Hydrogen and oxygen only react at significant rate at high temperatures. In a hydrogen explosion, the rate initially starts off low, but as the reaction is highly exothermic, heat from the reaction increases the temperature and causes the reaction to accelerate until reactant concentrations are sufficiently depleted.

This means that hydrogen and oxygen can only react quickly if the concentration of hydrogen in a volume is high enough that the increase in temperature from the exothermic reaction is faster than the dissipation of heat, so the temperature will rapidly expand, causing an explosion. For this reason, hydrogen tends to build up and then explode, rather than burning continuously.

It might be possible to trap hydrogen in a small volume and burn it in a series of small explosions; however, I don't think the reactors are designed that way.

Couldn't you stick a piece of platinum at the mouth of a vent and just vent it steadily?

IANANS, however from what I understand the problem is this: -> The reactor is a BWR type, hence even though it is scrammed it is still generating some amount of power from fission products. -> The ECCS failed, meaning that the core can't maintain sufficient cooling. -> So what water was left in the core is then exposed to very high temperatures and pressures. -> That apparently then creates conditions sufficient to cause the water to react with the zirconium alloy which is used in the fuel rods, forming zirconium dioxide and hydrogen. (2H2O + Zn -> ZnO2 + 2H2). -> As reactors generally don't create much gaseous waste (The only waste products that I can think of would be, for example, radon, and that would be in small quantities), and the core isn't expected to be above the autoignition temperature of whatever gas that is being vented. What could be happening is that the gas is being vented but is being concentrated at some point where it comes into contact with something hot enough to ignite it thus causing an explosion. For example, if the gas is being vented out of the core, but into a secondary containment vessel, if the vessel is filled with air and if it contains say a piece of metal hot enough to ignite the hydrogen, it would simply fill with gas until it reaches a point where the mixture is explosive, before then it then explodes.

Disclaimer: I'm not a nuclear scientist, any corrections/additional information would be very welcome.

As I understand it, anything that gets vented from the outer structure first goes through an activated charcoal scrubber to remove radioactive contaminants. This may limit the rate at which gas can get vented.

Oxygen and hydrogen form explosive mixtures over a very wide range of relative concentrations, which doesn't help.

Any amount of hydrogen vented into the outer building could be burned with environmental oxygen, regardless of how much of it available in the exhaust. Can't be that hard.

Can't be that hard, really? I can't say one way or another but I'd kinda expect the experts to have thought of this at some point if it was a trivial possibility.