It's not random detonation that is the problem, the issue is containment failures, or anything that causes one. A collision, or stress fracture for example will allow oxygen into the tank. In an atmosphere with enough oxygen in it like ours, that's always a risk.
Oxygen leaking in is not an explosion risk. As noted, it takes 25% O / 75% H before it can ignite. Above 75% H it cannot.
With positive airflow around possible leaks, sufficient H2 to sustain a flame can be prevented. In a collision, gas released goes up and out of trouble.
Exactly, the lower limit, when hydrogen has escaped into the air, not when air has leaked into the contained hydrogen. The article makes clear that above 75% H2, ignition is impossible (but only if you read it).
"Traveling along grain boundaries" can only leak trace amounts. With, as I said, positive airflow, trace leaks can be constantly diluted below the concentration where ignition is possible.
> In a collision, gas released goes up and out of trouble
Except for the cases when there's an open flame nearby or sufficient heat, as is the case in a non-insignificant number of collisions. What if a hydrogen engine collides with a standard petrol engine? Or if the gas becomes exposed to hot surfaces during the oxygen mixing? The gas expansion is also a major cause of explosion. You essentially pull oxygen into the fuel, and rapid fuel-air mixing is a known cause of detonation.
What leads you to think that anything would "pull oxygen into the fuel"? H2 under pressure moves outward, and in a containment breach tends to, instead, increase the H2 fraction above the 75% that can sustain a flame until most of the gas has escaped and leapt skyward. H2 explosions are a concern in enclosed, unventilated spaces.
"This is probably the most absurd myth about the Hindenburg disaster, yet it is frequently promoted by hydrogen fuel advocates."
In addition, hydrogen tanks are not just sitting there in a vacuum. In order for the hydrogen to "leap skyward", you have to have it, again like I said earlier, not come in contact with anything that will induce combustion. Also, you need a perfectly open environment. Your point about the mixture being above 75% just shows a somewhat naïve understanding of the physical chemistry at play here. All it takes is for an area near the fuel supply to hit the right fuel-air mixture ratio, which leads to combustion, as in common elementary school science experiments (the hydrogen 'pop' demo). This event leads to a pulling in of the air and fuel around it, sustaining the exposition.
The Hindenberg disaster, in fact, involved no exploding hydrogen. It was a big fire. Fire can be very destructive. There is no need to invent imaginary explosions to explain destruction from fire.
It is a fact that fewer than half who were aboard died. Those who did die obviously could have been burned, and, despite your absurd and repeated accusation, I never claimed otherwise. Nobody who was not aboard died.
Burning hydrogen moves upward the same as non-burning hydrogen. Gas in contact with flame does not magically become non-buoyant.
We are not, in fact, discussing "hydrogen cars", so whatever "sealed environment" you picture in them is wholly irrelevant.
Had my house once fill up with natural gas for many weeks.
Much to the surprise of the gas company it did not explode.
Of course everyone has severe long term effects from that.
Still apparently hydrogen fires in daylight are basically invisible (the produce mostly infrared radiation). NASA engineers used to walk with a broom in front of them so as not to accidentally walk into a thousand+ degree invisible fire when testing hydrogen equipment.
So perhaps a decent enough oxygen barrier, say mylar, combined with something like iron oxide to absorb any oxygen contamination, might provide a decent safety profile? Interesting to think about.
