Monomer Hazmat Incident

Earlier this week there was a hazmat situation outside of Mobile, AL involving acrylamide monomer overheating in a delivery tank wagon. According to news reports (here, here and here) the incident was resolved without injuries or damage to property. While far from a potentially catastrophic incident, it could have been much worse.

Full Disclosure: I used to work for the company involved in the incident, but I was never employed at, worked at or visited the facility in question.

Background

Acrylamide (C3H5NO) is an industrial chemical known as a vinyl monomer. This class of chemicals reacts with itself to form polymers. The vinyl polymerization process is exothermic, it produces more heat energy than required to initiate the reaction. The acrylamide polymerization reaction produces about 19.8-kcal/mole of heat energy.

While pure acrylamide is a white powder, it is most commonly used in industry as an aqueous solution of 35 to 50% acrylamide. According to news reports, the tank wagon contained about 45,000 lbs of 50% acrylamide.

Since acrylamide is a self-reactive chemical there are always some reactions happening. The rate of reaction is temperature related. Generally speaking, when stored at less than 90°F, the rate of reaction is not a safety concern and only a limited quality concern. At temperatures above 90°F the rate of reaction begins to generate enough heat to produce a noticeable temperature increase in the material. And the higher the temperature the faster rate of reaction becomes. This causes a ‘self-accelerating’ reaction’. At temperatures above about 115°F it becomes very difficult to control the temperature rise with cooling systems and the rate of temperature rise increases dramatically.

The Hazards

Even so, a 50% acrylamide solution is a relatively safe monomer. The high-water content means that there is a heat sink to soak up much of the heat of reaction. In monomers that are liquids in their pure state, water (or other solvents) are seldom if ever added, so there is no heat sink. Additionally, a pure monomer shipment of 45,000-lbs contains 45,000-lbs of reactive material; the acrylamide tank wagon only contained 22,500-lbs of reactive material.

Even so, there was a potential for the tank wagon to rupture fairly violently. The heat of reaction could have theoretically caused a heat rise of about 150°F. This would certainly have brought the water to a boil and produced some pressure in the tank wagon. Typically, these tank-wagons are not really high-pressure vessels so there is the possibility of a steam-pressure rupture of the tank. There is a pressure relief valve to (with usually about a ½” diameter discharge line). With a slow, steady-state heat increase this safety system would certainly provide adequate relief for a slow, but steady increase steam pressure.

Unfortunately, the last 30 to 40° of temperature increase takes place in about 30 seconds (I have had the unpleasant honor of measuring this adiabatic heat rise on a number of occasions in the lab and in production vessels. The resulting pressure rise after 212° is also quite dramatically quick, if limited in scope. Depending on the maintenance history of the tank wagon, the pressure relief system would probably be adequate if only steam pressure were involved.

Unfortunately, the acrylamide conversion to polyacrylamide does not provide for that simple a problem. As the polyacrylamide chains begin to lengthen they stop being water soluble and becomes a very viscous liquid that is not miscible (mixable) with water. The resulting two liquid phases take up more volume than did the 50% acrylamide solution. This increases the pressure in the tank-wagon and if any of the polyacrylamide reaches the pressure relief valve, it is very likely to plug the valve and make it inoperable. A tank wagon breach is nearly inevitable.

A major acrylamide supplier in the 1990’s had an interesting safety training film for their acrylamide customers. One of the things that was included was a description of the results of an acrylamide rail-car that had a runaway reaction on their site. Since railcars are generally capable of handling higher pressures, the violence of the final catastrophic failure was probably higher than what we would see with a tank wagon, but I vividly remember the narrator describing finding the top bell of the rail car hundreds of yards away after the incident.

Fortunately, the gas phase discharge from this type of catastrophic pressure release event (NOT an explosion) is steam contaminated with some amount of unreacted acrylamide in suspended droplets. This means that there is no fire or secondary explosion hazard. You just have to deal with possible contamination with acrylamide which is a bioaccumulating neuro-toxin. Oh, and of course, the impact hazard from various high-speed bits of metal and plastic in the vicinity of the event.

Incident Response

Any time a self-reactive chemical is delivered to a facility at a temperature above (or even below) the accepted storage temperature, it is a matter for concern. For vinyl monomers like acrylamide, the immediate response should be sparging the contents of the tank wagon with air. The oxygen in the air reacts with a chemical added to the monomer to inhibit the polymerization reaction. The air sparge also serves as a mixing aid for the contents of the tank wagon (or rail car).

One good way to effect this sparge is to open the top manway on the tank wagon (if the temperature is still below 115°F) and stick a long pipe into the liquid with the upper end attached to an air-line; open the valve and sparge away. REMEMBER: “Air-Line”, not nitrogen, not carbon dioxide, they will just stop the inhibition reaction and the acrylamide runaway reaction will initiate at a lower than normal temperature. NOTE: The open manway also acts as a much more effective pressure relief device, but please put a shade shield over it; you can get UV light polymerization started by sunshine which will defeat the whole purpose).

Applying cooling is probably not going to be effective if the temperature is anywhere near 115°F. Applying a monitor (fixed or semi-fixed fire hose) stream to the tank wagon is not generally going to be very effective. There is a layer of insulation between the pretty, shiny, outer-skin of the tank wagon and the actual container holding the acrylamide. If the tank wagon is set up with internal coils (typically used for heating, not cooling), then running cold water through those lines may be effective, but only if there is some sort of mixing. That air sparge will provide some mixing but moving the truck very short distances back and forth will provide the most effective mixing.

The biggest problem for incident management is the problem of monitoring the temperature and pressure in the container. Tank wagons used for carrying temperature sensitive materials are equipped with a temperature gauge. While we are starting to see some tank wagons equipped with electronic gauges with remote reporting capabilities, most have old style analog gauges. In my experience, these are seldom accurate (I have seen one that was apparently operational that ended up being off by 20°F on the high side) and frequently inoperable or unreadable. Pressure gauges are generally more reliable, but they are located on the top of the tank wagon near the manway, the last place you want to be in an incident where high pressure failures are expected or even possible.

Off-Site Response

The Al.com article raised issues with the community response to this incident. Looking at the Google Maps® for the facility we can see that there is a residential community less than 2,000 feet away from the front gate to the facility. At least a portion of this residential community was under a mandatory evacuation for about 2½ hours during the incident. Complaints were made about the amount and accuracy of the information shared with that community and some apparent (unspecified in the article) confusion amongst first responders as to what was going on.

This is a common complaint heard in chemical incidents around the country. Even where the chemical industry does a good job communicating with local government and even local activist organizations, during an incident the flow of communications to affected individuals off-site usually leaves something to be desired.

There are a number of reasons for this, but probably the most important is that the facility management team is concentrating on managing the incident so as to keep the effects, on and off site, as minor as possible. Managing the information flow to the residential level is not a high priority and it is not something easy to practice.

While communications systems (reverse 911 and social media campaigns) can be put into place to provide information to the local community, actual communication will only come if the facility management team puts into place processes to educate the community (before any incidents take place) about the actual potential hazards associated with the facility, what is being done on a daily basis to mitigate those hazards, and what actions community members might have to take in the event of a serious incident. Only then will the incident communications provide actionable information. Only then will the instructions to ‘evacuate’ or ‘shelter-in-place’ mean enough to the affected individuals to gain effective compliance and not blind panic or studied indifference.

That pre-incident communication requirement is even more important for all of the first responders that might be involved in incident response. They need to know in advance what kind of support they might be called upon to provide to the facility in the event of an incident. While they may not need to know the reaction profile characteristics of acrylamide, they do need to understand the problems associated with any self-reactive chemicals on site, including actions to take and actions to avoid.

The law enforcement personnel that are normally called upon to conduct on-site notification of evacuation or shelter-in-place activities need to fully understand the potential hazards associated with the releases from a facility and they must have the personal protective equipment (and the training to use it) necessary to operate in a contaminated environment while they conduct those notifications.

It all comes down to effective prior planning.