Some rate equations for hydrogen diffusion through Pd-Ni-Cu alloys.

One of the more remarkable things one hears when one begins to look into the element palladium is the remarkable fact that hydrogen gas diffuses right through the metal.

Being aware of this property when I first heard of the now notorious announcement of Fleichmann and Pons that they had "discovered" "cold fusion" - reported by credulous journalists with bad science educations - it was the first question that I asked myself. Were this true, how come it had never been observed in diffusion experiments using palladium - I assumed that they must have been done if for no other reason than to investigate facile fast isotope separations - using tritium or deuterium in other settings? Surely many similar experiments had been done over the years, and no one had produced a huge amount of heat in this way.

We now all know that "cold fusion" as explained by Fleischman and Pons was probably worse drivel than the sort of stuff that dribbles out of the mouth of Amory Lovins, mystic natural gas salesman.

But palladium diffusion of hydrogen is a real effect.

Most of the world's hydrogen today is made from dangerous natural gas, although about 1% is made as a side product of the chlorine industry by electrolysis.

The dangerous natural gas reaction is a reformation reaction, in which steam (or supercritical water) is heated with methane to give either a mixture of CO gas and hydrogen, or carbon dioxide and hydrogen, the latter through water gas shift reactions. If one desires pure hydrogen, say to run one of Amory Lovins' hydrogen HYPErcars http://news.nationalgeographic.com/news/2001/10/1016_TVhypercar_2.html">that will be in showrooms by 2005, one will use the shift reaction, strip off the carbon dioxide with a suitable stripping agent like monoethanolamine, and dump the dangerous fossil fuel waste carbon dioxide into the waste dump (aka "the atmosphere") before shipping the hydrogen to the HYPErcar fueling station, or one of Arnie's Hydrogen Hummer Stations that are funded for Arnie's pleasure in the California budget.

(One hopes the fires don't get there before the Hummer.)

So if one desires pure hydrogen separation from side products is not particularly difficult - as long as one is not required to provide a permanent carbon dioxide waste dump other than the atmosphere - making it clear that the use of the relatively expensive and relatively rare metal palladium for hydrogen separations is not of much industrial interest, except maybe for exotica like, well, isotope separation.

Palladium, by the way, although expensive, is a constituent of so called "nuclear waste," although such palladium as is obtained - unless obtained from the decay of isolated ruthenium-106 - is mildly radioactive.

Nevertheless there is a claim in the literature that the use of palladium - possibly in a type of membrane process known as "pressure swing absorption" might be so used because of a perceived energy advantage, particularly if one can use a cheaper alloy of palladium.

A paper in the Journal of Membrane Science that talks about this matter and gives equations for flow rates across such membranes.

Here's the abstract: http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6TGK-4JFH6T1-3&_user=10&_coverDate=09%2F01%2F2006&_alid=1002614176&_rdoc=1&_fmt=high&_orig=search&_cdi=5257&_sort=r&_docanchor=&view=c&_ct=13&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=6a492bc92f8543f69cc352ad7444b321">Journal of Membrane Science
Volume 280, Issues 1-2, 1 September 2006, Pages 883-888>

It turns out that the rates of diffusion unsurprisingly depend on pressure differentials and of course the nature of supports and surface area.

Diffusion rates turn out to be on the order of magnitude of 0.01 mol-m² sec^-1, not particularly impressive for solids although possibly of use in a pressure swing system with a high support surface area particles.

But like I said, monoethanolamine works just fine for this sort of thing in any case.

Whatever.

Membrane and extraction separations for other gases are of commercial and industrial importance however, and this kind of membrane science is very important for many applications.