Event
- Event ID
- 944
- Quality
- Description
- A 50-litre standard hydrogen gas cylinder was temporarily placed and used in a laboratory. The hydrogen gas was used for a flame ionization detector (FID) in gas chromatography (GC) instrument.
The safety relief valve on the pressure regulator blew open and released of about 340 g of hydrogen into a laboratory. The gas cloud did not ignite so there was no injury or damage.
In Henriksen et al (see references) a full investigation is presented to verify the cause of leakage and estimate the gas concentration of the dispersion and gas cloud, including the modelling of a likely explosion. If the gas cloud had ignited, the explosion pressure would most likely cause significant structural damage. - Event Initiating system
- Classification of the physical effects
- Unignited Hydrogen Release
- Nature of the consequences
- Leak No Ignition (No additional details provided)
- Macro-region
- Europe
- Country
- Norway
- Date
- Root causes
- Root CAUSE analysis
- The INITIATING CAUSE was the degradation and leaking of a pressure regulator.
The temporary safety setting of the lab functioned as designed, and the detector activated on hydrogen presence before further escalation.
However, due a misunderstanding within the organisation, a bigger hydrogen cylinder was installed than the one agreed during risk assessment. Moreover, the regulator had never been the object of an inspection and maintenance plan. All this suggests a ROOT CAUSE in shortcoming of operative and safety procedures.
Facility
- Application
- Laboratory / R&d
- Sub-application
- hydrogen laboratory
- Hydrogen supply chain stage
- Hydrogen Storage (No additional details provided)
- All components affected
- hydrogen pressure regulator, 200 bar compressed hydrogen bottle,
- Location type
- Confined
- Operational condition
- Pre-event occurrences
- The condition were abnormal because of the temporary decision to isntal compressed gas in a lab.
Emergency & Consequences
- Number of injured persons
- 0
- Number of fatalities
- 0
Lesson Learnt
- Lesson Learnt
- According to Henriksen et al. (see references), the were:
1. Under the abnormal condition of operation of the lab, the planned reduction of the total hydrogen quantity to a small 10 l bottle would have reduced the released hydrogen by 55% even in the case of complete depleted.
2. The presence of a hydrogen detector and alarm may have saved the laboratory from a severe accidental explosion. Without it, the whole hydrogen gas cylinder would probably have been released into the room, creating a 1.2 meters high gas cloud with 15 % stoichiometric hydrogen.
3. The age of the hydrogen gas regulator used was difficult to determine by reading the numbers printing on the back. The age was first established after contacting a supplier. This is an important information to determine the end of life of lab instrumentation.
4. Related to point 3., there is a need for regular maintenance/check of the gas regulators . Following the incident, a service has been established.
5. A pressure safety valve on the gas regulator is not a safety feature when the safety valve releases the gas into an environment that contains several ignition sources. To work as a safety feature the gas needs to be routed to a safe ventilation point. A similar event happened at a commercial facility in 05/03/2013, where a failure of the pressure safety valve occurred. One of the conclusions from the investigation was that if the safety valve shall function as a safety feature the gas needs to be released in a safe location.
Event Nature
- Release type
- gas
- Involved substances (% vol)
- H2 100%
- Actual pressure (MPa)
- 20
- Design pressure (MPa)
- 20
- Presumed ignition source
- No ignition
References
- Reference & weblink
M. Henriksen et al, <br />
"Accidental hydrogen release in GC-laboratory; a case study", <br />
Int. Conference on Hydrogen safety, 2015 October 21-25, Yokohama, Japan
JRC assessment
- Sources categories
- Scientific article