 Industrial catastrophes are not very frequent events, but are more common than many would like and every decade has had a multitude of shocking disasters. The 1980s was no exception and definitely had more than its fair share, like for example the well-known ones in the form of Chernobyl, Piper Alpha and Baupoil to name a few. These are the ones that are so devastating that the ramifications of such events have travelled well beyond the borders of the country of origin, but what I personally find more worrying are the smaller localized events, the ones that are not often reported across the globe, getting forgotten to time. These disasters are the ones caused by seemingly mundane reasons such as incorrect routine maintenance, on the face of minor adjustments to procedures or just a failure to check things properly. An explosion in a chemical facility can easily become part of the white noise background of the daily news cycle and in some regards allows the event to be downplayed. A disaster would rack up a death toll of 23 employees, injuring 314 and the company in charge would get a modest $5 million fine, even though safety on the site was inadequately managed. As such, today's subject will get a rating of 7 on my plainly difficult disaster scale and also a 7 on historical legacy scale as the event helped improve process safety. Welcome to a brief history of the 1989 Philips Petroleum Chemical Plant Explosion and Fire in Pasadena, Texas. Philips Petroleum was a petrochemical company with its history tracing all the way back to 1917 and the forming of the company by brothers Lee Elders Phillips and Frank Phillips based out of Oklahoma. Quickly the company worked in all parts of the oil industry including fuel production, crude oil pipelines, refineries and marketing of petroleum products. Like many companies during this period, the demand for fuel boosted the organization's coffers, but the growth would really speed up during the big old dumpster fire that was the Second World War. By the late 1940s, the ever-growing Philips Petroleum had become active in the chemicals business, creating Philips Chemical Company, which set up and began operating a large plant on the Houston Ship Channel. In the 1950s, the plant began producing polyethylene, plastic, trademarked as Marlex. This was used in blow molded and rotationally molded components. This would be the facility that we will be focusing on today and would be the epicenter for a disaster nearly 40 years after it began operation. By the 1980s, the site was being used to produce polyethylene, which is one of the most common plastics used in bags, food packaging and pretty much anything you use day to day. Making this material can be a dangerous process as highly flammable gases are given off, which would be the cause for the disaster. But first let's look at the process used by Philips Chemical. High-density polyethylene is manufactured at plants 4 and 5 at the Houston Chemical Complex at high temperatures and high pressures. The process uses a thing called a loop reactor, which is an arrangement of 30-inch diameter pipes mounted vertically in 150 feet tall continuous circular shaped structures. This heats and pressurizes the chemical mixture to 600 psi and to a temperature range of 82 to 110 degrees centigrade. Reaction is one of condensed polymerization. Of around 95% ethylene dissolved in isobutane, hydrogen and hexene were also added to achieve targeted product quality specifications. As the reaction product gained mass in the loop reactor, it would eventually become heavy enough to drop out of the circulating reaction mixture. At the bottom of the loop reactor were setting legs 6 in total and 8 feet long, but are used for collecting the polyethylene particles. Each leg had two valves, one at the top and one at the bottom. The upper consisted of an air-operated 8-inch ball. These were used to isolate the section from the reactor after production. The bottom valves were take-off ones that were fed into a flash tank. Once a leg was full or clogged with product, the 8-inch ball valve would be closed. This was to stop the highly flammable gases inside the reactor from venting to atmosphere. The setting legs were the interface between the high pressure reactor and low pressure flash tank process sections. Once the process was complete, the polyethylene fluff was then taken from the tank to be turned into the finished product. But the way the reactor and setting legs were designed to work had a problem and that was everything had to be shut down once all 6 became full or clogged. This necessitated the need to be able to empty the legs during full power production to reduce downtime. But its petroleum had a policy for safe equipment isolation, meaning any maintenance would have backup isolation called double blocking. This can be done by a double valve setup. Obviously, this would only work if the area you want to isolate had two valves. If this wasn't possible, then a bolted flange setup could be used to have two sets of protection. This had, in the case of Loop Reactor, a problem. That installing the blind flange was time-consuming, making maintenance whilst the reactor was running pointless. So as we see very often in these videos, a local solution was devised. This involved the 8-inch ball valve being closed to isolate the plug setting leg from the Loop Reactor. Then the 8-inch ball valve stem was physically locked in its closed position. Finally, the inlet and outlet air hoses actuating the 8-inch ball valves from the remote valve switch control panel. This was double locked but definitely not double blocked as any failure of the ball valve could mean the reactor was exposed. After this locking was complete, the setting leg could be emptied by removing the release valve and having a contractor reach their arm up into the leg to put out the fluff. This leads us to the 22nd of October 1989. Three of the six setting legs in Reactor 6 in Plant 5 were plugged. As had been done before during production, operations personnel isolated the setting legs according to the alternative isolation procedure. Closing the 8-inch ball valve, locking them and disconnecting the air hoses. The first was disassembled and had its fluff removed relatively easily by the contractor. This happened whilst the reactor was still in operation. The crew wouldn't tackle the second leg until the next day when the contractor attempted to remove the blocked fluff. Only three quarters were set free, leaving the leg partially blocked. The morning hadn't gone to plan but the contractor attempted to get out the remaining material around lunchtime. At some point, the air hoses to the valve had been reconnected and lockout device was not properly installed. Somehow, the valve opened releasing the highly flammable gases out of the reactor into the air. At around 1pm, the reactor lost its entire contents of around £85,000 of the heavily flammable process materials after which the escaping chemicals then ignited. The explosion created a force equivalent to 2.4 tonnes of TNT, registering as a 3.5 magnitude earthquake on the Richter scale. Several other explosions were felt as chemicals stored on site also ignited. 23 people were killed and 314 were injured. The fire raged at the 14 storey high site. First response was provided by Philips Petroleum Company's contractor Fire Brigade, which was soon joined by members of the Channel Industries Mutual Aid Association. Calling water from the fire apparatus was trained on the area of the fires, but the intensity and the fuel load was such that the responders were only able to contain the fire. The efforts were also hindered by low water pressure caused by pipe rupturing during the explosion. Eventually control over the fire was achieved by approximately 1900 hours, although smaller blazes continued throughout the night. Rescue and recovery of the victims from the explosion was severely hindered during the fire and even afterwards due to fears of structural collapse. A 1 mile no fly zone was set up around the site to prevent anything being disturbed by aircraft downwash. With a heavily damaged site reaching a cost to be reported over $700 million, a question is left. How? Well, investigators quickly found that the valve had been improperly locked out, but a lot of guesswork is left as to why. This is because the operators working at the reactor were killed in the explosion. It shouldn't have really happened as a process for locking the valve was very well known and established on site. Investigators also discovered that the valve was capable of being locked closed as well as worryingly open, which could have led to confusion. Coupled with the operators not removing the airlines could have been the root cause. Regardless the procedure should never have been allowed to happen without a blind flange, which would have been the form of double protection. And this was reflected in the investigation into the disaster, which found that Philips petroleum had a lack of process hazard analysis employing inadequate standard operating procedures. The legs came into question for not having a failsafe block valve and an inadequate lockout tag procedure in the form of double locked but not double blocked procedure. Furthermore, there was an onsite lack of combustible gas detection and alarm system, which didn't give workers warning of an impending incident. Also it was found that the fire wasn't helped by the presence of ignition sources and inadequate ventilation systems for nearby buildings. Philips was fined $4 million for the infractions that led to the explosion, as well as improving working practices across the company and its sister enterprises. The ramifications from the event would spread widely across the whole industry, helping improve safety although things are never as safe as they should be. Check out my Twitter for all sorts of photos and odds and sods, as well as hints of future videos. I've got Patreon and YouTube membership as well, so check them out if you fancy supporting the channel financially. And all that's left to say is thank you for watching.