CSB Updates for Majestic Paints Explosion – 4-13-21

Last week there was an explosion and fire at a paint and polymer plant in Columbus, OH. The Chemical Safety Board announced that they would be deploying an investigation team to the incident that involved on death and large scale destruction at the facility. The next day they published their first informational report on the incident and this week followed up with their second update. The investigation is still being conducted and it will likely be some time before the CSB issues its final report.

It looks like this will be the new standard for incident information sharing during incident investigations, since they were also proactive about the information flow on the liquid nitrogen release incident in Georgia in February.

The Incident

According to news reports (here, here, and here) last week the explosion and fire occurred after midnight in one of the units at the facility. Eight people were taken to area hospitals, with two in critical condition and six in stable condition. One person was killed. There were apparently no injuries off-site and only minor damage at neighboring facilities. One of the news reports stated that there were ‘additional explosions’ while fire crews battled the blaze.

CSB Initial Report

The initial deployment update from the CSB provided some basic background information on the facility. It noted that the explosion and fire took place in the OPC Polymer unit at the facility. They reported that there were 21 employees at the site at the time of the incident with one being killed, five hospitalized and four ‘non-serious injuries’. They also reported that the “fire and explosion occurred at approximately 12:30 a.m. EDT”.

Follow-up Report

This week’s report from the CSB provided more information about the incident. The new information includes:

• A review of security video allowed the investigation team to develop a more precise time of the incident. The first event (loss of containment) occurred at 12:02 a.m. EDT (system time) and the video cameras stopped recording at 12:04 a.m. EDT (system time).

• Yenkin-Majestic provided the investigation team with an overview of the resin manufacturing process.  The process is called alkyd resin manufacturing which utilizes the company’s own technology.

• Materials are mixed in a metallic kettle which contains an agitator and is heated by a furnace before it goes to a scale tank and then to product storage. There are a total of six kettles: four are heated by gas furnaces and two are steam heated.

• The incident occurred during the batch production in Kettle #3.

• Kettle #3 has a rated capacity of 20,900 lbs. and is heated by a gas furnace.

• The fire suppression system in a portion of the paint production facility was activated during the event. The investigation team verified there was no fire event in the adjacent paint production facility.

• The east wall of the paint production facility sustained damaged as a result of the explosion event in the adjacent OPC Resin facility.

Alkyd Resin Manufacturing

Alkyd resins are long chain molecules made up of repeating units. Instead of self-reacting monomers making up those units as is seen in polymer manufacturing, alkyd resins are typically made with glycols (molecules with alcohol groups on each end) and poly-basic acids (molecules with carboxylic acid groups on each end). As the molecules grow longer their viscosity increases significantly. To keep the viscosity manageable solvents are typically added to the mixture.

The condensation reaction between the glycol and carboxylic acids requires the application of heat and produces water as a byproduct. The water must be removed from the system, coming off as steam and usually includes significant amounts of solvent as it leaves the vessel. That gas is forced through a condenser and the solvent is returned to the vessel and the water is collected in separate vessel. That produced water is contaminated with solvent and is either disposed of as waste or processed to remove the residual solvent.

All of the alkyd resin production that I have been associated with used steam as the heat source driving the reaction. Steam is easy to use, and the flames used to heat the boiler to produce the steam are typically isolated from the production area in a separate building. According to the CSB, the process involved in the incident used a gas furnace instead of steam. This is almost certainly because the OPC Resin process called for higher temperatures than are typically available from steam.

Commentary

The CSB has a lot of investigating to complete before an official cause of the incident can be determined. Having said that, based upon the very limited information available to date, I can imagine a scenario that could lead to an explosion and fire like that seen at the Columbus facility.

A release (or loss of containment) between the reaction vessel and the condenser would lead to a steam cloud containing solvent vapors. The temperature at the point of release would almost certainly be above the flash point for the solvent. If the vapor/steam cloud reached the flames in the furnace (or another ignition source) an explosion could result.

As we currently understand the facts surrounding this incident, the above scenario is one readily possible explanation for the explosion and fire. We will have to wait and see what other facts the CSB uncovers in their investigation.

One final point that I would like to make is that I greatly appreciate this new information sharing activity on the part of the CSB. The CSB has always done an excellent job of preparing final reports on their investigations, but these preliminary facts help the process industry look at their operations with an eye toward making them safer. It is always better if we can learn from someone else’s mistakes.