Podcast: Lessons Learned From The 2013 Williams Olefins Explosion

Welcome to Process Safety with Trish and Traci, the podcast that aims to share insights from past incidents to help avoid future events. This podcast and its transcript can be found at chemicalprocessing.com. I'm Traci Purdum, editor in chief of Chemical Processing, and as always, I'm joined by Trish Kerin, the director of the IChemE Safety Centre. Hey Trish, how are you?

Today we're going to talk about the Williams Olefins Plant explosion, which occurred on June 13th, 2013. So we're coming up on the 10-year anniversary. It's a petrochemical plant located near Baton Rouge, Louisiana. During that incident, two workers were killed and numerous others, 167 others were injured. The effects of the explosion were felt several miles away, a shelter in place order was issued to residences and businesses within a two-mile radius of the plant. The US Occupational Safety and Health Administration, and the US Chemical Safety Board launched in investigations. Can you give us some insight into the incident itself and the final determined cause?

Trish: Yeah, so this particular incident was a heat exchanger where they had inadvertently blocked in its pressure relief system when the heat exchanger was brought into service. Now, when they actually did bring the heat exchanger into service, they did so by putting the hot fluid through the heat exchanger without realizing that over many years of the heat exchanger sitting there out of service, propylene had built up inside of the exchanger as well. So they had the perfect situation there once they started heating that propylene for the pressure to rise significantly. And as a result when they did start to heat it up with no pressure relief it did cause a catastrophic failure and then subsequent explosion and fire.

Traci: I was reading about this and I read that there were two shell and two heat exchangers, and they were originally designed to be used concurrently. A modification was performed in 2001 to allow each heat exchanger to be used independently from the other one allowing a filed heat exchanger to be cleaned without shutting the unit down. Is this advisable, and if so, is there a proper way to do this? Did they do it properly?

Trish: So I don't think there's anything wrong with having duty and standby equipment and being able to do crossover and shut down one piece of equipment while you clean it and bring the other one back into service while you're doing that. That's quite a standard acceptable way to operate. The catch is you actually need to make sure that you do that safely. And so that means you've got to have a appropriate isolation so that you are positively isolated from the in-service unit. There's no way that any contamination can flow back in. You also need to make sure that you've got the in-service unit is protected by pressure relief at all times, and that was one of the key failings that happened here. It was not protected by pressure relief because that valve had been locked closed, and so there was no relief way for the pressure to go.

You also need to make sure that when you do bring a standby into service, that you understand the condition that standby is in. So as I mentioned, it had accumulated hydrocarbon in it over a period of years. It wasn't an empty vessel at the time it was brought into service. And so there's a couple of different steps that you've got to make sure you cover. Make sure you understand the condition the equipment's in, make sure you have positive isolation from the operating plant and make sure that you have adequate pressure protection on every piece of equipment that's actually functioning.

Traci: This facility was fined for several violations. I'm going to list a few of these and I want to get your thoughts a little bit on this. One of the violations inadvertently mixing hot quench water with propylene. Another one, failing to provide appropriate pressure protection for a pressure vessel. Failing to complete a process hazard analysis to address the opening of hot quench water flow into a pressure vessel. Failing to properly document workplace training. Failing to promptly correct deficiencies related to process safety management discovered by an internal compliance audit team and the willful violation develop clear written procedures for how to change and put idle pressure vessels into service.

You and I have talked about many of these issues and these are the basis of making sure that you have proper safety features in your plant. How does a facility learn from these violations? Do they learn from these violations or is it just a matter of hoping for the best and dealing with the worst when it happens?

Trish: No, I think we have to recognize these. These are an opportunity to prevent a future incident happening rather than just trying to respond to it when it does. So if I could just go back to the idea that an MOC was performed when they first made the change because initially in the design they had two heat exchanges both operating in parallel to run the unit. They then decided through failing issues that they would take one out of service and have it as a standby and use a main one and only switch over when they needed to defoul the main service at that point in time.

Now, they did do a management of change on this change; however they did the management of change after they had already made the change. So straight away you are already on the back foot. You are already trying to justify why it was the right thing to do rather than justify if it is the right thing to do. And they're two very different perspectives to come to a management of change with. The idea of management of change is we do it before the change happens because it is about determining whether it is safe to do it and whether we have all the controls in place. The simple fact that they actually did the change and then did the paperwork afterwards suggests that they weren't perhaps viewing it in the appropriate way. They were looking to just sign it off rather than to really determine whether they had everything in place.

And they did have isolation valves on the pressure relief system. And again, that's not uncommon, but they're meant to be locked open and they're meant to be brought into service when they need to be if you are bringing the vessel into service. So certainly in terms of the written procedure for how to put the vessel into service, there was most definitely gaps in that. There was no pre-startup safety review that was done to make sure that the isolations and de-isolations were all done appropriately, and they certainly did fail to address the hazard adequately.

There'd been a number of PHAs done on that particular unit, and the PHAs had identified, one of them had identified that they needed to ensure they had the pressure relief valve accessible to the in-service vessel, and they did that for the operating vessel, but they didn't do it for the standby vessel. So when they brought the standby vessel in without the adequate procedures, without the pre-startup safety review, they were unable to remember to put the right valve in service and therefore ended up with this over pressure situation.

It's also interesting to note that their strategy for managing fouling was to basically run one vessel, one heat exchanger until it fouled beyond hope and then switch over as opposed to potentially doing other preventative maintenance activities, say switching between the two pressure vessels more frequently. And that may have led to more familiarity with what they were doing and certainly a procedure for something that frequent as well. So I think there's a few different learnings out there for not only for this site, but for a lot of other sites around, because a lot of facilities do have duty and standby. It's a very common way to run a plant, but you need to make sure you've got the right procedures, you've done the right risk assessments, and you understand what's going on.

Traci: Now, you talked a little bit about the MOC afterward. Is that common to do it afterward? Obviously not correct, but are more facilities doing that just to get the paperwork done?

Trish: Look, I don't have any evidence to make a claim, but my intuition suggests that yes, that is probably common. People will do something and then potentially realize afterwards that there was a change and they need to go back and complete the paperwork because it will be found in an audit or it was found in an audit to have not been done. So I'd say it's not uncommon for that to happen, but it's not the intent of the process.

As I said, it changes your mindset once you've done the change, you are actually doing an MOC to justify why the decision was right. When you do the MOC beforehand, the MOC is to justify if the decision is right, not why the decision was right. And they're two very different ways to approach a problem. One of them is it almost predetermines the answer because you're not likely to go back and reverse the change. So you've got to make sure that you do it ahead of time. You have to be ahead of the curve on these things to make sure that you've got the right controls in place before you ever make that change.

Traci: What can other facilities, what can operators at other facilities do to prevent not only this incident but similar administrative things happening like they did with this facility?

Trish: So that's where adequate written procedures for doing specific tasks, particularly if they're infrequent tasks, need to be developed and verified by the people actually doing the work. So we don't just write procedures by engineers sitting in a room together by themselves. It actually needs to be procedures written with the operator, with the maintainer right there, because they're the ones that actually do the work. They know what really happens. Make sure that they're involved in the development of those procedures.

And the use of checklists and checklists are often one of these things that people feel very strongly about. They either love them or they hate them. Checklists are really good ways to focus and remember when specific steps need to be done. But if all you are doing is filling in a checklist and using it as a tick box exercise, you're not going to get value out of it. But if the checklist is actually helping you to remember critical steps in the right order and make sure that you sign off or you check off that you've done those steps, then the use of procedures can actually be very, very positive. And certainly the medical field and the aviation field use procedure checklists extensively to make sure that when they're making critical safety decisions, they've got the right information in front of them and they make the right decisions on the spot. Really important in a high stress situation.

Traci: And sometimes it feels like the redundancy is ridiculous, but the redundancy is necessary.

Trish: Yeah, absolutely. That's why we have the redundancy because things go wrong, equipment fails, people make mistakes. The redundancy helps us mitigate the impacts of those things occurring.

Traci: Trish, is there anything that you'd want to add that we didn't touch on?

Trish: I think it's interesting that the CSB have also just released their latest investigation into the Husky Wisconsin Refinery Fire. And one of the comments or the key leading area they're talking about there is transient operations. And to a certain extent, this incident is also a little bit about transient operations or certainly unusual operations. Switching from one in-service vessel to the backup vessel is actually somewhat of a transient operation that's going on. And we need to be very careful to make sure we adequately risk assess those operations because they're not steady state. And the moment we have a steady state plant doing anything that is not steady state, we increase the hazards and the risks that we face at that point in time.

 So shut down and start up are two of the most hazardous times in a steady state operating plant because that's when things are not in their normal parameters, and that's when we need to be extra vigilant in what we're doing.

Traci: Well, Trish, you are ever vigilant on making sure that we have the correct information or at least information that we can make better choices on to make sure that our operators go home after every shift. And I appreciate that. Unfortunate events happen all over the world, and we will be here to discuss and learn from them. Subscribe to this free podcast so you can stay on top of best practices. You can also visit us at chemicalprocessing.com for more tools and resources aimed at helping you run efficient and safe facilities. On behalf of Trish, I'm Traci, and this is Process Safety with Trish and Traci.

Trish: Stay Safe.