CSB releases update on Philadelphia refinery fire

Article by Amanda Doyle

THE US Chemical Safety Board (CSB) has released its factual update on the refinery fire that occurred at Philadelphia Energy Solutions (PES) on 21 June.

The fire broke out at the refinery at 04:02 local time on 21 June and was extinguished at around 08:30 on 22 June. Five workers received minor injuries. The substantial damage to the refinery, which had a production capacity of 335,000 bbl/d of crude oil, was enough to cause PES to permanently close the refinery and file for bankruptcy.

The CSB factual update found that a corroded pipe elbow ruptured in the refinery’s alkylation unit, which released process fluid that included around 2.3 t of hydrofluoric acid (HF). This formed a ground-hugging vapour cloud which ignited two minutes later causing a fire and explosions. The third, and largest, explosion occurred when the V-1 Treater Feed Surge Drum ruptured. This launched a 17.2 t fragment of the drum across the Schuylkill River – a distance of around 640 m – and two other large fragments landed within the refinery.

Just after the cloud ignited, the control room operator activated the Rapid Acid Deinventory (RAD) system which prevented the release of additional HF by draining the HF to a storage vessel. The HF in the RAD drum was later neutralised by adding a base to produce water and salt. This process took place over three weeks in August.

Kristen Kulinowski, Interim Executive, said: “Since 2015, the CSB has investigated three major incidents at refineries that utilise HF for alkylation. Incidents in Superior, WI, and Torrance, CA, fortunately did not result in a HF release. That was not the case here in Philadelphia. Though the main tank holding HF was not breached, HF was a component of the process fluid released from the alkylation unit. We are lucky there were no serious injuries or fatalities.”

Piping is susceptible to corrosion from HF. The pipe thickness had been monitored periodically, however the pipe elbow was not monitored. The pipe elbow had been corroded to a thickness of 0.03 cm, despite the default retirement thickness being 0.45 cm. The failed pipe had a high nickel and copper content, and the CSB noted that various industry publications have found that this type of steel will corrode faster when HF is used.

CSB Supervisory Investigator Lauren Grim said: “Corrosion is not a new issue for the CSB. In its prior investigation of a 2012 Chevron Refinery fire we determined that corrosion caused the rupture of a piping component. Similarly, the 2009 Silver Eagle refinery fire was also caused by the failure of piping that had thinned due to corrosion.”

The CSB investigation is still ongoing.

Article by Amanda Doyle

Staff Reporter, The Chemical Engineer

Recent Editions

Catch up on the latest news, views and jobs from The Chemical Engineer. Below are the four latest issues. View a wider selection of the archive from within the Magazine section of this site.