Project Examples

World Class Training



Landfill Gas

Helpful Tech Books



Contact Us


Investigators Present Final Report in Last Year's Explosion at D.D. Williamson and Co., Citing Improper Installation, Lack of Emergency Relief Valves, Damage History of Caramel Processing Vessel.

Louisville, KY, March 12, 2004 - Investigators from the U.S. Chemical Safety and Hazard Investigation Board (CSB) said the April 11, 2003, explosion and resulting ammonia release at the D.D. Williamson and Co. plant in Louisville were caused by over-pressurization of an eight-foot-tall food additive processing tank. The CSB said the accident could have been prevented had the company installed an emergency pressure relief valve on the tank. The CSB also noted that the tank that exploded had a history of prior damage.

The explosion took the life of an employee who had worked for five years at the plant. It caused extensive damage to the facility, which makes caramel coloring for use in food products such as soft drinks. The explosion, which occurred around 2:10 a.m., blew the top of the tank some 100 yards to the west. The tank shell struck a nearby ammonia tank, knocking it off its foundation. This resulted in the release of an estimated 26,000 pounds of aqua ammonia (ammonia gas in a water solution) over a five-hour period, forcing the evacuation of 26 residents and requiring 1500 others to remain sheltered in their homes.

Lead CSB investigator David Heller said, "The accident was avoidable. In the 1980s, the company shipped two used tanks, including the one that exploded, from out of state to the Louisville facility. The tanks had not been inspected, certified, or registered as pressure vessels prior to bringing them into Kentucky - a requirement of the state's Boiler and Pressure Vessel Safety Act."

Mr. Heller said the company routinely heated liquid caramel in the vessels to 160°F and then used compressed air to help push the caramel out to a dryer. Mr. Heller said, "Since the vessels were operated above pressures of 15 pounds per square inch, the company should have classified these tanks as pressure vessels as required by law. The tanks should have been equipped with emergency pressure relief valves, pressure and temperature alarms, and automatic systems to shut down the process in case of over-pressurization. In the absence of these safety measures, operators had to rely on visual inspection of temperature and pressure gauges to keep the process under control."

Investigators determined that on the night of the incident, two workers, who were brothers, filled the tank with liquid caramel and turned on the heating steam to the vessel. Meanwhile, they were occupied in another room re-labeling some product boxes that had been mislabeled. Returning later to the tank room, the second operator noticed that the caramel was leaking from the top of the vessel and called in the lead operator. A metal insulation band snapped in two as the tank expanded under the increasing temperature and pressure inside. The lead operator then sent his brother to locate a mechanic. Moments later the vessel exploded, killing the lead operator.

Investigators said the lead operator likely had attempted to open the tank's air vent to release the excess pressure. But the vent was not designed for emergency pressure relief and was not adequately sized for the vessel. In any event, investigators later found that the vent pipe had clogged with solidified caramel product.

The CSB concluded that it was "improbable" - based on the temperature of the heating steam - that the pressure inside the tank ever exceeded 130 pounds per square inch (psi). Drawings show that the tank was built with a maximum working pressure of 40 psi, and CSB investigators estimated that the tank, as originally designed, was probably capable of withstanding pressure up to 180 psi. Therefore, the CSB said, the "more likely cause of failure" was that the tank had been weakened sometime earlier. The report noted that the tank had been deformed on two occasions prior to being installed in Louisville when it was subjected to excessive vacuum, and was subsequently repaired. The repairs were not inspected or certified.

CSB Board Chairman Carolyn Merritt said, "The tragedy that befell this worker is another example of why plant owners and managers must have effective engineering oversight and hazard analysis systems in place. They should be regularly analyzing various scenarios that could lead to accidents and put into place safety systems that result in extra layers of protection."

Investigators cited several root causes, noting that the feed tanks were installed without a review of their design or fitness for service. Investigators concluded that D.D. Williamson and Co. did not have effective programs to determine if equipment and processes met basic engineering requirements. The company also lacked effective systems for assessing the hazards of its processes. Finally, the company did not instruct workers on the hazards of overheating or over-pressurizing the caramel vessels.

CSB investigators proposed several recommendations be issued to the D.D. Williamson and Co.: examine all vessels at company facilities and ensure that each pressure vessel has adequate pressure relief systems and alarms. The CSB also recommended the company upgrade operating procedures, train its operators, and implement a hazard evaluation procedure to determine the potential for catastrophic accidents. The CSB recommended that the Kentucky state government inform pressure vessel owners, mechanical contractors, engineering companies, and insurers that used pressure vessels must be inspected and registered before being placed in service in Kentucky.

The U.S. Chemical Safety and Hazard Investigation Board (CSB) announced it will conduct a root-cause investigation of the April 12, 2004 chemical accident at the MFG Chemical Inc. facility in Dalton, Georgia, which triggered a release of allyl alcohol and hydrochloric acid into the surrounding community.

Board investigations typically take 9 to 12 months to complete.

The accident occurred as operators were preparing for the first full-scale batch of a new chemical product intended for use in the rubber industry. Seven of the facility's approximately 35 employees were present at the time.

To begin the process, operators added allyl alcohol, cyanuric chloride, and a catalyst into a 4,000-gallon processing vessel. After addition, the internal temperature in the vessel began rising rapidly. Increasing pressure burst the vessel's rupture safety disk. Heated chemicals were then forced through the vent line onto the first floor of the facility, forming a gas cloud that traveled out into the surrounding community. Hydrochloric acid was generated inside the vessel and was released, together with allyl alcohol.

Although no fire was observed, emergency responders applied a significant quantity of water in an effort to control the gas cloud and cool the processing vessel. Workers and residents within about one-quarter mile of the facility were evacuated until the following day due to the chemical release. Based on initial figures, 184 people were seen at a local hospital for possible chemical exposure, including 31 who were transported there by ambulance. Six people were hospitalized overnight on Monday, and all were subsequently released. Vegetation in the vicinity of the plant site was damaged by chemical exposure, and fish in a nearby stream were killed by contaminated water runoff from the emergency response.

The U.S. Chemical Safety and Hazard Investigation Board (CSB) is investigating a spill of approximately 12,500 tons of molten sulfur from a storage tank located at the Port of Tampa, Florida.

The spill was reported to have occurred on Saturday, April 3, 2004 when the tank failed following the offloading of sulfur from a cargo ship.

Washington, DC, March 30, 2004 - U.S. Chemical Safety and Hazard Investigation Board (CSB) Chairman Carolyn W. Merritt praised the U.S. Occupational Safety and Health Administration (OSHA) for its initiative in forming a new alliance to address reactive chemical hazards, saying the effort "can help save lives, protect businesses, and spare communities from the devastating impact of reactive chemical accidents."

Reactive hazards exist when a single chemical or a mixture of chemicals has the potential to undergo a violent, uncontrolled reaction when processed or handled improperly.

Such chemical reactions can release large quantities of heat, energy and gases, causing fires, explosions or toxic emissions. A 2002 investigation by the Chemical Safety Board found 167 serious reactive incidents in the U.S. over the previous two decades. These incidents caused 108 fatalities as well as extensive injuries and property and environmental damage.

Baton Rouge, LA, March 30, 2004 - Investigators from the U.S. Chemical Safety and Hazard Investigation Board (CSB) provided a community update on the ongoing investigations into three accidents that occurred at Honeywell International in East Baton Rouge Parish over a 24-day period in 2003.


A chlorine release on July 20, 2003, injured eight Honeywell employees with four of them hospitalized. It was caused by a failure of a chlorine cooler, which allowed liquid chlorine to enter the refrigerant system.

Chlorine gas escaped through a hole in a chlorine cooler tube. Investigator Lisa Long reported that operators first detected the release at 3:05 a.m. Local authorities activated a siren system at 4 a.m. advising residents to remain in their homes for safety. The company attempted to contain the release by spraying the chlorine cloud with water, but it was not until 7 a.m. that the all-clear signal was given."

Honeywell's testing of chlorine levels, beginning at 5:30 a.m., found concentrations of 1.2 ppm. Concentrations of 1 to 3 ppm can cause mucous membrane irritation; much higher concentrations cause chest pains and death. By 7 a.m. the company reported concentrations below 0.35 ppm.


Investigator Mike Morris reported that a worker, Delvin S. Henry, was fatally injured after he was sprayed with spent, or used, antimony pentachloride, a highly corrosive chemical that can cause serious chemical burns and lung damage.

The operator was preparing a large used refrigerant cylinder for off-site testing. Five years earlier, the cylinder, originally labeled as antimony pentachloride, had been relabeled as a refrigerant before shipping it to the Baton Rouge plant. The re-labeling was performed by a specialty company in Denver after consulting with Honeywell operators in California.

Though he had worked in the plant for about 3 years, the operator was new to this particular job, and evidently believed all of the cylinders he was working with were empty or contained only small amounts of refrigerant and were labeled as such.

Unable to properly purge, or clear, the contents of the cylinder by hooking up hoses to the cylinder's drain valve, the operator removed a plug from the back of the cylinder. The spent antimony pentachloride sprayed out under pressure, enveloping the operator, resulting in his death the next day.


CSB Investigator Johnnie Banks reported that the Honeywell plant had been rapidly shut down following the July 20, 2003, chlorine accident. After the July 29 incident the company ordered the facility to remain shut down until procedures and equipment could be checked to ensure the facility could be operated safely. On Aug. 13, liquid hydrofluoric acid splashed onto a worker.

The rapid shutdown after July 20 had left significant amounts of the HF, a colorless chemical that can quickly destroy human tissue, in an internal piping system. A decision was made to remove it. Operators began draining it into a sewer through a one-inch pipe where flowing water siphoned the HF into the wastewater sewer system.

On Aug. 13, the system was suspected of being clogged, and an operator opened and closed two valves in an attempt to clear the blockage. The operator then reopened the valves, resulting in a pressure surge that ran through the one-inch pipe, called a "venturi stick."

Pressurized by nitrogen, hydrofluric acid and water escaped from the sewer, splashing onto the worker. CSB investigators found the pipe had not been properly secured, and was suspended by a single rope.

The operator was aided by a maintenance supervisor, who experienced a coughing spell from the fumes. The operator was treated and released at the local hospital; the maintenance supervisor was released from the hospital the next day.

Investigators from the U.S. Chemical Safety and Hazard Investigation Board (CSB) have determined that the October 2003 fatal accident at the Hayes Lemmerz automotive parts plant at Huntington, Indiana likely involved an explosion of aluminum dust that originated near an aluminum chip melting furnace.

A secondary explosion occurred some minutes later in the dust collection equipment, and there were also smaller blasts involving gas cylinders and tires located outside of the building in an environmental contractor’s trailer.

Two Hayes Lemmerz employees were severely burned during the explosions. One died the following day, and the other remains hospitalized. A third employee was treated at a hospital and released. Approximately 80 workers were at the plant on the evening of October 29 when the incident occurred.

The furnace involved was used to melt aluminum chips, which were a byproduct of machining auto wheels cast at the plant. The chip processing produced aluminum dust, which like many other finely divided metals is flammable when mixed with air. Dust that had accumulated on surfaces in the work area was ignited by a flash fire that escaped from the dust collection hood over the furnace.

The initial blast occurred as employees were restarting the chip processing operations after a temporary interruption caused by a small duct fire earlier in the day. Investigators believe that fire which occurred in a separate duct system had been fully extinguished and was not the ignition source for the later explosions.

The aluminum reprocessing area remains damaged and shut down in the aftermath of the accident, although other parts of the facility are back in operation. The explosions also destroyed external dust collection equipment and the environmental contractor’s trailer, as well as causing some damage to the roof of the main building.

Board member Dr. Gerald V. Poje, who accompanied the field team, said: "We are deeply concerned by the number of serious dust explosions that have occurred at various companies in recent months. The Board will work expeditiously to determine the root causes of the accident at Hayes Lemmerz and recommend new safety measures to save lives at this and other firms around the country."

The Board is currently investigating two dust explosions that killed a total of 13 workers at unrelated facilities in North Carolina and Kentucky earlier this year. The Kentucky plant, which made fiberglass automotive parts, was heavily damaged, while the plant in North Carolina was destroyed and is being rebuilt at another site.

(Washington, DC, Aug. 12, 2003) U.S. Chemical Safety Board Chair Carolyn Merritt today commended the recent action by the New Jersey Department of Environmental Protection which adds reactive chemicals to the list of extraordinarily hazardous substances that trigger the risk management planning requirements of New Jersey’s Toxic Catastrophe Prevention Act. The action is designed to provide greater protection for residents living near industrial facilities.

Following an extensive, two-year study of reactive hazards in the chemical industry, the 5-member CSB Board last fall approved recommendations to the federal Environmental Protection Agency (EPA) and the Occupational Safety and Health Administration (OSHA) on the need for additional regulation of reactive hazards. The CSB study identified 167 reactive incidents which caused 108 deaths over a twenty year period. Dr. Gerald Poje, on behalf of the CSB, earlier this year had urged New Jersey officials to take action.

Chemical products often are made through the process of chemical reactions. However, when chemicals are improperly mixed or improperly exposed to heat, pressure or incompatible substances, they can explode and cause death and destruction of property.

We have found a need for plant management to study potential reactive hazards before they make changes to their processes because accidents occur when inadequate safety and control measures are in place, Chairman Merritt said.

New Jersey announced it would require companies handling reactive chemicals to prepare accidental release prevention plans and examine safer technologies to prevent industrial incidents like the tragic ones that occurred at Napp Technologies in Lodi in 1995 and at Morton International in Paterson in 1998.

Both explosions were the result of the improper mixing of certain chemicals.

For help in analyzing the potential for hazardous reactivity in your processes, NST/Engineers suggests reference to "Screen Your Facilities for Chemical Reactivity Hazards" by Johnson and Lodal, page 50, August 2003, Chemical Engineering Progress (the AIChE monthly).

Combustible dust, ignited by a malfunctioning oven, caused an explosion and fire that took seven lives and caused over thirty injuries. The incident occured at the CTA Acoustics plant in Corbin, KY, in Feb. 2003

The initial explosion and fire occurred on a production line that was partially shut down and being cleaned at the time of the incident. During the cleaning, a thick cloud of dust dispersed around the line. The dust was likely ignited by a fire that spread from the production line's oven, which was still operating, investigators said.

"The plant's four production lines had a history of small fires erupting in the ovens," said lead investigator Bill Hoyle. "Plant operators routinely put out these fires. However, during the cleaning operation, no one was present in the immediate area of the oven who could have detected a fire."

Dr. Gerald Poje, a CSB Board Member has said, "Dust explosions are a significant hazard in manufacturing operations. This accident happened only a few weeks after the terrible tragedy in Kinston, North Carolina, which was also caused by the ignition of dust in the plant and which claimed six lives. OSHA has safety standards to prevent dust explosions in grain elevators, but not in other types of manufacturing facilities."

The CSB preliminary report to the community found that one or more oven doors on the production line where the incident began had been left open to cool down the oven because temperature controls had been malfunctioning for several days. Flames likely escaped from the oven door and ignited the dust cloud.

"The fire quickly spread over a wide area of the plant," according to Investigator Hoyle. "Dust that had accumulated on flat surfaces throughout the plant was disturbed and became airborne, providing more fuel for the fire. The initial explosion stirred up more dust and led to secondary explosions," Mr. Hoyle said.

The dust was composed primarily of phenolic resin, a raw material used in the production process. The resin is similar in consistency to talcum powder.

(NST/Engineers note: However, inorganic talcum powder is not the potential combustible fuel that phenolic resin powder, an organic compound will be.)

CSB Chairman Carolyn Merritt said, "This accident has led to one of the most significant and far-reaching CSB investigations to date. The tragedy at CTA resulted in deaths and serious injuries that everyone - in industry, labor and government - wants to avoid. Our investigation and recommendations will help save lives in the future by getting industry managers to increase their focus on the insidious dangers of combustible dust." Merritt noted that the U.S. Occupational Safety and Health Administration (OSHA) does not have specific standards for controlling combustible dust hazards in many industrial facilities. National standards are in place for dust hazards in coal mines and grain handling facilities. As a result of the accident at CTA Acoustics and other fatal dust explosions in 2003 in Kinston, NC, and Huntington, IN, the CSB is also conducting a nationwide review of the number and severity of dust explosions at U.S. facilities in recent decades. The dust explosion January 29, 2003, at West Pharmaceutical Services in Kinston killed six workers and destroyed the facility.

In New York City the U.S. Chemical Safety and Hazard Investigation Board (CSB) has presented preliminary findings from an investigation into the April 2002, explosion at Kaltech Industries, which injured 31 people seriously enough to seek hospital treatment, including 14 members of the public.

Lead CSB investigator Steve Selk reports Kaltech employees improperly mixed hazardous waste chemicals in an operation that lacked basic safe handling procedures. "The investigation found the chemicals that most likely produced the explosion were nitric acid and lacquer thinner, which workers combined in a 55-gallon drum," Selk says. "An uncontrolled chemical reaction between the two materials produced significant amounts of gas or vapor that exploded within the basement, causing significant damage as high as the fifth floor."

The The Kaltech chemical explosion seriously injured many workers and members of the public and disrupted the lives of New Yorkers, "We are very concerned with any chemical accident that has a significant impact on the public. Our investigation can help prevent similar accidents in the future."

Mr. Selk told the five-member board in his report, "Kaltech did not maintain a list of chemicals in the workplace. The company did not provide its employees with required safety data sheets on the chemicals or train workers on the hazards of the chemicals. Apparently management was simply not aware of the federal safety regulations requiring those things."

Selk said the blast also caused a secondary fire by knocking over a drum of alcohol, which spilled and ignited, possibly by contact with nearby unapproved electrical equipment.

The CSB's preliminary findings indicate that Kaltech failed to follow a number of requirements of the U.S. Occupational Safety and Health Administration (OSHA) which likely would have prevented the accident. However, the investigation found that OSHA had never inspected the company or its predecessor sign business in at least ten years.

Investigator Selk said that many cities and states institute their own codes on handling hazardous materials. Although New York City's fire regulations include various controls, Selk noted, "Other fire codes require that businesses maintain material safety data sheets, label all containers, and submit hazardous materials management plans before permits are issued. These codes parallel federal OSHA regulations, and they afford local inspectors the authority to enforce safe practices."

Washington, DC, March 19, 2004- Carolyn Merritt, Chairman of the U.S. Chemical Safety and Hazard Investigation Board (CSB), said the Board is pleased with the progress of the City of New York in responding to the CSB's recommendation to revise its antiquated fire code to better control the storage and use of hazardous materials. The New York City Council announced in a March 5, 2004 news release that the city's fire department has decided to revise the code and has allocated substantial funding in the department's new budget to support the revision.

The recommendation to revise the code was contained in the Sept. 30, 2003 final report into the explosion at Kaltech Enterprises in the Chelsea district of the city, which occurred on April 25, 2002. Investigators found that employees improperly mixed hazardous waste materials at the company, and that the NYC fire code did not adequately cover such hazards.

(Washington, DC - November 22, 2002) The U.S. Chemical Safety Board approved the final staff report into the root causes of the January 16, 2002, hydrogen sulfide gas leak at the Georgia-Pacific Naheola pulp and paper mill in Pennington, Alabama, which took the lives of two workers and injured another eight.

The Board concluded that plant management had not followed good engineering and process safety practices when in 1995 they connected a drain from a tank truck unloading area into an acidic process sewer system. The plant was then owned by the James River Corporation, which later merged with Fort Howard and was acquired by Georgia-Pacific.

On the day of the incident, sodium hydrosulfide -- a process chemical that had spilled in the unloading area - reacted to release deadly hydrogen sulfide gas when it contacted acidic material in the sewer.

The toxic gas vented from the sewer through a nearby fiberglass manhole cover and engulfed the workers. The two deaths and all but one of the injuries occurred among employees of Burkes Construction who were working in the vicinity on an unrelated construction project.

The Board concluded that neither Georgia-Pacific nor the previous plant owners adequately analyzed or controlled the hazards of the sewer system, including the potential for hazardous chemical reactions.

In September 2002, the CSB completed a special investigation which recommended that the U.S. Occupational Safety and Health Administration (OSHA) and the Environmental Protection Agency (EPA) develop more comprehensive regulatory coverage for reactive hazards.

The Board recommended that Georgia-Pacific Corporation review sewer system safety at all its plants to prevent the inadvertent mixing of potentially reactive chemicals - including those that can form toxic gases.

The Board also requested that Georgia-Pacific identify plant areas (such as truck unloading areas) where there is a risk of hydrogen sulfide release and require appropriate safeguards and training for all workers in those areas.

"This tragedy raises a number of significant safety issues that likely warrant further study," according to Board Chairman Carolyn W. Merritt, herself a veteran of the paper industry. "We need to examine how many other incidents of this kind have occurred: is this an isolated event or a symptom of a larger problem?

I am directing the CSB staff to look into past events involving toxic gas releases from acid-reactive materials, like sodium hydrosulfide and other sulfide and cyanide salts. We are particularly interested in releases or exposures involving sewers and other disposal systems.

We need to study how the paper industry and other industries manage these reactive materials, and whether current practices, training, standards, and guidance are adequate to ensure the safety of workers and the public.

Also of interest are the recommendations from sodium hydrosulfide manufacturers for safe handling, spill control, and disposal of these materials: are they consistent, are they sufficient, and are they being followed?

I have asked the staff to report back to the Board by December 11 with an outline for the study. During this process we will solicit significant input from the American Forest and Paper Association and other interested trade and labor groups."

(Sept. 20, 2002 - Washington, DC) Meeting before a public audience in Houston on September 17th, the U.S. Chemical Safety and Hazard Investigation Board (CSB) unanimously approved a total of 18 new recommendations to reduce the number of serious industrial accidents caused by uncontrolled chemical reactions and called on the Occupational Safety and Health Administration (OSHA) and the U.S. Environmental Protection Agency (EPA) to issue new mandatory safety standards.

CSB investigators told the Board members and the public that inadequate controls of chemical reaction hazards are responsible for continuing deaths, injuries, and environmental and property damage around the country. Three workers who were severely burned on March 27, 2000, in a reactive accident at Phillips Chemical Co. in Pasadena, Texas, spoke at the public meeting prior to the Board vote.

For the first time the Board called on OSHA and EPA to extend their process safety regulations - known as the Process Safety Management standard and the Risk Management Program rule - to better control hazards associated with chemical reactivity. Under the terms of the Clean Air Act, OSHA and EPA must respond within no more than 180 days to the Board's recommendations.

"This report and its recommendations are a landmark for the Board," said Carolyn W. Merritt, CSB chairman, after the final vote was announced. "In time, I hope the recommendations will be seen as a landmark for the progress of chemical safety as well. Since the process safety regulations were first promulgated a decade ago, there has been a notable hole in the coverage of reactive hazards. The Board's recommendations today mark a first step in closing that hole and implementing new standards that ultimately will save lives."

The approval culminates a two-year special CSB investigation into hazards at U.S. sites that manufacture, store, or use potentially reactive chemicals. The study examined 167 serious chemical accidents in the U.S. over the last 20 years that have involved uncontrolled chemical reactions. These accidents caused 108 deaths as well as hundreds of millions of dollars in property damage. More than half of the accidents involved chemicals that are exempt from OSHA and EPA process safety rules.

The Board requested that OSHA broaden the application of the PSM standard to cover both individual chemicals and combinations of chemicals that can undergo hazardous reactions under specific process conditions. The standard currently applies to only 137 listed chemicals, plus a class of flammable substances (there are estimated to be thousands of chemicals in common industrial use). Just 38 "highly reactive" chemicals are currently covered by the PSM standard. But the CSB study documented numerous examples where chemicals considered to be less reactive nonetheless caused runaway reactions, explosions, fires, or toxic gas releases, often with fatal consequences.

The Board made a separate but similar recommendation to the EPA, which currently does not specifically regulate reactive hazards under its RMP rule. The Board investigation pointed to numerous examples where reactive accidents had a significant public or environmental impact. For example, the 1999 Concept Sciences explosion outside of Allentown, PA, killed a member of the public and damaged nearby businesses. The chemical involved, hydroxylamine, is not covered under the EPA rule. Likewise, large public evacuations were necessary in May 1990 after a significant toxic gas release at Dow Chemical's Freeport (TX) facility - caused by a reaction between two incompatible chemicals that remain unregulated under the RMP rule.

The Board further called on OSHA to modify the PSM standard by requiring companies to evaluate the potential for hazardous reactions in each covered process. Companies would also be required to consult a wider array of scientific and technical literature on reactivity in compiling process safety information - information that is critical in designing safe processes and in protecting employees from workplace hazards. The Board cited deficiencies in process safety information as a root cause of the 1998 Morton explosion in Paterson, NJ, a reactive accident which injured nine workers and gave rise to the Board's reactive hazard investigation.

EPA and OSHA were also requested to collect additional reporting information on reactive accidents within their respective jurisdictions. Board investigators said that progress on preventing reactive accidents was hampered by a general lack of reliable data - including information on root causes and lessons learned. They also noted that the tally of 167 reactive incidents and 108 fatalities was almost certainly an underestimate due to data deficiencies.

Citing inadequacies in existing industry guidance on reactives, the Board called on the American Chemistry Council (ACC), the National Association of Chemical Distributors (NACD), the Synthetic Organic Chemical Manufacturers Association (SOCMA), and the Center for Chemical Process Safety (CCPS) to develop new voluntary codes and standards for controlling reactive hazards. Two of those groups - ACC and SOCMA - were also called on to cooperate with the National Institute of Standards and Technology (NIST) in developing a new national database of reactivity test information. This public database of industrial test data would complement existing knowledge on reactive hazards available from the published literature.

The CSB is an independent federal agency established in 1998 with the mission to protect workers, the public, and the environment by investigating and preventing chemical accidents. The CSB determines the root causes of these accidents and makes safety recommendations to government agencies, companies, and other organizations. The CSB does not issue fines or citations or apportion responsibility for accidents. Additional information is available from www.chemsafety.gov.


New York City Scaffolding Disaster Kills 5 and Injures Many Others

October 15th, 2001

A 14-story combination frame and tube and clamp scaffolding erection collapsed and crushed five workers to death in New York City's Gramercy Park neighborhood on October the 15th. NYC Licenses and Inspection Authorities were quoted as saying that the scaffolding was erected without a permit and was never inspected by the city. Also injured in the incident were 13 workers, a pregnant woman who was a bystander and several FDNY and NYPD rescue workers.

Workers erected the scaffolding, which was set up to repair the facade in a courtyard behind 215 Park Avenue South. It was reported that both the construction firm and the scaffold erectors had had run-ins with both regulatory and legal authorities in the recent past.

According to Rayman and Lefkowitz, police report indicated that a heavy concrete load placed on upper levels of the scaffolding caused swaying and buckling at the eighth-level of the structure. The scaffolding in use was about two years old, police said. OSHA and the NYC Building Department are investigating to find the exact cause of the catastrophe, one of the worst scaffolding accidents since the Philadelphia accident several years ago.

OSHA's scaffolding regulations are found at 29 CFR 1926, Subpart L. They state clearly that all scaffolding erections must be designed by a competent person and that only trained personnel may erect scaffolding. The regulations are clear as to heights and means used for securing scaffolding to the working structure. Further, they are clear, in 29 CFR 1926.451 as to the construction and load bearing capacity of different types of scaffolding.

Whether the scaffolding was improperly erected or overloaded remains to be seen. OSHA is also clear that scaffolding must be inspected daily and all deficiencies corrected prior to work commencing, which apparently had not taken place here.

This is the most serious construction related incident in the recent history of New York City.

Another Ammonium Nitrate Disaster, 29 people killed, 34 'gravely injured', and 650 others hospitalized.

September 21st, 2001

The French city of Toulouse ( 1 million inhabitants) suffered from a tremendous explosion at an Atofina fertilizer plant. Neighboring buildings were destroyed and half the window glass in the city was broken. The explosion is believed to have been caused by ammonium nitrate.

The most damaging non-nuclear explosions worldwide have been caused by ammonium nitrate. The most disastrous was the 1947 explosion of an ammonium nitrate loaded ship in the Texas City, Texas Harbor.

Ammonium nitrate is an amazing fertilizer. You can write your name in your lawn by carefully sprinkling it. However, its great hazard is that it is also both an explosive and a monopropellant. That is, the molecule contains both oxidizing (nitrate) and reducing (ammonium) components.

Ammonium nitrate generally requires a detonator to initiate explosion. However, self-heating accidents have often occurred when small-scale procedures have been scaled up without considering the effects of size. 'Critical diameter' measurements have been made for explosives and monopropellants. That means the hazard of storage increases with the volume of material stored. There is a critical diameter that varies with the purity of the stored material. An amount stored that measures larger than the critical diameter can spontaneously explode should some decomposition internally provide the necessary initiation.

There are also critical temperatures for these materials, above which rapid decomposition and explosion occurs. Safe amounts can be estimated. Accidents have occurred when additional materials were added to previously safe operations.

This accident is still under investigation and we will continue to cover it as it unfolds.

Freight Train Carrying Hazardous Materials Catches Fire in Railroad Tunnel Beneath Downtown Baltimore

July 17th, 2001

The incident began about 3:10 p.m. CSX Transportation (CSX) officials, and the train's two-man crew (Engineer and Freight Conductor), said the first sign of a problem was actuation of the train's emergency air brake. That halted the northbound tanker and boxcar train in the process of exiting the 1.7-mile long Howard Street tunnel. The train crew was unable to repressurize the brakes. Whenever the air brake system loses pressure for any reason, the brakes are automatically applied. The engineer cannot re-start the train until the air sensor on the last car in the train acknowledges that it detects sufficient pressure. Due to the curvature of the Howard Street tunnel, the line-of-sight radio communication required could not be completed by the engineer. Following proper emergency procedures, the crew uncoupled the engines and pulled free of the train to a safe distance. The crew acted properly according to their HAZMAT training, by uncoupling from the train and removing the locomotives and themselves from harm's way. They immediately radioed to their CSX employer, as required, describing the nature of the emergency as they understood it (heavy smoke and uncoupled train) and the steps they had taken.

As the Baltimore City Fire Department responded, the Freight Conductor supplied the 'Rail Waybill', including the 'HAZMAT Chart' to the On-Scene Incident Commander. The Conductor availed himself to the Commander for further assistance for many hours.

Sixty Car Train Carrying Paper, Acids and a 'Highly Flammable'

According to the Waybills and the HAZMAT Charts, many of the freight cars in this 60 car train were carrying wood pulp and large rolls of paper; typical combustibles. As reported, nine cars were carrying chemicals including five acid tank cars. Two contained fluorosilicic acid, two hydrochloric acid, and one of glacial acetic acid. Other materials on the manifest were ethylhexyl phthalate (a plasticizer reducing embrittlement in plastics), tripropylene (a detergent raw material), and propylene glycol (an antifreeze ingredient). The Emergency Response Guidebook (prepared by the transportation departments of Canada, U.S., and Mexico) lists tripropylene as "highly flammable...vapors heavier than air...vapor explosion hazard...may irritate or burn skin or eyes".

The Rail Waybills and the HAZMAT Charts

These documents identify all cars that make up the train. Cars are numbered in order from car number one, immediately behind the last (in this case second) locomotive, to the last car in the train. This allows emergency responders to know the exact location of each car, and to know its contents. Between the Rail Waybills and the HAZMAT Charts, the Incident Commander then has a computer printout providing full details of all cars in the train, especially identifying all cars carrying hazardous materials. The IC then knows car positions in the train, what each is carrying, and the proper emergency response procedures for fires, spills, explosions, and other releases. The IC knew the acid, chemical, and combustible hazards of the freight.

Serious Fire and Derailment

By the time the Fire Department was notified at 4:15 p.m., black smoke was rising through manhole covers on the street above the tunnel. Realizing a serious fire emergency was developing, the Fire Department sounded five alarms bringing 125 firefighters and equipment to the scene. At some point derailment was discovered. Drivers were trapped for hours on grid locked streets above the tunnel. The Coast Guard closed the Inner Harbor to boat traffic at 5 p.m.

Difficulties Encountered by Firefighters

Firefighters aimed water cannons from each end of the tunnel in an attempt to extinguish a fire they could not see. The heavy black smoke totally obscured vision inside the tunnel. Portable lighting was nearly useless. Firefighters entering the tunnel lost all vision as the smoke deposited a black film on facepieces and goggles. The use of Self Contained Breathing Apparatus (SCBA) was essential. 'Gas masks' and Air-Purifying Respirators (APRs) were useless.

About a half-dozen firefighters wearing standard turnout gear went in the southern end with hoses, trying to reach the train about three-quarters of a mile to the north through smoke and intense heat. They got within about 300 yards of the cars, but had to retreat due to a burning sensation on their necks that they believed was due to chemical attack.

Hazards Restricting Human Intervention

Any incident involving the potential for vaporized hazardous materials, including the possibility of acid vapor attack and splashing, requires full body protection in the form of Level A Chemical Protective Clothing. That is quite different from standard firefighter 'turnout gear'. So with the combined requirements for flame protection, supplied air respiration, and protection from chemical attack; and in addition, facing the hazards of the confinement of an enclosed space with explosion imminent, and vision totally obscured, Firefighters were presented with a situation too dangerous for human intervention. There had to be a reduction in some of those hazard factors, over a period of many hours, before human intervention inside the tunnel was safe. Hazards continued for at least three days.

The heroic actions taken to abate a potential center city disaster were that firefighters had to physically enter the tunnel with a tow cable. In the total blackout conditions, lighting was of no assistance. The responders were able to 'work blind' by following the tracks until they reached the cars. They coupled the tow cable to the last car, exited the tunnel and the process of extracting the smoldering and burning cars to a position where they could be extinguished by waiting firefighters began. This is the kind of life threatening service that HAZWOPER certified, First Responders are expected to provide for the overall safety of the public. Should the tripropylene tanker have caught fire, a BLEVE (Boiling Liquid Expanding Vapor Explosion) could have resulted. The tremendous shock wave of a BLEVE in these circumstances could have collapsed the tunnel and killed all of the responders.

Who Is To Blame?

It has been reported that Baltimore City officials are blaming CSX for causing a break in the city's water main. They are blaming CSX for not properly maintaining their track and roadbed inside the tunnel. They claim that the train derailment and subsequent fire caused the water main break.

Our source at CSX says that there has been a major water leak from a water main line in the tunnel for several months. That water leak could have possibly caused erosion of the roadbed, and actually caused the derailment.

The NTSB (National Transportation Safety Board) is conducting a full investigation of the incident.

Difficulties in Pin-Pointing All Atmospheric Contaminants

Hazardous materials experts from the Maryland Department of the Environment tested the atmosphere repeatedly during the response at both ends of the tunnel. They did not detect acid content or other compounds of concern. They did detect significant wood-ash content in the heavy smoke from the burning freight of combustibles.

At times we have found it difficult to detect all of the atmospheric pollutants that can affect individuals.


The catastrophic failure of a storage tank containing spent sulfuric acid that followed with a fire killed one contract employee and injured eight people.

The incident occurred at the Motiva Enterprise's Delaware City, Delaware refinery.

Maintenance operations were ongoing near the tank, which was close to other tanks. Five contract employees were welding on metal catwalks above the tanks.

Over 600,000 gallons of acid emptied into the Delaware River. As of the first week of August, the dead contractor had not been found.

Employees reported that the tank had holes in it before the incident "that you could put your arm through". Possibly the practice was to not fill the tank to those levels that would cause an immediate leak. Whatever berm there was around the tank to contain any spills apparently failed rapidly and allowed most of the stored acid into the river.

Some of the neighboring tanks appeared to be in questionable condition. There was a report by the CSB that the Material Safety Data Sheet for the spent acid did not indicate a fire or explosion hazard!! ---------

PATERSON, NEW JERSEY - July 18, 2000
Investigators from the U.S. Chemical Safety and Hazard Investigation Board (CSB) have concluded that the April 1998 explosion at Morton Specialty Chemical's (now Rohm and Haas) Paterson, New Jersey facility likely could have been prevented had the company's safety program for reactive chemicals followed recommended industry safety practices.

The blast injured nine workers and released chemicals into the neighboring community. It has been characterized as a “runaway chemical reaction”.

This represents staff conclusions for public review. The findings of fact and the recommendations to prevent similar incidents are not finalized until they have been reviewed and accepted by vote of the Board at a later date.

The Paterson plant manufactures a series of dye products. The explosion and fire occurred during the production of Yellow 96 Dye, which was used to tint petroleum fuel products. A 2,000-gallon kettle being used to produce the dye experienced an uncontrolled rapid temperature and pressure rise (runaway chemical reaction) which resulted in the explosion, injuries and release of material into the community.

Yellow 96 Dye was produced by the mixing and reaction of two chemicals, ortho-nitrochlorobenzene (o-NCB) and 2-ethylhexylamine (2-EHA).

An investigator said the safety programs that were used by Morton for managing reactive chemical hazards did not uncover the potential for the catastrophic runaway chemical reaction in the production of Yellow 96 Dye.

CSB investigators also found that important safety information and recommendations about the hazards of the Yellow 96 Dye process discovered by Morton's United Kingdom research facility were not made known to development and production people at the Paterson facility. This resulted in design flaws and omissions in the kettle and operating instructions used to produce the dye.

For example, investigators found that the kettle did not have the cooling capacity to safely control the temperature of the reaction if reasonably foreseeable upsets occurred.

The kettle also was not equipped with safety equipment, such as a quench system or a reactor dump system, to stop the process in the event of a runaway reaction. In addition, a high-pressure relief device was far too small to safely vent the contents of the kettle in the event that an uncontrollable runaway reaction took place.

Investigators further determined that company training and operating procedures did not prepare operators to safely operate the process to produce the dye or when to evacuate the facility. Some of those injured had stayed near the kettle even while pressure was building uncontrollably and the vessel was rumbling and showing other signs of an impending explosion. Procedures and training did not give sufficient direction to ensure that personnel evacuated prior to an explosion.

Investigators proposed that the Board adopt a recommendation to the U.S. Occupational Safety and Health Administration (OSHA) and the U.S. Environmental Protection Agency (EPA) that they issue joint guidelines for the control of reactive chemical process hazards. They noted that existing federal safety standards do not provide sufficient guidance for reactive chemical process safety.

Unlike incident investigations conducted by regulatory agencies, federal law prohibits the CSB from affixing blame or levying fines or other penalties. Instead, Congress created the CSB to contribute to improved chemical safety by conducting incident investigations, determining all the possible causes, and producing high-quality reports and special studies. From the results of investigations and studies the CSB ultimately adopts recommendations to prevent future incidents, submitting them as appropriate to agencies, organizations, companies and other bodies. ---------


Employee Lawsuit Against Tosco

On January 19, 2000 the surviving victim of a February 23, 1999 explosion at Tosco Corp.'s refinery in Avon, CA near San Francisco and the family of an operator killed in the blast filed a wrongful death lawsuit against the TOSCO. The explosion, occurred when repairs were made to a leaking naphtha 'drawline' in a crude fractionator unit. Four contract workers were killed. Tosco staff operator Steven Duncan, the lead plaintiff in the case was severely burned. Duncan reportedly saved his life by jumping 58 feet down from the scaffolding that supported the repair crew.

Tosco Policy Vs. Lawsuit Claims

Tosco claims its corporate policy states that operators are empowered to shut down any units considered unsafe. However, the suit charges, Avon supervisor Mike Kenney requested that staff repair the leak without first shutting down the fractionator, or, at least, assuring that all flammable chemicals had been drained from the pipeline.

According to the plaintiffs' complaint, several staffers then warned the plant's maintenance supervisor, Bill Scott, that the repair would be too dangerous. They claim they were coerced to proceed. A flash fire started. Operators working on the pipe sustained third-degree burns. Although they were stranded on scaffolding beyond the reach of the plant's emergency response team, the suit alleges Tosco managers waited before contacting local firefighting authorities with adequate rescue equipment.

State of California Charge and Fine
On January 7, the state of California charged the company with criminal negligence for the Avon explosion, and imposed a $2-million fine, one of the highest to date for a CPI plant. Last August, Tosco agreed to pay California's Div. of Occupational Safety and Health a $810,750 fine.

Post-Explosion Arthur D. Little Investigation
Problems at the Avon plant have been the focus of an ongoing investigation by the U.S. Chemical Safety Board (CSB, Washington, D.C.) as well as an April 1999 report by Arthur D. Little, Inc. (ADL, Cambridge, MA).

According to the ADL study, organizational changes, impending layoffs and rapid changes in procedures contributed to a "serious cynicism" about new safety-related programs and initiatives. In the majority of units, operators received 'mixed messages" about safety when it threatened output or production efficiency, the study found.

Other flaws found by the study included an openly hostile relationship between labor and management, and lack of awareness of human factor issues. Key process equipment, for example, was hard to operate and poorly located. ADL recommended 72 steps, and has checked on their implementation at the Avon facility every month.

TOSCO Implimenting 72 Safety Recommendations
As a goodwill gesture, Tosco shut down the Avon facility for four months following the blast, paying workers full wages. Since that time, the company has implemented all 72 steps recommended by ADL, says John Peirson, manager of the ADL project based in Santa Barbara, Calif.

DuPont Style Safety Training Being Instituted
The company has hired DuPont to train staff in safety audits. All employees will be trained by the third quarter of this year. DuPont has long been known as a U.S. leader in chemical industry safety. In addition, Tosco spokesperson Jeff Lyon says, worker contracts have been amended to guarantee the formation of a health and safety committee and to guarantee additional staffing. Over 33 new operator trainees have already been hired, as well as additional health and safety staff, Lyon says.

Contract Workers Have Lawsuit Option
This "wrongful death" lawsuit could have far-reaching impact on California companies in the chemical process industries. Under the current law, full-time staff cannot sue employers for damages when they are disabled on the job, although contract workers can.

Tosco Corporation Announced a New Corporate Office for Safety, Health and Environment on February 16th
Also according to PRNewswire, Thomas D. O'Malley, Tosco's Chairman and Chief Executive Officer announced at its Stamford, Connecticut headquarters the formation of a new senior corporate office for Safety, Health and the Environment. Duane B. Bordvick will head the office as Tosco Corporation's Senior Vice President of Safety, Health and Environment. O'Malley said, ``This new office and new senior position for Tosco is an essential expansion of our Corporation given our continuing growth and our strong commitment to protect our people, our neighbors, our customers and the environment.''
According to O'Malley, each individual within Tosco is empowered to take the appropriate action to ensure the safety of all and the protection of the environment.

Words From a Survivor
The survivor, victims' families, worker advocates and community activists gathered on February 23, 2000 for a Memorial event on the one year anniversary of the fatal San Francisco area Tosco fire that claimed the lives of four workers. Steve Duncan, the sole survivor of the Tosco Refinery fire on February 23, 1999, stated, ``Tosco has paid millions of dollars to agencies in the last few weeks to make it look like the slate has been wiped clean, but they still refuse to compensate the people who were most affected by this disaster.'' Duncan states: ``The truth is Tosco has a lot more work to do to do right by us and clean up their act.''

PSM-W3 (What Went Wrong?)

If the allegations brought in the suit against the Avon refinery are correct, several questions come to mind.
  1. Was there a PHA conducted of the maintenance job that was to be performed?
  2. Was proper training conducted for all concerned?
  3. Were Supervisors and Workers aware of their rights and responsibilities?
  4. Were Supervisors and Workers empowered by the company to stop work until all legitimate safety and engineering concerns were addressed?

PSM and PHA Fundamentals

As a vital part of Process Safety Management, Process Hazard Analysis must take place for any assignment where there is a reasonable likelihood that a hazard does or could exist. The PHA must address any reasonably foreseeable event, act, situation that could result from the action of carrying out the assignment.

In the Avon case, accepting the allegations as true, it would appear that some members of the work force saw a hazard in the assignment but were coerced to proceed. If true, this would be a violation of the most basic building block of PSM-that the employee doing the job often has the most detailed insight into the circumstances and possible outcome.

Experience tells us that before any work is to be performed on any part of a system, the system must be properly prepared before proceeding. Years of accident analysis have taught us: De-Energize, Lock-Out/Tag-Out, Drain, Blank/Blind/Disconnect, Atmospheric Testing and a host of other measures must take place prior to proceeding.

PHA check-off/Sign-Off lists are an excellent tool to assure the proper procedures were taken. The fact that the system contained highly flammable Naphtha should have raised instant red flags.

What Would Deming Say?

Training is another vital building block to an effective PSM program. Although it appears clear that some training had taken place, the ADL study reports that the worker looked on the program with great skepticism. Although the level and quality of training is not stated, it may not have been adequate.

As W. Edwards Deming said repeatedly in his landmark works on quality systems: there must be top down support and enthusiasm for any project or initiative to succeed. Deming would certainly include worker safety.

The fact that contractors were conducting some of the work raises another red flag. Companies often have difficulties assessing the level of training of contract employees. The company must have documented proof of the contractor's training prior to commencing work, or conduct the same training for all employees and contractors alike. This is a major stumbling block in all CPIs.

Was Empowerment in Place?

OSHA states that no "employee" (meaning worker, and including "employee worker" and "contractor worker") can be coerced into a job that that "employee" deems as unsafe. An "employee" is never allowed to put themselves or others, property, or the environment in jeopardy for any reason. Were the workers aware of their empowerment to stop work on a job and seek guidence when they considered that job to be unsafe?


Union Executive Reacts to Energy Secretary's Comments

As reported in PRNewswire from Nashville, Tennessee, on February 17th, Executive Vice President Robert E. Wages of PACE International Union (Paper, Allied-Industrial, Chemical and Energy Workers) reacted angrily to February 16th statements by U.S. Energy Secretary Bill Richardson. Richardson had urged U.S. oil refiners to delay scheduled maintenance work in order to boost production.

Mr. Wages said ``delays in scheduled maintenance in an industry that already stretches its maintenance schedule to the outer limit poses an imminent danger to all involved.'' Robert Wages heads the PACE national oil bargaining program and handles the union's day-to-day relations with oil industry employers.

Maintenence Delays Blamed for Other Refinery Incidents

Richardson's request, made in a statement on February 16th, was announced as a move to help keep oil prices down by boosting production. ``This is indefensible,'' said Wages. ``The increased risk to the health, safety and environment of our members, other refinery workers and PACE communities is unacceptable to us.'' Wages cited two other serious refinery accidents in recent years, one at a Unocal facility in Lemont, Ill., and another at a Shell refinery in Norco, La., both of which involved the employer's failure to perform scheduled maintenance. The February 1999 accident at a Tosco Corp. refinery in Avon, Calif., involved maintenance performed on high temperature and pressure units while they were operating to minimize production losses.

``Delayed maintenance at oil refineries is not something the government needs to be supporting,'' said Wages. ``The price in death, destruction, pollution and lost production when one of these refineries explodes because maintenance wasn't done more than cancels out any possible marginal benefit to the consumer gained by delaying scheduled maintenance.

``Secretary Richardson should disavow this action immediately, and this union has asked him formally to do so,'' added Wages.

...and it still happens...


An explosion and fire at CF Industries' Donaldsonville, La., fertilizer plant killed one worker and injured 10 others. Five of the injured workers and the one killed were contractors employed by Catalyst Process Specialists Inc.; another injured worker was with Turner Industries Services. Both companies had employees at the site conducting routine maintenance. The other four injured were CF Industries employees.

The company says the cause of the explosion in its ammonia No. 3 unit is still being investigated.

The site was expanded in 1998 and, according to the company, is the largest and most modern nitrogen fertilizer facility in the U.S., with an annual capacity of 4 million tons. CF Industries reports that, based on the plant's safety record, the National Safety Council ranked it first in the U.S. and third in the world among similar facilities in 1999. ---------

An explosion at a hydroxylamine plant in Gunma Prefecture in Japan destroyed the facility on June 10, 2000. The blast killed four people and injured 28.

The plant was operated by Nisshin Chemical, a small producer of purified hydroxylamine. The company's president is reported to have said that purified hydroxylamine, which has an explosive power similar to TNT, had exploded. The material, which is used as a cleaning agent in the manufacturing of semiconductors, becomes unstable when heated.

However, heating is one of the steps in the distillation of unrefined hydroxylamine. A smaller blast reportedly took place at the same facility in November 1998.

(See the report that follows on the Concept Sciences explosion in Pennsylvania in February 1999.) Japanese police are reportedly investigating criminal negligence by Nisshin. The blasts at Nisshin and Concept Sciences have made BASF the only remaining supplier of purified hydroxylamine. ---------


The federal Occupational Safety and Health Administration has cited Concept Sciences, a chemical manufacturing company near Allentown, Pa., for safety violations that allegedly led to a catastrophic explosion in February, 1999. OSHA proposed penalties of $641,200. The explosion killed five workers, including a father and son, and injured two others.

OSHA Findings

OSHA cited Concept Sciences, Inc. (CSI), a specialty chemical manufacturer, for 20 alleged violations, including 11 willful violations for failure to protect employees from the explosive potential of hazardous chemicals. The explosion occurred Feb. 19 at CSI's manufacturing facility in Hanover Township, Pa. Four of the workers killed were employed by CSI; one worker from an adjacent business was also killed in the explosion.

"This is precisely why OSHA established standards to prevent catastrophic incidents involving hazardous chemicals," Secretary of Labor Alexis M. Herman said. "CSI management was clearly aware of the requirements of those standards, but failed to take adequate safety measures prior to producing a chemical known throughout the industry as potentially explosive."

"What's most troubling," added Charles N. Jeffress, OSHA administrator, "is that CSI management did not provide pertinent information to their employees on the hazards involved in the production process, or the explosive nature of the chemical."

Runaway Reaction Of A Scaled-Up Process

CSI employed 21 workers at two locations in the Allentown area. Employees were involved in the company's first production run of a concentrated solution of hydroxylamine in water.

The solution has unique properties that make it useful in a variety of industries, as described later.

OSHA's inspection revealed that the explosion occurred at a 2,500-gallon fiberglass reinforced charge tank containing approximately 750 pounds of hydroxylamine. The tank was being used in a distillation process. Workers at the plant were producing 50 weight% hydroxylamine distillate (with a proprietary stabilizer) package using the following process.

Solid hydroxylamine sulfate was neutralized with potassium hydroxide, filtered to remove potassium sulfate precipitate, and then vacuum distilled at 122degF. Pure hydroxylamine has explosive energy roughly equivalent to that of TNT. The building was completely destroyed by the explosion and has since been demolished.

OSHA Violations Cited

Of the 20 violations cited, 11 were alleged willful violations of OSHA's Process Safety Management (PSM) standard, as well as provisions of the Hazard Communication standard. The PSM standard establishes requirements to prevent, or minimize, the potential for fire or explosion caused by dangerous chemicals, while hazard communication addresses the potential hazards of chemicals and establishes procedures to communicate those hazards to employees.

The 11 alleged, willful violations, with a total proposed penalty of $616,000, are composed of various groupings of individual requirements of both standards. They include:
*failure to compile process safety information;
*inadequate process hazard analysis and operating procedures;
*failure to train employees on operating procedures and the physical hazards of chemicals;
*lack of a pre-startup safety review;
*process equipment deficiencies; and
*failure to develop mechanical integrity procedures.

Provisions of the hazard communication standard cited as alleged willful violations related to deficiencies in both employee training and material safety data sheets(MSDS).

OSHA also issued nine alleged serious violations to CSI, totaling $25,200, citing:
*the lack of employee participation in a PSM program,
*failure to adopt safer work practices,
*no injury and illness logs for contract employees,
*inadequate mechanical maintenance training,
*deficiencies in chemical hazard evaluation procedures, and
*improper labeling of chemical containers.

OSHA defines "Willful violations" as those committed with an intentional disregard of, or plain indifference to, the requirements of the Occupational Safety and Health Act and OSHA regulations. A "serious violation" is defined as one in which there is a substantial probability that death or serious physical harm could result, and the employer knew or should have known of the hazard.

Frequency and Severity of Chemical Incidents

The Chemical Safety & Hazard Investigation Board reports that, on average, there are roughly 60,000 "commercial - chemical" incidents per year. Nearly half of the incidents are involved in transportation. At plant sites there have been about 30 deaths per year and about 1,000 injuries. Gasoline was reported as the "chemical" involved in most incidents. An average of about 250 people die in these accidents each year.

Properties of Hydroxylamine

Concept Sciences had been purifying and concentrating free-base aqueous hydroxylamine solutions at 50% and 30% concentrations. "SAX's Dangerous Properties of Industrial Materials, Ninth Edition, 1998" (by Lewis, Wiley Environmental Sciences), has characterized hydroxylamine. It is a "colorless or white, thermally unstable, hygroscopic material occurring as liquid needles. Decomposition occurs rapidly at room temperature. Melting point: 93deg.F, boiling point: 230deg.F, flash point: explodes at 265deg.F. Thermal decomposition in air results in fumes of oxides of nitrogen."

The Aftermath and an Estimate of the Explosive Equivalency

A four-foot deep by sixteen to eighteen foot diameter crater remains where the plant's chemical distillation equipment once stood. An investigator for the Ames Laboratory division of the Los Alamos National Laboratory has calculated that it appears theoretically possible that a hydroxylamine explosion could resemble that of the same quantity of TNT. Determining the cause of the explosion is expected to be a lengthy process of many months.

According to information from Concept Sciences, their 50% free-base hydroxylamine aqueous product, 50-HA (TM), avoids problems introduced by use of hydroxylamine salts, making their product an ideal raw material for the production of aldoximes and ketoximes.

Some of the uses of their hydroxylamine, either as a raw material for the production of product compounds or directly because of its chemical properties, can be important to a wide spectrum of chemical and allied industries according to Concept Sciences.

A Specialty Chemical with Wide Applications

Pharmaceutical Industry - Synthesis of antibiotics (cephalosporins), sulfa drugs, tranquilizers (clorodiazepoxide hydrochloride) analgesics (acetaminephine) spasmolytics, and antitubercular drugs.
  Agricultural Chemicals - Hydroxylamine is a precursor in the production of insecticides, herbicides and plant growth regulators. It is also a raw material for the production of oxime carbamate insecticides and herbicides.
  Specialty Strippers for Integrated Circuit Fabrication - 50% aqueous free-base hydroxylamine is employed as a critical component in specialty formulated strippers used for the removal of photoresist and etch residues in the manufacture of integrated circuits.
  Plastics Industry - This represents the overall largest use for hydroxylamine which is generated in-situ from one of the hydroxylammonium salts (i.e. ; sulfate, chloride, nitrate, etc.). More than 90% of hydroxylamine used globally is for the production of cyclohexanone oxime (i.e.; the direct precursor to e-caprolactum which is required for the manufacture of Nylon-6); also used as a "regulator" and inhibitor for various other polymerizations.
  Photography - Used as a stabilizer for developers and as an additive for color emulsions.
  Paints and Coatings - Used for the manufacture of specialty oximes which are then employed as components in anti-creaming agents.
  Rubber Industry - Used as a regulator in the polymerization of synthetic rubber and as an anti-oxidant for natural rubber. It is also used as an "accelerator" in the vulcanization of synthetic rubber.

How Much Risk Management Planning Was There?

There have been questions as to rules and regulations covering the hydroxylamine at the plant with regard to the surrounding community. Hydroxylamine is listed in 29CFR1910.119 (PSM) Appendix A "List of Highly Hasardous Chemicals, Toxics, and Reactives", with a "Threshold Quantity for Coverage Under the Standard" of 2,500 pounds. The charge tank in which the explosion occurred was estimated to contain 750 pounds of hydroxylamine. The total amount present in the at the time of the explosion is uncertain.


OSHA and EPA lists are used by companies to determine their need for a Risk Management Plan (RMP). RMPs are studies, required by law to determine what risk stored or process chemicals pose to the surrounding community in the event of an unintentional release. Hydroxylamine is also on the EPAs Toxic Substances Control Act(TSCA) list and is classified as a mutagen. Materials on the TSCA list are regulated in terms of storage, handling and disposal.

It must be understood that hydroxylamine has not yet been determined to be the cause of this tragic incident. The U.S. Chemical Safety & Hazard Investigation Board (CSB)is continuing their full-scale investigation of the factors contributing to the blast. ---------

Herbicide Exposure Contributes to Fatalities

On October 12, 1998 there was a fatal exposure to herbicide components in a Michigan factory. The incident occurred at the Dow Chemical AgroSciences plant in Midland, Mich. The worker's hands and feet were exposed to 2,4-dichlorophenol and were undergoing decontamination when he suffered a fatal cardiac arrest. ---------

Six Die in Explosion During Recovery from Power Outage

An explosion occurred at about 1:30 p.m. Pacific Standard Time Wednesday (11/25/98) at the Equilon Puget Sound Refining Co. in Anacortes, Wash.

Company officials report that at the time of the explosion the plant was recovering from a total electrical power failure which began Monday evening as a result of recent wind storms. They say the explosion and fire occurred in the delayed coker unit, which is part of the refining process.

According to Equilon, of the six men killed, four were employees of a contractor, Western Plant Services, Inc. and two worked for Equilon.

Equilon is reported to have been formed in January 1998 as a joint venture of Texaco's and Shell's regional refining and marketing businesses. ---------

Seven Die in Natural Gas Well Explosion

The CSB is investigating the Bryceland, La. site of a natural gas well explosion which claimed the lives of seven workers. Bryceland is about 45 miles east of Shreveport.

The explosion occurred at about 2 p.m. Saturday October 24, 1998 at a well owned by Sonat Exploration Company. The CSB already has an investigation underway of a March 4, 1998 explosion at a Sonat oil separation facility, in Pitkin, La., which killed four men working for the company.

All seven of the most recent fatalities worked for two subcontractors employed by Sonat. In the March explosion three of the four fatalities were employees of a subcontractor; the fourth worked for Sonat. ---------

35 Injured in Processing Tank Incident

An explosion and fire ripped through a Point Comfort, Texas chemical plant on October 12, 1998. The incident at the Formosa Plastics Corporation injured at least 35 employees and contract workers and at least two members of the public.

  Reports from the company indicate that the incident occurred when a processing tank exploded for unknown reasons. The force of the blast sent a shock wave that blew out the windows of a maintenance building 300 yards away. Most of the injuries were reportedly to workers who were struck by flying debris, and none were considered life-threatening. ---------


Nitrogen is odorless. Natural gas is almost odorless. Most people are familiar with the odor of the mercaptans added universally to consumer natural gas so that concentrations of natural gas can be detected by the disagreeable mercaptan odor. That odor has proven very effective in warning of natural gas leaks.

OSHA reports that in the 1990s there have been three to four deaths per year caused by asphyxiation due to nitrogen. Individuals have tried briefly entering nitrogen atmospheres while holding their breath. One slip-up, and a breath or two in that atmosphere can mean death. Just don't try it!

A March 1998 asphyxiation due to a worker entering a pipeline blanketed with nitrogen has prompted the CSB to request that NIOSH investigate the feasibility of adding an odorant to nitrogen. NIOSH has commented that such a study will be involved and will take awhile.

Meanwhile, all workers must be properly trained about the hazards of entering confined spaces, and must be trained in the importance of sampling for a safe breathing oxygen level, and must know the use of PPE.

More recently: Fatal Accident at Valero Refinery, Delaware City, DE, November 5, 2005

Two contract employees were overcome and fatally injured by nitrogen as they performed maintenance work near a 24-inch opening on the top of a reactor. One of the workers died attempting rescue.

For expert assistance in training your workers, click on TRAINING HELP.

For a fee-based consultation call with a Licensed Professional Engineer between 3:00pm and 5:00pm Eastern Daylight Time, call (302) 239-2700.

For information at any time on our Hazardous Waste Operations and Emergency Response (HAZWOPER) and other safety training, click here.