When the deal was made five years ago, officials were proud to announce it was the first refinery expansion project in the U.S. in 30 years. Motiva Enterprises' CEO Bill Welte called it a "momentous occasion" for his firm and its owners Royal Dutch Shell and Saudi Aramco. The final product would be the largest refinery in U.S. It was projected to produce more than 12 million gallons of gasoline per day from crude oil shipped initially by tankers from Saudi Arabia to the Port Arthur, TX site.
Fast forward to the grand opening ceremony on May 31, 2012 where five executives including Shell's CEO Peter Vosser and CEO of Saudi Aramco Khalid Al-Falih turned a wheel valve to symbolize the refinery going on-line. Shell's news release noted
"a prestigious group of guests gathered to celebrate this milestone event, which was followed by an open house and site tours."
The expansion ultimately cost about $10 billion.
While the oil giants were still sweeping up the confetti, a caustic contaminant seeped into the new 30-story high crude distillation unit, described by Scientific American's Erwin Seba as "the heart of the refinery process." In "Within hours, caustic vapors wreaked quiet ruin on biggest U.S. refinery," Seba explains:
"While harmless when mixed with crude, the undiluted caustic vaporized into an invisible but devastating agent of corrosion as the chamber heated up to 700 degrees Fahrenheit (370 Celsius); the chemical gas raced through key units, fouled huge heaters and corroded thousands of feet of stainless steel pipe."
Workers at the plant were tipped off that something had run seriously amok
"when two fires broke out and a heater ruptured."
Unlike the public fanfare less than two weeks earlier at the grand opening ceremony, Seba writes:
"Motiva has said little about the incident. ...11 days after it occurred, the company confirmed for the first time that the unit might remain shut for 'several months.' Sources say officials are telling workers that the unit could be idle for as long as a year."
Details are dribbling out about the chain of events that preceded the meltdown, and Seba offers some in his piece. They include criss-crosssed wires on a temperature gauge and cracks in pipelines. Seba's article also remind us of the complexities that correspond with start-ups and shut-downs of volatile petrochemical processes. This latter point, in particular, should be well recognized by any refinery. These operations have a history of catastrophic incidents and anyone involved in the industry or observers of it know the importance of process safety management.
I'm really having a hard time wrapping my mind around what happened at this Motiva plant. A few things puzzle me. First, it seems that an effective process safety management system would have prevented this corrosive ruin from occurring. The petrochemical industry has developed its own consensus standards to prevent unintended releases of toxics and/or fires, explosions or other incidents that may affects workers, the community, the environment or property. Process safety management is a systems approach to identifying, eliminating and controlling hazards. As the American Petroleum Institute says, when talking about process system:
"While petroleum operations involve risks, they can be effectively managed and/or eliminated when appropriate measures are taken. Operating safely is the most basic feature of business efficiency and has proven to improve productivity."
Motiva Enterprises knows this. They even have a page on their website about their safety management system. The Occupational Safety and Health Administration (OSHA) has a regulation on the books (since 1992) specifically addressing hazards associated with highly-hazardous chemical processes. OSHA's PSM standard requires employers to conduct a process hazard analysis, defined as:
"a careful review of what could go wrong and what safeguards must be implemented to prevent releases of hazardous chemicals." The PSM rule also requires prestartup safety reviews, evaluation of mechanical integrity of critical equipment, and written procedures for managing change.
This "managing change" phrase refers to the exact kinds of start-ups and shut-downs that were anticipated with the grand opening of the Motiva plant.
I'm also puzzled by the plant's massive meltdown because this Port Arthur refinery has been recognized for its safety record by the American Fuel & Petroleum Manufacturers, (formerly the National Petroleum Refiners Association.) The worksite has also been a member of OSHA's Voluntary Protection Program (VPP) since 2006. OSHA officials call VPP worksites:
"models for effective employee protection," saying "the most obvious evidence of the program’s success is the impressive reduction in occupational injury and illness rates, as well as reduced workers’ compensation costs and decreased employee turnover."
A key feature of these VPP sites is their:
"exemplary achievement in the prevention and control of occupational safety and health hazards and the development, implementation and continuous improvement of their safety and health management system."
An exemplary (i.e., effective) safety management system for chemical processes should have been able to prevent this cataclysmic event. Some knowledgeable insiders, including experts within the federal government, think this Motiva event may be the worst process safety incident in the U.S. since the 2005 BP Texas City explosion that killed 15 workers. Thankfully, no one was killed or injured in this most recent event at the Motiva Port Arthur refinery. Two workers have died on-the-job at the expansion-construction site, one in 2010 and one in 2011, but according to OSHA these incidents were not considered part of the VPP Motiva site.
A spokesperson for the U.S. Chemical Safety and Hazard Investigation Board (CSB) told me they are not investigating this most recent major incident. Maybe that's because the CSB investigators would likely find the same organizational and systems-management problems they've identified and described in previous disaster investigations. A Labor Department spokesperson told me that OSHA has also not opened an inspection at the Motiva Port Arthur refinery. OSHA typically will not conduct a post-incident inspection unless a worker is killed or three or more individuals are hospitalized. Given however the Motiva site's designation as a VPP site, I have to ask why OSHA is not doing so. Investigations by GAO in 2009 and in 2004, and the Center for Public Integrity (here, here, here) strongly suggest that OSHA needs to look more closely at serious incidents at VPP sites to determine whether the metrics the agency uses really mean the worksites have exemplary safety programs. The primary metric used currently by OSHA is an injury and illness rate that is at or below the nation average for their respective industry.
I've written previously (here, here) that by relying heavily on "OSHA recordable" injury and illness cases as the measure of safety performance, we're missing the factors that probably better assess the risk of a catastrophic incident. In a workplace like a refinery, with a toxic soup of highly hazardous compounds and hundreds of miles of pipes carrying them, it's as important to consider problems found during routine maintenance, unscheduled repairs and unintentional fires, releases, and ruptures, as cases that cause twisted ankles and sprained lower backs. The Baker Panel commission investigating the 2005 BP Texas City disaster made a similar conclusion about the limitations of using injury rates as a measure of safety performance:
“BP’s reliance on injury rates significantly hindered its perception of process risk. …BP’s corporate safety management system for its US refineries does not effectively measure and monitor process safety performance.”
The catastrophic event last month at the OSHA VPP Motiva Port Arthur facility should be used by the agency to develop more informative metrics of safety and test them. Failing to do so, OSHA and all of the VPP community will have lost a golden opportunity.
I suspect the basic take-away here is that processes that aren't yet producing anything haven't themselves been tested yet.
No...as part of PSM you are expected to evaluate this specific type of oops event and address countermeasures Before any work....the PSM standard and subsidary documents have extensive guides on how to do this. Also, as the process is probably not "brand new" there is what is known as history from other facilites.
It is doubtful we will get full information to understand and learn from this event for many political and economic reasons but, the more public questioning that occurs will facilitate fact finding.