Sunday, June 3, 2007

Beware celebrated scientists bearing good news

It has become my custom of late to listen to improving podcasts during my daily constitutional as opposed to the more rhythm-sustaining loud music I was won't to do. Just to clarify the meaning of constitutional, I am using the English vernacular which means exercise; in a recent conversation with a native I learned that constitutional might be taken to imply a much more intimate activity normally associated with one's early morning preparations. Let it be known that I am very much with Kenny Williams, the White Sox GM, on this one: I need my maximum focus in that department, and improving podcasts would constitute, so to speak, a substantial distraction. In any event, I appear to be digressing from the main thrust. One of my favourite podcasts was Scipod, produced for more than a year by New Scientist, long my preferred scientific publication. The podcast shared much in common with the parent publication, being informed, engaging, accessible and witty. There was one particularly memorable discussion on sources of domestic injuries and trousers, it appeared, presented a greater source of injuries than kitchen knives. So, imagine my disappointment when it recently signed off with the message that it was shutting up shop.

As an alternative I have been road-testing Science Friday hosted by someone with a name that sounds like Ira Playdough, but perhaps I am hard of hearing. One of the topics this week was about hydrogen - a very timely business, particularly since my summer research at the mighty Argonne is on the very subject of fuel cells. In this segment, Jerry Woodall, a highly decorated pioneer of semiconductor electronic devices. Success in that field does not necessarily prepare one for ground-breaking progress in alternative energy. The premise, I thought, was mistaken in that the biggest problem for hydrogen-propelled fuel cell cars was the danger of the hydrogen fuel tank. I would have agreed with the biggest problem being the generation of the hydrogen in the first place economically, assuming all other issues with fuel cells are sorted, which is far from being the case. So, Prof. Woodall has a cunning solution, although by no means an entirely novel one based on other work I have seen: generate the hydrogen insitu by adding a reactive metal to a tank of water. There is a certain magic sound to this: the use of water as a fuel. I have learned from some cursory research that there is much myth, hype and outright mischief associated with the pursuit of the water powered vehicle - more of that to follow.

The chemistry here is trivial and the metal selected is aluminium (Al), suitable for both its reactivity and low molar mass. Of course, as any devotee of canned beverages can attest to, Al cans are not at all reactive in fact due entirely to a very thin but impervious passivating layer of oxide that forms rapidly on any fresh Al surface. Without it, the widespread application of this supremely abundant and low density metal throughout society could not have happened. Woodall's trick was to recognize that an alloy of Al with a heavy member of group 3, gallium (Ga), did not possess this passivating layer, and that it reacted readily with water to generate hydrogen.

All very well I thought, but in his enthusiasm to promote this putative automotive "technology" a few questions remain unanswered. One of the great attractions of hydrogen as a fuel is its energy density - a mammoth 34.2 kcal/g compared with 8.7 kcal/g for gasoline, which is no slouch in its own right on that account. So, I have to ask, why would you want to replace compressed hydrogen, for which there is no real evidence of any danger greater than with conventional fuel, with a tank of water and metal to react with it. Consider the numbers. In the water molecule (H2O) the H accounts for only 1/9 of the mass. Now throw the metal into the equation. 54 grams of Al would be required to obtain 6 grams of H2 from 54 grams H2O. So now we are down to the fuel being less than 6 % of the total mass. The energy density is now down to less than 3 kcal/g - almost five times worse than conventional fuel. And I am not even factoring in the gallium which only makes matters worse. Woodall spoke of using a little gallium but his website shows graphs with Ga contents as much as 80 %.

There are other questions. Woodall is definitely correct in saying that Al is abundant; but back in the day when I was working on alumino-silicates like zeolites, we were substituting the Al by Ga but knew that this was not really an industrial possibility because of the limited availability of this element. What gives one to think that something unsuitable for widespread deployment in catalysis could be suitable for transportation?

The reaction of Al with H2O will produce the H2 fuel for the fuel cell; but it will also produce a quantity of heat. It is not clear how this heat will be utilized if at all.

The process is said to be easily recycled. "Cheap" electricity will reverse the aluminum oxide product to the metal for reuse using a "highly efficient" electrolytic process. Of course recycling of Al is one of the most economically viable processes in the pantheon of recycling but that is largely because of the high cost of producing virgin materials. It is an energy-intensive process because of the high stability of the oxide. Factor in also the transportation costs and logistical problems associated with recycling massive quantities of the Al2O3 residue. After all this "cheap" electricity will only reside near the windfarms, waves, nuclear plants or solar panels that will generate it, even presupposing that these sources will indeed be cheap. I am not entirely sure that will be the case. Has nuclear ever been cheap? Maybe I am missing something, but I cannot see how this approach gets off the ground compared with using pure hydrogen off the bat.

I mentioned earlier an abundance of nonsense associated with water as a fuel. I intend to pursue some of this in future notes.

1 comment:

Snipper "Book Zeller" said...

Perhaps a experiment of this nature can be performed on one of the all day wednesday lab fests.

Indeed, it is a very interesting concept. I am still stuck on how a thin oxidized layer would not form. My previous experiments with my favorite element have often left a oxide layer and while different in concept... it still makes me ponder.

I may have to bust out and consult Munowitz