Kanellos made a small error in his article (or Trulite did when they briefed him), describing the device as using a chemical reaction involving "sodium hydride" to extract hydrogen from water.
That surprised me, since sodium hydride is extremely caustic (it's a base, the opposite of an acid), strong enough to be dangerous if exposed to air or water.
That put me back onto familiar ground, since Medis Technologies, another company developing fuel-cell products for consumer and commercial applications, also uses sodium borohydride, and I've looked into that company's technology before. In both systems, the hydrolysis of sodium borohydride releases pure hydrogen, which is run through a fuel cell. The hydrogen reacts with oxygen from the air, producing power and water.
Sodium borohydride is safer than sodium hydride, as I said; it's also generally safer than gasoline, which is the most common fuel in the small generators that these fuel-cell systems are designed to replace. But it's only relatively safe; it's still what most people would call a hazardous chemical. An MSDS (material safety data sheet) for sodium borohydride (here) says:
DANGER! CORROSIVE. CAUSES BURNS TO ANY AREA OF CONTACT. HARMFUL IF SWALLOWED, INHALED OR ABSORBED THROUGH SKIN. FLAMMABLE SOLID. DANGEROUS WHEN WET.
So for safety reasons, when sodium borohydride is used in consumer products, it must be provided in some kind of sealed canister, which probably receives the waste from the process. If not, there must be another canister involved, since there's definitely some waste to deal with.
Trulite and Medis don't talk about the waste produced in their systems; ordinarily it stays sealed up in its canister so users don't have to worry about it. But I did a little research, and it looks like the waste is sodium metaborate tetrahydrate, a relative of good old borax, the stuff Ronald Reagan used to advertise on TV (like here). It isn't very dangerous, but users will still need to be careful about these canisters, in part because the original sodium borohydride may not be completely consumed when the fuel cell can no longer generate useful amounts of power.
I suggest that users should check with their local garbage companies to see if they're allowed to discard these canisters in household trash. (And of course you know most garbage companies are going to reply with a immediate "no!"; it may take months to get a more carefully considered response.)
It seems to me that Trulite and Medis want people to focus on the fuel cell part of the product because fuel cells are thought of as very "green"-- to the extent consumers know anything about fuel cells, they probably know the traditional refrain that "fuel cells emit only water."
But with these things, the result is water plus a canister full of gunk you won't want in the house and can't throw in the trash. That is not so green.
My other objection to sodium borohydride-based fuel cells is that they're not very efficient over the total cycle. Sodium borohydride has to be manufactured, and the manufacturing process takes a lot more energy than the user will ever get out of the fuel cell. Then there's the canister itself, and the costs of shipping the canisters around.
The Trulite KH4 described in Kanellos's article weighs 20 pounds and generates not more than 200 watts of power. The company's website doesn't talk much about the energy capacity of the system, but a December 2005 SEC filing (here) says the KH4 (apparently that stands for Kitty Hawk 4) can produce 100 watts for 4 hours. (I should note that the KH4 may have been substantially improved since then, but since the company isn't saying, I have to go with the information I can get.) That's as much energy as eight average laptop batteries-- a significant amount-- but charging those batteries would cost the user about a dime. Trulite's replaceable Hydrocell canisters surely must cost at least 100 times that much. (And I wouldn't be surprised to hear they cost 1,000 times that much, given the high price of the KH4.)
The Medis 24-7 Power Pack, just now coming to market, is a 6.5-ounce cellphone-size gizmo that produces 20 watt-hours of energy and costs about $30. That's about the size, weight, cost, and energy capacity of 8 AA NiMH batteries, but those batteries can be recharged hundreds of times. When the Medis device runs out of energy, it has to be thrown away.
If these devices become popular, manufacturing costs will come down, and so will retail prices, presumably. But for now, you'd have to have a really extraordinary reason to want one.