PROPER STORAGE THE KEY TO ACETYLENE TANK SAFETY
BY STEVEN P. MASLANSKY AND JOHN M. CULLEN
At 0759 hours on March 24, 1995, the White Plains (NY) Fire Department was dispatched to a reported explosion at The Westchester, a large (2.5 million square feet–one million square feet of retail space and 1.5 million square feet of parking), recently opened mall that was still under construction in some areas.
On arrival, police and fire units found no fire, a light smoke condition, and a plumber lying adjacent to a demolished tool chest. The victim, a 21-year-old male, was transported to the hospital with severe leg injuries. Surgeons were able to save his leg.
The Fire Investigation Unit of the White Plains (NY) Fire Department, the Detective Division of the White Plains Police Department, the Westchester County Department of Public Safety Bomb Squad, the Westchester County Office of Fire Services Hazardous Materials Response Team, and the U.S. Department of Labor Occupational Safety and Health Administration (OSHA) investigated.
BACKGROUND
According to witnesses, the plumber`s tool chest (gang box) was locked prior to the explosion. This was confirmed by an examination of the padlock, which was still on the lid. The box lid had to be retrieved from a ceiling girder some 18 feet directly above the box, where it had been blown. It was also reported that an acetylene tank was found in the damaged box after the explosion, but workers had removed it to a “safer” location immediately after the explosion. A considerable quantity of copper fittings, rolled solder, and various small containers of pipe compounds were also stored in the box.
Although the box was severely damaged, particularly at the welds and corners, it showed no perforation or fragmentation damage. A piece of the box was blown approximately 90 feet away, striking a gypsum board wall. The acetylene tank was remarkably undamaged except for a slightly bent valve stem that also showed abrasive damage. It is not known if the damage was caused by the explosion or prior use. The valve stem damage suggested that it had been opened and closed with a pair of pliers or a wrench instead of a valve key. The cylinder and the tool chest and its contents showed no evidence of fire damage; there was no soot on any of the surfaces.
The cylinder was found with the valve in the closed position; it was checked for leaks with a soapy water solution and a gas detector capable of detecting acetylene leaks in the low parts-per-million range. No leaks were found when the valve was in the closed position. A leak immediately developed around the valve stem and the gland nut when the valve was turned to any open position. Escaping gas then was measured with a percent lower explosive limit (LEL) meter. In a few sections, concentrations in excess of 100 percent of the LEL were obtained directly over the valve. The torch and hose connected to the cylinder were tested, and no leaks were observed.
THE EXPLOSION
There was no evidence of flammable liquids or flammable gases (other than the acetylene cylinder) in the area of the incident. There was nothing to suggest that flammable liquids or gases were stored in or around the tool chest. No evidence of an explosive device was found. It is reasonably certain that the cylinder leaked acetylene and was the source of the explosive material.
SOURCE OF IGNITION
Several scenarios that follow might explain what caused the acetylene to detonate:
chemical decomposition reaction with self-detonation,
chemical decomposition reaction with detonation by shock,
flammable concentrations within or around the tool chest with static or friction as an ignition source, and
flammable concentrations within or around the tool chest with smoking materials as an ignition source.
PROPERTIES OF ACETYLENE
To better understand the above scenarios, the physical-chemical properties of acetylene must be discussed. Acetylene (C2H2) is a colorless, flammable, odorless gas when 100 percent pure, but it exhibits a characteristic garlic-like odor when of ordinary commercial purity (95 to 99 percent). It is slightly lighter than air and can autoignite at temperatures between 571° and 824°F. The flammable range for acetylene is between 2.5 and 82 percent by volume. Acetylene burns (even without a supplemental oxygen source) at 4,200° to 6,300°F, which is much greater than most flammable gases.
Acetylene vapors can decompose explosively when present at concentrations above 81 percent and at pressures as low as six psig, depending on container size and shape. Large-diameter pipes are the most dangerous. The wide flammable range and low detonation pressures make acetylene an extremely dangerous gas unless it is handled properly.
Although the acetylene tank in question had a working pressure of 250 psig when full, this pressure usually poses no hazard because the cylinder-gas system is designed to keep the gas stable. Acetylene is always dissolved in a solvent such as acetone to make it less susceptible to low pressure detonation; the cylinder contains a porous filler that does not allow free acetylene gas to collect. When released from the cylinder into small-diameter hoses and accessories, acetylene is regulated to pressures of less than 15 psig. In the event of fire or other situations that could heat the cylinder, a fusible plug with a melt temperature of approximately 212°F prevents excessive pressures from developing.
Acetylene typically burns in air with a hot, luminous, smoky flame and can spontaneously explode when it comes in contact with chlorine. Acetylene can form spontaneously explosive compounds with unalloyed copper, silver, or mercury. Shock-sensitive copper acetylide can form when acetylene comes in contact with copper oxide coatings. Acetylides can also form on copper in the presence of caustic materials such as ammonia.
ANALYSIS
The acetylene tank in question was a standard 40-cubic-foot-capacity tank, commonly referred to as a “B” tank, that showed a gauge reading of three-quarters full. It was the investigators` understanding that the tank was close to full when it was secured the night before the incident. Assuming the gauge was correct, no more than 10 cubic feet could have leaked into the box. The box in question has a volume of approximately 16 cubic feet. Given a lower explosive limit of 2.5 percent by volume and an initial gas concentration of 98 percent, approximately 400 cubic feet of flammable gas could have been formed during the night and leaked out of the nonairtight chest, producing flammable mixtures around the chest as well as maintaining a flammable concentration within the chest.
The static discharge or spark by friction as a source of ignition does not appear reasonable in light of the room conditions and the chest`s construction. The copper acetylide formation theory, although very intriguing, does not appear reasonable in light of the lack of oxidized surfaces on the copper fittings, although some formation of this compound should not be dismissed. An acetylene concentration in excess of 81 percent in the tool chest ignited by a shock or through self-detonation because of a slight increase in internal pressure also probably is not reasonable. Although it was reported that the injured worker forcefully tried to lift the lid, there was a delay in the explosion. It was reported that he was bending over beside the chest when it detonated. It is also unlikely that the gas could be released fast enough to increase an internal pressure greater than the atmospheric pressure due the construction of the tool chest. It was also noted that the box had two small saw holes from a previous sawing error, further decreasing the potential for a buildup in gas pressure within the chest.
The most reasonable explanation for a source of ignition was that the injured worker was smoking or lighting a cigarette while bending over, and the source of ignition came in contact with a flammable concentration just outside the box. The reaction was propagated to the inside of the chest, which then detonated the remaining gas within the tool chest. The injured worker, although a smoker, denied he was smoking or preparing to smoke at the time of the incident.
LESSONS LEARNED
On April 25, 1995, OSHA issued two “serious” citations against the plumbing contractor–for failure to use a special wrench and keep it in place on the cylinder [29 CFR 1926.350(d) (2)] and to place fuel gas in a well-ventilated and accessible location [29 CFR 1926.350 (e) (2)].
Fire officers and inspectors must understand that although acetylene gas is used on a regular basis, they need to ensure that it is handled and stored properly.
Acetylene is a flammable gas subject to detonation or explosion if cylinders are not handled properly. Cylinder keys or wheels should always be used to properly open and close gas cylinders. Acetylene cylinders should always be stored upright in an area where gases cannot collect. Smoking should be prohibited in areas where flammable gas cylinders are stored or used.
When conducting an investigation into an explosion, all causes must be examined. Use the experts in the various fields (bomb squad, haz-mat, industrial, for example). There is no room for guesswork in this type of investigation. n
References
Braker, W. and A.L. Mossman. Matheson Gas Data Book, Sixth Edition, 1980.
Handbook of Compressed Gases, Compressed Gas Association (Van Nostrand Reinhold, 1990).
 |
 |
(Top) The area of the explosion. Note the box lid, which has wrapped itself around the ceiling beam. (Bottom) A gang box with an improperly stored acetylene tank identical to the one involved in this incident. (Photos courtesy of the White Plains Police Department.)
STEVEN P. MASLANSKY has more than 23 years of experience in hazardous materials. He is deputy fire coordinator for hazardous materials in Westchester County, New York, and president of GeoEnvironmental Consultants, Inc., also in White Plains. He has consulted and lectured on hazardous materials in North America and Europe and has a bachelor`s degree in geology and a master`s degree in environmental science and civil engineering.
JOHN M. CULLEN, an 18-year veteran of the fire service, is a lieutenant and fire investigator in the White Plains (NY) Fire Department and a member of the Tarrytown (NY) Volunteer Fire Department. He has an associate`s degree in fire science and a bachelor`s degree in public safety
