Event
- Event ID
- 498
- Quality
- Description
- An explosion occurred at a compressed gas storage facility of a research laboratory, during filling hydrogen cylinders from a hydrogen tube trailer.
The system consisted of a bank of valves supplying different pressurised gases through a common manifold.
When workers started the operation of filling trailer tubes with hydrogen, they manifold was still containing residual oxygen and was connected to a tube trailer of oxygen. There was only one valve separating the two gases. When attempting to purge oxygen from the manifold using low-pressure hydrogen, the hydrogen-oxygen mixture ignited. The fire damaged the valve, allowing high-pressure oxygen to flow into the hydrogen tubes of the trailer.
A composition of 42%vol. hydrogen – 58%vol. oxygen was estimated present in the tube at ignition, at an initial pressure of 4 MPa. Following. Following ignition, the flame transitioned to detonation at a pressure of approximately 40 MPa. However, this would not have been enough to break the tube, which had been previously tested at 90 bar. Therefore, a peak pressure possibly 40 time higher than the initial pressure could be assumed. The tubes ruptured, and large pieces of shattered steel became shrapnel. A 20-kg piece of the hydrogen tube was found about 425 m from the site of the explosion. A significant cloud of hydrogen was released to form a fireball of short duration.
The consequence were two injuries and extensive damage to equipment. - Event Initiating system
- Classification of the physical effects
- Hydrogen Release and Ignition
- Nature of the consequences
- Macro-region
- North America
- Country
- United States
- Date
- Main component involved?
- Valve (Shut-Off)
- How was it involved?
- Rupture & Formation Of A Flammable H2-Air Mixture
- Initiating cause
- Inadequate Or Wrong Design
- Root causes
- Root CAUSE analysis
- The INITIATING CAUSE was the failure of the shut-off valve, which allowed oxygen flowing into the hydrogen tube.
The IGNITION SOURCE was reported by some source as rushing sand particles in the tubing producing sparks. However, . Peterson and J. G. Weisend reported that forensic analysis determined that " as the injected oxygen mixed with the hydrogen the higher oxygen pressure adiabatically compressed the mixture, raising its temperature until it exceeded the autoignition temperature".
They also reported that an existing isolation valve had been removed, to reduce operative time. This modification, confirmed by H2TOOLS, had not been authorised.
Investigators determined also that the fittings for the hydrogen and oxygen delivery systems were similar, allowing the connection of incompatible gases to the same manifold. In addition, workers were not following procedures and did not understand the serious hazards of mixing hydrogen and oxygen. Therefore, ROOT CAUSES were design and installation errors, and a lack of safety standards, training and risk management.
Facility
- Application
- Hydrogen Transport And Distribution
- Sub-application
- CGH2 tube trailer
- Hydrogen supply chain stage
- Hydrogen Transfer (No additional details provided)
- All components affected
- shut-off valve, tube trailer
- Location type
- Open
- Operational condition
- Pre-event occurrences
- According to H2TOOLS, "an employee, without authorization, fabricated and installed an adapter to connect a hydrogen tube trailer manifold to an oxygen tube trailer manifold".
- Description of the facility/unit/process/substances
- DESCRIPTION OF THE TRAILER
From the photos provided, it appears that the trailer consisted of approximately 36 tubes, containing hydrogen at approximately 38 bar (a surprisingly low value, provided only by one source).
Emergency & Consequences
- Number of injured persons
- 2
- Number of fatalities
- 0
- Environmental damage
- 0
- Property loss (offsite)
- 0
- Post-event summary
- Two workers were severely burned and were absent from work for a total of 4 months. They received first and second-degree burns over about 30% of the bodies.
The hydrogen tube trailer experienced major damage, with almost all tubes deformed and/or ruptured. Minor damage was caused to a portion of the facility. Pieces of the tube flew approximately 700 yards away (approx. 640 m).
Lesson Learnt
- Lesson Learnt
- As pointed out by Peterson and Weisend, this incident highlighted the following:
• Changes to any system require careful review, including thorough risk assessment. Even minor alterations to a hydrogen system or operation can lead to drastic changes in outcome.
• Combustion in confined volumes, in this instance the tubes of the trailer, can lead to high overpressures capable of shattering heavy-duty steel.
• Management responsibility extends to all aspects of hydrogen operations, and certainly those involving hazardous operations. Ultimately, in this accident, plant management was held accountable.
• Appropriate training of staff is critical. A deep knowledge of the hydrogen properties and field experience is required to foresee all consequences and perform a proper risk assessment. - Corrective Measures
- The following modifications were implemented:
(1) The gas facility’s management structure was changed;
(2) Standard operating procedures were rewritten;
(3) Worker training requirements were improved;
(4) Engineering changes were made, including changing the fittings and piping to make it impossible to mistakenly mix hydrogen and oxygen.
Event Nature
- Release type
- gas
- Involved substances (% vol)
- H2 100%
- Actual pressure (MPa)
- 100
- Design pressure (MPa)
- 3.8
- Presumed ignition source
- Auto-ignition
References
- Reference & weblink
Peterson and Weisend, <br />
Cryogenic Safety - A Guide to Best Practice in the Lab and <br />
Workplace, 2019, in International Cryogenics Monograph Series, Springer Nature <br />
https://link.springer.com/chapter/10.1007/978-3-030-16508-6_1?fromPaywa… />
(accesed December 2025)Extract from Glass “ Pressure Safety Orientation”, Los Alamos report LA-UR-17-24065, (2017) available at <br />
https://www.osti.gov/servlets/purl/1358153<br />
(accessed October 2025)Jakupca, Hydrogen Systems Risk and Reliability Workshop, University of Maryland, 13 September 2024<br />
UMD HyTERP Hydrogen Risk and Reliability.pdf<br />
https://ntrs.nasa.gov/api/citations/20240011501/downloads/UMD%20HyTERP%… />
(accessed October 2025)Some quantitative data are provided by a NUREG/CR-3551<br />
https://www.nrc.gov/docs/ML2012/ML20126D044.pdf<br />
(accessed October 2025)Also uptaken (withour date), by H2TOOLS<br />
https://h2tools.org/lessons/hydrogen-tube-trailer-explosion<br />
(accessed January 2026)
JRC assessment
- Sources categories
- Investigation report