How chemists should learn to deal with spills in the lab and in the plant
As chemists, we should be appalled at what happened in East Palestine, Ohio, US. If you’re not aware, the Norfolk Southern Railway had a derailment accident, and a number of chemicals stored in railcars began burning. Some of the railcars contained vinyl chloride, which is a hazardous and toxic reagent. Authorities at the scene decided to vent the vinyl chloride into a trench and burn it, which provided literally hellish pictures. Rightly or wrongly, people will be thinking of these pictures for the rest of their lives, and associating chemicals – all chemicals – with this incident.
I don’t think it’s useful for me to talk about the aftermath of a major release to the environment. It’s not in my nature to second guess people who are trying to solve an emergency under pressure, so I have no interest in speculating as to what they should or should not have done. I look forward to the reports from authorities and experts that will be released no sooner than a year from now, and that’s when we will really learn what happened, and why.
Instead, let’s think about a situation that we know about – leaks and spills in the laboratory or the plant. Maybe you haven’t shipped a railcar containing 120,000 litres of hazardous chemicals across the country, but you’ve carried a litre of solvent down the hall. What should you be doing?
Leaks and spills happen to chemists all the time
Let’s start with your best tool – preparation. You should know what you’re transporting, and what the safety data sheet has to say about it. You should have the PPE ready to go. Secondary containment (like those little rubber buckets) will save you, especially if you’re a bit clumsy like me and might trip and send a bottle flying. Knowing where the nearest spill kit is (and what its contents are, be it absorbent pads or diatomaceous earth) is helpful as well.
There are plenty of places for a reactor in the plant to potentially leak: it’s essentially a stainless steel bowl with holes drilled in it. That said, the likelihood of a leak is probably low – I certainly haven’t seen one before, but I’ve only been working around larger reactors for a little over a decade. One way to check the ability of the system to hold liquid is to hold some gas instead. You can seal the reactor and pump some nitrogen and see if the reactor will hold pressure. If the reactor holds 10psi positive nitrogen for 30 minutes without it leaking away, that’s a good sign that you have a reactor or a bottom valve that can handle the weight of 6000kg of solvent.
I don’t think we have any data for the most common cause of leaks or spills in either the plant or the laboratory. In the lab, I suspect that gravity is the biggest cause of spills. In the plant, the most common cause is probably the forklift. After all, it’s a truck with knives on the front. Another aspect that makes forklifts dangerous with drums is that the insertion point is on the bottom – if you hit the drum or bag, chances are you’re going to drain the entire container. If you’re not careful as you’re lining up the forks, you could be cleaning up 200 litres of solvent (and if you’re unlucky and you hit two drums, you’ll be mopping up 400 litres).
That’s the dramatic kind of spill, but I believe leaks are far more often discovered when you open the door to the warehouse and you smell a funny smell. It either tickles your nose or punches you in the face – either way, you’ve got your first sign that you’re walking into a spill situation. A chemist or a chemical operator will often identify that whiff and know what they have to deal with. Another sign is a puddle on the floor. That’s when you can begin the process of getting your PPE on, assessing the scene, preventing a leak to the environment, finding another container or an overpack, sealing the drum and cleaning up the mess. I have a feeling this is a situation where chemistry plays a role – many of the containers are metal, even if the inner linings are plastic. With enough water, heat or jostling, a sufficiently volatile acid or a base will begin eating away at the metal, and you’re off to the races.
Again, I have no good answers for the problem of ‘how do you ameliorate the spill of tonnes of vinyl chloride?’ ‘Don’t be in that situation’ is my overly facile answer. But leaks and spills happen to chemists all the time, and we can be better prepared for far more common situations. We can learn to prevent those situations, and to solve them when they happen.