Is Hydrogen really so dangerous?

My mum hasn’t been feeling too well recently, and she’s had to start using Oxygen. Thus when I was over in Ireland a couple of weeks ago I had to get familiar with the use of oxygen tanks. It led me to ask the question, which is more dangerous, oxygen or hydrogen?

Figure 1: The Hindenburg [Source: Lakhurst, 1937 ]

Figure 1: The Hindenburg [Source: Lakhurst, 1937]

Now most people would say hydrogen, after all look what happened to the Hindenburg. However, this ignores the fact that in order for a fire to start you need three things, a fuel (which admittedly a football field sized tank of hydrogen adequately provides!), a source of ignition (its believed that the actual fire started in the Hindenburg’s skin, likely as a result of static electrical discharge, which then spread to the hydrogen) and of course an oxidiser….And there’s the problem, pure oxygen is a about as good as you can get as oxidisers come!

Figure 2: Nitrous Oxide tanks, there’s a reason why boy racers use them, as injecting NOx into a car’s engine provides more oxygen and thus more combustion [Source: ]

Figure 2: Nitrous Oxide tanks, there’s a reason why boy racers use them, as injecting NOx into a car’s engine provides more oxygen and thus more combustion [Source:]

Inject oxygen into any environment, or leave a tank of the stuff leaking into your home can therefore create all sorts of fire hazards. This is not helped by the fact that oxygen (being slightly denser than nitrogen that makes up the bulk of air) tends to linger, while hydrogen (being lighter than air) tends to be extremely buoyant (i.e. it just leaks straight up and escapes into the atmosphere).

The hazards presented by a concentrated oxygen environment were vividly illustrated by the Apollo program. You’ve probably heard the joke that during Apollo, NASA spent millions developing a pen that worked in zero gravity (actually it was a private contractor), while the Russians just used pencils. Well that wasn’t very funny for NASA at the time. They were terrified of the consequences of introducing graphite or anything else that could cause a spark into the pure oxygen environment of an Apollo capsule. A paranoia that was justified when an idle spark (from an electrical system) set alight Apollo 1 and killed 3 astronauts.

Figure 3: The Apollo one fire [Source: ]

Figure 3: The Apollo one fire [Source:]

Another vivid illustration of the dangers of oxygen was the loss of Valuejet 592 over Florida in 1996. A fire started in a sealed hold compartment. The fire went undetected by the flight crew as the hold in question did not have any means of detecting a fire (given that it was sealed and there wasn’t supposed to be anything particularly combustible inside ( it was assumed any fire would burn itself out before threatening the airplane). However a set of oxygen generators (the sort that make oxygen for passengers in the event of a loss of cabin pressure) were inadvertently placed in the hold. These not only provided the source of ignition (as they are exothermic and generate lots of heat) but also fed the fire with oxygen, quickly creating an inferno that brought down the plane.

Furthermore you also have the issue of tank safety to consider. My mum’s oxygen cylinders are pressurized to 350 bar and are made of cast iron, while hydrogen (at least in vehicles) is generally stored at similar pressure, but in carbon fibre reinforced tanks. I know which of the two I’d least want in the back of a car in an accident!

Indeed a number of studies have attempted to study and compare the effects of a hydrogen leak and fire in a vehicle. One of the earliest of these tests, conducted by the University of Miami (see below) involved breaching and then setting alight a hydrogen tank and the tank of a conventional petrol fuelled car.

Figure 4: Breach Test hydrogen tank (left) compared to petrol powered car (right) [Source: Swain (2001) ]

Figure 4: Breach Test hydrogen tank (left) compared to petrol powered car (right)
[Source: Swain (2001)]

The results showed that while the hydrogen did produce a spectacularly large flame, this carried much of the fuel and heat away from the car, meaning there was little if any secondary ignition and not a lot in the way of toxic fumes. Within a few minutes the pressure in the tank had fallen to the point where it could no longer supply enough gas to support combustion and the fire essentially burnt itself out. The car with the petrol tank however, just kept on burning and burning and burning until there was nothing left but a burnt out shell!

The Zeppelin Menace
Indeed, Zeppelins like the Hindenburg’s predecessors of World War I create an interesting analogy. As discussed in a recent Channel 4 documentary, initially the RAF found the Zeppelins very difficult to shoot down. They would spend out aircraft to spray bullets into the Zeppelins….and nothing happened! The bullets when in one side of the gas bags and straight out the other. This caused a few leaks of hydrogen, although given that the gas was under relatively low pressure such leaks seldom led to any serious loss of lift, and normally the Zeppelin crew could compensate by simply releasing their bombs early or dumping water ballast.

Figure 5: The Zeppelins of World War I initially proved to be fairly difficult to shoot down

The RAF tired again, this time with incendiary bullets. And they also went straight through the gas bag and out the other side. Again returning to our fire triangle, you need fuel, ignition and an oxidiser. As the air outside was unable to mix with the hydrogen (as it could not leak in against the pressure of the hydrogen trying to escape) the incendiaries could not start a fire. The RAF then tried using explosive rounds. These proved a bit better, as they tended to blast large holes in the gas bags causing more serious leaks. Potentially, if they fired enough rounds, they could cause enough hydrogen leaks to force the Zeppelin into a crash landing, however most still made it back over the channel before coming down.

Finally the RAF hit on a winning combination, incendiary bullets and explosive rounds loaded 50/50 into the same ammunition drum. This blasted holes large enough to cause serious leaks, where hydrogen and oxygen could mix, while the incendiaries provided the all important spark. The results were immediately successful and pretty soon being in a Zeppelin over England was tantamount to suicide and the Germans soon switched to using bombers instead.

In Context
I’m not arguing that hydrogen is the panacea to all of our problems, because it isn’t. Notably Hydrogen is just an energy carrier not a source, and it’s the thorny issue of where you get the energy from that’s the issue. Also while I can see hydrogen being used as a replacement for Natural gas (e.g. winter heating, power stations, CHP plant, etc.) I’d argue the jury is still out as to whether it’s the best choice for vehicles.

I’m also not arguing that Hydrogen is a perfectly safe gas, far from it! It is dangerous (just look at the photo above of the Hindenburg if you doubt that!), but it’s a danger we need to put in the proper context. Especially when you consider that most homes have a pipe delivering highly combustible natural gas (otherwise known as methane) into them. Or that we have many hundreds of millions of “self-moving petrol bombs” running around the roads.

Any future hydrogen economy does come with technical issues to be resolved as well as some risks and safety concerns. But the bulk of these risks arise from the fact that hydrogen simply behaves differently to the fuels we are used to using, not because there is anything intrinsically dangerous about hydrogen.


About daryan12

Engineer, expertise: Energy, Sustainablity, Computer Aided Engineering, Renewables technology
This entry was posted in climate change, efficiency, energy, renewables, sustainability, transport. Bookmark the permalink.

4 Responses to Is Hydrogen really so dangerous?

  1. Jason Bascom says:

    The late Dan Simmons had thought to store energy from remote wind turbines, through the conversion of water into ammonia using electrolysis. Hydrogen has storage problems ammonia doesn’t have.

    • daryan12 says:

      As I seem to recall Ammonia has a boiling point of around -30’C, so in the absence of cryogenic storage it would be a gas, which would present a number of the same problems as hydrogen, quite apart from being toxic!

      Further, it would depend on what you were trying to do with it, e.g. make fertilizers, run a car engine, central heating, etc.

      • chrispedohansen says:

        But NH3 is much bigger than H2, so it can’t escape through tank nanopores.

      • daryan12 says:

        I don’t remember mentioning Ammonia in this article. Either way the leakage rate of hydrogen is often overestimated, as it involves rates based on materials (such as certain plastics) which are very porous to H2, while ignoring the fact that many other materials (such as metals or carbon fibres) are much more capable of containing hydrogen.

        I have some past experience with hydrogen cars and we had cars out of service for over a year and yet when we checked the H2 tank, there was still stuff inside at a good few bar of pressure. Keeping in mind our sensor couldn’t record accurately below 20bar, so it had to be something higher than this. Hence we always had to treat the car and the fuel tanks as “filled” except when we were sure they’d been fully emptied or drained.

        In fact, fully emptying a H2 tank ain’t easy you know, the little buggers (h2 atoms) are very buoyant and very good at cramming into nooks and crannies and difficult to flush out!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.