Hydrogen Explosions in Nuclear Reactors and a Passive Way to Prevent Them

By: Robert E. Buxbaum, REB Research & Consulting co

The explosions that happened recently at the Daiichi nuclear plant were hydrogen explosions. As president of a hydrogen products company, I thought it worthwhile to write an essay on where the hydrogen comes from, and on a retrofit product we’ve been selling for the last ten plus years for passive nuclear safety in this regard.

The Daiichi reactors, like most nuclear reactors, are made of metal and cooled by water. This reactor is mostly zirconium alloy; many other reactors are made of stainless steel. This plant was cooled by light water, H₂O; many others, particularly in Canada, are cooled by heavy water, D₂O. Whatever the metal, and whatever type of water is used to cool it, all metals of nuclear construction corrode in water at some rate – usually slow -- in the presence of intense radiation and hot water or steam. This corrosion results in the formation of a metal oxide, usually adherent, and hydrogen gas that enters the liquid or steam coolant. Corrosion is worse during a nuclear accident where cooling power is diminished and temperatures are higher, but hydrogen-producing corrosion is a normal part of all reactors, occurring even when the reactors are shut down.

For most reactors, most of the time, this hydrogen generation is merely a nuisance. The hydrogen contributes to hydrogen embrittlement and swelling of the nuclear materials – shortening the reactor life over time. It also induces excess pressure in the steam line equal to the partial pressure of the hydrogen. In boiling water reactors, the presence of hydrogen interferes with heat transfer process from the steam and thus must be vented. This is a normal part of normal operation, relieved by venting a small amount of steam hydrogen-steam mix from a cooler part of the line. The ideal is to capture this steam before it reaches the environment, but for a light water reactor, this steam is only mildly radioactive, and most plant licenses allow a small amount of steam venting directly to the environment without further cleanup. Occasional, small explosions are not uncommon since the steam contains a significant content of hydrogen, especially when vented at temperatures, below the condensation point of water, but above the flash-point of hydrogen.

The explosions that removed the roofs of Daiichi reactors 1, 2, and 3 are larger scale versions of these, normal explosions. The fuel and water heated up beyond their normal limits when the cooling pumps shut down; corrosion sped up beyond its normal limits because of the high temperatures; the hydrogen output increased beyond its normal limits because of the increased corrosion. And, when released, this larger amount of hydrogen exploded with a lot more force than is normally seen. At Daiichi reactor 4 there has not been an explosion, but instead two hydrogen fires. It is likely that low-pressure, corrosion-hydrogen is what’s burning.

In a sense, these hydrogen release processes are very good for reactor safety. As with the lancing of a boil, the release of these hydrogen-streams reduces the high pressure that could otherwise destroy the reactor vessel. Also, this hydrogen removal improves the heat transfer properties of the steam inside the reactor, lending a greater degree of passive cooling. Still, there are less explosive ways to remove the hydrogen and reduce the pressure inside the reactor vessel. One of these is a high-pressure membrane that our company has been making for the last ten years or so.

We hold a patent on a high-pressure low-temperature, hydrogen extraction membrane that’s found some use in nuclear reactors to extract hydrogen from steam at pressures up to 1500 psi, and temperatures down to room temperature. These have been tested in a variety of nuclear reactor simulation tests and have been found to extract hydrogen well from the steam inside the reactor, both during normal operation, about 250 C, and for the conditions expected during mild to severe upsets. Hydrogen in the steam diffuses through the membrane material at a controlled rate and (during normal operation) oxidizes to water on the Pd-coated downstream side of the membrane. Since no steam leaves with the hydrogen the radioactive release rate is diminished, and since the hydrogen removal is continuous and small, there is never an explosion.

During a loss of cooling power accident, as happened in Japan, the rate of hydrogen removal through the membrane would increase, and it would be possible to have a hydrogen fire at the exit of the membrane. While such fires are never welcome, they are less damaging than an explosion would be. Removal of the hydrogen decreases the pressure in the nuclear reactor vessel and increases the passive cooling capabilities of the reactor vessel. This is because the heat transfer capabilities of a hydrogen mix is vastly less than the heat transfer capabilities of the same amount of condensing steam. Condensing steam transfers heat through a passive mechanism that’s sometimes called heat piping, where the energy of the heat actually the flow of heat transfer fluid (steam). When hydrogen is present, it blocks the steam flow to the cooler parts of the reactor vessel, decreasing the rate of passive heat removal; the steam in the reactor ends up hotter than it would be otherwise.

Membranes of this sort are sold in the specialty section of our catalog.