The Storm Reached the Chlorine

▶️ Prefer to watch? Every story’s a short video → https://www.youtube.com/@GetSafetyIntel · 📣 Got a near-miss? Share it anonymously (QR + link at the bottom).

Six years ago this morning, a hurricane peeled the roof off a Louisiana chemical plant and let the rain in — and the rain is what nearly killed a town. Over a million pounds of pool chemical, a leaky building, and a storm everyone saw coming for days. Today: how “just add water” became a chlorine cloud over Interstate 10, and two things on safety leaders’ desks right now. Five minutes. Let’s go.

At about 7:48 in the morning on August 27, 2020, Hurricane Laura came ashore near Cameron, Louisiana as a Category 4 storm — 150 mph winds — and drove straight up over Lake Charles. At the Bio-Lab plant in Westlake, the wind did something simple and catastrophic: it tore the roofs off the buildings where the company stored its product.

That product was trichloroisocyanuric acid — TCCA, the stuff that makes pool chlorine tablets. On any normal day it’s stable in a dry building. Drop it in a swimming pool and it dissolves slowly and releases chlorine into the water, exactly like it’s supposed to. But there’s a trap built into the chemistry: if TCCA gets a little water — rainwater, humidity, a wet floor — instead of dissolving, it reacts. That reaction makes heat. The heat drives it to decompose. And what comes off is chlorine gas.

On the morning of the 27th there were over one million pounds of TCCA on that site. The wind opened the roofs, the rain got in, and the reaction started. It generated enough heat to start a fire, and a large plume of hazardous gas — including toxic chlorine — lifted off the plant and drifted across the parish. Officials shut down a stretch of Interstate 10 for more than 28 hours and issued a shelter-in-place order for everyone downwind. In the middle of a hurricane evacuation, people who’d stayed were now told to seal their windows against a chemical cloud.

Here’s the part that should stop you: none of this was a surprise. Three years earlier, during Hurricane Harvey, the Arkema plant in Crosby, Texas flooded and its organic peroxides cooked off and burned — same script, different chemical, and the U.S. Chemical Safety Board investigated it and published guidance on preparing chemical sites for extreme weather. Bio-Lab didn’t implement it. The CSB’s final report on Bio-Lab (released April 24, 2023) found the company had even done a 2010 hazard analysis that flagged the question of whether the buildings and roofs could take hurricane winds — and never acted on it. The roof was the whole ballgame, and the answer was already written down.

Name it and cite it, and here’s where it gets uncomfortable. The rule you’d reach for is 29 CFR 1910.119 — OSHA’s Process Safety Management standard — which forces a facility running highly hazardous chemistry to analyze its hazards, keep its buildings and equipment sound (mechanical integrity), and manage change before conditions shift. Its environmental twin is EPA’s Risk Management Program under the Clean Air Act (40 CFR 68), written to stop exactly this kind of offsite chemical release. But TCCA is a reactive hazard, and reactive chemicals sit in a gap: TCCA is covered by neither PSM nor RMP. So a plant holding a million pounds of a chemical that turns rainwater into chlorine gas wasn’t required to run the baseline process-safety program that might have hardened that roof. The CSB has been asking OSHA and EPA to close that reactive-chemical gap for years — and reiterated it again in this report.

Translate it to the floor: your chemical doesn’t have to be on a regulatory list to be able to kill the neighborhood. The people closest to that warehouse knew water and oxidizer don’t mix — the hazard wasn’t secret. What was missing was a system that treated “the roof might not hold in a storm” as a process-safety problem worth money and a deadline, instead of a note in a 2010 file. HOP says it plain: error is normal and blame fixes nothing — but the flip side is that a known hazard left unfixed for a decade isn’t an error, it’s a decision. How leadership answered that 2010 question is the whole story.

Most of what we cover is the “S,” but this one is Environmental, Health, and Safety in a single event — you can’t pull them apart. (E) This was a textbook offsite release: a chlorine plume left the fenceline, closed a federal interstate for over a day, and forced a whole community to shelter in place. The Clean Air Act’s Risk Management Program (RMP) exists precisely to prevent chemical clouds from crossing into where people live — and the reactive-hazard gap is exactly why this one wasn’t on that program’s radar. (H) Chlorine is a respiratory poison: even low concentrations burn the eyes and throat and trigger coughing; higher doses flood the lungs (pulmonary edema) and can kill. The plume put responders and residents downwind at risk on top of a hurricane. Honest and narrow, because facts-only: the CSB did not attribute deaths to the Bio-Lab release itself — the danger here was the exposure and the near-miss on a massive scale, not a body count. (S) And the safety failure was the environmental release — the same unhardened roof that let the process run away is what put the town under a cloud. One weak point, three hazards. Fix the roof and the storage, and you answer all three at once.

1. The CSB just closed the book on a coke-plant explosion — and the cause was where people were standing. On August 10, 2026, the Chemical Safety Board issued its final report on the August 11, 2025 coke oven gas explosion at U.S. Steel’s Clairton Coke Works outside Pittsburgh — the largest coke plant in the Western Hemisphere. Two workers were killed, eleven were injured, and the blast did an estimated $52.5 million in damage during a maintenance operation. The CSB’s findings: an ad hoc, unapproved maintenance procedure, personnel buildings sited too close to the hazard, and a process-safety management system that wasn’t doing its job. OSHA proposed $118,214 for seven serious violations. So what for safety leaders: two of those three causes — informal procedures and facility siting — are things you can walk out and check today. Where do your people take breaks, fill out paperwork, and wait out a job? If the answer is “right next to the thing that can let go,” that’s not bad luck waiting to happen, it’s a decision you can still un-make.

2. OSHA is waving the flag on storm season — read it as a direct callback to today’s story. In its August 20, 2026 QuickTakes, OSHA reminded employers that September is both peak hurricane season and National Preparedness Month, and asked a blunt question: does your workplace actually have an emergency evacuation plan? The rule behind it is 29 CFR 1910.38 — the Emergency Action Plan standard — which wants activation conditions, a chain of command, routes and exits, a way to account for everyone on site, and a plan you’ve actually walked with your crew. So what for safety leaders: Bio-Lab is what “we’ll deal with the storm when it comes” looks like six years later. The forecast gave days of warning. The plan is the thing you build now, in calm weather — not the morning the roof comes off.

Shared blamelessly, because that’s the only way anyone learns from it.

A distribution warehouse stocks pool and spa chemicals for the season — pallets of chlorine tablets and shock, shrink-wrapped and stacked. A slow roof leak from an overnight storm has left a dark patch on the concrete in the back corner, and a couple of totes of oxidizer are sitting in it. It’s the busy stretch before a holiday weekend, everybody’s picking orders, and the wet floor reads as a housekeeping problem — somebody’ll mop it. The newest picker on the crew, closest to that corner, notices something the veterans have walked past twice: a faint bite in the air near the wet pallets, and the shrink-wrap on one tote fogged with condensation from the inside. He says “hey, that’s warm, and it smells like a pool in here.” They move the wet totes outside, open the bay, and call it in. Turned out one tote had taken on enough water to start warming. Caught it before it cooked.

Here’s the HOP read, and it matters: nobody was careless. A wet floor is a housekeeping problem a hundred times out of a hundred and one — context drives behavior, and the veterans’ brains filed it correctly based on every other day. The defense that worked wasn’t a rule on the wall; it was a new guy close enough to the work to trust his nose and feel safe saying “this is off” — and a crew that moved instead of finishing the pick first. The lesson isn’t “be more careful with the mop.” It’s that oxidizer plus water is a chemistry problem wearing a housekeeping costume, and the person closest to it saw through the disguise.

Got a fail or a near-miss? Hit reply, or scan the QR below — we’ll feature it anonymously — no names, no company, no blame. Just the lesson, so the next crew doesn’t learn it the hard way.

Before the next storm in the forecast — or today, if you store anything that reacts with water — walk your worst-weather scenario out loud with the crew. Ask the two questions Bio-Lab’s file answered wrong: if the roof failed and rain got in, what do we have that would react? And who decides when we move it, shut it down, or get out — and by what trigger? Then check the boring stuff that actually saves you: is the roof over your oxidizers sound, are the drains clear, is there a real Emergency Action Plan your people have walked, not just a binder. A hurricane gives you days of warning. The only question is whether anybody spent them.

🎬 Rather watch than read? Every Safety Intel story is a short video on YouTube → https://www.youtube.com/@GetSafetyIntel — subscribe so you never miss one. 📸 Share YOUR near-miss (100% anonymous): https://scantoconfess.com — or just scan the QR above. 📤 Forward this to a crew member who needs it today.

Please stay Safe & Hydrated!!!