Process Safety Management (PSM) Module 1 - Review of Industrial Catastrophes Related to PSM July, 2010 [Adapted from State of Ohio – Dept. of Commerce, Div. of Labor & Worker Safety]
A series of catastrophic releases of chemicals leading to fires, explosions and fatalities have occurred in chemical processing plants around the world over the years. These incidents lead to the passage of the Process Safety Management Rule in 1992. History of PSM Catastrophes
Purpose of the PSM Standard • This standard contains requirements for preventing or minimizing the consequences of catastrophic releases of chemicals that are: - Toxic, • Reactive, • Flammable, • Explosive • These releases may result in toxic, fire or explosion hazards • A number of catastrophic accidents have occurred resulting in loss of life and great property damage.
Ammonium Nitrate Explosion Oppau, Germany – Sept. 21, 1921 Photo shows crater and destruction at plant following explosion. At 7:30 a.m. on September 21, 1921, two powerful explosions occurred at the BASF plant in Oppau, Germany. Hazardous material: Ammonium sulfate & ammonium nitrate (50/50) Facility type: fertilizer manufacturing Deaths: 430-530 (differing numbers on reported fatalities)
Ammonium Nitrate Explosion Texas City, Texas – April 16, 1947 This aerial photograph , looking south over Monsanto Chemical Co., was taken about 30 minutes following the blast of the ship S. S. GRANDCAMP during loading of ammonium nitrate. The accident damaged more than 90% of the city's buildings and killed nearly 600 people. Source: Fire Prevention And Engineering Bureau Of Texas
Major PSM related incidents1974 - 2001 Source: Wharton Risk Management and Decision Processes Center of the University of Pennsylvania.
Seveso, Italy - 1976 The Seveso accident happened in 1976 at a chemical plant manufacturing pesticides and herbicides. A dense vapor cloud containing tetrachlorodibenzoparadioxin(Dioxin) was released from a reactor, used for the production of trichlorofenol. Commonly known as dioxin, this was a poisonous and carcinogenic by-product of an uncontrolled exothermic reaction. Although no immediate fatalities were reported, kilogram quantities of the substance lethal to man even in microgram doses were widely dispersed which resulted in an immediate contamination of some ten square miles of land and vegetation. More than 600 people had to be evacuated from their homes and as many as 2,000 were treated for dioxin poisoning. This lead to the European “Seveso Directive” to try to prevent similar incidents.
Cyclohexane Release & ExplosionFlixborough, England – June 1, 1974 20” bypass piping fabricated on-site from shop stock. This pipe ruptured and released cyclohexane which exploded. Source: UK Health and Safety Executive, Hazardous Installations Directorate On June 1, 1974 the Nypro Co. site at Flixborough, England was severely damaged by a large explosion. Twenty-eight workers were killed and a further 36 suffered injuries. It is recognized that the number of casualties would have been more if the incident had occurred on a weekday, as the main office block was not occupied. Offsite consequences resulted in fifty-three reported injuries. Property in the surrounding area was damaged to a varying degree.
Methyl Isocyanate Tank Rupture and ReleaseBhopal, India – Dec. 2-3, 1984 Photo Source: Indian state government of Madhya Pradesh Source: United Nations Environment Programme On the night of December 2-3, 1984, a sudden release of about 30 metric tons of methyl isocyanate (MIC) occurred at the Union Carbide pesticide plant at Bhopal, India. The accident was a result of poor safety management practices, poor early warning systems, and the lack of community preparedness. The accident led to the death of over 2,800 people (other estimates put the immediate death toll as high as 8000) living in the vicinity and caused respiratory damage and eye damage to over 20,000 others. At least 200,000 people fled Bhopal during the week after the accident. Estimates of the damage vary widely between $350 million to as high as $3 billion. Report – The Accident in Bhopal: Observations 20 years later
Bhopal Aftermath Water entered tank 610 in foreground of picture reacting with MIC and caused an uncontrolled release of a vapor cloud. Bhopal flare tower with corroded and missing section of pipe.Plant enclosed by fence in foreground.
Phillips 66 Houston Chemical Complex Pasadena, Texas – Oct. 23, 1989 On October 23, 1989, at approximately 1:00 p.m., an explosion and fire ripped through the Phillips 66 Company Houston Chemical Complex in Pasadena, Texas. Twenty-three workers were killed and more than 130 were injured. Property damage was nearly three-quarters of a billion dollars. The accident resulted from a release of extremely flammable process gases that occurred during regular maintenance operations on one of the plant's polyethylene reactors. The evidence showed that more than 85,000 pounds of highly flammable gases were released through an open valve. A vapor cloud formed and traveled rapidly through the polyethylene plant. Within 90 to 120 seconds, the vapor cloud came into contact with an ignition source and exploded with the force of 2.4 tons of TNT. This event and the Bhopal disaster triggered the development of the PSM standard
Summary of Chemical Accidents 1994-1999 Source: Kleindorfer, P., Belke, J., Elliott, M., Lee, K., Lowe, R., Feldman, H.,"Accident Epidemiology & U.S. Chemical Industry: Accident History & Worst-Case Data from RMP*Info", Risk Analysis, Vol.23, No 4, 2003, pp 865-881.
It was the terrible accident at Flixborough that led to major changes in the UK regulations and the Seveso Italy catastrophe that led to the EU Seveso directive, but both of these events had little impact on laws and regulations in the USA. The Bhopal tragedy reinforced by the 1989 Phillips accident were needed before the USA the OSHA Process Management Standard (PSM) and EPA Risk Management Program (RMP) and industry trade associations adopted the Canadian Responsible Care program. Accidents continue to occur even after the advent of legislation aimed at curbing PSM related catastrophes. The following slides outline three recent accidents, one in Washington state. Recent Accidents
Accident occurred exactly 80 years after the Oppau, Germany disaster! Ammonium Nitrate ExplosionToulouse, France – September, 21 2001 A huge explosion ripped through AZF (Azote de France) fertilizer factory in an industrial zone on the outskirts of Toulouse, southwest of France, at 10:15 am, Friday September 21, 2001. Immediately after the accident, 30 people were reported dead The total number of injuries was said to be 2,442. More than 350 people were in the plant at the time (266 AZF employees and 100 subcontractors). The explosion had occurred in a warehouse in which granular ammonium nitrate was stored flat, separated by partitions. The amount is said to be between 200 to 300 metric tons of ammonium nitrate, which is used to make fertilizers. A spokesman for the Interior Ministry in Paris ruled out a criminal attack, saying the explosion had been caused by an accident following an "incident in the handling of products". The exact cause remains unknown. Source: United Nations Environment Programme
BP American Refinery Explosion – Texas City, Texas March 23, 2005 At approximately 1:20 p.m. on March 23, 2005, a series of explosions occurred at the BP Texas City refinery during the restarting of a hydrocarbon isomerization unit. Fifteen workers were killed and 180 others were injured. Many of the victims were in or around work trailers located near an atmospheric vent stack. The explosions occurred when a distillation tower flooded with hydrocarbons and was overpressurized, causing a geyser-like release from the vent stack. Link to Chemical Safety board final report
Equilon Refinery AccidentNovember 25, 1998 • Equilon Oil Refinery – Anacortes, Washington, now owned by Shell Oil. • Multiple Fatalities – 6 workers killed This was the worst worker fatality incident in the history of Washington State until the April, 2010 Tesoro Oil Refinery explosion where seven workers were killed. The Tesoro incident is currently under investigation as of July, 2010 Link to short history of Equilon incident
Events Leading Up to Incident On Tuesday, Nov. 24, 1998, high winds cause power outage & complete refinery shutdown A large vessel known as “Drum A” was about 1 hour into a routine charging cycle 46,000 gallons of hot coke hydrocarbons @ 900 F became trapped in Drum “A” Coke Drum “A”
Coke Producing Process Heavy oil from crude oil processing called vacuum residuum is heated and pumped into an on-line coke drum. Heavy long-chain hydrocarbon molecules are “cracked” under high temperature and pressure. The lighter hydrocarbons produced are carried to the top of the drum and over to a fractioning column for further processing. Since no reaction is 100%, the remain long chain molecules combine to form a high carbon material called “coke” which ends up at the bottom of the coke drum and must be removed at the end of the cycle before the process can be repeated. Coke is a heavy, dense black material similar to a charcoal briquettes.
Normal Coke Removal Operation • Drum A is cooled first with steam followed by water. • When acceptable temperatures are reached “safe work permits” are issued and acknowledged by the coke cutting contractor to unhead the drum. • A high pressure water wand is lowered into the drum, and coke is “cut” into chunks which flow out of the bottom of Drum A into a pit below. • The cooling steps are extremely important to ensure the drum contents are adequately cooled prior to vessel unheading.
Drum A Schematic In this situation workers needed to get this coke out of the vessel. But, they couldn’t move any steam or water through the charge line (“DCU” line) to cool the hot coke in the drum because it was clogged with coke material that cooled and hardened during the power outage. 46,000 gallons of hot coke
Attempts are made to clear the DCU heater lines when steam is restored around 10:00 AM. Operators are optimistic that steam made its way through the heater and up into the bottom of the “A” Drum. Instead, it is likely the pressure relief valves were lifting and simply diverting steam to their blowdown system November 24th - During The Day
The “Night Orders” Unit foreman writes the night orders and discusses at a 3 PM managers meeting on November 24th. He says “…drum is cooling without water. Do not put water into drum. Day shift will un-head Wednesday morning.” Little activity during the night shift, drum sits idle. Several impromptu meetings held the next morning between unit foreman and operators. Additional attempts made to clear the line into the drum without success.
Actual temperature of hot coke unknown This picture shows the approximate locations of the available temperature indicators - one at the gooseneck, one on the overhead vapor line and one skin temperature sensor about a third of the way up (above the level of material in the drum). Employees had no way to actually determine the temperature of the material inside the drum. By all accounts no technical assistance was requested or provided to even estimate the temperatures. Later estimates were that it would take over 200 days for the coke to cool down to a safe temperature. On November 25, unit foreman and operators review available drum parameters and conclude drum contents sufficiently cooled to un-head. Top head of drum was removed without incident or indication of temperature of coke at bottom of Drum A.
The Tragic Results Using hydraulic controls, employees lower the bottom head. Coke spews out in a 360 degree fashion. Coke is still at auto-ignition temperature. 46,000 gallons dumps in about 6 seconds and ignites enveloping six workers in flames. = employees
The Cost Six Lives Lost Equipment Damage Lost Production DOSH Citations Third Party Lawsuits Worker Morale Standing in the community DOSH citations ultimately cost the company 4 million dollars in the form of a settlement agreement. Lost production and equipment damage estimated to be 10 to 20 million dollars. Third party lawsuits totaled about 45 million dollars.
Lessons Learned Management of Changes Emergency Preparedness Operator Training Procedural Development Permit Systems MOC- It’s not enough to just have a system to manage changes. The system has to be alive and well. At the first suggestion of a procedural change, the system should have been triggered. Audit your own programs to ensure your managing changes (this includes equipment, procedures even manpower issues) Emergency Preparedness- Review your emergency operating procedures-do they cover all the credible events. (PHA’s should cover this in detail and flush out the deficiencies) Operator training - Is the training adequate. Do operators know the true hazards of the process they’re operating. What about responding to upsets and unusual operating conditions. Procedures-Are they current and accurate? Have they been effectively communicated to operations? Permit Systems- If you don’t have them, develop them. In this case the system was seriously flawed. It was acknowledged that they thought drum parameters were adequate for deheading when they really had no way of knowing the temperature.
To Do List Audit your facilities !! Encourage participation from the trenches. Document deficiencies and establish target dates for correction. Investigate the small stuff and fix it. Review operation procedures to ensure as many “what if” scenarios are included. Ensure upper management is included in the decision-making for unusual situations.
Resources for More Information American Institute of Chemical Engineers International Institute of Ammonia Refrigeration The Chlorine Institute AcuSafe – Internet resource for safety & risk management information DOSH Directive 24.25 – Chemical Facility Process Safety Management NEP DOSH Directive 24.10 – Process Safety Management Your Own Industry Organization Click here for videos of recent chemical plant explosions and accidents U.S. Chemical Safety Board