Richmond refinery fire (2012): what investigators found about sulfidation
The Richmond refinery fire was the August 6, 2012, rupture and fire of a corroded crude unit pipe at a refinery in Richmond, California. The US Chemical Safety and Hazard Investigation Board (CSB) found that sulfidation corrosion had thinned a low-silicon carbon steel section that thickness monitoring did not cover.
By Jun Zhong, Editor·Researched and drafted by Beacon's AI newsroom·Checked by Jun Zhong, Editor·Last checked ·Next review
Why does it matter?
In the decade before the fire, staff recommended inspecting every component of the line that failed or upgrading its steel, and its replacement was turned down at the 2007 and 2011 turnarounds 1. The thickness data used to justify running on came mostly from fittings with more silicon than the component that ruptured 1. Nobody was killed, but 19 workers were caught in the vapor cloud and about 15,000 people from nearby communities later sought medical treatment 1.
What happened on August 6, 2012?
The line carried light gas oil, about 10,800 barrels per day, from the crude unit's atmospheric column; this stream was called the 4-sidecut 1.
| Date and time | Event |
|---|---|
| 2007 and 2011 turnarounds | Requests to replace the 8-inch 4-sidecut piping are rejected under the turnaround scope criteria; much of the 12-inch section of the same stream is replaced in 2011 1 |
| August 6, about 3:50 p.m. | An operator on rounds finds drips from the insulated line about 14 feet (4.3 m) up; it cannot be isolated 1 |
| Shortly after 4:00 p.m. | The plant fire department is called and sets a hot zone of 20 by 20 feet (6 by 6 m) 1 |
| About 5:00 p.m. | Feed to the line is cut by 5,000 barrels per day; about 40 people gather at the leak over the afternoon 1 |
| Early evening | Firefighters strip sheathing and insulation from the hot pipe; soaked insulation briefly catches fire; hose streams knock more insulation off and the leak becomes a spray; managers decide to shut the unit down 1 |
| About 6:30 p.m. | The pipe ruptures; a vapor cloud engulfs 19 workers and ignites at 6:33 p.m., about two minutes after the release; 18 get out before ignition and a firefighter in a fire engine escapes through the flames 1 |
| 6:38 to 11:12 p.m. | Shelter-in-place advisory for Richmond, San Pablo and North Richmond 1 |
| January 28, 2015 | The CSB releases its final report, after an interim report (2013) and a regulatory report (2014) 1,2 |
Sources: CSB Report 2012-03-I-CA, 2015 1; CSB investigation page 2.
Why did the pipe rupture?
The CSB found that sulfidation corrosion had made a 52-inch-long (1.3 m) component of the 8-inch line extremely thin 1. The line was built to a carbon steel pipe specification with no minimum silicon content; the failed component contained 0.01% silicon by weight, while the elbow just upstream, at 0.16%, had thinned far less 1. Thickness readings had mostly been taken on fittings with more silicon, so the data said the line could last until 2016 1. More on the mechanism is in What is sulfidation (high-temperature sulfur corrosion)?.
Further CSB findings cover how the line was managed and how the leak was handled 1:
- A 2009 internal metallurgy report ranked this line top priority for inspecting every component, but that inspection had not been done by the time of the rupture 1.
- Experimental guided wave sensors covered only a small part of the line, not the failed component, and when they proved unreliable no manual inspection replaced them 1.
- After the 2011 turnaround, the line's alert thickness was lowered on the strength of a minimum calculated for one small component and applied to the whole circuit, with no formal multi-person review 1.
- On the day, some firefighters believed the line ran at about 130 °F (54 °C) when it was near 640 °F (338 °C), and there was no leak response protocol; the CSB judged that a proper evaluation would likely have led to an immediate shutdown 1.
What did the CSB recommend?
The recommendations below are the CSB's, addressed to the refinery's owner, to state and local government and to industry bodies 1:
- To the refinery's owner: damage mechanism hazard reviews at all its US refineries, an auditable route for denied or deferred turnaround inspection work, and a technical review before any minimum alert thickness is lowered 1.
- To California and local governments: require documented damage mechanism hazard reviews in the process hazard analysis cycle, report process safety indicators, and set up a joint multi-agency refinery safety program 1.
- To the American Petroleum Institute (API): revise RP 939-C, RP 571, API 570, RP 578 and RP 574 so that carbon steel circuits that may contain low-silicon parts are found and either fully inspected or replaced, and revise RP 2001 to require a site leak response protocol 1.
What changed afterwards?
- California rule. Section 5189.1 of Title 8, Process Safety Management for Petroleum Refineries, became operative on October 1, 2017 5. It requires a damage mechanism review for each refinery process where a damage mechanism exists, with half of the first reviews due within three years and all within five, and revalidation every five years 3. The CSB closed its recommendation to the state as "acceptable action" on May 30, 2018 3.
- API RP 939-C. API published a second edition in January 2019 4. The CSB reviewed it and in January 2021 accepted it as an alternative, noting that it addresses silicon content in carbon steel piping and treats low-silicon components as permanent condition monitoring locations, though it keeps advisory "should" wording 4.
- Other industry documents. The CSB lists its recommendations on RP 571, API 570, RP 578, RP 574 and RP 2001 as closed with acceptable or alternative action, and one to the American Society of Mechanical Engineers (ASME) on leak repair as still open 2.
Where integrity problems show up
| Element or step | Integrity link | In the Library |
|---|---|---|
| Placing thickness monitoring points | The 8-inch line had 19 condition monitoring locations (CMLs), most of them historically on higher-silicon fittings and none on the 52-inch component that failed, so their readings did not show how fast the low-silicon pipe was thinning 1 | What are condition monitoring locations (CMLs)? |
| Setting the alert and minimum thickness | Alert thickness lowered using one component's calculation, without multi-person review 1 | Minimum required thickness |
| Setting turnaround scope | Replacement and full inspection requests rejected under scope criteria in 2007 and 2011 1 | How is inspection scope set for a turnaround? |
| Relying on new monitoring technology | Guided wave sensors left the failed part uncovered and were not backed up 1 | Guided wave testing |
| Verifying material in old carbon steel | Silicon varies by component and is hard to check in the field 1 | What is positive material identification (PMI)? |
| Responding to a leak | No protocol; responders lacked the line's temperature and corrosion history 1 | What is process safety management (PSM)? |
Sources: CSB Report 2012-03-I-CA, 2015 1.
Common questions
- Did anyone die in the Richmond refinery fire?
- No. The 19 workers in the vapor cloud all escaped; six had minor injuries, and about 20 members of the public were admitted to hospital.
- Was this corrosion under insulation?
- No. The CSB found that the metal was lost from the inside, as sulfur compounds in the hot oil reacted with the steel; the insulation only hid where the leak was. See [What is corrosion under insulation (CUI)?](/library/corrosion-under-insulation/) for that separate mechanism.
- Was the pipe expected to last?
- Yes. Inspectors told the response team that readings taken two months earlier showed enough wall for the line to reach the 2016 turnaround.
Beacon explains these concepts in its own words for education. It does not reproduce or replace the standards cited, and Winda is not affiliated with or endorsed by the organizations that publish them. Engineering decisions need the current edition of the standard and a qualified engineer.
Sources
- 1.Final Investigation Report: Chevron Richmond Refinery Pipe Rupture and Fire (Report 2012-03-I-CA) · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 2.Chevron Richmond Refinery Fire (investigation page with recommendation status) · US Chemical Safety and Hazard Investigation Board (CSB)
- 3.Recommendations Status Change Summary, Recommendation 2012-3-I-CA-R09 (California State Legislature, Governor of California) · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 4.Recommendations Status Change Summary, Recommendation 2012-03-I-CA-R26 (American Petroleum Institute) · US Chemical Safety and Hazard Investigation Board (CSB) ·
- 5.California Code of Regulations, Title 8, Section 5189.1, Process Safety Management for Petroleum Refineries · California Department of Industrial Relations