Many chemicals safely transported and used, can turn dangerous if their engineered controls are exhausted or ineffective. Because these chemicals are not often seen without these controls, their true hazardous nature can be forgotten.
Emergency Response Specialists, Inc.. (ERS) responds across North America to problem chemicals and accidents. In 1997, ERS was tasked with disposal of an older cylinder of uninhibited 1,3-butadiene. If handled with an inhibitor under specific guidelines it is safe. However, outside of these specifications, explosive peroxides and pyrophoric "popcorn" polymers can form. The cylinder's contents not only dated back to 1984, but likely had been exposed to heat, rust and oxygen. These are conditions which increase peroxide and "popcorn" formation. Given the probability of considerable danger, the company asked ERS to devise a safe method to handle the 48- by 14-inch cylinder. ERS's team extensively researched the reactions of butadiene under the conditions present.
THE EXPLOSIVE CAPABILITY OF 1,3-BUTADIENE
Butadiene peroxide was the greatest concern. Concentrated butadiene peroxide can explode more powerfully than nitroglycerin or TNT. Older material is also up to 80 times more sensitive to detonation from shock energy than nitroglycerin. A shock force of only one pound dropped from one inch was estimated to give a detonation probability of 50 percent.
To assess the extent of the hazard, the total product volume was needed. Not surprisingly, no one volunteered to weigh it. Instead, ultrasound and x-ray analysis were used so the cylinder would not be disturbed. The ultra-sound determined the liquid height and density. X-rays were taken to visualize gross popcorn polymer around valving, which could block valves or provide seed for explosive polymerization during handling. The NDT analysis revealed about 75 pounds of liquid and clear valve tubes.
Given the large volume of extremely unstable material, two options were presented; treat or detonate the cylinder in place with shielding of surrounding structures, or move it to an open field, over a mile of rough gravel road.
Controlled detonation in place was considered due to the peroxide's extreme sensitivity to shock and heat. Also, unlike other organic peroxides formed during storage, it's minimally soluble in, and denser than its original product. So instead of remaining dispersed, it concentrates at the bottom. It can also self detonate if a sufficient mass radius is reached. For a "slab", the critical radius is about five centimeter at 27o Celsius.
Since only a block wall separated the cylinder from $34,000,000 of equipment and neighborhoods were close by, the client opted to move and treat the cylinder. ERS's team was faced with two serious problems by this decision, how to safeguard personnel and how to control the blast if it detonated in transit.
First, Response Directors Moore and Haywood, with team members Riley, Staggs, Kirby, Briggs and Keck, designed and built a remote system. This system can selectively operate numerous valves, add chemicals, purge, flare and monitor reactions. Second, for blast protection during transport, ERS had the largest mobile blast suppression vessel in the US, designed and built in 10 days. It stands 10 feet tall and eight feet in diameter with a six foot inner chamber, all of 2-inch thick cold rolled steel. Sand fills all voids. A blast mat woven of 5 /8 -inch cable and with 16 attachments effectively minimized projectile hazards. The vessel is rated for 100 pounds of high explosive.
HOW THE JOB WAS RUN
Nearly the entire operation was to be performed remotely except for moving the cylinder 100 feet from the cold room to the blast vessel outside. This would be the most intense phase. Since the cylinder would be exposed briefly to dangerous summertime heat, it would be packed in ice once outside.
Warning of a reaction, potentially leading to explosion, would be provided by a thermocouple attached to the cylinder. Besides explosion from shock, it was possible from either a thermal decomposition of the peroxide or rapid "popcorn" polymerization. Both events would be signaled by heat.
Because any mistake or delay could be disastrous, a dry run, then practice run with an old, but inhibited, cylinder of butadiene was performed to refine each step.
After planning, building and practicing, we were ready to tackle the job. The plant's team provided invaluable support, including securing the area, analysis and public coordination.
Operations began by remotely lifting the cylinder with tripods, ropes and pulleys. As it was lifted, a drum dolly rolled precisely underneath. Once it was lowered onto the dolly, the temperature was checked via closed circuit TV. We entered and very gently rolled it outside.
Remote controls and hoses were quickly attached to the cylinder valves. As the lifting straps were secured, the cylinder's temperature began rising. With time short, the core team lifted it with a boom truck into the drum. The temperature rise was now beyond that accountable by ambient heat. We moved quickly to pack it in ice and stabilize the internal reactions.
The ice packed cylinder was then lifted into the blast vessel with practiced care and secured. Cameras were used to guide it into the vessel's core with bumping. Four vessel-mounted cameras showed the cylinder valves, scale and thermocouple readouts. The scale would monitor volumes added and removed.
After the blast mat was secured, the low boy began its two m.p.h. journey down the gravel road. Two connections were made between the cylinder's hoses and the staged treatment system. From a bunker 250 feet away, Haywood, Moore and Keck could now open and close each valve, add or remove cold or hot water, caustic, solvent, nitrogen and remove product as vapor or liquid. The system also allowed remote sampling, pressure readings and flaring. Monitors in the bunker displayed views from five operation cameras.
The first hurdle was the remote opening of the cylinder's vapor line to monitor pressure. Next, the liquid valve was opened and a sample taken 200 feet away to verify the contents as uninhibited butadiene. This was confirmed by the plant's team.
The remote flaring of the liquid followed. To ensure the flame didn't flashback through the oxygen carrying peroxide, mechanical and water checks were used. Air was added to the flame for a smokeless burn. The flame brightened intensely, periodically confirming peroxides. Hot water added to the vessel's core melted the exterior ice as the flare proceeded. The ice stabilized the peroxide but kept the pressure too low for flaring. As it melted the temperature again began a rapid rise above ambient. The hot water was stopped and cold water added back. Nitrogen was then used to gently raise the cylinder's pressure to continue.
The burn took about four hours through 1 /4 -inch lines. The cylinder was then filled with water, flushed and filled again to prevent any pyrophoric polymer from igniting. With the main hazard gone, removing the polymer would be saved for the next day.
Polymer destruction was also done remotely since peroxides could be entrained in it. Popcorn polymer is insoluble and difficult to remove from surfaces. We chose to destroy it by a controlled burn of the cylinder's interior. This was done by drilling a two inch hole and lowering a flare inside. The water level was slowly dropped so that the popcorn polymer's burn would be controlled.
After interior was fully exposed to flame, the cylinder was finally safe. It was removed from the vessel and the interior steamed. Upon inspection, only a pea size amount of popcorn polymer remained inside a rusty, but clean cylinder. What had threatened to destroy a facility had safely gone through extensive team work and innovation.
?Emergency Response Specialists, Inc. has facilities in Birmingham, Houston and Chicago. If a situation arises at your facility where these capabilities are needed or for more information call 1-800-647-4377.