There was an interesting article on the Apalachtimes.com this week about a leak of a chlorine cylinder at a water treatment plant. A leaking 150-lb chlorine cylinder was detected by a local chemical alarm. The local area was evacuated, the cylinder was isolated and removed, and the area remediated. No injuries were sustained and the water treatment facility continued to operate as normal during the incident.
So, what made the incident interesting? What first caught my eye was the following:
“According to Manuel Aguilar, who oversees the water treatment plant, he arrived for work at about 6:30 a.m. and soon found the plant’s alarm, which detects the presence of chlorine in the air, was going off.”
Any facility that manufactures, stores or uses toxic inhalation hazard (TIH) chemicals typically uses chemical alarms to notify personnel in and around the area of the release of the TIH chemicals. By definition, TIH chemicals like chlorine can cause serious injury or death at relatively low concentrations. Even when chemicals like chlorine have distinctive odors the human nose cannot distinguish between relatively safe or unsafe levels of the chemical in the atmosphere.
Ideally chemical alarms used in TIH containing facilities should perform two functions. First they should act as a warning to those in the local area that are not wearing the personal protective equipment (PPE) necessary to operate in an exposure situation to evacuate the area as quickly as possible. Second it should serve as notification to initiate emergency response activities. In facilities that are not manned 24-7, there should be an off-site notification system to a response agency that is operational 24-7.
According to the news report Aguilar checked the chlorine tank in active use to see if it or any of its connections were leaking; no leak was found in the active system. Then:
“When he walked to where the chlorine gas canisters were secured, he could smell the pungent odor of chlorine, and noticed one of the 150pound canisters had ice builtup [sic] around an apparent leak at its base.”
Now there are a couple of things wrong with this picture. First off, since the active cylinder and the stored cylinder were stored in separate locations, the gas alarm should have indicated which location had the gas leak. It appears that there were sensors in the two locations, but the alarm indicator did not apparently indicate in which of the two areas that a leak was occurring. This would make it more difficult for emergency responders to locate the leak.
The second problem is that there is no way that Aguilar should have been able to smell the chlorine leak since he should have been in proper PPE when investigating a leak of unknown size and origin. And that PPE should have included a respirator. The fact that he looked for a potentially leaking tank without PPE probably indicates that either he was inadequately trained about the hazards of chlorine gas or that small leaks were relatively common and he had stopped taking the training that he had received seriously. The former is quite common.
Fortunately for the unprotected Aguilar the leak in the tank had essentially been stopped by the formation of chlorine gas hydrates. The evaporation of the chlorine liquid leaking from the bottom of the tank (nice picture here) created ice from the water vapor in the air which encapsulated much of the chlorine gas and ended up at some point blocking further leakage. If the full 75-lb (estimated) of chlorine lost from the cylinder had been in vapor form the concentration in the air could have been enough to cause serious injury or perhaps death to Aguilar when he entered the area.
Here is probably what should have happened. Sometime in the middle of the night (when the leak started) the person on duty at the Public Works Department should have received an alarm notification from the water treatment plant. The alarm would have indicated the location of the alarm (the cylinder storage area) and the chlorine concentration in that area. Calls would have been made to the local police and fire department and the facility manager. Based upon the location of the alarm and the chlorine gas concentration the facility emergency response plan may have called for local evacuations or shelter in place for the houses and business adjacent to the facility.
Depending in the training available (this is a small facility and full hazwoper training is typically not provided to facility personnel in such facilities) a team from the facility, local fire department and/or regional HAZMAT team would have suited up and entered the facility to investigate the leak.
There is nothing in the article about security at the site, and that is to be expected since this was a leak incident not a security incident. Having said that, you can look at the facility on Google Street View and see that the cylinders not in use are being stored outside (which, to be fair, reduced the risk to Aguilar in this situation) with the only security being a chain link fence with two locked gates very close to the cylinders. The only thing securing the cylinders to the facility are the cloth safety straps designed to keep them from falling over.
If this were a facility covered by the Chemical Facility Anti-Terrorism Standards (CFATS) program those security measures would be totally inadequate. Unfortunately, Congress in its infinite wisdom exempted water treatment facilities from CFATS coverage because the EPA had its own water security program in place. This facility apparently meets those (obviously very lax) standards.
The CFATS program treats 150-lb chlorine cylinders as a theft-diversion risk. The idea is that the cylinders like these could be used by terrorists to attack indoor gatherings of people (like shopping malls, churches or schools). Or they could be used in conjunction with an explosive device to make an improvised chemical bomb (like we have seen used in Syria and Iraq by ISIS). In any case, preventing the theft of this type of cylinder takes a lot more than just a chain link fence with gates that can be opened with a pair of bolt cutters.
KUDOS: I would like to mention that the news report from David Alderstein was very professionally done. I was particularly pleased to see the description of the chlorine hydrate that was provided; a very nice touch. It is not often that we see chemical incident news reports handled this well in the press.