by the next morning and pumping would be
completed then. A small pool area not draining to the ditch was of
concern because of concentration. The ditch area was fairly diluted.
It was thought that the intake hose on the smaller pump would have to
be moved to remove this.
July 16, 1979
The pond area looked good and it was decided to send the 10-inch
pump back to the STP.
The floodgates at the pump station were closed. This would cause
some accumulation, dilution, and slow discharge to Nonconnah Creek.
A backhoe would be on-site the next morning to dig out the dam.
Cleanup of the building was going well. Mr. Shockey and Mr.
Smith were satisfied with the progress of the cleanup.
July 17-18, 1979
The dam was broken on July 17. Some sediment remained, which
caused the overflow dam to bleed material to the pump station.
Nonconnah Creek was septic again. No yellow color effect from
contaminated water was noticeable.
Progress slowed at the plant site because of work in the Hot
Room.
No rain had occurred. There were no problems at McKellar Lake.
2-14
-------
3. PHOTOGRAPHS
3-1
-------
CHEMICAL PLANT AFTER FIRE
-------
CONTAMINATION ALONG RAILROAD TRACKS
3-3
-------
STORAGE AREA AFTER FIRE
-------
CONTAMINATED RUNOFF WATER
-------
CONTAINMENT AREA FOR CONTAMINATED WATER RUNOFF
-------
CONTAINMENT AREA AFTER MOST OF THE WATER
HAD BEEN DISPOSED OF
-------
PUMPS USED TO MOVE CONTAMINATED WATER
-------
PORTABLE BIOASSAY UNIT
-------
CREWS REMOVING CONTAMINATED METAL FROM SITE
-------
CREWS REMOVING CONTAINERS FOR RE-DRUMMING
-------
-------
STATIC BIO ASSAY TESTING
-------
4. SUPPORTING MATERIAL
4-1
-------
-------
1
2
3.
4
5
6
7
8
9
10
11
12
13
DREXEL CHEMICAL COMPANY
MEMPHIS, TENNESSEE
INVENTORY
DNBP
- Formulated as A1kanolamine Salt - 3 lb DNBP/gal 1,290 gal
- Formulated in Xylene Large Solvent - 5 lb DNBP/gal 8,760 gal
Paraffinic Oil plus Emulsifier 18,795 gal
Maleic Hydrazide
- Technical 90-98% 21,602 lb
- Formulated as DEA Salt - 3 lb MH/gal )
- Formulated as K-Salt - 1.5 lb MH/gal )... 16,179 gal
- Formulated as K-Salt - 2.25 lb MH/gal )
Fatty Alcohols (Cg + C^q) P^us Emulsifier
6.06 lb Alcohol/gal 10,024 gal
Sulfur
- 98% Manufacturers Grade 67,500 lb
- Formulated as 6 lb/gal - packaged 12,735 gal
- Formulated as 6 lb/gal - bulk tanks 13,000 gal -
(May not have been released)
Sodium Chlorate - 3 lb/gal with 1.5 lb Urea/gal 29,810 gal
Hydrazine Hydrate - 54.4% Solution 1,144 gal
Hydrated Lime 33,750 lb
Maleic Anhydride Less than 40,000 lb -
(Truck was being unloaded)
Atrazine
- Technical 95-98% 14,650 lb
- Formulated 4 lb/gal 1,110 gal
EPN 93% Technical 39,600 lb
Methyl Parathion 80% Technical 39,600 lb
Various Surfactants & Emulsifiers 85,717 lb
4-3
-------
INVENTORY
Size
No. Containers Total
Dynamite 3
5
gal
258
1,290 gal
Dynamite 5
30
gal
291
8,730 gal
Dynamite 5
5
gal
6
30 gal
Peptoil
4
x 1
gal
3,065
12,260 gal
Sucker Stuff
4
x 1
gal
765
3,060 gal
Peptoil
5
gal
1,307
6,535 gal
Super Sucker Stuff
4
x 1
gal
1,453
5,812 gal
Super Sucker Stuff
4
x 1
gal
248
992 gal
Sucker Plucker
4
x 1
gal
2,506
10,024 gal
Riverside Sulfur.
5
gal
210
1,050 gal
Barber Sulfur
5
gal
99
495 gal
Drexel Suffa
5
gal
1,650
8,250 gal
Surfac 820
4
x 1
gal
1,520
6,080 gal
Helena Sulfur
5
gal
588
2,940 gal
Defol
55
gal
542
29,810 gal
S.S. Stuff
4
x 1
gal
1,980
7,920 gal
Atlox 804
4,522
4,522 lb
Anti Foam
3,279
3,279 lb
Clay
50
lb
23
1,150 lb
CR273C Acid
100
lb
14
1,400 lb
DOWCIL
100
lb
13
1,300 lb
Hydrazide Solution
440
lb
26
11,400 lb
Kelzan
300
lb
21
6,300 lb
LS-300
16,039
16,039 lb
Lime
50
lb
675
33,700 lb
Lomar
50
lb
240
12,000 lb
LO 529
450
lb
34
15,300 lb
Maleic Hydrazide
44
lb
320
14,080 lb
PVP 30
100
lb
35
3,500 lb
Maleic Anhydrazide
50
lb
50
2,500 lb
Drexel H.H. Tech
7,522
7,522 lb
Paraformaldehyde
50
lb
40
2,000 lb
Petro-D-425
60
lb
71
4,260 lb
PolyFon 0
50
lb
55
2,750 lb
Sulfur
50
lb
1,350
67,500 lb
Soda Ash
640
640 lb
Urea
100
lb/bag
138,037
13,803,700 lb
Unico Atrazine
100
lb
222
22,200 lb
Super P-150
10,367
Super Sucker Stuff HC
10,656
EPN 93% Tech
550
lb/drum
72
39,600 lb
Methyl Parathion Tech
00
o
550
lb/drum
72
39,600 lb
Atrazine Tech
50
lb
293
14,650 lb
4-4
-------
Inventory - Pg. 2
The following containers were empty:
Size No. Containers
Super Sucker Stuff HC
4
X
1
804
Sucker Plucker
4
X
1
4,573
Super Sucker Stuff
4
X
1
389
Sucker Stuff
4
X
1
1,647
Peptoil
4
X
1
253
Surfac 820
4
X
1
2,232
Ansar
4
X
1
507
Unsleeved
4
X
1
1,671
Plain
2
X
2-1/2
1,944
5 Gal Pails (black)
1,214
5 Gal Plastic
126
5 Gal Open Head
7,533
5 Gal Old Open Head
798
5 Gal White Metal
59
New 55 Gal Drums
114
New 55 Gal Plastic
768
30 Gal Drums
469
Helena SMH
4
X
1
48
Plain
48
144
Black P/P Caps
14,200
2" Seals
4,396
PI ugs
9,000
2-1/2 Gal Caps
101,000
4x1 Gal Caps
254,380
Flex Spouts
15,665
4-5
-------
TELEGRAPHIC MESSAGE
NAME Of AGENCY
EPA, Region IV
Atlanta, Georgia
PRECEDENCE
action* PRIORITY
INFO:
SECURITY CLASSIFICATION
UNCLAS
ACCOUNTNG CLASSIFICATION
date prepared
07/09/79
TYPE OF MESSAGE
0 SINGLE
~ »OOK
|~1 MULTIPLE-ADDRESS
FOR INFORMATION CALL
NAME
John R. Moebes
PHONE NUMBER
(404)881-3973
THIS SPACE FOR USE OF COMMUNICATION UNIT
MESSAGE TO 8E TRANSMITTED (Use doubU spacing and all capital Utters)
TO:
MAYNARD STILES, DIRECTOR
DIVISION OF PUBLIC WORKS
CITY OF MEMPHIS
125 NORTH MAIN
MEMPHIS, TENNESSEE 38103
RE: MEMPHIS SOUTH WASTEWATER TREATMENT PLANT
NPDES PERMIT NO. TN0020729
THE ON-SCENE COORDINATOR OF THE ENVIRONMENTAL EMERGENCY CAUSED BY
THE FIRE AT THE DREXEL CHEMICAL COMPANY HAS REQUESTED THAT
CONTAMINATED WATERS RESULTING FROM THAT FIRE BE PUMPED TO THE
MEMPHIS SOUTH WASTEWATER TREATMENT PLANT. THE CITY HAS AGREED TO
ACCEPT THAT CONTAMINATED WATER IF EPA AGREES NOT TO TAKE ACTION
FOR VIOLATION OF THE SOUTH WASTEWATER TREATMENT PLANT NPDES PERMIT
CAUSED BY THE CONTAMINATED WATER. EPA, IN CONSIDERATION OF THE
EMERGENCY THAT EXISTS, AGREES NOT TO TAKE ACTION FOR ANY PERMIT
VIOLATION CAUSED BY THE CONTAMINATED WATER BEING PUMPED TO THE
SOUTH WASTEWATER TREATMENT PLANT.
SANFORDyW. HARVEY,UR.V) DATE SIGNED:
DIRECTOR, ENFORCEMENT DIVISION
EPA, REGION IV
PAGE NO.
NO. OF PCS.
SECURITY CLASSIFICATION
UNCLAS
{STANDARD FORM 14
tfVISCO AUGUST 1967
QSA PPia (41 CFR) 101-15.306
6U.S. OowrniMnt Prhitlne Offie*: 197ft— 2G1-447/344S
14-102
4-6
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
DATE.
July 16, 1979
SUBJECT.
FROM
TO
The Analysis of Water Samples Taken at the
Drexel Chemical Co., Memphis, Tennessee
After the Explosion and Fire
George J. Sand
Pesticides Monitoring
NSTL Station, MS 39529
Mr. Jan Rogers, Enforcement Division
Hazardous Materials Section, Region IV
Federal On-scene Coordinator
Thru: Dr. Han Tai
7^,
The water samples that were taken at the Drexel Chemical Company
on July 7, 1979, and delivered to our laboratory by the U.S. Coast
Guard, have been analyzed for pesticides, herbicides, and toxic
chemicals. The analytical results are listed as follows:
Analytical Results of Water Samples Taken at
Drexel Chemical Company, Memphis, Tennessee
Lab Acc.
No.
9-73-018
9-73-019
Sample Site
Above
Earthen dam
Methyl
Parathion
(ppm)
0.07
Drainage ditch 1.0
below plant,
between RR
bridge
Atrazine
(ppm)
8.3
101.5
EPN
(ppm)
ND
0.41
Dinoseb
(DNBP)
(PPm)
8.9
25.3
p-nitrophenol
(ppm)
1.7
7.1
Note: ND - None detected
Minimum Detectable Level
Organophos
Toxic Chem
Dinoseb
Atrazine
.05 ppb
10 ppb
10 ppb
.10 ppb
Sample extraction and
analysis by:
Gerald Gardner
Milas Blaylock
Ray Shaw
If you should have any questions about the above results, please contact
us.
We will store these water samples for a short period of time in the event
that you may have need for these samples or may request some additional
analysis.
cc: Sam Wastler
EPA Form 1320-6 (Rev. 3-76)
4-7
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
DATE: August 13, 1979
SUBJECT: Drexel Chemical Company
FROM: John M. Gilbert
Environmental Response Team
TO: J. Stephen Dorrler, Leader
Environmental Response Team
1. Place Visited: Memphis, Tennessee
2. Dates: July 5, 1979 through July 13, 1979
3. Facility Involved: Drexel Chemical Company
2487 Pennsylvania Street
Memphis, Tennessee 38109
(901) 774-4370
4. Type of Spill: Explosion and fire at a pesticide plant
5. Products Involved: Methyl Parathion, DNBP, EPN, and Atrazine
6. OSC: Jan Rogers, Region IV
7. Remarks:
This memo will be a brief summary of the incident. Jan and I are
collectively writing a complete report on the incident. E&E will
assist in printing the report when completed.
SUMMARY
On July 5, 1979 an explosion and fire completely destroyed the Drexel
Chemical Company's manufacturing facility at 2487 Pennsylvania Street.
Drexel produced a wide range of pesticides and herbicides. The explosion
occured in the "hot room" where two class B poisons were stored. The
class B poisons handled at the facility were:
1. Methyl Parathion
2. EPN
3. DNBP (not stored in the hot room)
Runoff from the fire fighting effort flowed through a drainage ditch
into Nonconnah Creek. Nonconnah Creek empties into McKellar Lake.
During the fire the drainage ditch was dammed by the City Public Works,
however, some runoff reach Nonconnah Creek prior to construction of the
dam.
4-8
-------
Ifre company added soda ash, NaOH and KOH to the runoff water in order
to hydrolize the methyl parathion, EPN, and DNBP. Soda ash was also added
to Nonconnah Creek. Literature and experience indicated both methyl
parathion, and EPN will hydrolize readily. DNBP is slower to hydrolize.
When I arrived, around 1900, the fire was out and the runoff water con-
tained. Jan had arrived around 1600 hours and had already talked to
the company. Clean up fell into the following:
1. Plant site - contaminated metal, drums (full & empty) and standing
water.
2. Contaminated runoff water
The company engaged OH Materials to do the clean up.
Plant Clean Up
Due to the configuration of the facility it was impossible to contain
all standing water at the plant. To contain as much as possible, the
OSC had the company plug all storm sewers and construct earthen dikes
around the facility. The drainage ditch running from the plant to
Nonconnah Creek could not be contained close to the plant because the
ditch handled runoff from the surrounding neighborhood in addition to
the plant.
The material at the plant was handled in the following manner:
1. Metal was decontaminated with an acid wash and hauled off by a metal
reclaimer.
2. Contaminated water was pumped into the company's storage tanks.
3. Containers with product remaining in thera were emptied into drums.
4. Debris was hauled to a local landfill - permission given by Tennessee
Solid Waste.
Contaminated Water
The dike constructed by the city where the drainage ditch goes under Horn
Lake Road stopped all additional flow into Nonconnah Creek. The OSC had
the dam reinforced to contain additional runoff from rainfall. The dam
contained a large volumn of contaminated water that had to be disposed of.
After close coordination with EPA, Tennessee Water Quality and the city
of Memphis, the contaminated water was pumped into the South Sewage Treat-
ment Plant. The effleunt from the STP flows into the Mississippi River.
Initially, the pumping rate was such that a 5% solution was entering the
plant. When it became evident that this rate would require the pumping
for many weeks, the rate was pushed to full pumping capacity. This caused
problems at STP and the plant's entire flow had to be bypassed. To keep
the STP alive, the bypass was not in one shot, but staggered. When the
4-9
-------
piant went back on line the pumping rate of contaminated water was
rfeduced to a level (5%) that the plant could handle. After sufficient
time to stabilize the plant, the flow as upped and the plant bypassed.
This continued until the contaminated pond was dry.
TASK FORCE
To evaluate the entire situation, the OSC held a meeting with all interested
organizations. As a result of the meetings, the following was agreed
upon:
1. FDA would handle contamination of commercial products.
2. Health Department would check gardens, yards, and public fishing.
3. Poison Center would handle all complaints about human effects.
4. EPA and State would handle the clean up.
SAMPLE ANALYSIS
The following samples were analyzed:
Drexel - initial runoff samples
EPA(Bay St. Louis) - runoff water samples
State - runoff, Nonconnah Creek, bioassay, STP effluent
OH - metal
City - STP bugs, STP effluent
The bioassay test was the first indicator on how toxic the runoff was
and proved to be invaluable by providing quick (4 hour) results. The
decision not to discharge the contaminated water into the Nonconnah Creek
and the initial pumping rate was based on the bioassay results.
After Jan and I left, the company and the Poison Control Center began
a cooperative effort to investigate fall-out on soil and vegetation.-
The work is being done by the Poison Control Center, a local university
is doing the analysis, and Drexel is paying the'bill.
cc: Jan Rogers
4-10
------- |