Oil Company Refinery Fire in 1984 - Case Study Example

THE UNION OIL REFINERY FIRE IN 1984 1. Introduction Accidents by definition are unpleasant events. They occur without warning and by irregular way. Technological accidents strike us as particularly frightening because they symbolize an unexpected collapse in an efficiently operating, inanimate world, which appears before to stand apart from human agency and the possibility of human error. The failure of a bridge, the busting of a dam, the malfunctioning of a tested medical device, the escape of lethal radiation from a nuclear power plant or a chemical factory all appear superficially to be by chance, freakish, and illogical, opposing the common order of things. However a second look often uncovers a completely different picture, in which not only things but people and institutions played a critical and destructive role. The fire at Union Oil Company’s refinery in the evening of July 23, 1984, was reported as”the most expensive refinery accident in history|” (Schmer 20). The oil company is located in Romeoville, Illinois. The company employs more than 700 workers and processing around 151,000 barrels of oil a day. The explosion was caused by the ignition of a large cloud of flammable gas, a mixture of propane and butane, which had leaked from a ruptured amine-absorber pressure vessel. The gas ignited in a massive explosion which sends the upper part of the tower high in the air and landing kilometres away from the site. The blast was felt by people 20 kilometres away and the consequent fire sent flames over 150 meters high in the sky. 2. The Union Oil Refinery Fire -1984 2.1 What happened? In the evening of July 23, 1984, at around 6 in the evening, a pressure vessel at the Union Oil petroleum refinery near Chicago exploded, killing 17 refinery employees. The explosion and subsequent 2200oC fire caused extensive damage to nearby plant, estimated at $100 million. The failure had significant implication, and the accident and subsequent failure investigation have been widely reported. At around six in the evening of July 23, 1984, a worker assigned near the absorber tower saw gas leaking from a horizontal crack about 150mm long near the bottom of the vessel. The worker tried to close the main inlet valve but the crack gradually extends further to 600mm thus the worker initiated the evacuation of the area. A few minutes later, the crack at the absorber grew further and large amount of gas began to leak rapidly (TWI 1). The accident killed at least 17 and injured 23 others. Two of the injured suffered burns of at least 70% of their body. Blistered trees and scorched grass marked the area where the incident occurred. There were reports of a 5 foot piece of sheet metal struck a house two half kilometres away from the refinery. A number of reports mentioned broken windows and nearby residents who suffered head injuries. The dead were taken to funeral homes in Joliet (Beitler 1)."It was like Hiroshima” (Illinois Fire Service Institute 1), as Lieut. Ed Smith of the Romeoville Fire Department describes the blast. The size of the accident brought support from more than 30 cities. The fires burned throughout the night without further explosion or injuries. 2.2 How it happened? The vessel was a 2.6m diameter, 18.8m high amine absorber tower, used to strip hydrogen sulphide gas from the propane and butane process stream. The failure initiated from a crack in circumferential weld, which resulted in the upper 30 ton portion of the vessel being propelled like a rocket, landing about 1 km from its original site. The pressure vessel was fabricated from 25mm thick ASTM A516 Grade 70 as-rolled plate, to Section VIII of the ASME Boiler and Pressure Vessel Code. The vessel welds were not stress-relieved after welding. The operating pressure was fairly low or about 1.4MPa, with an operating temperature of 38oC. The vessel was put into service in 1970, after which it suffered a series of problems relating to corrosion. Several courses had to be placed after hydrogen blisters and delaminations were detected in 1972 and 1974. The repair welds were not subject to post-weld heat treatment. In addition, a nickel alloy liner was installed on the bottom head and most of Course 1. On the day of the explosion, at around 6 PM, a worker noticed vapours coming out from a hairline crack in high-pressure, 55 ft monoethanolamine or MEA tower. He attempted to isolate the feeds to the tower but a spark from an unknown source, ignited the vapours, causing the 34 tons tower to explode. The tower rocketed over 1,000 metres and downed a 130kV power line. Nearby towers and tanks were ruptured, including an LPG tanks that BLEVEd resulting in a second explosion (Hyatt 7). Fire fighters from the Union Oil Fire Brigade and Romeoville Fire Department responded immediately. Many towers, tanks, and other refinery structures began to rupture or collapse and the site’s fire hydrant system was damaged. Consequently, water used to fight the fire was taken from a nearby sanitary canal. Just moments before fire fighter attack the flames; a tank full of LPG exploded and created a huge fireball that rose thousands of feet into the air. The explosion was felt 15 miles away and an airplane was reported hit by debris form the blast (Groves 1). 2.3 Why it happened? When the section of the vessel were examined in one of NBS’s laboratories in 1985 by a multi-disciplinary group headed by Harry McHenry who was the Deputy Chief of the Fracture and Deformation Division of NBS, magnetic particle inspection revealed hundred of cracks in the inner surfaces along the weld (see Fig. 2.2.2) between Courses 2 and 3. Moreover, indications of delamination damage were found in Course 1 through Ultrasonic measurement. However, result of thickness test showed the Courses 1 and 2 wall thicknesses was within the standard allowance for pressure vessels. In addition, test on all initial and replacement components reveals that they are all within standard specifications. It was only when these materials when metallography results were combined with stress corrosion cracking, hydrogen embrittlement test, and fracture mechanic analysis that the cause of failure became clear. “It seems that a pre-existing crack had extended through more than 90% of the wall thickness and was about 800 mm in length” (Siewert 350). The failure investigation concluded that all the steel plate and welds used for the initial construction and subsequent repairs met the specified requirements for chemical composition and mechanics properties. However, the repair welding procedure used for the circumferential welds resulted in a hard brittle heat affected zone or HAZ, near the inside surface of the vessel, which was not tempered by subsequent weld passes, and which was susceptible to hydrogen stress corrosion cracking. The relatively low applied stresses or 35MPa, combined with high residual stresses within the weld which as mentioned earlier was not stress relieved, provided sufficient driving force for cracks to propagate in the hard repair weld HAZ. When the deepest of the pre-existing cracks reached 90% of the wall thickness the remaining ligament tore through to provide a leak path. The crack then tore circumferentially around the vessel to a final length of 800mm prior to catastrophic failure (Royston 115). The main failure investigation focused on the embrittling effect that the hydrogen environment had on the weld metal toughness properties. Good Charpy impact energy levels were measure for the weldment, with the 20J transition temperatures ranging from negative 51 degrees centigrade for the weld metal, to negative 40 degrees centigrade for the HAZ, to 0 degrees centigrade for the parent steel. However, fracture toughness test in terms of crack-tip opening displacement or CTOD, showed that the HAZ toughness at 38 degrees centigrade reduced from 0.17mm to 0.064mm when test specimens were soaked in hydrogen. This represents a 62% decrease in toughness for the hydrogen charged specimens. The failure investigation reported did not account for the influence of tensile welding residual stresses, which can be very significant in fracture mechanic assessment (Royston 116). 2.4 Consequences Aside from an eight months reduction in gasoline production and hazardous materials released in the environment (Hyatt 7), the Union Oil failure had significant consequences in the petro-chemical industry. In addition to the $100 million cost of the damaged caused by explosion, the Occupational Safety and Health Administration or OSHA accused the firm of safety violations and fined it $31,000. Specifically, Union Oil was cited for failure to have an effective maintenance and inspection programme for pressure vessels, for not providing protective equipment for workers, failure to adequately train a fire brigade, and lack of effective emergency procedures. On a decision contained on OSHRC Docket No. 85-0111, the Secretary of Labour issued Union Oil citations for violations of Occupational Safety and Health Act of 1970 with the maximum penalty of $10,000. The accident also resulted in procedures for fabrication, inspection and repair of pressure vessels being reviewed across industry. Data from the Union Oil failure continue to be use, and have been subsequently used to support validation of the PD 5493 procedures (Royston 114). Seventeen employees of the oil company were killed including ten members of the fire brigade. The burns were so severe that the coroners had to rely on dental records to identify some of the fatalities. The estimated cost of the fire is $500 million and in terms of damage property and loss of life, it can be compared to the 1910 Union Stockyards Fire and the 1893 World’s Columbian Exposition Fire (Groves 1). From the very nature of crude oil, its refining and the processes relating to its operation provide an extremely hazardous situation. Above all is the inherent danger of fire. For this reason, it is very likely that the Union Oil Refinery explosion and subsequent fire released several harmful substances in the air that may present unreasonable risks to human health or the environment. Toxic gases can be easily inhaled by people particularly those that are in a closed proximity. Inhalation of products of combustion can produce a range of injuries to the airway, including irritation of the upper airway. Smoke contains sensory irritants in the vapour phase and the particulate phase. Moreover, fire can cause extensive and profound changes in an environment, sometimes even the elimination of living organisms (Sullivan & Krieger 632). The consequences of the explosion and subsequent fire from the Union Oil Refinery fire in 1984 not only claim the lives of its employees, fire fighters, and injured many but undoubtedly affected people’s health around the site and the environment. 2.5 Lessons Learned The Union Oil failure showed that it is not sufficient to just meet the standard requirements of materials and fabrication codes. Some detailed knowledge is also required concerning the intended operating environment, especially for corrosive environments such as that of the Union Oil amine-absorber vessel. In particular, special consideration is required concerning the structural integrity of weld repairs. A more proactive engineering approach, such as the fitness-for-purpose assessment methods in PD 6493, enables these issues to be quantified and managed effectively. The fitness-for-purpose principles state that a weld flaw is acceptable provided the conditions for failure are not reached in the service lifetime. In 1980 the British Standards Institution issued a published document PD 6493 for the derivation of acceptance levels for flaws in fusion welded joints. Moreover, extensive revisions were published in 1991, incorporating CEGB R6 procedures which serve as a guide on methods for assessing the acceptability of flaws in fusion welded structures. The fitness-for-purpose concept is now accepted engineering practice, and the procedures are routinely applied in a range of industries, including pressure vessels, pipelines, offshore structures, storage tanks, buildings, bridges, and other structural components (Royston 116). Our technological society requires much of metals, and they are generally performed so well that we have a tendency to expect them to maintain their specifications. Over time, we grow careless, and we may not pay sufficient attention to the demands placed upon building materials (McHenry 104). First and foremost, we learn then we must plan for failure. Engineers need to design adequate fail-safe mechanisms, so that when metal does wear out, its effects will not be catastrophic. For instance, a chemical storage tank can be designed so that it will leak noticeably long before it explodes. The direct results of carelessness are so well know that they need no consideration. They are anything in the way of loss of material, damage to property, and personal injury. Many people tend to think of carelessness almost entirely in terms of accidents. According to author Larry May, carelessness, thoughtlessness, and insensitivity are all attitudes that are usually involved in negligent behaviour, in the sense that they are often responsible for a person’s failure to act reasonably. And given that attitudes are states of mind, negligence usually involves states of mind, in the sense the negligent acts are cause by, or intimately involved with, various attitudes. These attitudes and dispositions are not merely “negative’ for they are more than descriptions of what is lacking in someone’s state of mind (96). In the occupational health and safety, the success of occupational hazard prevention depends on the proper identification of all sources and causative factors which can adversely affect the workplace environment and health of workers (Asian and Pacific Regional Centre for Labour Administration 67). Apparently this did not happen in Union Oil refineries since a violation of these provisions were cited by the court. 3. Work Cited List Asian and Pacific Regional Centre for Labour Administration, Labour inspection skills in the petroleum industry: proceedings and training material ofILO/ARPLA/CLI Regional Training Course on Labour Inspection in Oil Refineries, Bombay, 16 October to 3 November 1989, Switzerland: International Labour Organization, 1991 Beitler, Stu, Romeoville, IL Refinery Explosions, July 1984, April 23, 2009, http://www3.gendisasters.com/illinois/6895/romeoville-il-refinery-explosions-july-1984 Groves, Adams, Union Oil Company Refinery Fire, Romeoville: July 23, 1984, April 23, 2009, https://www.ideals.uiuc.edu Hyatt, Nigel, Guidelines for Process Hazards Analysis, Hazards Identification and Risk Analysis, Canada: CRC Press, 2004 Illinois Fire Service Institute, Incident Summary, April 23, 2009, https://www.fsi.illinois.edu/content/library/IFLODD/search/Firefighter.cfm?ID=46 May, Larry, Sharing Responsibility, US: University of Chicago Press, 1996 McHenry, Harry, Catastrophe, Popular Mechanics Aug 1988, 130 pages, Vol. 165, No. 8, U.S. Hearst Magazines, 1988 Royston, Kenneth Penny, Risk, Economy and Safety, Failure Minimisation and Analysis: Failures '96 : Proceedings of the Second International Symposium on Risk, Economy and Safety, Failure Minimisation and Analysis, Pilanesberg, South Africa, 22-26 July 1996, Netherlands: Taylor & Francis, 1996 Schmer, Robert, The Motor Gasoline Industry: Past, Present & Future, US: DIANE Publishing, 2000 Siewert, Tom, Analysis of the Catastrophic Rupture of a Pressure Vessel, April 23, 2009, http://nvl.nist.gov/pub/nistpubs/sp958-lide/350-352.pdf Sullivan, John Burke and Krieger, Gary R. Clinical environmental health and toxic exposures, US: Lippincott Williams & Wilkins, 2001 TWI, Union Oil amine absorber tower: Case Study no. 180, World Centre for Materials Joining Technology, UK: 2001 Read More