Okay, so does this mean,hydrogen is generated at 50% solar effieciency so it can be burned in an IC engine at 50 %? giving an overall effiency of 25%?
This is vs electricity and solar efficiency of 20% and charge to wheels of an EV car of 70%, giving an overall eff of 14%!.
THEREFORE, It is better to solar generate hydrogen than electricity!
I guess America will become a hyrogen economy!

IC engine efficiency is only about 30% not 50%. So overall its 15% not 25%. Unless on one sunny day fuel cells suddently become cheap and affordable.

And add, how much the catalyst cost? How long does it lasts? Lets wait them to put it into a hydrogen factory first.

Interesting ... but what's this business about a "sacrificial electron acceptor"? Doesn't sound like a catalyst to me. Also could wish that its quantum yield spectrum had better overlap with the solar spectrum, but it's still early days -- maybe that will change.

Interesting. And I just found this.

Combine water with an electrolyte, and run current through the solution, forcing the water molecules to split into hydrogen and oxygen gases. But electrolysis-formed hydrogen has long been hampered by the high capital cost of the metals used in the process, around “thousands of dollars per kilowatt,” says Richard Bourgeois, GE’s electrolysis project leader. GE’s breakthrough comes from a proprietary material called Noryl, a highly chemical- and temperature-resistant plastic developed by the GE labs, that lowers the cost of hydrogen production to hundreds of dollars per kilowatt, according to Bourgeois.

Although GE has only built a prototype in their lab, Bourgeois believes that demonstrations can come as soon as the end of next year, and commercialization will follow that. The goal of the project, according to Bourgeois, is to bring down equipment costs enough to take the cost of hydrogen from $8 per kilogram to $3 per kilogram—comparable in energy and price to a gallon of gasoline.

And remember, when Honda announced the 2009/2010 next-gen FCX, it claimed a fuel efficiency of 70 miles/kgH. (Miles per kilogram of hydrogen. Yikes! What a unit!)

If GE can get the cost of hydrogen down to even $5/kg, that yields a fuel cost of 7.1 cents/mile, slightly better than the 7.5 cents/mile for a 40MPG car running on $3 gasoline.

I think we're going to see one heck of a footrace between EV's and hydrogen fuel cells in the coming years. To me, EV's feel like the eventual winner, but if hydrogen can do the job, great. Plus it would have the benefit of making all the people saying hydrogen will never work look like the Y2K doomers...

Well, Lou, that would be a switch, given that we currently appear to be on track for having the hydrogen supporters look like the Y2K doomers...

You have misquoted the article regarding the use of the Silver Nitrate in a way that changes the meaning. The actual quote:

"If silver nitrate is used as a sacrificial electron acceptor, the quantum efficiency for oxygen evolution rises to 51% at 420–440 nm, which is 20 times higher than that for overall water splitting."

Essentially they are burning Silver Nitrate.

Note that mother nature long ago figured out how to split water into hydrogen and oxygen at low temperatures: it's a key component of the photosynthesis cycle. That's why biofuels have a positive energy balance - admittedly, some more so than others - and efforts based on silver nitrate do not.

While I am skeptical about hydrogen, I am not sure that itis completely relevant whether or not it has a positive energy balance any more than whether or not electricity stored in a battery inside a vehicle has a positive energy balance.

Hydrogen is, in essense, a storage medium, comparable to a battery that stores electricity. Hydrogen can be created with off peak wind power, for example, and perhaps should be compared to the electricity that would be created at night to recharge a PHEV. The question for the future, I think, is which process is more efficient.

Hydrogen has problems with respect to range regarding how much hydrogen can be practically and econonomically stored in a tank and how much leakage there will be from that tank. But then batteries have similar issues.

Hydrogen is, in essense, a storage medium, comparable to a battery that stores electricity.

This is of course correct. A catalyst cannot change the thermodynamics of the system, only the kinetics. We'd still take a 2nd law efficiency hit with hydrogen. But at least this way it uses energy that's being radiated no matter what we do.

Note that mother nature long ago figured out how to split water into hydrogen and oxygen at low temperatures

Also correct, and more interestingly it happens over a much wider wavelength range. Even more interestingly, if you consider the light energy required to reduce CO2, you need a wavelength of no more than ~250 nm, deep into the UV. No deep UV reaches the surface, so nature figured out how to do this with multiple lower-energy photons (seriously nontrivial doesn't even begin to come close to describing the process.) Pretty clever for blue-green algae!

"Pretty clever for blue-green algae"

Although algae is not really dirt, perhaps we should retire the phrase "dumb as dirt". Also, maybe we should have respect for pond scum.

Zinc doesn't have the storage problems of hydrogen and potentially could have a higher energy density (kg/l) than gasoline.

If I read it correctly, the 51% efficiency is only for a small part of the spectrum. The efficiency of solar panels is determined against the full solar spectrum. So their yield (Watt per square m) is probably better. That's the only thing you should be interested in.

how exactly do you split hydrogen and oxygen?