A bubble erupted from the column with a violent pop! I hit the floor of the laboratory so hard the bruise on my chin was noticeable for several days. I was lucky — and I knew it. Detonation, according to the scant safety report, would be preceded by the liberation of hydrogen fluoride. Unfortunately, the upper temperature limit was defined too low. All week, I had been attempting to distill the solvent from a plasticizer I needed for my propellant research. Gumstock tests showed a polymer filled with bubbles, which meant that the plasticizer was too impure. With the vacuum pump at 100%, I pushed the electric heater past the detonation limit. After the first shock, I lowered the temperature a degree or two and continued the distillation a little over the detonation temperature. The gumstock tests passed this time. My research culminated in two successful papers. And yet, I was uncomfortable with my violation of the safety rules. No wonder complacency is such a problem!
Danger is real in chemical engineering. The question is: “How to put it foremost in the minds of those among us — those of a “Casey Jones” mentality?” Consider the following suggestions:
- enforce safety rules with strict discipline;
- encourage peer pressure;
- quickly identify when safety is cumbersome;
- develop your own monitoring system; and
- push for accurate, complete, and quantified values in safety standards.
A well-disciplined team should exhibit a careful balance between initiative and control. When workers are punished it should involve a peer review, if possible, but it must be done swiftly. The goal is to change behavior while memories are still strong. A points system seems to work but it should be separate from other personnel systems.
No one willingly wants to work with an unsafe worker. Peer pressure is highly useful. It can be developed by establishing regular, informal meetings among the team. These should be held off-site, away from management, to encourage free conversations. Where this is difficult, consider a luncheon to bring people together. If your presence might interfere, excuse yourself.
At one company, they have a saying, “If you find yourself looking behind you to see who’s watching — you’re probably doing something you shouldn’t.” Violating safety rules is more common than you think. I have seen companies that are accustomed to venting a column using an emergency vent or bypassing a gas sniffer to open a valve. Eventually, bad luck catches up with them. I am sure the guy at Pacific Engineering Company who welded on the side of an ammonium perchlorate tank thought it would be alright — he’d done it before and survived — but not that time!
Sometimes, safety rules seem to prevent successful completion of a task. When this occurs, do a safety review — quickly implement suggestions and make the task safe. More times than we’d like to admit, equipment cannot be operated within the safety guidelines provided by vendors. When I worked in the TiO2 industry, we regularly exceeded the limits on our oxygen heaters during start-up. So did our competitors! High coil failure was common and predicted. Again, the best approach is to re-engineer the task to reduce, or eliminate, the danger. A joint effort involving the manufacturer should have been called to develop reasonable guidelines.
Regardless of the diligence of your safety department, it is smart to bring your own canary into the mine. I recommend British Petroleum’s “safety footprint” for its flexibility and simplicity (p. 300, “Managing Risk and Reliability of Process Plants,” by Mark Tweeddale). Some original thinking is required. Setting the indices is tedious but once established they provide graphic insight into problems. Typical questions for a footprint are: “How often has the relief system operated?” or “How many times have we had a spill?” The answers give an indication of plant safety. To be effective, operators should be encouraged to maintain separate maintenance logs and shift logs. These can be used to complete the footprint. As with any activity, audits should be conducted periodically, by someone outside of the team, to assure that complacency has not settled in.
Dirk Willard, senior editor