There are several concerns with “snuffing” a hydrogen fire from a vent stack. Most importantly, snuffing a hydrogen fire before the hydrogen is isolated can lead to the buildup of a hydrogen vapor cloud, which may then re-ignite, especially with hot surfaces available from the previous fire. The largest hazard is an explosion of the vapor cloud caused by delayed ignition.  It’s always better to isolate the hydrogen at its source to extinguish the fire as fuel runs out. 

Snuffing systems have been used in the past for vent system outlets mainly due to the negative   perception of a visible hydrogen flame at the top of the vent stack, particularly at night.  The success of these systems was marginal since high and sustained rates of inert gas were required to snuff the flame and sufficiently cool the piping outlet to prevent the venting flow from reigniting.  Generally, it’s preferred to design the vent system such that it can withstand a worst-case continuous fire on the outlet without affecting its integrity or surrounding exposures.  If those criteria are met, then it’s inherently safer to allow the vent to burn than to try to snuff it. 

Codes and standards to address issues like this one are under development, along with applied research and field trials. As with any new application, appropriate codes and standards must be developed to meet public risk targets.

No, this is not a common or preferred approach. Isolating the source of hydrogen is the best safety practice. Water systems could extinguish the flame but allow the gas to continue leaking and result in an explosion if reignited.

Sprinkler systems and other fire suppression means are prescribed per building and fire codes to limit fire spread to other materials. In the case of a hydrogen leak and fire, it is best practice to isolate the hydrogen source, and let any residual hydrogen gas burn out. Even if the initial fire is extinguished, additional leaking hydrogen may accumulate and ignite with the potential for an explosion. 

The potential of an explosive atmosphere is inherent with any vent system and must be addressed through adequate design. Purging for most vent stacks is impractical due to availability or cost. In addition, and particularly for LH2 systems, the purge gas can cause potential safety issues. The primary way that explosive atmospheres are addressed is through ensuring that the design of the vent system can withstand an internal deflagration or detonation. This is not that difficult for smaller systems (less than 6”) but can be challenging when vent systems are larger and/or operate more as ducting than pipe. Where the vent system can’t be built strong enough for the potential internal overpressure, purging can be a necessary and prudent safeguard.

Additionally, the amount of O2 in the vent stacks is typically small (i.e. 1.22 scf /.1 lbs. in a 3” dia/25 ft tall vent stack). As hydrogen flows into the stack the time that there is a flammable (between 4 and 74%) region within the vent stack is also small.

For a detonation there must be the correct amount of hydrogen and oxygen. In a 3” vent stack, 25 ft tall there is ~ 1.25 cu ft of oxygen at atmospheric pressure. (=.1 lbs/.0032 lbmoles). The flammable range of H2 is 4-74% H2. At the stochiometric ratio, there is ~.0064 lbMoles of H2 that can react with the O2 in the vent stack. This represents ~.013 lb of h2 that can react. This is quite small amount energy release.

Calculations

Radius – 1.5”
Piping Volume = (1.5/12)^2*3.14*25 ft = 1.22 scf
Weight – 1.22 scf/12.08 scf/lb =.1 lb
Moles - .1 lb/32 lb/lbmole =.0032 lbmoles

H2 + ½ O2 = H20
.0064+.0032 = .0064
.0064 lb moles H2 X 2lb/lb mole = .0128 lb H2

Several organizations published a paper together on this topic in 2017 (see attached). Based on comparisons with tests and CFD simulations, the following conclusions were drawn:

1. The gas concentration for vapor cloud explosion blast load calculations for H2 jets can be limited to approximately 10% to 75%. Note that testing for H2-air VCEs in congested environments has been performed by organizations such as Baker Risk and concluded that 10% is the lowest H2 concentration that needs to be considered. This published this as well.
2. For ignition of the H2-air jet at 30%H2, a mass release rate of about 0.5 kg/s is needed to get above a TNO Muli-Energy Severity Level of 4 (i.e., where VCE blast load perspective starts getting significant with a maximum overpressure of 0.1 bar). This corresponds to a flame speed of about 100 m/s and is shown in Figure 13 of the attached paper.
3. Ignition of the H2-air jet at 60%H2 (worst-case ignition location) requires a mass release rate of about 0.1 kg/s (100 g/s) to get above a TNO Muli-Energy Severity Level of 4. More testing on this has been done and is being done, so these might get refined in the future, but it is not expected that there will be major changes in the “threshold” mass release rate
   needed to produce a jet that (if ignited) can represent a VCE hazard. Of course, the blast loads from a hydrogen jet won’t extend a long distance because the explosion energy (i.e., flammable cloud size) is limited compared to traditional VCE cases (e.g. where a large flammable cloud fills all of a refinery process unit). Lastly, if a facility owner defined a hazard level of concern (e.g., greater than 0.5 psig at 100 feet), then a mass release rate of concern could be calculated.

1.    As of January 2024, we are not aware of any public data on incidents or investigations where a hydrogen fired steam boiler exploded.

2.    The potential for detonations within a boiler tube would depend on both the equivalence ratio of the hydrogen present and the diameter of the boiler tube.
a.    At a minimum, if the circumference of the tube is smaller than the detonation cell size, then a detonation cannot propagate in the tube (experimentally, the critical diameter might be significantly larger).
b.    If you have a very large diameter tube, but the concentration of hydrogen is below the limit for fast flame acceleration (something like ~10-12% vol. H2), then the hydrogen-air mixture cannot run-up to detonation.

3.    There’s a good database of detonation cell sizes and critical tube diameters at: https://shepherd.caltech.edu/detn_db/html/db.html
a.    Here’s an example of critical tube diameter data for hydrogen-air mixtures: https://shepherd.caltech.edu/detn_db/html/H2-Air11.html
b.    Here’s an example of detonation cell size for hydrogen-air mixtures: https://shepherd.caltech.edu/detn_db/html/H2-Air1.html

4.    For information on pressure loads in tubes resulting from a detonation, there’s information in NFPA 67.
a.    The peak pressure would be related to the CJ detonation pressure of the mixture that forms.  Not applicable to a fire tube boiler, but for other geometries there could be regions where pressures significantly higher than the CJ detonation pressure could develop due to shock reflection at end caps/elbows.
b.    The pressures would be significantly higher in the region where the deflagration transitions to a detonation.
c.    The CJ detonation pressure of a mixture can be calculated with tools like the Shock and Detonation Toolbox: https://shepherd.caltech.edu/EDL/PublicResources/sdt/
d.    Even without a detonation, a fast flame propagating within a tube can generate maximum pressures on the order of the constant volume explosion pressure of the mixture, which can be estimated by a chemical equilibrium solver like Cantera or NASA CEA.

5.      For the pressures where a DDT occurs (i.e., where the pressure can be significantly higher than the CJ pressure), we have seen this in incident investigations, and put out a paper illustrating this. These loads extend over several pipe diameters and have significant associated impulse (i.e., the structure containing the mixture is likely to respond to the peak pressure).
Geng, J. and J.K. Thomas (2012) “Pressure Distribution Inside Pipes Due to DDT,” PVP2012-78590, ASME 2012 Pressure Vessels and Piping Conference, Toronto, July 15-19, 2012.

6.      If you fill the boiler with an H2-air mixture, a DDT can occur.  A fairly applicable example would be a test we ran at very low congestion, which may be representative of the congestion in a fire tube boiler, within our DLG test rig 48 ft long x 24 ft deep x 12 ft high (15 m long, x 7 m deep x  4 m high), with one long face open as a vent.  We got a relatively strong deflagration at 20%H2.  We got a DDT at 22.5%H2.  A paper describing these tests:
Horn, B.J., O.A. Rodriquez, D.R. Malik and J.K. Thomas (2018) “Deflagration-to-Detonation Transition (DDT) in a Vented Hydrogen Explosion,” 14th Global Congress on Process Safety (52st Loss Prevention Symposium), AIChE Annual Meeting, Orlando, FL, April 22-25, 2018.

7.    The DLG tests described above were performed with the entire test rig filled with a relatively uniform and quiescent mixture.  In an accidental scenario, the boiler could have a non-uniform concentration and, depending on the scenario, only a portion of the boiler may be filled with a flammable mixture.  In this case, we would normally turn to computational fluid dynamics (CFD) analysis using the FLACS code.  We have developed a criterion for evaluating the FLACS results to determine if a DDT would occur.  An example of the application of this approach for a H2-air explosion within a vaporizer set is described in:
Thomas, J.K., J. Geng, O.A. Rodriquez, et al. (2018) “Potential for Hydrogen DDT with Ambient Vaporizers,” Mary Kay O’Connor Process Safety International Symposium, College Station, TX, October 2018.
Relative to the point above, please note that some experts do not concur with using FLACS for DDT analysis. That being said, we have gotten a reasonable match to our VCE test data using this approach.

8.    Relative to natural gas fired steam boiler failures due to internal explosions, some work we did relative to reformers is somewhat applicable, although we did not establish the type of frequency information he is looking for:
Maxwell-Shaffer, D.F., A.G. Sarrack and J.K. Thomas (2014) “Unusual Reformer Events and Modeling,” 2014 AIChE Safety in Ammonia Plants and Related Facilities Symposium, Vancouver, September 2014.

See attached files for several references.

Yes, these would be ignitable mixtures.  In this case, it does not appear complicated geometry is involved, so 1200 psig pipe should be more than adequate to protect against internal deflagration. The most likely scenario is a "backfire," similar to a car, where ignition occurs too soon and shoots out the open end of the pipe. Consider using an inline deflagration flash arrestor on the supply line to protect upstream piping and equipment. Also, make sure pipe welds, fittings, and instruments have a comparable rating. The pipe needs to withstand an over-pressure resulting from an ignition at the mixing point of H2-O2. This is the issue that necessitates a high pipe pressure rating, preferably high enough to withstand a detonation.

The nominal H2-air detonation peak pressure is about 10-20 times the initial pressure, but there are also possible reflected pressures, a comfortable margin is needed to withstand shock wave pressures above the nominal detonation peak pressure. It’s possible for reflected pressures to get a little higher, but these generally remain within 20X. If a pipe is at atmospheric pressure, 300 psig will result in 20X the pressure, so that is the basis for the 1200 psig pipe having sufficient margin.