BY David F. Peterson
Crescent City. Kingman. Waverly. Memphis. Do these cities mean anything to you? They should! They are locations of disasters involving liquefied petroleum gas (LPG) that killed 37 people combinedincluding 17 firefightersand injured more than 200 people. How about Warwick, Quebec, Canada, and Burnside, Illinois? These cities also witnessed tragedies over the past eight years involving LPG tanks that failed because of flame impingement, killing six firefighters and injuring five.
Imagine you are settling into bed on a cold evening in early spring and you get called to a propane tank fire a couple of miles out of town. The dispatcher tells you the fire started close to five minutes earlier when an all-terrain vehicle hit some propane supply lines coming from a large tank. Somehow the propane ignited, prompting calls to the fire department.
You arrive on-scene nearly nine minutes after the page with 19 personnel to find an 18,000-gallon steel propane tank engulfed in fire. Two pressure relief valves (PRVs) on top of the tank are roaring like jet engines, so loudly you can hardly hear yourself think. You and your crew are directed to spray water on nearby buildings to save them from the intense heat of the propane fire. Members set up portable water tanks and stretch hoselines while the fire roars only 100 feet away. Suddenly, a huge, ear-piercing explosion occurs!
Large pieces of the propane tank are hurled in all directions. Two firefighters are struck by one large piece. Another large piece rockets through one of the nearby buildings. Other firefighters are severely injured from flying debris and from the explosion’s fireball. You are thrown to the ground by the force of the explosion, and you feel the intense heat from the vaporized propane burning. Your thoughts immediately turn to helping your comrades after you regain your senses.
The fireball is consumed, and the scene becomes immediately dark and silent. The only noises you can hear are the engines of the fire trucks and your buddies screaming from their injuries. Other firefighters are frantically digging for a missing firefighter in a crater left by a rocketed piece of tank. In all, six firefighters receive thermal burns, bumps, bruises, and some fractures; two firefighters are killed instantly.
This incident actually happened. The above information was taken from reports by the National Fire Protection Association (NFPA), the U.S. Chemical Safety and Hazard Investigation Board (CSB), and the U.S. National Institute for Occupational Safety and Health (NIOSH) as a result of their investigations into a 1998 BLEVEboiling-liquid, expanding-vapor explosionin Iowa.
Over the years many BLEVEs have occurred and have taken many civilians’ and firefighters’ lives. Most of the earlier tragedies involved large containerswith 18,000- to 34,000-gallon tanksboth fixed storage and transportation. Interestingly, all of the firefighter deaths occurred in rural or farm settings involving smaller, noninsulated steel tanks.
The responders at these incidents may have overlooked several critical risk factors. Also, some fallacies regarding fires involving pressurized compressed gas containers may have come into play. Examining the risks and fallacies concerning BLEVEs may help prevent firefighter deaths in the future.
THE QUESTION OF TRAINING
After much effort has been concentrated on training firefighters about BLEVE dangers, why are firefighters still making the supreme sacrifice at these emergencies? Some notable fire service personnel offer some possible conclusions along with their own hypothetical questions. Francis L. Brannigan, a frequent lecturer; author of Building Construction for the Fire Service, Third Edition; and Fire Engineering editorial advisory board member, asks if teaching firefighters to approach and shut off the valve to a propane tank under fire conditions may be giving personnel a false sense of security. In a recent Fire Engineering column, he asks, “Do such tactics give firefighters a false impression of the hazard?” Similarly, Vincent Dunn, a retired Fire Department of New York deputy chief, asks, “Where is the propane tank explosion zone?” which many responders may be hard pressed to answer. Stephen L. Hermann, an Arizona hazardous materials specialist who responds to many statewide transportation incidents, asks, “Shouldn’t we train according to what we will encounter?” He questions whether responders are being trained to properly respond to compressed liquefied gas incidents.
Are firefighters being killed at these types of container fires because of other factors, such as forgetting past tragedies, having to relearn old lessons, high personnel turnover and the loss of the “experience factor,” being in the “invincibility mode,” having tunnel vision, playing the odds, or just plain ignorance? Whatever the reasons for these deaths, far too many responders are being killed by BLEVEs. It is readily apparent that more education and training need to be disseminated concerning liquefied petroleum gases and container fires to help promote firefighter safety. When was the last time you attended a drill on recognizing BLEVEs and taking appropriate response actions?
Training efforts in the 1970s and 1980s by major railroads brought to light the dangers of LPG tank fires and the phenomenon known as BLEVE. The National Fire Academy was also involved in developing these early courses geared for all first responders. The NFPA developed a videotape in 1977 titled “BLEVE” that details how it occurs and appropriate tactics to maintain safety. The National Propane Gas Association (NPGA) has developed a training program geared to firefighters concerning propane safety. This program, developed by Michael Hildebrand and Gregory Noll of Hildebrand and Noll Associates, has been shipped to every fire department in the United States free of charge.
BLEVE is an acronym that was first coined by three Factory Mutual (FM) researchers. On April 24, 1957, a substantial container failure occurred as a result of overheating a mixture of formalin (a solution of formaldehyde gas in water) and phenol in a chemical reactor at an FM research facility. In the analysis that followed the incident, it became evident that the physical model of the container, due to overpressurization, was also applicable to any liquid that was at a temperature well above its normal boiling point at the moment of vessel failure. Liquefied gases are good examples. Because of this observation and to avoid using excessive wording, the term “BLEVE” was introduced by FM researchers James B. Smith, William S. Marsh, and Wilbur L. Walls.
Walls went on to work for the NFPA as a fire protection engineer and made many presentations on BLEVEs. In his November 1978 NFPA Fire Journal article “Just What Is a BLEVE?” Walls refers to the NFPA definition: “a major container failure, into two or more pieces, at a moment in time when the contained liquid is at a temperature well above its boiling point at normal atmospheric pressure.”
While this definition is a broad one, Walls points out that it is an exact one. He suggests that some confusion over the type of BLEVE has led to misapplication of the term itself. For instance, the actual mechanism of a BLEVE is a physical reaction in which the material rapidly, and instantaneously, converts from a liquid to a gas. It is merely a change of state that yields pressure. This reality is in contrast to a chemical reaction in which the material is converted to other materials, especially gases, chemically, such as with high explosives. BLEVE is actually a misnomer because it is not technically an “explosion.” Make no mistake, however; the pressure involved with a liquefied gas changing state can be extreme and violent.
Walls also points out that the definition of BLEVE is independent of the cause of the container failure. For a BLEVE to occur, the container has to be under pressure, the pressure has to exceed the strength of the container, and the container has to be weakened in some way (impact, corrosion, fire). Walls goes on to discuss different types of BLEVEs such as containers failing from flame impingement. If the liquefied gas is flammable and released because of a BLEVE, the important and dangerous dimensions of fireballs and ignition of vapor clouds have to be considered. Walls warns that the impression that BLEVEs are solely restricted to flammable, liquefied gases is false. BLEVEs occur with many types of liquefied gases, flammable and nonflammable.
A. Michael Birk, a professional engineer and a professor with Queens University at Kingston, Ontario, Canada, offers this viewpoint on BLEVEs: “You get a BLEVE when a vessel holding a “pressure” liquefied gas fails catastrophically. It does not matter how the container fails. It can be by fire impingement, impact, corrosion, etc. The BLEVE is the boiling liquid expanding vapor explosion that happens when the tank opens up fully.” He goes on to say, “A BLEVE is a physical explosion of compressed vapor and rapidly vaporizing liquid. Upon vessel failure the vapor space sends out a shock wave from the liquid flashing to vapor. If the material is flammable, a fireball may follow it. The rapid explosion can also cause projectile effects.”
In the General Hazardous Materials Behavior Model (GHMBO) by Ludwig Benner, the BLEVE would be considered a “release event” as a result of the container’s failing. As above, the manner in which the container fails, or breaches, is independent of the BLEVE. The release event occurs at the time the container fails and the energy is unleashed. According to Benner, the magnitude of the release depends on the characteristics of the product, the state and quantity of the product, the flow or release rate, the propulsion force, and even weather conditions. Benner stresses that analysis of the events at a hazardous materials event, and especially an understanding of the release event, can help assess risk and better predict outcomes at potential BLEVE incidents.
David Lesak, a nationally known hazardous materials author, lecturer, and course developer, defines a BLEVE as “a pressure release from catastrophic container failure.” The result of a BLEVE is total devastation in the immediate area with potentially large loss of life and property.
As far as the types of BLEVEs is concerned, an October 1995 article in Chemical Business entitled “Boiling Liquid Expanding Vapour Explosion (BLEVE) An Overview” by V.K. Singh and A.D. Kharait cites research by Richard W. Prugh. A survey of 50 notable BLEVEs from around the world over a 60-year span (1926 to 1986) was compiled by Prugh, who chronicles the types of BLEVEs that have occurred (see Figure 1).
The size of the BLEVE is dependent on the size and weight of the container along with the amount of liquid that remains inside the container at the moment of the BLEVE. Generally speaking, the bigger the container, the bigger the BLEVE. Most flame-induced liquefied gas BLEVEs occurred when there was approximately one-half to three-fourths of the liquid remaining in the container. Essentially, the destruction of the container produces rockets that can be propelled great distances as a result of the remaining liquid’s vaporizing. According to the NFPA, deaths from these projectiles have occurred as far as 800 feet from the BLEVE.
Additionally, the material inside the container may not completely vaporize at the time of the BLEVE; instead, it may also be propelled away from the scene. Some firefighters who have survived BLEVEs have commented how they have been cooled by the rapid evaporation of liquids that have passed in their vicinity.
A.M. Birk’s research concerning safe distances for personnel from BLEVEs suggests that four times the fireball radius for a specific size tank would be appropriate. As an example, a container of 1,000 liters would require a safe distance of approximately 100 meters. A minimum distance of 100 meters for any size container impinged by fire is suggested.
Fireballs several hundred feet in diameter have also been observed as a result of BLEVEs, and deaths and injuries have been documented as far as 250 feet from the container. Birk’s data show that a 1,000-liter container BLEVE with LPG would result in an approximately 25-meter radius for a fireball. Birk has also found that the shape of fireballs vary significantly. If a container fails rapidly, it will produce classic, spherical fireballs. If the liquid within the tank is relatively cool at the time of a BLEVE, a large ground fire may result. According to Birk, these differences can change the risks to responders profoundly.
RISK FACTORS AND FALLACIES
Keeping in mind 10 risk factors at pressurized container fires can help you avoid deadly situations and stay safe. The basis for these risk factors is a 1978 flowchart by Gene Carlson that appeared in Fire Engineering. Carlson addressed the potential for tragedy at BLEVE situations with a six-question flowchart that advised personnel to either evacuate or work to effectively cool the tank with water with each subsequent question. Note that assessing each factor may not lead to a decision to evacuate but in combination with other factors may indicate that “getting out of Dodge” is indeed the best course of action.
Factor #1: What material is involved in the container fire? While unstable, reactive, and explosive materials in any container are dangerous, if these commodities are impinged by fire it may be time to back off quickly to a safe distance. A quick check of the 2000 Department of Transportation (DOT) Emergency Response Guidebook (ERG) will lead you to Guide #112 for explosives such as bombs and artillery. This guide states that if a railcar or trailer of explosives is involved in fire, isolate for a 1,600-meter perimeter. Unstable and reactive materials call for similar guidelines.
It is equally important to realize that compressed liquefied gas containers in fire situations are also dangerous. ERG Guide #115 also states that if a railcar or highway trailer is involved in fire, isolate for 1,600 meters in all directions. This distance is recommended, of course, because of the potential for a BLEVE. Projectiles can travel thousands of feet from the blast site. A notable event occurred near Murdoch, Illinois, in which a portion of a rail container traveled nearly 5,200 feet after a BLEVE.
Also keep in mind that many containers can fail violently either as a BLEVE or a pressure-induced rupture. A Houston firefighter and experienced haz-mat responder said that the most spectacular container failure he had ever witnessed was from a properly operating water heater that failed because the owner had removed and plugged the pressure relief valve on the container.
You must identify the product(s) accurately as soon as possible and conduct a thorough and rapid risk/benefit analysis. The sooner you recognize a “loser” situation and react appropriately, the greater your chance of survival.
Fallacy: Only flammable materials in a container can BLEVE. Remember, any container, especially a pressurized one, can BLEVE in fire situations. A young man in Vermont threw a nearly empty quarter barrel of beer into a bonfire in 1988. Soon after, the keg failed catastrophically, and the ensuing shrapnel killed him. Basically, any materials that can convert to a vapor and overcome the container can BLEVE.
Factor #2: How long has the fire been burning? At most incidents, it will be hard to determine when the fire started and when the container became involved in the fire. Still, the time element is crucial when it comes to the BLEVE potential. While it is difficult to say exactly when a BLEVE will occur after flame impingement on a container, numerous case studies indicate the 10- to 20-minute range has produced tragic results (especially with noninsulated tanks). The NFPA reports that 58 percent of the documented incidents in which pressurized containers failed from flame impingement occurred within 15 minutes. Research by Birk and the U.S. Department of Transportation has indicated smaller containers can fail from direct flame impingement in as little as 21/2 minutes.
Remember, in these types of emergencies, the clock does not start when you arrive on-scene. Rather, the “lead time” includes the time it took for someone to become aware of the problem, the time it took to call your local communications center, the time it took for the communications specialist to notify you, and finally your reaction and travel time to the scene. Obviously, if the emergency is in the country and several miles away, the lead time can be quite lengthy. You may be fighting a “loser” before you even get on location.
Factor #3: Where is the flame impingement on the container? Direct flame impingement on the vapor space of the container is a major causal factor in BLEVEs. Since there is no liquid inside the tank above the liquid level to absorb the heat being applied and to keep the steel cool, the steel will eventually begin to heat up. This thermal absorption will weaken, elongate, and thin the steel container and, if the heat is continually applied and adequately intense, the container has a high likelihood of failure. This failure, in combination with increased internal pressure, can cause the tank to violently tear with horrific results. The flame impingement is also a factorit may be a very intense blowtorch effect or a relatively gentle flame contact on the container.
Remember, too, that the vapor space on the tank will always be at the top of the tank because gravity causes the heavier liquid to settle on the bottom.
Fallacy: A BLEVE can occur only when the flames are on the container’s vapor space. Many thermal stressor-induced BLEVEs have occurred from direct flame impingement on the vapor space of the container. This occurs most often because the vapor in the container cannot effectively remove the heat from the container. This leads to the container’s overheating and weakening as described above. This process caused the tragedies in Kingman, Arizona, and Iowa.
While it may be rare, a BLEVE could occur as described above with a critical temperature stressor. The contents in the pressurized container are heated so fast that they instantly turn into vapor within the tank, causing immediate container failure. This is what happened in Houston, Texas, as the result of a water heater BLEVE.
Factor #4: What type of container is impinged by fire? If the container has blown-on or layered insulation, the insulation may protect the tank from flame contact longer than a noninsulated container would. Insulated containers are required in railroad transportation, but noninsulated containers are most common in highway transportation and fixed sites. Essentially, the potential for failure from direct flame impingement is higher with noninsulated containers than with the insulated containers on railcars. Since the advent of the BLEVE-preventing engineering controls such as thermal insulation, head shields, and shelf couplers in the early 1980s, the NFPA reports that there have been no thermally induced BLEVEs from rail containers since 1980. More importantly, no firefighters have been killed from rail containers since these changes were implementedthe eight firefighters killed since 1993 have all died as a result of BLEVEs of noninsulated containers.
Factor #5: What is the orientation of the container? If the container is upright, the PRV should operate properly, allowing excess pressure to vent. If the container is on its side or even upside down, the PRV may not perform adequately. Furthermore, if the PRV is facedown, it may be inoperable from lying in dirt or debris at the scene. The risk of BLEVE may be increased because the tank may not be capable of adequately venting itself of excessive pressure to prevent a BLEVE.
Factor #6: Is the relief valve activated? Nearly all pressurized containers have PRVs that are designed to relieve the excess pressure and maintain the integrity of the tank. When the PRV activates, the tank is talking to you! It is saying that “something has caused my internal vapor pressure to increase and in order to stay together I am attempting to reduce my internal pressure.” If the source of energy is from direct flame impingement, the tank’s spring-loaded PRV should open at a set pressure. If it operates properly, it will reseat when the pressure is reduced below the activation pressure.
Although the activation of a relief device is a warning, especially if induced by flame impingement, it does not necessarily indicate the container will fail. Nevertheless, heed an activated PRV. It will roar with a pitch that indicates the pressure. ERG Guide #115 states, “Withdraw immediately in case of rising sound from venting safety device or discoloration of tank.” If you hear the pitch of the sound continue to rise after the relief device activates, the container is enduring greater pressures. Most relief devices are set to activate at 75 percent of the test pressure for the container. For many fixed site containers, this pressure will be approximately 270 pounds per square inch (psi).
Note: If the discharge of vapors from the PRV has been ignited from nearby flames, how far is the visible flame from the end of the relief device? Usually the greater the distance, the greater the internal pressure.
Fallacy: It is safe if the pressure relief valve is activated. Nearly all pressurized containers have PRVs that activate at a preset pressure. These spring-loaded valves are designed to relieve the container of excess pressure and then reseat. They are designed to prevent the container from failing. However, if the heat on the container is too great or the flame impingement is too intense, the PRV may not be adequate, and the container may fail.
An activated PRV is a warning. If the container continues to be heated, the PRV may not be adequate. Cooling measures may be the only tactic to keep the container intact and reseat the PRV, but this tactic would be employed under very dangerous conditions!
The NFPA warns, “Don’t assume that operating relief vents will prevent a BLEVE.” The National Propane Gas Association says, “This may mean pulling personnel out of the area whenever a tank’s relief valves are working. As soon as LP-Gas tank relief valves begin shrieking, it’s too late to apply cooling water to the tank. The next action should be immediate withdrawal of all personnel, as a BLEVE is imminent.”
Factor #7: What are the exposures? Consider the physical setting of the emergency in terms of risk assessment. For instance, there is a big difference in risk between a pressurized container fire in the desert and one in the downtown area of a large city. The desert setting will probably have few people in jeopardy, whereas the city setting will probably have many people needing evacuation. Also, the desert would most likely have a poor water supply to cool the container while the city will have a strong water supply. And a BLEVE in the desert would most likely impact very little property; the city would most likely have very expensive exposures. Ask, Is our involvement worth what we are attempting to save?
As a result of the tragedy in Iowa, the NFPA’s Ted Lemoff, a principal gases engineer, states, “Firefighters need to step back when responding to LP-Gas fires and ask themselves whether they’re willing to risk being caught in a BLEVE to save property. Firefighters have to question their approach, especially when no human life is at stake.”
Fallacy: Avoid the ends of the containers that are involved in fires. Positioning firefighters away from the ends of tanks is no guarantee of safetywhen a container BLEVEs, there is no way to predict how the tank will fail, how many pieces will be produced, and where the pieces will travel. Some containers have failed and did not break into pieces, remaining where they were originally. Other containers have separated into pieces that flew thousands of feet. Also, containers can swivel or separate and travel at different angles from their original axis. This was the case in Iowa where the two firefighter fatalities were struck by the fourth largest piece of the tank at approximately a 457 angle from the original tank axis.
Early instruction on safe responses to pressurized container fires stressed avoiding the tank ends. Some instructors compared viewing the tank ends to looking down the barrel of a shotgun. Structurally, the ends of the pressurized tank are the strongest components. A container will usually fail where the heat is most intense, and it frequently manifests into a rapidly developing tear on the shell of the container. What remains is the elliptical tank end that quite often becomes a projectile. The container can and has flown in all directions at the moment of BLEVE.
Factor #8: What is the water supply? Assess whether you will have enough water to cool the container adequately to prevent a BLEVE. Generally, the larger the tank, the higher the flow needed to cool a tank involved in a fire. The NFPA has recommended a minimum of 500 gallons per minute on each point of impingement for fires involving rail containers. Smaller containers may require substantially less water to cool them adequately. If you cannot supply the required water flow, consider rapid withdrawal.
Factor #9: Can water be applied to each point of impingement? The cooling water has to reach the point of the container that is being heated. It may be difficult to apply water because of obstructions such as buildings, trees, power poles, other vehicles, terrain, and even weather. Consider withdrawal if you cannot apply the water from unstaffed master streams.
Fallacy: BLEVEs can only occur from direct flame impingement on a container. According to David Lesak, BLEVEs are caused by one of four stressors to the container:
Thermalcaused by direct thermal stress to the vapor space of the container. Over time, the flame will weaken the container, causing it to soften, elongate, or thin and thereby reduce its capacity to contain the increasing pressure within the tank. Metallurgists say that carbon steel loses its strength very quickly when heated above 4007F. Unconfirmed reports also indicate the possibility of a polymerization BLEVE of a vinyl chloride monomer (VCM) railcar following a 43-car train derailment in Livingston, Louisiana, in September 1983.
Chemicalcaused by the thermal stress induced by a chemical reaction within the container such as corrosion, oxidation, or even by-product polymerization. One noteworthy incident in which a railcar catastrophically failed from product polymerization occurred in Deer Park, Texas, in July 1988. A railcar of meth-acrylic acid (MAA) was found to be overheating internally in a rail yard, and the relief valve was activated. Quick-thinking railroad employees moved the car to a remote area away from exposures. The car was surrounded by empty railcars on all sides to act as shielding in the event of failure. Eventually, the railcar did fail from enormous internal pressure, completely obliterating the container. Fortunately, no one was injured or killed because the area had been evacuated.
Critical temperaturethe liquid within the container is heated above its critical temperature, causing an extreme pressure that overcomes the container integrity. Informal reports of a railcar failure in Moline, Illinois, in 1986 point to the possibility of a critical temperature BLEVE as the cause. In an incident to which I responded, a 30,000-gallon container of LPG was struck by lightning during a fast-moving summer storm. Eyewitnesses at the propane facility stated that the relief valve was activated and ignited simultaneously as a result of the lightning strike. On our arrival nearly 10 minutes later, we observed the relief valve activated and still burning with a gentle, roaring sound. Within a few minutes, the relief valve closed. Container specialists surmise that the tank and its contents were heated very quickly from the lightning strike, causing the overpressurization of the tank to activate the relief valve. It is possible that a defective PRV was activated by the lightning strike. Fortunately, the overpressurization did not cause container failure.
Mechanicaldamage to the container, perhaps from collisions or some other traumatic event, is sufficient to cause structural weakening. The internal pressure from the product overcomes the tank integrity. This type of BLEVE does not involve thermal impingement from a fire, though increases in ambient temperature may cause the product’s vapor pressure to rise and impact the weakened container.
Perhaps the most famous mechanical BLEVE occurred in Waverly, Tennessee, in the late 1970s. A butane railcar experienced a BLEVE two days (40 hours) after a train derailment in which it received severe damage to its shell; 16 people were killed. Prior to that, a little-known mechanical BLEVE occurred in Cumming, Iowa. This incident received very little notoriety because no one was injured or killed. Still, it was significant because it was possibly the first incident in which a container failed from a physically damaged pressurized container. On April 29, 1969, a train derailment of 29 cars created a large scrape on the bottom of a 33,000-gallon anhydrous ammonia railcar. On May 1, the car was righted at noon and, at approximately 2:00 p.m., the car ruptured. The cause of the container failure was mechanical impact of the car and subsequent pressure rise from an ambient temperature increase. One-third of the car was propelled 300 feet as a result of the car failure.
The most recent documented mechanical BLEVE occurred during a Union Pacific train derailment in Eunice, Louisiana, on May 27, 2000, in which a mechanical breach to a pressure car during the derailment of 33 cars caused the car to fail violently. No one was killed or inured.
Factor #10: How quickly can water be applied? If you are considering the application of water, move quickly, as time is of the essence. Every second that goes by increases your risk of injury or death; rapid thinking is imperative.
Assess these risk factors and fallacies at every container fire. You can put them into a checklist form until they become routine at every container fire.
CLOSE CALLS AND SUCCESS STORIES
New York: On Long Island in 1988, firefighters were confronted with a propane fire from an overturned 2,500-gallon bobtail tank truck. They successfully applied water from several master streams for 24 hours to cool the tank and allow product to burn away. The eight-lane highway was rerouted, and several thousand people were evacuated until the danger passed.
Utah: On a night in October 1986, a tractor-trailer with 70 head of cattle veered off a highway and into an 18,000-gallon steel propane tank at a propane facility. The truck also struck and dislodged pipes underneath the tank, and an immediate fire ensued. Responders tried to get near the truck to rescue the driver, but the heat was too intense. Firefighters stretched attack lines to cool the tank that was filled to 80-percent capacity. As the roar of the gas escaping from the pressure relief valve increased, the rescue attempt was abandoned. Firefighters were evacuating to a safe area when the tank failed and a BLEVE ensued. The main portion of the tank rocketed through a building, destroying two pickup trucks, a propane delivery truck, and a fire department tanker truck. This section landed 2,674 feet from its origin but not before causing $1.5 million in damage. Fortunately, no firefighters were killed because they were protected by a 30,000-gallon propane tank that sat adjacent to the tank that BLEVE’d. It is theorized that this big tank deflected the energy of the BLEVE away from responders. The big tank was knocked 35 feet off of its pedestals, and piping was severed. Consequently, propane leaking from the pipes was ignited and burned throughout the night. Approximately 25,500 gallons of propane burned away as excess flow valves opened off and on. One responder stated, “It is a miracle we were not killed.”
Wisconsin: In 1981, I responded to an overturned, 9,500-gallon propane tractor-trailer incident on a highway. It was approximately 3:00 a.m. on a summer day. When we arrived, the area was well lit from the sheriff’s deputies who arrived first. Not only did headlights illuminate the scene, but road flares had also been employed. As my shift commander and I viewed the propane tank trailer lying across the road, I was asked to walk around the tank and report back my observations. I noticed the tractor was steaming from leaking radiator fluid; the driver had been ejected from the tractor and died from being partially crushed under the trailer, but there was no propane leaking. The scene was secured for hours without further incident, and eventually the contents were off-loaded to another tanker; a crane removed the damaged tank. Needless to say, that incident left a huge impression on me. Mainly, we needed comprehensive training to better handle those types of events.
In 1998 in Walworth, Wisconsin, a suicidal man drove his automobile into the piping under a stationary 30,000-gallon propane tank and immediately crawled into the trunk. The engine of the automobile soon ignited the leaking propane, and flames impinged on the liquid portion of the tank. The local fire department responded by cooling the tank with a master stream within five minutes of the fire’s inception. Soon after, four additional master streams supplied 2,500 gpm until the product burned away nearly 24 hours later. More than three million gallons of water was used to cool the tank. Unfortunately, the man succumbed to the heat and toxic gases within the car; his body was recovered after the fire. However, the fire department’s actions saved the tank and the city from a BLEVE. In fact, they cooled the tank so successfully that the pressure relief device was never activated.
Even though these types of emergencies are fairly rare, we must be prepared to handle them safely and effectively. The tactics range from rapid application of adequate amounts of water at each point of flame impingement on the container to rapid evacuation of the area. Critical aspects of leaving the scene safely are quick recognition of the problem and the potential of the situation and what resources can be deployed in a timely manner. Remember your resources, and train on these concepts frequently to maintain and increase your safety at potential BLEVE incidents.
Thanks to the following for their assistance: Gregory Noll, Hildebrand & Noll, Associates; Charles J. Wright, Union Pacific Railroad; Bill Hand, Houston Fire Department Hazardous Materials Response Team; David Ghormley, Rohm and Haas; and David Lesak, chief of the Lehigh County (PA) Haz Mat Team.
- Propane Emergencies, developed for the National Propane Gas Association by Michael S. Hildebrand and Gregory G. Noll, Q 1999 by Red Hat Publishing.
- United States Chemical Safety and Hazard Investigation Board Investigations and News, www.chemsafety.gov/reports/1999/herrig.
- National Institute for Occupational Safety and Health Fire Fighter Fatality Investigation and Prevention Program #98F-14, www.cdc.gov/niosh/firehome.html.
- NIOSH Fire Fighting Hazards During Propane Tank Fires, www.cdc.gov/niosh/hid7.html.
- BLEVE: Boiling Liquid Expanding Vapor Explosion LP-Gas BLEVE’s Result in Fire Fighter Fatalities, www.nfpa.org/Research/ Fire_Investigation/Alert_Bulletins/ BLEVE/bleve.html.
- “BLEVE Kills Two,” Alisa Wolf, NFPA Journal, Nov./Dec. 1998.
- Lesak, David M. Hazardous Materials: Strategies and Tactics (Prentice-Hall, Inc.: Upper Saddle River, NJ, 1999).
- BLEVE information and photographs, Queens University at Kingston, Ontario, Canada, http://conn.me.queensu.ca/birk/bleveinng.htm.
- “Boiling Liquid Expanding Vapor Explosion (BLEVE)An Overview,” by V.K. Singh and A.D. Kharait, Chemical Business, Oct., 1995, 9:3, 71-76.
- Benner, Ludwig. Hazardous Materials Emergencies, 2nd Edition, Q 1978, Lufred Industries, Inc., Oakton, Virginia.
- NFPA Fire Protection Handbook, 17th Edition, National Fire Protection Association, Boston, Massachusetts.
DAVID F. PETERSON, a 22-year veteran of the fire service, is a lieutenant with the Madison (WI) Fire Department and the operations and training coordinator for the Regional Level A Haz Mat Response Team. He is the owner of Americhem Safety & Environmental, LLC, a haz-mat training and consulting firm in Janesville, Wisconsin. He is also an IAFF Master Trainer, an adjunct instructor for the National Fire Academy and the Emergency Management Institute, and an FDIC presenter. He is a member of the NFPA Classification and Properties of Hazardous Chemical Data Committee and the founder and past president of the Wisconsin Association of Hazardous Materials Responders, Inc.