~

nruerai




-------
                                   EPA-430/9-80-012
                                   August 1980
ILLINOIS CENTRAL GULF  TRAIN DERAILMENT

        CLAXTON, KENTU'CKY
  Environmental Protection Agency
     Oil and Special  Materials
         Control  Division
      Washington,  D.C.  20460

-------
                              PREFACE
     This report reflects the activities of the Environmental



Emergency Branch (EEB) Region IV, EPA, as it was concerned with



the captioned event.  The report outlines the key actions of the



Federal Ch-Scene Coordinator, the Regional Response Team (RRT),



the state, local, and Federal members of the RRT.  It reflects



the complexity of responding 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 Contingency Plan



and to help others learn from our experiences.
                                       V. Smith, Chief



                                  Environmental Emergency Branch

-------
                          TABLE OF CONTENTS
Section                                                    Page

  1.      SUMMARY                                           1-1

  2.      THE SITUATION                                     2-1

  3.      PHOTOGRAPHS                                       3-1

  4.      PROBLEMS ENCOUNTERED                              4-1

  5.      ANALYSIS OF AIR, WATER, AND SOIL SAMPLES          5-1

          a.   Report on Air Monitoring
          b.   Laboratory Analysis of Water
              and Soil Samples

-------
                             1.  SUMMARY
     Eighteen railroad tank cars containing vinyl chloride, hydro-
fluoric acid, ethyl acrylate, sodium chlorate, chlorine, caustic
potash, and butadiene were derailed near Claxton, Kentucky.  Caustic
potash, ethyl acrylate, and vinyl chloride were discharged from the
ruptured tank cars onto the ground about a quarter of a mile from
Montgomery Creek.  The creek was dry at the time of the derailment.
A hopper car of sodium chlorate overturned and mixed with the spilled
ethyl acrylate.  The reaction between the two chemicals created a
fire and polymerized some of the ethyl acrylate.
     The primary problem associated with this derailment was the
burning of two vinyl chloride tank cars.  One of the vinyl chloride
cars flared from a large rupture at one end of the tank car, the
other in the vicinity of the dome.
     None of the other tank cars was leaking.  However, the potential
for damaging these tank cars in the removal operation, thereby creat-
ing serious environmental problems, existed for seven days.
     Meanwhile, the vinyl chloride tank cars were burning and would
have done so for several more days.  It was decided to accelerate the
burning of the vinyl chloride by cutting holes in the tank cars with
plastic explosives.
Air Quality Problems

     During the combustion phase of vinyl chloride, it is possible
that hydrochloric acid, phosgene, and carbon monoxide can be re-
leased.  All of these products are highly toxic when released to the
atmosphere.  Also, vinyl chloride is very reactive or explosive at
high temperatures or pressure.

                                 1-1

-------
      Thus, it was imperative for the vinyl chloride tank cars to
 continue burning.  No effort was made to use dry chemicals and carbon
 dioxide to extinguish the fire.  By putting the fire out, an explo-
 sive condition could develop and endanger the integrity of the other
 tank cars.
      EPA and the Kentucky Department of Natural Resources and Envi-
 ronmental Protection set up air surveillance crews to monitor for
 toxic chemicals on a 24-hour basis.  These crews responded to all
 complaints from residents in the surrounding area.  In addition, air
 sampling was conducted on a routine basis.
      Vinyl  chloride  was detected in the air on several occasions near
 the burning tank cars.  An acid mist formed during rainy periods and
 settled  in the valley near the wreck site.  The acid mist irritated
 the skin and made it difficult to breathe.  Work was halted during
 these periods.
      Air samples were collected and tested for phosgene.  However, it
 was not  possible to  detect phosgene in  the atmosphere while the vinyl
 chloride tank cars were burning.

 Venting  of the  Burning Vinyl  Chloride Tank Cars

      For seven  days,  the ruptured vinyl  chloride tanks burned while
 the other  derailed tank  cars  were removed.   One of the vinyl  chloride
 cars  developed  a frost line.   It was estimated that 25% of the vinyl
 chloride remained  in  this  tank  car.   The  other tank car contained
 about  70%  of  its volume.   The capacity  of each tank car was 20,000
 gallons.
      A decision  had  to be  made  whether  to let  the  vinyl chloride burn
 out or vent the  tank  cars  with  plastic  explosives  and  accelerate the
 burning  process.   By  letting  the tank cars  burn out on their  own,
 people would  have  to  remain evacuated from  their homes for  a  long
 period of time.   In  addition,  Illinois  Central  Gulf continued rail
traffic  through  the area.  Thus,  the potential  for  another  incident
existed  while the  tank cars were burning  adjacent  to  the tracks.
                                1-2

-------
      It was decided  to vent the tanks  and  assume  a  short-term  risk  of
polluting the  atmosphere, rather than  risk  another  explosive-type
incident.  An  Army Explosive  Ordnance  Disposal  (EOD)  team  was  brought
in by the On-Scene Coordinator to  implement a plan  for cutting  two
holes in each  tank car,  using plastic  explosives.   This  allowed the
vinyl chloride to flow out of the  tank cars into  a  large diked  area.
The vinyl chloride was then ignited by means of an  incendiary  device.
     The charges were set by  the EOD team  on both tank cars  so  they
could be vented at the same time.  However, the plastic  explosives
did not cut through  the  egress hole on one  of the tank cars.   Another
plastic charge had to be set  to puncture the tank car.
     Once the tank cars  were  punctured, the vinyl chloride burned for
about one hour.  This created a huge,  black plume that was monitored
by EPA air personnel.  It was raining  at the time.   The mixing  of the
rain with the gases  released  from  the  burning of  the  vinyl chloride
created an acid mist.  EPA advised the residents, through  the
Kentucky State Police, to remain in their  homes and out  of the acid
mist.

Analyses of Water and Soil Samples

     Soil samples were collected by the Kentucky  Department  of Nat-
ural Resources and Environmental Protection in the  area  where  ethyl
acrylate, caustic potash, sodium chlorate,  and vinyl  chloride  were
discharged.  In addition, surface  waters,  wells,  and  cisterns  were
sampled.  Toxic chemicals were not detected in these  samples.
     Montgomery Creek was dry at the time  of this incident.  However,
there was concern that,  after a rain,  the  chemicals could  enter the
creek via runoff.  A retaining structure was constructed to  prevent
this occurrence.
     In conclusion,  this operation was conducted  in accordance with
the Regional Contingency Plan; the plan worked and  the operation was
a complete success.
                                1-3

-------
                          2.  THE SITUATION
     At 11:50 p.m., on October 17, 1978, a dispatcher for  Illinois
Central Gulf Railroad (ICG), reported, via the National Response cen-
ter, a train derailment near Claxton, Kentucky.  The initial report
indicated that five tank cars containing sodium chlorate,  vinyl chlo-
ride, and hydrofluoric acid were derailed.  One of the tank cars was
burning and some of the nearby residents were evacuated.
     EPA official Fred Stroud forwarded the information to Jack
Stonebraker, also of EPA, who would become the On-Scene Coordinator
(OSC) for this spill event.  Arrangements were made for the OSC and
Jim Littell of EPA's Air Emergency Branch to take the earliest pos-
sible flight to Nashville, Tennessee, in order to respond  to this
incident.
     During the early morning hours of October 18, EPA continued to
receive updates on the number of tank cars derailed and their con-
tents.  Some of the derailed cars contained chlorine, ethyl acrylate,
and caustic potash.  The nearest stream was approximately  one-half
mile from the wreck site.  In addition, two tank cars of vinyl chlo-
ride were burning.
     EPA requested assistance from the Gulf Strike Team (GST)  and the
Public Information Assistance Team (PIAT).  The Gulf Strike Team was
to bring personnel safety equipment and a mobile command post.  Both
teams would arrive on-scene later in the day to assist the OSC.
                                2-1

-------
 Response and Observations:  Chronological Events
 October 18. 1978

      Jim Littell and the OSC arrived on-scene at 11:30 a.m., met with
 the Kentucky State Highway Patrol and Edward Clark of the State Fire
 Marshall's office and obtained the following information:
      Eighteen cars were involved in derailment.
      Tank cars containing ethyl acrylate, caustic potash, and vinyl
      chloride and a hopper car of sodium carbonate had been ruptured
      and material discharged on the ground.
      The other tank cars containing butadiene,  hydrofluoric acid, and
      chlorine apparently were not leaking.
      144 houses were evacuated in a two and one-half mile radius of
      the wreck site.
      Two tank cars of vinyl  chloride were still burning.

      The OSC met George Bradel  and Don Hayes of the Kentucky Depart-
ment  of Natural  Resources and Environmental  Protection (KDNREP).
They  had just returned  from  inspecting Montgomery Creek for possible
environmental  damage due to  the chemical substances discharged from
the ruptured  tank cars.   However,  Mr.  Bradel reported that the creek
was dry.  There  wasn't  any evidence of chemical  substances in the
streambed.   If  it rained,  there was a  possibility that some of the
chemicals  could  enter the stream via runoff from the wreck site.
      This  information was  forwarded to the EPA  regional office.   Joe
Lafornara  of  the Emergency Response Team (ERT)  was  enroute to observe
the operation.
      Ben Eason and  Richard Griggs  of the PIAT arrived on-scene.   The
OSC discussed the  situation with them  and  made  plans for  responding
to questions  from  the media  and the public.   This part of the opera-
tion would begin on  the  following day.
      Early in the  evening, Jim  Littell  and the  OSC  noticed an inver-
sion taking place  in  the  low  areas  around  the wreck  site.  The ther-
mal inversion caused  the  plume  from the  burning  vinyl chloride to
stay close to the ground.  There was a distinct  odor in the  air.   The
                                2-2

-------
air was tested at several  locations outside the evacuated zone  for
vinyl chloride and hydrochloric acid.  All of the  instrument readings
were negative for these two chemicals.

October 19, 1978

     The mobile command post arrived on-scene at 12:03 a.m.  The
State Highway Patrol called South Central Bell to  have telephones
installed  in the command post.

1:00 a.m.  - Bob Tittle of  the State Fire Marshall's office  reported
that the plume from the burning vinyl chloride cars was much larger
than it had been in the afternoon.  He requested that  EPA monitor for
toxic chemicals downwind.  Jim Littell and the OSC took readings for
the two parameters mentioned above at several locations around  the
wreck site and in Dawson Springs, Kentucky, six miles  away.  The
instruments did not detect any hydrochloric acid or vinyl chloride.

2:47 a.m.  - From the command post, a flare-up at the wreck  site area
was noticed.  The flames illuminated the sky and burned for about 45
minutes.   There was speculation as to what caused  the  flare; some
throught the tank car fractured or exploded, while others thought the
safety valves released vapors that were ignited.

3:00 a.m.  - While the tank cars were burning at an accelerated  pace,
air samples were taken downwind at several locations.  Although the
odor in the air was unpleasant, no toxic chemicals were detected.

7:10 a.m.  - The State Fire Marshall reported toxic fumes  along  High-
way 672 and requested Disaster Emergency Services  (DES) to  evacuate
this area.  The air in the area was monitored but  no harmful chemi-
cals were  detected.  The air was very noxious and  the  inversion kept
smoke close to the ground.

8:20 a.m.  - The OSC took an overflight in a helicopter.   There  was  a
tank car of chlorine lying on a burning coal car.  The tank car of
butadiene  appeared to be upright and on rails.  The ethyl acrylate
                                 2-3

-------
 was discharged on the north side of the tracks, had run down the
 ballast, and mixed with sodium chlorate.  Apparently, this mixture
 caused a violent exothermic reaction, sending up a huge fire ball.
 One of the vinyl chloride cars was burning from a rupture at one end
 of the tank car.  The other vinyl chloride car was burning at the
 dome.

 8:40 a.m. - The OSC held a meeting in the command post to discuss the
 situation and review future plans.  Agencies represented were as fol-
 lows:  EPA, GST, State Fire Marshall  Office, State Highway Patrol,
 DES,  Hulcher Emergency Service, Illinois Central Gulf, and Kentucky
 Department of Natural  Resources and Environmental Protection.
      The representative  from Illinois Central Gulf provided a com-
 plete list of derailed cars and their contents;  six vinyl  chloride
 cars,  one sodium chlorate  car,  one liquid  caustic potash car, three
 chlorine cars,  two  cars  containing hydrofluoric  acid,  two ethyl
 acrylate cars,  one  butadiene car,  and two  coal  cars.   Plans were to
 maintain a status quo  since conditions were basically stable.  No
 attempt  was made to  put  out the fires from the vinyl  chloride cars
 because  extinguishing  them would create an explosive condition.
      The  question was, how long would the  vinyl  chloride burn?   Esti-
 mates  ranged from one-to-several  days.  It was decided that Art  Prof-
 rock of  Hulcher  and  a  representative  from  the State Fire Marshall's
 office would take a  closer look at  the burning cars.   Self-contained
 breathing  apparatus  would  be utilized  in the  investigation.

 9:30 a.m.  -  The  news  media arrived at the command  post  and asked for
 information  about the  other  two tanks  that had exploded.   The OSC
 asked  them  where  they  got  their information.  Their reply was that
 DES had dispatched a news  release from Frankfort,  Kentucky.   It  was
explained that the same  two  vinyl  tank cars were burning and  no  other
cars had exploded.   Evidently,  the flare-up that  occurred  in  the
early morning hours  had  been construed as  an  explosion of  other  tank
cars.

10:15 a.m.  - The OSC talked  to  Debbie  Hockensmith of the KDNREP  about
setting up a sampling program for wells, surface  water,  and  soil.
It was necessary to determine how far  the  spilled material  had
                                 2-4

-------
penetrated  the  soil  and  the direction of  the migration.  There was
also a possibility of contamination of cisterns and ponds from the
plume.  Disposal  sites were needed for the contaminated  soil.

11:00 a.m.  - Mr.  Profrock returned from inspecting the vinyl chloride
tank cars.  He  reported  that  a frost line had formed  on  one of the
tank cars.  His estimate of the volume of vinyl chloride remaining
was 20%, or 4,000 gallons.  A frost line  had not  appeared on the
other tank  car.   Perhaps 75-80% of the liquid remained in the tank
car, or about 15,000 gallons.

12:30 p.m.  to 3:00 p.m.  - During this time period, Debbie Hockensmith
and the OSC conducted joint interviews with the news  media.  They
wanted to know  what had  transpired and EPA's plans for monitoring the
air and collecting water samples.
     The news media were interviewing nearly every agency present.
It was obvious  that a lot of misinformation had been  given to the
media.  Ben Eason suggested that a news conference be held on October
20, 1978.   This would give the media an opportunity to ask questions
of a representative from DES, the State Fire Marshall's  office,  EPA,
Illinois Central  Gulf Railroad, and the State Highway Patrol.  The
OSC agreed, and the PIAT made plans for a news conference at 9:30
a.m. on October 20, 1978.
     Later, Ben Eason informed the OSC that DES was the  coordinator
for the state agencies,  and all news releases had to  come from their
office in Frankfort, Kentucky.  Experience with this  type of proce-
dure demonstrates that it doesn't work.   Procedures for  responding to
questions from  the news media by the Federal OSC  are  clearly outlined
in the National Contingency Plan.  Plans  for the  news conference pro-
ceeded, regardless of whether DES and the other state agencies parti-
cipated.

3:30 p.m. - Mr. Randolph Jensen, Engineer for Rohm and Haas, visited
the command post.  Rohm and Haas was the  shipper  for  the two ethyl
acrylate tank cars.  Mr. Jenson was helpful in the discussion of the
properties  of ethyl acrylate.  In his opinion, the mixing of the
strong oxidizer sodium chlorate with the  ethyl acrylate  started  a
fire.  The  burning of the ethyl acrylate  or exposure  to  heat spon-
sored polymerization to  a noxious resin.

                                2-5

-------
      The safety engineer for Illinois Central Gulf came into the
 command post looking for Mr. Jensen.  He wanted Mr. Jensen to locate
 a clean tank car for the transfer of ethyl acrylate remaining in the
 ruptured car.  The ruptured car was located across the track upwind
 from the burning vinyl chloride cars.  It was possible that the ethyl
 acrylate could be safely removed by a pumping operation.
      Mr. Jensen said that the clean cars were located on the east
 side of the wreck site.  The damaged tank car was on the west side.
 The OSC suggested that they get some tank trucks and load them on
 flat cars.   It was estimated that 10,000-12,000 gallons of ethyl
 acrylate remained in the tank car.  Therefore, only two tank trucks
 would be required.  Mr. Jensen  said that tank trucks could be on-
 scene the next day.
      Approximately 250 feet of  hose and a diaphragm pump, which could
 be  run  off  the train engine, were needed.  Arrangements were made by
 Mr.  Jensen  to begin  the pumping operation on October 20, 1978.

 5:00 p.m. to Midnight - Air monitoring  teams were set up to respond
 to  any  complaints.  During  this time period, several complaints came
 in  about fumes  in  certain downwind areas.  The instruments were un-
 able to  detect  any harmful  chemicals.

 October  20,  1978

 Midnight to  7:QQ  a.m.  - Air complaints  were responded to,  and the air
 was  monitored on  a routine  basis.   Most  of the complaints  of a  bad
 odor  came from  the citizens in  Dawson Springs, Kentucky, about  six
miles from the  wreck  site.
      The OSC  talked  to  the  chief of police of Dawson Springs.  He
 indicated that  the people really didn't  know what  was involved  in
the  train derailment.   Most of  them wanted to evacuate  their homes,
especially when they  could  smell  the fumes in the  air.
     The air  was sampled  at several  locations in  Dawson Springs, but
no harmful chemicals  were detected.
                                 2-6

-------
9:00 a.m. - Four television stations and  several newspaper reporters
were present for the news conference.  Representatives from  DES, the
State Fire Marshall's office,  Illinois Central Gulf  Railroad,  Hulcher
Emergency Service, KDNREP, and EPA participated  in the conference.
It went well, and afterward, the news media were allowed to  go  on-
scene to take pictures of the derailment.

11:00 a.m. - An air monitoring crew consisting of Jim Littell  and two
members of the Gulf Strike Team went on-site.  There were some  fire-
men on-site putting water on a chlorine tank car, which was  lying on
a burning coal car.  Mr. Littell took an  air reading for vinyl  chlo-
ride and detected about 50 ppm in the smoke.  After  receiving  this
information, the firemen and Hulcher's personnel immediately departed
from the area.

2:00 p.m. - The removal operation of the  ethyl acrylate from the
damaged tank car to tank trucks on flat cars began at this time.  Two
Gulf Strike Team members monitored the transfer operation.   The oper-
ation was completed at 6:00 p.m.

3:00 p.m. - Ben Eason made arrangements with the mayor of Dawson
Springs for the town meeting.  The high school gymnasium was reserved
for 7:00 p.m.  Mr. Eason traveled to Dawson Springs  to assist  in
publicizing the meeting.  All participants were notified.

7:30 p.m. - 50-100 people attended the town meeting  in Dawson
Springs, Kentucky.  Members of the panel  were as follows:

     Jack Stonebraker, EPA OSC
     Jim Littell, EPA
     Derris Kirkman, State Highway Patrol
     Debbie Hockensmith, KDNREP
     Rod Raby, Deputy State Fire Marshall
     Ed Clark, State Fire Marshall's Office
     Jim Walker, Governor's Aide
     Mark Holcomb, State Fire Marshall's  Office
     Mayor Dickson, Dawson Springs, KY
     Jannis Rice, Chief of Police, Dawson Springs, KY
                                2-7

-------
      The people asked questions about the environmental  problems
 associated with the spill  incident.
      The director of a nursing home was present at the meeting.  She
 expressed a deep concern for the patients and whether or not they
 should be evacuated.  Mr.  Raby stated that the city had  a contin-
 gency plan for evacuation.   If something happened, the plan would be
 implemented immediately.  The people of Dawson Springs would have
 adequate time  for evacuation.
      The meeting was very successful.  There were no air complaints
 during the rest of the night.  However,  the air was still  being
 monitored on a routine basis.  Sample tests revealed no  detectable
 levels of harmful  chemicals.

 October 21,  1978

 6:00  a.m.  -  Hulcher's  equipment for  re-railing the tank  cars began to
 arrive on-scene.   They were  unable to work  in the area because  of the
 thermal  inversion  in the valley.  Jim Littell and an air monitoring
team checked the  area  for vinyl  chloride,  hydrochloric acid,  phos-
gene,  and  chlorine.  No  harmful  chemicals  were detected.   Work  could
not begin  until the thermal  inversion lifted.

9:00 a.m.  - A meeting  was held  at the command post with  the OSC,  the
Kentucky  State  Fire Marshall, KDNREP,  the  Department of  Transporta-
tion,  and  Illinois Central Gulf.  Plans  for this  day were  as  follows:

     Remove chlorine and caustic potash  cars.
     EPA & KDNREP would  collect  soil  samples  and  water samples  (both
     well and surface waters).
     Area would be cleared of all personnel  except Hulcher's  work
     crew and State Fire Marshall.
     Hulcher would provide safety equipment  for their  personnel  and  a
     decontamination station.
                               2-8

-------
 10:00  a.m.  - Soil  samples  were  collected  from  several  areas  where
 spillage had occurred.   Duplicate  samples were taken,  one  to  be
 analyzed by the  EPA  laboratory, the  other by  state  laboratories.
     Another sampling crew collected  samples  from wells, cisterns,
 and  surface waters.

 Noon to 6:30 p.m.  -  The  caustic potash  and  two chlorine  cars  were
 removed and put  on trucks.   The vinyl chloride and  undamaged  ethyl
 acrylate cars were moved down the  slope toward the  tracks  on  the east
 side.  All  of the tank cars  were checked  for  leaks  by  Gulf Strike
 Team personnel.
     we received a report  that  the health care center  in Dawson
 Springs, Kentucky, was being evacuated.   We checked with the  chief  of
 police and  the Health Care Center  and found they were  evacuating only
 those patients with  serious  health problems.   The state  police,  State
 Fire Marshall, and the OSC all  agreed that  this  was not  necessary.

 7:00 p.m. - During the day,  several  inquiries were  received  from
 people who  had been  evacuated from their  homes.  They  were concerned
 about damage to  crops and  livestock.
     Mr. H.W. Dorough, toxicologist  for the University of  Kentucky,
 visited the command  post to  obtain information about the chemicals.
 Mr. Dorough received available  information  from  the Chemical  Hazards
 Response Information System  (CHRIS)  and Technical Assistance Data
 System (TADS) files.  He would  evaluate the situation  and  release  a
 report through one of the  state agencies.

 October 22, 1978

 6:00 a.m. to 6:00 p.m. - The coal  cars, chlorine car,  butadiene  car,
 and one of the hydrofluoric  acid cars were  moved and re-railed.   The
 hydrofluoric acid car was  not put  on  trucks.   Ramps were prepared  to
move the burning vinyl chloride cars  down the  slopes on  the  north  and
 south sides of the tracks.
                                 2-9

-------
      A representative from Penn Walt Chemical Company talked to the
 OSC about the spilled sodium chlorate.  He said that sodium chlorate,
 a strong oxidizer, is used as a herbicide; if mixed with any organic
 material, it would ignite.  He was concerned about Hulcher's person-
 nel smoking in the area.  Also, sodium chlorate is very soluble in
 water.   Therefore, any water applied in the area would allow the
 sodium chlorate to migrate through the culvert under the tracks.  The
 culvert was blocked on the south side of the tracks but open on the
 north side.  Unless the culvert were blocked on both sides, the
 sodium chlorate could run off into the culvert and be trapped under
 tracks.
      After receiving this information, no smoking was allowed in the
 area,  and the  culvert was blocked on both sides.
      Debbie Hockensmith and the OSC were concerned about the spilled
 material  entering Montgomery Creek via runoff from the wreck site.
 This  could be  controlled by sumps, trenches, or a retaining struc-
 ture.   Hulcher's  crew agreed  to construct a retaining dike to control
 the runoff.  The  structure was completed at 6:30 p.m.

 8:00  p.m.  -  There was a report that a resident of Dawson Road had
 been  overcome  by  toxic fumes.   Joe Lafornara and  a trooper for the
 State Highway  Patrol  responded to the complaint.   They found that it
 was a prank  call.

 October 23,  1978

 7:30  a.m.  -  The OSC called Al  Smith,  Chief of the Environmental  Emer-
 gency Branch,  EPA,  Region IV.   Mr.  Smith was  informed that the two
 vinyl chloride cars were still  burning  and probably would  burn for
 several days.  He  suggested that  holes  be  cut in  the  tanks using
 plastic explosives,  and  that the  vinyl  chloride be ignited in a con-
trolled area.
     The OSC talked to  Rod  Raby about Mr.  Smith's  proposal.   The
 State Fire Marshall's office was  considering  this  possibility.   Some
time ago, they had  used  an  Army  EOD team to  cut  holes in a butadiene
tank car near Louisville,  Kentucky.
                                2-10

-------
 10:00  a.m.  - Mr.  Raby  said  that  his  office  was  having  trouble getting
 an Army  EOD team.   He  stated  that  it  could  take four-to-five  days  for
 his  agency  to  get  a team  on-scene.   He  asked  if the OSC  could speed
 up the process  and  get  an EOD team out  of Fort  Campbell,  Kentucky.
 Mr.  Raby was told  that  the  OSC would  have to  make  a request to the
 Chairman  of the National  Response  Team  (NRT), Mr.  Ken  Biglane.

 10:20  a.m.  - The OSC called Ken  Biglane, EPA  Headquarters, Washing-
 ton, DC,  and requested  an Army EOD team.  Plans were to  cut two holes
 in the tank car that was  burning at  the dome, and  a single hole in
 the  other burning  tank  car.   This would allow the  liquid vinyl chlo-
 ride to  drain  out  into  a  large diked  area for re-ignition.  Mr.
 Biglane  said that  he would make  the  request through the  EOD represen-
 tative on the  NRT,  Col. Sadler.
     The  removal of the remaining derailed  tank cars continued.  The
 two  tank  cars of hydrofluoric acid would be the last cars to  be put
 on the trucks.

 2:50 p.m. - Col. Sadler informed the  OSC that he could set up direct
 liaison with the 17th  Explosive  Ordnance Detachment at Fort Campbell,
 Kentucky.

 3:10 p.m. - The OSC contacted Sergeant  Hobbs  of the EOD  and explained
what needed to be done.   The  EOD team would arrive at  the command
 post at 6:00 p.m.

4:00 p.m. - A  light rain  had  been falling most  of  the  day and a white
cloud began to settle  in the  valley  around  the  wreck site.  Jim
 Littell and the OSC took  the  air monitoring equipment  to the  site.
 The rain mixing with the gases from  the burning vinyl  chloride cars
formed an acid mist.  This was very  irritating  to  the  nasal passages,
which made  it difficult to breathe and  unpleasant  to work.

Hulcher's crew stopped  the removal operation  for the day.
                               2-11

-------
 6:00 p.m. - The EOD team arrived on-scene, met with the OSC and the
 State Fire Marshall, and was given an explanation of what needed to
 be done.  They had worked with vinyl chloride before, but not burning
 tank cars.  They were confident, however, that the job could be done.
 It was further explained that the holes would not be cut in the
 burning tank car until all  of the other tank cars were removed from
 the area.
      On the following day,  the EOD team would conduct a practice shot
 on a piece of scrap metal in an isolated area.

 7:30 p.m. - Ben Eason and the OSC had a conversation with Rod Raby,
 Ed Clark, and Bob Tittle.  Mr.  Raby said that he was envious of EPA
 operations because of the command post, the public information group,
 and the response of the news media.   He claimed that the State Fire
 Marshall's office did not have any money or equipment.  They had to
 borrow  all of their equipment.
      The OSC asked Mr.  Raby what assistance he needed.  He wanted
 publicity in the news media to  get recognition for his program.  The
 OSC told him that Mr. Eason could set up a news conference so that
 the State Fire Marshall's office could  explain the procedure for cut-
 ting  the holes in the burning tanks.   Perhaps they could arrange for
 the news media to  get pictures  and film of the detonation of the
 burning  vinyl  chloride  tank cars.
      It  was  the OSC's understanding  that the State Fire Marshall's
 office  had to  coordinate  this type of activity through the DES coor-
 dinator.   Mr.  Raby indicated that  he  didn't want to comply with the
 state contingency  plan, primarily  because DES had  to  make news
 releases  from  the  Frankfort office.   He said it was a poor system.
     Mr.  Eason  made  several  calls  to  the news media and  arranged a
 news conference for  the State Fire Marshall's office  for 11:00 a.m.
 on  October 24,  1978.

 October 24,  1978

8:00 a.m.  - The  EOD team  arrived  and  requested the OSC to sign three
civil release forms.  They  proceeded  to  the  wreck  site for  the prac-
tice shot.  Hulcher's crew  continued  to  re-rail  the damaged  tank
cars.
                                 2-12

-------
11:30 a.m. - The news conference was conducted by  Rod Raby.   EPA did
not participate.

1:30 p.m. - The practice shot on a  piece  of  scrap  metal  was  con-
ducted.   It was a complete success.

2:00 p.m. - David Hill, EPA, arrived at the  command  post to  take soil
and water samples back to EPA's laboratory in Athens, Georgia.

5:00 p.m. - Major Salter of the EOD team  arrived at  the  command post.
He stated, in the presence of Rod Raby, that his instructions from
Col. Sadler were to coordinate all  of the team's activities  through
the On-Scene Coordinator.  He wanted to explain a  proposed plan for
the detonation of the tank car to both the OSC and the State Fire
Marshall.  The plan was as follows:
     Set charges in both tank cars  and let vinyl chloride run into
     the contained area and ignite.
     Obtain self-contained breathing apparatus, protective clothing,
     and backup unit.
     Obtain background data on air, water, and soil  just prior to
     detonation.

     The OSC told Major Salter that the personnel  safety equipment
was available, and that soil and water samples had been  collected for
analysis.  The air would be monitored just prior to  the  detonation  by
Jim Littell and state personnel.
     Mr. Raby was under the impression that  Hulcher  would have all  of
the damaged tank cars out of the area by  6:00 p.m.  He wanted to  set
the charges on one of the tank cars that  evening.  However,  Jim
Littell advised the OSC and Mr. Raby that a  thermal  inversion would
occur that evening.  He advised not to detonate the  tank car that
evening because the inversion would keep  the gases close to  the
ground.  This could cause some air  problems  in the area  of Dawson
Springs, Kentucky.
     The OSC told Mr. Raby that Mr. Littell's  advice had to  be con-
sidered.  The OSC agreed that the tank cars  had to be detonated and
                                 2-13

-------
 he would accept a short-term risk with air problems versus a  long-
 term risk of letting tank cars burn out.  However, if the conditions
 would be better at a later time then perhaps a waiting period was in
 order.
      Mr. Raby received word from Mr. Hogan that the tank car of
 hydrofluoric acid and sodium chlorate wouldn't be moved until 11:00
 a.m. the next day.  Thus, the decision was made.  The tank cars would
 not be detonated until  all of the other tank cars had been removed.

 October 25,  1978

 8:00 a.m. -  Mike Donohoe, Jim Littell, and Kenneth Yates were stopped
 at  the check point by the State Highway Police.  They were not
 allowed on-scene under  orders from the State Fire Marshall.

 9:50 a.m. -  Ben  Eason informed  the OSC that state personnel were con-
 cerned about questions  from Frankfort, Kentucky, regarding the treat-
 ment of the  state  in  the  press.

 10:00 a.m. - Mr.  Raby wanted  the EOD team to go directly to the wreck
 site as soon as  it arrived.   The OSC wanted to talk to Major Salter
 before he went on-scene about his plans and signing civil  release
 forms.
      When the EOD  team  arrived, Major Salter and the  OSC reviewed
 the  plan  for detonation.   The OSC also told him that  there were some
 problems  with the  State Fire  Marshall,  the reasons for which were
 unknown.   However, the  OSC assured  Major  Salter that  he would support
 his  plan  in  any  way possible.

 10:20  a.m. - The EOD  Team,  Jim  Paskewich,  and the OSC  went to the
wreck  site and met with Rod Raby.
     Once  again, Major  Salter reviewed  his plan.   Mr.  Raby agreed to
the  plan, with two exceptions:
     EPA  air  monitoring team  would  not  be  allowed on-scene to gather
     background data.
     The GST could provide backup  as  long  as  they remained outside
     the evacuated zone.   This  was one mile  away.
                                2-14

-------
      After  some  discussion,  Mr.  Raby said  that  he  would  allow the OSC
to observe  the detonation, but no one  else from EPA or the  U.S.  Coast
Guard.   He  would allow  Ben Eason to  set  up a  movie camera  in  the
bucket of a bulldozer.   Once this was  accomplished,  Mr.  Eason was to
leave the scene.   Prior  to the detonation, the  Army photographer, who
was part of the  "firing"  part, could  trip  the camera.
      It  was obvious  to  the OSC that  Mr.  Raby  wanted  to demonstrate
his authority.   The  OSC  chose to monitor the  radio in the command
post  rather than challenge his authority and  risk  getting someone
hurt.

1:00  p.m. to 5:00 p.m.  - The EOD team  set  charges  on both tank cars.
All of Hulcher's equipment was moved out of the area.  Just prior to
ignition of the  fuses,  Jim Littell,  Kenneth Yates, Charles  Dai ley,
and troopers for the State Highway Patrol  took  up  stations  on Highway
672 to follow and monitor the plume.

5:16  p.m. - The  fuses were ignited,  but  only  the charges on one tank
were  detonated.   A huge,  black plume evolved  from  the  ignition of the
vinyl chloride.   It  was  raining  at the time,  which held  the cloud
down.
      A report came in from Mr. Littell that if  the wind  shifted, the
fumes could affect several houses.   He recommended to the state
police that the  evacuation zone  be increased.   The rain  mixing with
the gases was creating  an acid mist.   Mr.  Littell  noticed a burning
sensation on his  skin.   He recommended to  state police that people
beyond Highway 62 should  remain  indoors  and stay out of  the acid
mist.

5:47  p.m. - The  EOD  team  reported that plastic  charges cut  a  hole in
the top of  the second tank car but not the bottom.  Another plastic
charge would have to be  placed in the  bottom  of the tank car.

9:29  p.m. - The  second charge on the tank  car was  detonated and the
vinyl chloride was ignited. The  same conditions that prevailed on the
ignition of the  first tank car existed at  this  time. Jim  Littell and
the state air people monitored the plume.
                                 2-15

-------
     Both of the tank cars were  successfully  ignited with complete
burning of vinyl chloride.  Only one complaint was received from  an
elederly man who was caught in the acid mist.  He complained  about a
burning sensation when he swallowed.   It was  recommended that the
rescue squad take him to the hospital  where he could be examined  by a
physician.
                                2-16

-------
3.   PHOTOGRAPHS
     3-1

-------
I .
                                      AERIAL VIEW OF 18-CAR DERAILMENT
                                          NEAR CLAXTON, KENTUCKY

-------


VINYL CHLORIDE TANK CARS ADJACENT TO HOPPER CARS
            CONTAINING BURNING COAL

-------
CO

I--.
                                 VINYL CHLORIDE TANK CARS. WHITE TANK CAR
                                       CONTAINS HYDROFLUORIC ACID

-------
OJ
(Jl
                                TANK CARS, TWO OF WHICH CONTAIN CHLORINE

-------

co
01
                               VINYL CHLORIDE TANK CAR (NOTICE FROST LINE)

-------
CLOSE-UP OF RUPTURED TANK CAR WITH BURNING
      VAPOR PHASE OF VINYL CHLORIDE
                   3-7

-------
U)

00
                                  HEAVY EQUIPMENT MOVING TANK CARS TO
                                          NORTH END OF TRACKS

-------
CO
I
UD
                                  CHLORINE TANK CAR IN FOREGROUND WAS
                                    REMOVED FROM BURNING COAL CAR

-------
I
I—-
o
                               TANK CAR OF HYDROFLUORIC ACID BEING PREPARED
                                             FOR RELOCATION

-------
CO
                                 HYDROFLUORIC ACID TANK CARS IN RELATIONSHIP
                                     TO THE BURNING VINYL CHLORIDE CARS

-------
I
K^
ro

                               TANK CARS Of CHLORINE AND VINYL CHLORIDE
                                       REMOVED FROM WRECKAGE

-------
DAMAGE TO OUTER SHELL OF TANK CAR
              3-13

-------
TANK CARS RE-RAILED ON NORTH END OF TRACKS

-------
co
i
                                 COMMAND POST FOR FEDERAL, STATE, AND LOCAL

                              OFFICIALS LOCATED ABOUT TWO MILES FROM WRECK SITE

-------
I
1—1
en
                                     ANOTHER VIEW OF THE COMMAND POST

-------
                       4. PROBLEMS ENCOUNTERED
     The cooperation between the participating state  agencies  and the
EPA On-Scene Coordinator was remarkable for the first  seven days.
Air, water, and soil samples were collected.  There was  an exchange
of information concerning the properties of the chemicals  involved in
the derailment.  The operation was smooth and efficient  up to  the
point just prior to the venting of the burning tank cars.  During the
last 36 hours, however, several problems arose.

October 24, 1978

     An incident occurred between Jim Littell and Rod  Raby just  prior
to the practice shot.  Jim Littell was with the EOD team which was in
the process of setting the charges.  Rod Raby advised  him  that he was
too close.   Mr. Littell explained that he had the radio for the
team.  Mr. Raby stated that the team didn't need  a radio and to  move
out of the area.  Mr. Littell said that he was a  big  boy now.  Mike
Donohoe, who was in visual contact with both parties,  told Mr.
Littell that maybe he was too close.

     The OSC, who was monitoring the radio  in the command  post,  heard
the conversation.  He advised Mr. Littell that he should move  out of
the area.  Littell reponded, "Okay, that's three  and  you are the
boss—I'm gone."

     From then on there was some friction between Mr.  Littell  and Mr.
Raby.
                                 4-1

-------
 October 25, 1978

      The State Fire Marshall's representative refused the entrance of
 EPA and Coast Guard personnel on-scene, giving the reason that  it was
 unsafe (however, at the time of the detonation, 60 people were  re-
 ported to be near the tank cars).  In addition, previous authority to
 enter the area granted GST personnel to set up backup equipment for
 EOD was suspended.

      By not allowing EPA and state air monitoring personnel on-scene
 prior to the detonation, it was impossible to collect data on air
 quality.  The OSC asked Mr. Raby his reasoning for not allowing EPA
 to  monitor  the air.   He said that Mr.  Littell  could not come on-scene
 because it  wasn't safe and that B.F. Goodrich could monitor air with
 their instruments and  test for vinyl  chloride.  The OSC told Mr. Raby
 that EPA could not  support any of the data taken by B. F. Goodrich
 unless  EPA  personnel  were  allowed on-scene.   In addition,  EPA wanted
 to  test for  hydrochloric acid and phosgene, as well as vinyl
 chloride.  Mr.  Raby  said that EPA would  have to monitor the air one
 mile outside the  evacuation zone.

      Privately, Mr.  Raby stated to the OSC that he had called EPA in
 Washington,  DC.   It  was  his understanding  that the OSC could take
 over  the whole  operation.   If that occurred, he would withdraw all of
 the  state resources.   The  OSC stated that  it was  his  intention to
 work with the  State  Field  Marshall,  not  take over the operation.

     Moreover, the Gulf  Strike  Team  had  been requessted  by  the EOD
team to provide backup with personnel  equipment,  so that, if one of
the EOD team members were  hurt, Gulf Strike Team  personnel  would
respond and  remove him to  a safe  location.   The  State Fire  Marshall
insisted that the backup unit remain outside the  evacuated  zone.
This was totally unacceptable to the On-Scene  Coordinator.
                               4-2

-------
     The basis for the State Fire Marshall's reactions  is  not  known;
it could have been a personality conflict or a breakdown  in communi-
cation.  Regardless, the success of the operation was jeopardized  by
actions taken by the Fire Marshall.

     Also on October 25, Ben Eason informed the OSC  that  state per-
sonnel were concerned about questions from Frankfort, Kentucky,  such
as, "Isn't anyone from the state of Kentucky down there?   All  they
see is EPA in the news media."  This was not true; the  news media
interviewed state personnel several times.  There was cooperation
with the public information officer for DES, Mr. Gordon Nichols, on
all news releases until he left.

     Mark Holcomb said that everything in the newspaper was about
EPA and that he wasn't going to allow that to happen again in  his
area.  The OSC reminded both Mr. Raby and Mr. Holcomb that there were
federal laws and state laws.   If a spill occurred again in Kentucky,
EPA would respond and utilize  all of its resources to protect  the
environment.

     As far as the newspapers  were concerned, the OSC stated  that  he
could not control what stories were written.  All he could do  was
give them factual information  about the spill event, as outlined in
the National Contingency Plan.

     A meeting took place between EPA and the various state agencies
in November, 1978, to discuss  the problems associated with the spill
incident.  The attached memos  present information on the  meeting and
its outcome.
                               4-3

-------
5.  ANALYSIS OF AIR, WATER, AND SOIL  SAMPLES
                    5-1

-------
     »Tt:


 SUBJECT:




    FROM:



     TO:
        UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                  REGION IV, Athens, Ga.  30605

December 5, 1978

Trip Report 10/18/78 - 10/26/78
Technical Assistance Request for Air Monitoring Support -
Train Derailment, Princeton, Kentucky
James K. Littell
Air Surveillance
Doyle T. Brittain,  Chief
Air Surveillance Branch
           SUMMARY

           On 10/18/78 we received a telephone call at 0020 requesting assist-
           ance involving assessing air quality conditions around a 18 car
           derailment in Princeton, Kentucky.  Mr. Jack Stonebreaker,  designated
           On Scene Federal Coordinator, requested our assistance to accompany
           him to Princeton, Kentucky for technical assistance pertaining to
           assessing air quality toxicity and explosive levels, evacuation perim-
           eters necessary to support environmental human and animal life in the
           wreck area, and possible long-term damage assessment affecting environ-
           mental surrounding caused by any air contaminant leakage from possible
           ruptured vessels involved in the train derailment.

           We assembled our necessary air sampling equipment and departed for
           the derailment scene at 0500 on 10/18/78.   Mr.  Stonebreaker and I
           arrived at the derailment at 11:00 on 10/18/78.  After visual inspec-
           tion of the scene we found the following condition at the derailment
           site:

               Eighteen cars were derailed on a 50 to 60 foot deep
               slope angling approximately 30 degrees from the rail-
               road tracks  downward.   The derailed cars were reported
               to  contain the following according to  the railroad
               manifest  supplied.   The  disposition of each was sur-
               veyed by  EPA representatives  on scene.

               6 Vinyl chloride tanks -  full - two burning - other
                4  intact -  liner  structure  unknown.

               1 Sodium  chlorate  tank -  contents  partially spilled
                and mixing  with ethyl  acrylate

               2 tanks ethyl  acrylate -  1  open and spilling -
                eventually mixed  and reacted  with sodium  chlorate

               1 liquid potassium  hydroxide  ruptured with  contents
                spilled  on ground.

              2 hydrogen floride - intact but in  a dangerous  position
                for possible rupture.
EPA FORM 1320-6 IRKV. 3-76)
                               5-2

-------
                                —2—

     1  butadiene -  intact - located adjacent to open ethyl
       acrylate car - possible explosive hazard

     3  chlorine tank cars - one tank ruptured on the outer
       core with a  coal fire burning around and under the
       exposed shell.  The other two chlorine cars were
       apparently intact, but positioned on their respective
       safety valves rendering the safety relief valves in-
       operative.

 I advised the Federal On Scene Coordinator to evacuate the wreck area,
 let  the scene be naturally flushed by the atmospheric air before con-
 tractors should be allowed to attempt any cleanup.  Also, I advised a
 2 1/2  raile evacuation radius based on the thermally explosive nature
 of the remaining cars and the potential toxic concentrations of chemi-
 cals that could potentially be emitted violently in the case of an on
 scene  explosion.

 No personnel other than EPA or State Fire Marshalls were allowed on
 scene  during this  environmental purging process.

 Five types of on scene air monitoring tests were conducted throughout
 the entire episode including chlorine, vinyl chloride, hydrogen chloride,
 phosgene and carbon monoxide.  Additional emergency sampling tubes were
 ordered and received on 10-20-78 at 0200 to refill the depleted supply
 used at the wreck  scene and surrounding neighboring areas.

 During the entire episode from 10-18-78 to 10-26-78 EPA Air Surveillance
 personnel and other Coast Guard and EPA personnel provided the following
 assistance for the On Scene Coordinator.

 1.  Checked the wreck scene each morning before fire and environmental
 personnel entered  to assure no lethal concentrations would overcome the
 entering personnel.  Example:  8 dead in the chlorine episode incident
 in Youngstown, Florida in 1978 when local populas rushed to the scene of
 that derailment.

 2.  Responded to all air complaints from law enforcement and local citi-
 zens involving obnoxious odors, clouds of smoke, etc. mostly occurring
 during evening hours between 2100 till 0900 in the morning when ground
 level  inversion conditions would lift.

 3.  Provided technical assistance to Mr. Stonebreaker, Federal On Scene
 Coordinator, by attending a two hour town meeting held on 10-21-78 in
 Dawson Spring, Kentucky.   This was requested by Jack Stonebreaker to
 squelch rumors of unsafe air conditions being allowed to accumulate in
 that town originating from the wreck area.

4.  Provided advice on the extc-nt of a safe evacuation perimeter to be
used.  This was constantly verbally updated depending on working condi-
 tions and the potential of hazards to vessles being moved at the wreck
                                 5-3

-------
                                 -3-

 sccne during cleanup operation being conducted by the Hulcher Corpora-
 tion and the Illinois Central Gulf Railroad crews.

 The following serious problems were encountered and should be addressed:

 State of Kentucky Fire Marshall personnel would not allow any environ-
 mental personnel on or near the scene of the wreck during the last two
 days of the incident.  The reasoning stated was not safety but they did
 not want Federal people being made aware of the progress of the cleanup
 because of publicity releases to the press constantly quoting EPA per-
 sonnel as to the disposition of the public safety involved and environ-
 mental damage to areas surrounding the derailment scene.  This caused
 unsafe working conditions to all civilian, local, state and army person-
 nel on the derailment scene during the final stages of the cleanup.

 The Kentucky State Fire Marshall's supervisor would not allow EPA air
 personnel on scene for safety inspections on the day of demolition,
 10-24-78.   They said that B.F.  Goodrich's personnel were monitoring
 for vinyl chloride using a continuous vinyl chloride analyzer.   This
 same analyzer was used by EPA and State Air Pollution personnel on
 10-23-78 and found to be unreliable and unstable.  Local hydrocarbon
 interferences,  either negative  or positive, were not addressed.   Also,
 Kentucky State  Fire Marshalls provided Army personnel with a lower
 explosion level  meter that,  as  like the B.F.  Goodrich analyzer,  was
 not calibrated.   The EOD personnel did not know this at the time.   On
 10-24-78 at  20:54 "batteries  went  dead" in this explosion meter  being
 used  to  monitor  the wreck scene  combustible gas concentration while
 Explosive Ordinance Demolition  personnel  were  preparing to detonate  the
 second of two vinyl chloride  tank  cars.   Fire  Marshall Rabys radioed
 instructors  to "hit it on its side."   These unprofessional sampling
 techniques could  have  resulted  in  the  loss  of  life  of the four Explo-
 sive Ordinance personnel  from Fort  Campbell, Kentucky plus all bystanders
 allowed  on scene  by  the  Kentucky State  Fire Marshall.

 The evacuator's perimeter for the wreck scene  set by the  State of
 Kentucky Fire Marshall's  Office was between 3/4  to  one mile.  EPA
 personnel requested  a minimum of two and  one-half miles due  to the
 severity of  the compounds.  No increase in  evacuation  perimeters
were requested by  the Fire Marshall's  Office during  the actual
 detonation of the vinyl chloride tanks.   This  resulted in the unneces-
 sary exposing of  local residences downwind  of  the detonation.  The
 evacuation perimeter was  enlarged when  EPA  personnel visually observed
plume or black smoke discharged from the  blast which was  approximately
300 to 350 feet high, 1/2 mile wide and of  unknown length, set down
on the downwind residential areas after the first detonation.  State
police offices,  local populous, and State and  local  Civil  Defense and
environmental personnel voiced considerable concern  about  the breathing
conditions and acid mist  forming from this discharged  plume.  The plume
                               5-4

-------
                                -4-

did not dissipate rapidly and citizens downwind of the detonation were
reportedly  taken to  the local hospitals for treatment from breathing
restrictions and burning skin.  EPA and State Air Pollution peronnel
downwind of the detonation experienced skin burning, clothes were
bleached with white  spots from the acid mist.  Breathing was laborous
after  the exposure but it was not known if it was fron the fumes or
lack of sleep.  No detectable readings of vinyl chloride, hydrogen
chloride, or phosgene were detected at approximately 5 miles downwind
of the detonation area and inside the black plume discharge.  Burning
of the skin was extremely evident.

The Princeton Fire Department was on scene during the demolition and
allerted to go into  the wreck area if a house located on a bluff over-
looking the blast area was to catch fire.  They were not made aware of
any of the dangers resulting from high temperature burning of vinyl
chloride resulting in phosgene, or from the vinyl chloride liquid
itself should they coine in contact with this principle substance.  I
talked with the Princeton Fire Chief about protective equipment and
made him aware of the danger.  This action could have caused unneces-
sary health problems or possibly a loss of life to one of the local
fire fighters or to  the 50 to 75 spectators positioned approximately
150 yards upwind of  the blast area.

ACTION

Action should be immediately taken to stop Kentucky State Fire Marshalls
from 1) Grossely and unnecessarily endangering the local populations and
road travelers from  inadequate and unprofessionally devised evacuation
perimeters.  Example:  3/4 to one mile during detonation of 30,000 gls.
of vinyl chloride 10-24-78.  2)  From hindering State and Federal air
pollution specialists from doing their duties of protecting cleanup
workers, State and Federal environmental and investigation personnel,
etc.,  and local population and fire fighting crews by illegally instruct-
ing State Police officers to keep these specialists away from the scene.
3)  From providing inadequate and uncalibrated air monitoring equipment
and unprofessional advice to personnel from the Explosive Ordinance
Demolition teams or  similar professional organizations called into the
area when they are assisting in disposing of this type hazardous material
containers.

A conference should be set up with representatives of State agencies
involved to determine the responsibilities of each agency during this
type emergency to eliminate this type dangerous situation happening
again.

BACKGROUND

Request from Mr. Jack Stonebreaker of 10-18-78 for assistance in respond-
ing to train derailment emergency of 10-17-78.
      K. Littell
                                5-5
     Al Smith

-------
    DATE:


 SUBJECT:
          UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
            Region IV, Surveillance and Analysis Division
11/2/78        College Station Road, Athens,  GA  30605

Analysis of Samples from Train Wreck at Princeton,  KY
    FROM:
     TO:
Tom B. Bennett, Jr., Chief
Analytical Services Section

David Hill, Chief
Ambient Monitoring Section
           SUMMARY

           Attached are the results  of  analysis  of  the Princeton, KY, train
           wreck samples (SAD Nos. 78C2943-46).

           The water sample (78C2944) was analyzed  by the purge-and-trap
           procedure with no vinyl chloride or other volatiles being detected
           with a minimum detection  limit of 5 ug/1.

           The sediment  samples were all analyzed by a head-space procedure
           for vinyl chloride and other volatiles.

           The sediment  samples were extracted with methylene chloride and
           the quantitations on the ethyl acrylate were made from this
           extraction.

           The gas  chromatograph/mass spectrometer was used for the above
           analyses.

           Five grams of  Sample No. 78C2946 were diluted in water and analyzed
           by  the automated  indophenol procedure.  Approximately 40 mg/kg of
           ammonia was found  in this sample.

          No  other  compounds were identified above a 0.5 mg/kg detection
          limit.

          ACTION

          Transmittal of data.

          BACKGROUND

          Request for analysis.

          Enclosure
EPA FORM 1320-6 IRETV. 3-76>
                          5-6

-------
Project   Train Wreck
                         DATA  REPORTING  SHEET
J- 1 lllCCLUU, &.I 	 • • 	 — ______ - . . ..... — __, _, , v>
ANALYSES TO BE RUM
AD //
3C
943
944
945
946

in
i
^sl










STATION //
1
SSS*'lf.
:/.*ri£ 1A
2
3

ND- None detected.
Tr- Trace
NA-- Not analyzed for.
i/ Tentative identifi
Minimum detection lira!
* - Minimum detection





DATE
SAMPLED
__
10/23/78
1535
—
—




nation
£s are in paren
Limit for water





Vinyl
Chloride
mR/kf>
ND(0.5)
ND(''0
ND(0.5)
ND(0.5)





heses.
sample is





Ethyl
Acrylate
ms/klR
4.4
NA
28,000
ND(0.5)






5 'yg/t.





Ethyl-nor-
Propanoatei
ms/kp.
ND(O.S)
NA
tr<0.5
ND(0.5)












r




































































-------
                              Bqjartmrnt ©f Agrtrulturp

10-2/1-78                             FRANKFORT


Gas Chroma tographic Procedure  for the Determination of Ethyl Acrylate Residues  in Soil


Reagents

                                                          7**^ ****  &*#'"
 Ethyl  Acryl ate Standard,


 Apparatus

 50 Ml  Screw-Top Erlenmeyer flasks (modified by boring a hole  in  the  plastic
 cap without puncturing the septum)

 5000 microliter gas- tight syringe

 Gas chrornatograph equipped with a flame ionization detector and  a  1.5% SP 2250,
 1.95%  SP2401 column

 Constant temperature water bath
Procedure:

Sample is ground to pass a 20 mesh sieve.   Sample  and  standards are weighed into
separate 50 ml flasks.  Flasks are capped  and  placed in a boiling water bath for
one hour.  At completion of the volatilization period, with flask remaining in the
water bath, the septum is pierced and the  sample of the volatile matter is
chromatographed under the following conditions:

              Column temp:              75 C
              Flow:                     24 ml/min
              Attn:                     10
              Inj  temp:                  105 C
              FID  temp                  250 C
              Slope Sens:               .05
                                         5-8

-------
       r%
         \
          '
         UNITED STATES
ENVIRONMENTAL PROTECTION AGENCY
      WASHINGTON. D.C. 20«6O
                        REPORT OF ANALYSIS
                                        t. SAMPLE NO.
                                           1,2  &  3
   3. REGION
                                                                               2. DATE COLLECTED
                       4. EPA REG. NO.
                                                           5. ESTABLISHMENT NO.
 6. DESCRIPTION OF SAMPLE
 	Three  soil samples in glass containers
 7. NAME AND ADDRESS OF ESTABLISHMENT WHERE SAMPLE WAS COLLECTEOOnc/urfc ZIP code)
                                                                        8. PRODUCT NAME
                                                                            Ethyl Acrylate
                                                                            Residues
                                                                        9. LOT OR CODE NUMBER(S)
 10. NAME AND ADDRESS OF PRODUCER (If different from 7 above)
 tl. RESULTS OF ANALYSIS
      Sample
      Soil-1
      Soil-2
      Soil-3
          Ingredient
          Ethyl Acrylate
          Ethyl Acrylate
          Ethyl Acrylate
Method
GC-FID
GC-FID
GC-FID
Amount Found
Negative
173  ppm
 55  ppm
 »2. LABORATORY COMMENTS
 13. SIGNATURE Oi  LAO SUPEnv'1'.OR
      Scott  Brynn
                                        t«. LABORATORY
                                        Kv-s.t Fed Meat La
                                                                               15. DATE
                          10-24-:
EPA Form 3540 i (f
-------
         Illinois
         Central
         Gulf
An tC Induslrie* Company'                                                Illinois Central
                                                                  Gull Railroad
                                                                  Two Illinois Cooler
  November  9,  1978    ;••.',  -  •        •  -.-                          233 North Michigan Avenue
                     ••-: •.  •   -,..     •'..'.                        Chicago. IL 60501
                     1 •• ,'.'•' v	"  -:..;;':                        (312)5651600
 Mr. Dame! R. Dolan
 Chief, Hazardous Materials & Waste  Management
 Division of Hazardous Material Waste  Management
 Pine Hill Plaza
 Frankfort, KY 40601

 Dear Mr.  Dolan:

 Reference is made to our meeting on November  7,  1978  at Earlington,
 Kentucky concerning possible environmental  contamination resulting from
 the train derailment at Mile Post 171, Claxtcn*  Kentucky.   In  an effort
 to determine the extent of any future environmental problems at the
 site,  we agreed to execute the following procedures:

 1)  On the north side of the track in the  area  where the  ethyl  acrylate
     and  vinyl  chloride compounds were diked,  qualitative  samples of
     the  soils  in the dike will be taken at weekly intervals to  determine
     the  rate at which dissipation occurs.

 2)  On the north side of the track, core samples of  embankment  soils
     will  be taken to determine the extent  to  which the ethyl  acrylate
     soaked into the embankment.

 3)  The  present dike on the north side will be  enlarged to insure that
     all  residual  materials in the dike will be  sufficiently contained.
     Additional  grading will  be used to insure the surface runoff will
     be directed away from the dike area.

4)   The  present dike on the south side of  the track  containing  the
     vinyl  chloride residuals,, will  be modified  to insure total  contain-
     ment.

5)   A  well will  be dug at the toe of the embankment  on the south side
     to monitor  the ground water  quality of the area.    Samples will be
     taken  at monthly intervals.

6)   An additional  soil  sample  on the south side at the toe of the
     embankment  will  be analyzed  qualitatively for vinyl chloride.
                                  5-10

-------
                                -2-
A11 results of these analyses will be forv/arded to you as soon as
possible.  I trust that this will meet with your approval, and if there
are any other questions please feel free to contact me on Extension
2384.
Very truly yours,
M. (7. McKinney
Environmental Engineer

cc:  Mr. John E. McClure, Jr.
     Envrfonmental Supervisor
     Hazardous Waste Section
     Pine Hill Plaza
     Frankfort, KY 40601

     Mr. Jack Vfatkins
     Environmental Specialist
     Division of Hazardous Material  & Waste Management
     Pine Hill Plaza
     Frankfort, KY 40601
                                5-11

-------
*• "'< OMITCO STATES
.? MA \ ENVIRONMENTAL PROTECTION AGENCY
5 ^X'^7 ~ KASMINCTON. D.C. 2046C
''H: "'-"•'' REPORT OF ANALYSIS
1. SAMPLE NO.
127376
3. REGION
2. DATE COLLECTED
10-27-78
4. EPA REG. HO.
S. ESTABLISHMENT NO.
IS. DESCRIPTION Or SAMPLE
I
        5 Soil &  1  Liquid samples  in glass containers
~i. NAVE AND ADDRESS OF ESTABLISHMENT Vr'HEKE SAMPLE WAS COLLECTED'/nc Jud«- Z1T code) B. PRODUCT NAME
Vinyl Chloride &
r 1 Ethyl Aery late
Residues
L J
9. LOT OR CODE NUM3EWIS)
12. NAME AND ADDRESS OF PRODUCER (II ailU-rcnt from 7 above)
11. RESULTS Or ANALYSIS
Sample I Ingredient
4 Vinyl Chloride
Vinyl Chloride
6 Vinyl Chloride
7 Vinyl Chloride
8 Ethyl Acrylate
9 Vinyl Chloride
Method
GLC-EC
GLC-EC
GLC-EC
GLC-EC
GLC-FID
GLC-EC
Amount Found
Positive
Positive
Trace*
Positive
Positive
Trace*
.12. LA-JORATORY COMMLNTS
    * Compounds could  be  showing contamination arid  not actual residues
''3. MC-N'-M UFT OT LAM 5'J f !:' ?< VI y
*   Scott  [;ryr;n        /-
                                            5-12
jl*. I A riOHAI O;/>i n
Ob- [ 11-1-
                                                                                 78

-------