Science

New technique for turning sunshine and water into hydrogen fuel

View 3 Images
Artist's conception of a commercial hydrogen production plant that uses sunlight to split water to produce clean hydrogen fuel (Image: University of Colorado Boulder)
Artist's conception of a commercial hydrogen production plant that uses sunlight to split water to produce clean hydrogen fuel (Image: University of Colorado Boulder)
A laboratory model of a multi-tube solar reactor at the University of Colorado Boulder that can be used to split water (Photo: University of Colorado Boulder)
Diagram explaining the process for producing hydrogen fuel from sunlight and water (Image: University of Colorado Boulder)
View gallery - 3 images

A new technique developed by a University of Colorado Boulder team converts sunshine and water directly into usable fuel. The technique involves concentrating sunlight in a solar tower to achieve temperatures high enough to drive chemical reactions that split water into its constituent oxygen and hydrogen molecules. In this way, the team says it should be able to cheaply produce massive amounts of hydrogen fuel.

The team's solar thermal system concentrates sunlight off a vast array of mirrors into a single point at the top of a tall tower to produce very high temperatures. When this heat is delivered into a reactor full of metal oxides, the oxides heat up and release oxygen. The reduced metal oxide now gains a chemical composition that makes it ready to bind with oxygen atoms. Introducing steam into the reactor, which can also be produced by heating water with sunlight, causes the compound to draw oxygen atoms out of the water molecules, leaving behind hydrogen molecules that can be collected as hydrogen gas.

While the concept of using an array of mirrors to concentrate sunlight into a single point at the top of a tall tower is nothing new, being the same technique used in solar thermal tower power plants, there are certain key differences here. Typically, sunlight is concentrated about 500 to 800 times in standard solar power tower designs to reach temperatures of about 500º C (932 º F) and produce steam that drives a turbine to generate electricity. However, splitting water requires temperatures of around 1,350º C (2,500º F), which is hot enough to melt steel.

"You need this high temperature both to give you the driving force to drive the chemical reactions and also the kinetics to make the reactions go fast enough to make the process practical," says Charles Musgrave, Professor of Chemical and Biological engineering at CU-Boulder.

To get those kinds of temperatures, the team added additional mirrors within the tower to further concentrate the sunlight onto the reactor and the active material. While it isn't too different in principle from using a magnifying glass to focus sunlight onto a piece of paper to get it to burn, this setup allows the reflected sunlight to be concentrated by up to 2,000 times. "We are trying to use sunlight to drive chemical reactions that require higher temperatures than combustion," says Musgrave.

The big breakthrough came about when the team discovered certain active materials that allowed both these chemical reactions (reducing the metal oxide and re-oxidizing it with steam) to occur at the same temperature.

Though there aren't any working models, conventional theory dictates that a change in temperature is necessary to make the two different reactions occur – a high temperature for reducing the oxide and a low temperature for re-oxidation. Instead, the introduction or the absence of steam is used to drive the different reactions and certain unique properties of the metal oxide compounds used makes this possible.

"We determined that both reactions could be driven at the same temperature of about 2,500° F (1,371° C)," Musgrave told us. "Even though we run at a constant and lower temperature we still generate more hydrogen than competing processes."

Alan Weimer, the research group leader at CU-Boulder says that eliminating the time and energy required for temperature swings lets them make more hydrogen in a given amount of time. To produce even more hydrogen fuel they'd only need to increase the amount of material in the reactor. "In many respects, our approach is out of the box where prior work was inside of the box using the temperature swing," he adds.

According to the team, huge solar plants spread across many acres could produce much more fuel per acre than biofuels for the same amount of acreage. Another advantage that this process has over other renewable technologies, such as wind and photovoltaics, is that it directs sunlight to directly drive chemical reactions to produce fuel for use in combustion engines or fuel cells. In contrast, photovoltaic processes first convert sunlight into electricity, reducing overall efficiency.

"Our objective is to produce hydrogen (H2) at $2/kg H2," Weimer tells Gizmag. "This is equivalent to about US$2/gallon (3.7 L) of gasoline based on mileage in a fuel cell car versus a combustion engine today." The team believes that a site with five 223 m (732 ft) tall towers and about two million sq m (21.5 million sq ft) of heliostats on 485 ha (1,200 acres) of land could generate 100,000 kg (222,460 lb) of hydrogen per day, which is enough to run over 5,000 hydrogen-fuel cell buses daily.

Though the technology has the potential to be a game-changer in pushing the hydrogen economy forward, commercialization might still be several years away thanks to continuing stiff competition from fossil fuels.

The National Science Foundation and the U.S. Department of Energy supported the research and a paper on the system was published in the Aug. 2 issue of Science.

Source : University of Colorado Boulder

Update: This story was amended on Aug. 7, 2013 to correct an error in figures given to us that related to land area required to generate 100,000 kg of hydrogen per day. We were originally told it was 120,000 acres, when the correct figure is 1,200 acres. Our apologies.

View gallery - 3 images
  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
27 comments
MQ
Even though, this system is still never going to provide more energy out than in (losses always occur).. Hydrogen (and petroleum) are just energy carriers, not an original source of energy, that would be the sun (in combination with the earth's gravitational energy providing the processing reactor for oil etc) .
It is the Cost per gallon equivalent which matters (price point).
Remember, we never have to manufacture Rock Oil, petroleum suffers from massive inefficiencies in its creation / manufacture, fortunately for us it was laid down in the strata a long time ago. (Sure we do make some synthetic fuels, as in synthetic diesel, and methanol, but that is still using a resource we didn't have to create (methane) ).
If this can produce more fuel than the equivalent acreage with agriculture, it may be a goer, as long as the backers are willing to invest.
Noting also, that places which are most suitable for solar thermal energy, may not be so good for intensive agriculture.
Just sayin.
Craig Jennings
"With the aid of a solar thermal plant, the team believes that they can generate 100,000 kg (222,460 lb) of hydrogen per day" Sigh... we gizmag readers do love projected yields with no area/cost information. Since we are gizmag readers, I suppose we could extrapolate how much land area that would take v time to make that much based on amplifying sunlight 2000 : 1. Any takers? I already have a headache.
Nairda
I see a lot of negative vibes floating around. 100,000kg or 10,000kg is irrelevant. The point is it can continue the process as long as there is light. Hydrogen is expensive. A plant like this would pay for itself within a year.
In simple principles, a solar furnace with stainless steel mirrors is only going to have a carbon footprint for smelting the steel and fabricating the various metal parts. From a monetary point, it can also be fabricated very inexpensively.
Beyond that, the run ongoing requirement will just be the water for the turbine and washing the mirrors.
In the case for the water splitting, concept, its just bumping up the temperature. All it will cost environmentally is the water required to make the product. Initial fabrication will be about the same.
Ongoing maintenance is chicken feed, and fabrication of hydrogen is also going to be as expensive as the water required to make it.
As far as discussions around land area required for the task, it really depends how well the lenses focus. Nothing stops the designer from incorporating a second set of lenses at the focal point for a refined focus.
Robt
@Nairda How do you know that it would pay for itself within a year? Do you have documenetd sources of build cost and ongoing costs versus output?
waleed Al Gharabally
One does not need all that acreage if you use 10 meter diameter water filled lenses and redirect all the focal points to one location.
Fretting Freddy the Ferret pressing the Fret
I'm curious where the water is drawn from. The sea? Aquifers? What water purification systems are needed before it goes into the system?
Considering the volume of hydrogen produced and the relatively inexpensive materials (although I'm curious what those metal oxides are) involved, I can see this project paying for itself in a few years.
This is the next step in concentrated solar power plants. Previously, pilot plants focused light to heat a medium (water, molten salts, oils). This goes through a heat exchanger to produce steam for the turbine and then produces electricity.
The capital investment for those CSP are large, because of heat exchangers, large medium storage tank, auxiliary heating systems, turbines. Not to mention corrosion problems with molten salts.
All in all, this is a positive step going from a CSP system that utilizes specific heat, to one utilizing chemical energy (skipping over a step that uses latent heat). It delivers an energy carrier which is energy dense per kg.
Dekarate
let's see - New York City has 5,602 buses and a land area of 195,000 acres. So if we replace all the buildings in the city with mirrors, we will be able to generate about enough fuel to run the buses for the people who are no longer there. Economics of scale - even with desert land prices are not favorable towards a cost effective solution
Ashtom1
Has hydrogen storage and haulage been worked out? Last I knew it was really hard and expensive to store hydrogen being the smallest element with ability to leak out. How does hydrogen storage and haulage factor into the cost?
yinfu99
I wonder if its any cheaper for the same amount of energy gain to instead of using mirrors to focus sunlight on a single tower, to rather use fresnel lens to focus light on a spot (not a tower), similar to what a magnifiying glass. Since the lense takes sunlight and focuses it to a smaller point, where as the mirrors take light and reflect it, which one is more energy efficient. I would think the fresnel len system would take less space and generate similar power levels. Perhaps someone could add more to this speculation.
WhoIsBramStoker
Something does not add up here. 120,000 acres is 188 sq miles!
Divide it all out and you get about 1000 ft squared ( 1 million sq ft) per bus.
Rocket bus?