« ELCAT Wraps; CO2 Conversion to Fuels Feasible, But Needs More Work | Main | Tesla Motors Completes $40M Bridge Financing to Support Continued Development of White Star »
Researchers Refine Aluminum Alloy to Enable Economically Viable, Large-Scale, On-Demand Hydrogen Production
19 February 2008
Researchers at Purdue University have further refined their aluminum-gallium alloy used in a hydrogen production process (earlier post) that they say is now economically competitive with conventional fuels for transportation and power generation.
The new alloy contains 95% aluminum and 5% of an alloy that is made of the metals gallium, indium and tin. Its predecessor in the research contained 80% aluminum and 20% gallium. Because the new alloy contains significantly less of the more expensive gallium than previous forms of the alloy, hydrogen can be produced less expensively, according to Jerry Woodall, professor of electrical and computer engineering at Purdue, who invented the process.
We now have an economically viable process for producing hydrogen on-demand for vehicles, electrical generating stations and other applications.
—Jerry Woodall
Aluminum reacts vigorously with water to produce hydrogen, alumina (aluminum oxide, Al2O3) and heat in the reaction:
2Al + 3H2O → 3H2 + Al2O3 + heat
However, air-exposed aluminum forms a passivating skin of alumina that protects it from further rapid oxidation. A viable aluminum-water hydrogen system must overcome the protective layer to allow the reaction to continue, while still meeting the other constraints for on-board hydrogen storage or generation. A number of efforts over the past decades have explored the potential of developing an amalgamated aluminum surface that can sustain the reaction with water.
Woodall took the basic approach of disrupting the passivating oxide skin with the gallium component of the alloy.
The aluminum oxide (alumina) resulting from the reaction can be recycled back into aluminum using the currently preferred industrial process called the Hall-Héroult process, which produces one-third as much carbon dioxide as combusting gasoline in an engine, Woodall said. Recycling aluminum from nearly pure alumina is also less expensive than mining the aluminum-containing ore bauxite, making the technology more competitive with other forms of energy production. The gallium-indium-tin alloy component is inert, which means it can be recovered and reused at an efficiency approaching 100%, Woodall said.
After recycling both the aluminum oxide back to aluminum and the inert gallium-indium-tin alloy only 60 times, the cost of producing energy both as hydrogen and heat using the technology would be reduced to 10 cents per kilowatt hour, making it competitive with other energy technologies.
—Jerry Woodall
The researchers developed the new alloy—which they call 95/5—in late 2007 and will present findings on it on 26 February during the conference Materials Innovations in an Emerging Hydrogen Economy (24-27 February in Cocoa Beach, Florida).
A key to developing the alloy for large-scale technologies is controlling the microscopic structure of the solid aluminum and the gallium-indium-tin alloy mixture. The mixture tends to resist forming entirely as a homogeneous solid due to the different crystal structures of the elements in the alloy and the low melting point of the gallium-indium-tin alloy, according to Woodall.
The alloy is said to have two phases because it contains abrupt changes in composition from one constituent to another.
I can form a one-phase melt of liquid aluminum and the gallium-indium-tin alloy by heating it. But when I cool it down, most of the gallium-indium-tin alloy is not homogeneously incorporated into the solid aluminum, but remains a separate phase of liquid. The constituents separate into two phases just like ice and liquid water.
—Jerry Woodall
The two-phase composition seems to be critical for the technology to work because it enables the aluminum alloy to react with water and produce hydrogen. The researchers had earlier discovered that slow-cooling and fast-cooling the new 95/5 aluminum alloy produced drastically different versions. The fast-cooled alloy contained aluminum and the gallium-indium-tin alloy apparently as a single phase. In order for it to produce hydrogen, it had to be in contact with a puddle of the liquid gallium-indium-tin alloy.
That finding showed that the alloy would react with water at room temperature to produce hydrogen until all of the aluminum was used up, Woodall said. The engineers were surprised to learn late last year, however, that slow-cooling formed a two-phase solid alloy, meaning solid pieces of the 95/5 aluminum alloy react with water to produce hydrogen, eliminating the need for the liquid gallium-indium-tin alloy.
That was a fantastic discovery. What used to be a curiosity is now a real alternative energy technology.
—Jerry Woodall
The slow-cooling technique made it possible to create forms of the alloy containing higher concentrations of aluminum. The Purdue researchers are developing a method to create briquettes of the alloy that could be placed in a tank to react with water and produce hydrogen on-demand. Such a technology would eliminate the need to store and transport hydrogen.
For the technology to be used in major applications such as cars and trucks or for power plants, however, a large-scale recycling program would be required to turn the alumina back into aluminum and to recover the gallium-indium-tin alloy. Other infrastructure components, such as those related to manufacturing and the supply chain, also would have to be developed.
Future research will include work to learn more about the chemical mechanisms behind the process and the microscopic structure of the alloy.
The research is partially funded by Purdue’s Energy Center at the university’s Discovery Park. The Purdue Research Foundation holds title to the primary patent, which has been filed with the US Patent and Trademark Office and is pending. An Indiana startup company, AlGalCo LLC., has received a license for the exclusive right to commercialize the process.
February 19, 2008 in Hydrogen Production, Hydrogen Storage, Materials | Permalink | Comments (35) | TrackBack (0)
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/t/trackback/22062/26292306
Listed below are links to weblogs that reference Researchers Refine Aluminum Alloy to Enable Economically Viable, Large-Scale, On-Demand Hydrogen Production:
Comments
Why do I get the feeling there's a devil in the details.
Posted by: Neil | Feb 19, 2008 2:22:03 PM
Don't these people know that hydrogen is a dead end? Everybody on the internet says so!
Posted by: Matthew | Feb 19, 2008 2:44:32 PM
What is the best application for this ?
Although the H2 storage and distribution problems are solved (may be), you still have to burn the H2.
I do not think it is very efficient to use an ICE and fuel cells are still far too expensive.
I wonder could you do some kind of Al battery to get direct electric drive.
Aerospace ? Is the fuel (Al) too heavy for a plane ?
Enriching diesel - or is this a con ?
Posted by: mahonj | Feb 19, 2008 2:45:37 PM
If they can avoid the formation of the oxydation layer, they could probably make a battery then, Al has the highest theoritical power density of all material, but it can't be used for a battery because of this Al2O3 layer formation that blocks the chemistry.
Asides I am higkly skeptical of a energy storing solution where you produce a waste that you have to recycle,it is not really easy to manage from an infrastructure point of view, the weight and volume of Al2O3 produce will 3 to 4 times the Al input.
Posted by: Treehugger | Feb 19, 2008 2:56:50 PM
One of the applications for this is importing and exporting electricity.
This already occurs to some degree. Some bauxite mined in New Zealand is smelted in Iceland because of the abundant, and cheap renewable energy there. When the refined aluminum leaves Iceland, they are in effect exporting electricity.
If you take that a step further, (and there are schemes out there to do so) you could have ships containing aluminum and alumina going back and forth between New Jersey and Reykjavik. It would be safer than liquified natural gas tanker ships.
Posted by: Healthy Breeze | Feb 19, 2008 3:04:23 PM
Instead of hitting "Regular" we hit "ALGAINSN" and pour liquid metal into our fuel tanks? And where would we dispose of the waste?
Posted by: | Feb 19, 2008 3:22:31 PM
"I am higkly skeptical of a energy storing solution where you produce a waste that you have to recycle"
Treehugger, I think you just described every rechargeable battery on the planet.
I have no idea how practical this will turn out to be but the idea of fueling a car with a garden hose is the stuff dreams are made of.
Posted by: Arthur | Feb 19, 2008 3:29:39 PM
I do not get why one needs to convert the aluminum energy to hydrogen energy? Cannot one just use the alumimum as the fuel in the fuel cell ? I though I heard somebody at the Fox Valley Electric Vehicle club was doing that back in the 80's or 90's ? He/They were burning pop cans for fuel.
Posted by: donee | Feb 19, 2008 4:06:57 PM
You still run into the problem that this requires a pound of aluminum alloy per mile of travel.
Not to mention the distribution of this stuff not existing.
Posted by: GreyFlcn | Feb 19, 2008 4:10:55 PM
That said, this might make some sense as a electricity grid storage mechanism.
But it certainly isn't going to be put directly inside a car.
Posted by: GreyFlcn | Feb 19, 2008 4:18:25 PM
donee: "they were burning pop cans for fuel".
Possibly using them as one pole of a battery?
A battery can be made from almost any two materials. Aluminum would undoubtedly work just as copper, zinc, and other metals do. But they don't work well enough.
Beer cans work better than cola cans. At least the experiments soon seem more agreeable.
Posted by: K | Feb 19, 2008 4:22:34 PM
@ Arthur,
You wouldn't refuel with a garden hose.
You would have to add water more often than aluminum alloy pellets, perhaps.
Essentially, you'd have two cannisters/resevoirs; one for Aluminum and one for Alumina. When the fuel pellets had all moved through the reaction chamber and over to the alumina resevoir, you'd go trade in both cannisters for a fresh ones. I don't know the math on how many kilowatt hours of energy could be held in the space of an 8-gallon tank. If memory serves, 1 liter of liquid H20 can create about 1,000 liters of H2 gas (about a cubic meter) at sea level, pressure. So, how far can your car go on about 30 cubic meters of H2?
Posted by: Healthy Breeze | Feb 19, 2008 4:22:41 PM
So a vehicle using this would need a water tank? How do you keep the water tank from freezing?
Posted by: james | Feb 19, 2008 5:32:32 PM
Mahonj, treehugger:
There is a lot of work going on to produce aluminum-air batteries. Like the invention described in this article, aluminum oxide is produced as a waste product. Incidentally, to avoid passivation of the aluminum, tin is added and other 'proprietary additives (gallium?).
So you might as well avoid a fuel cell and use an aluminum-air cell directly.
In addition, in order to generate 5 kg of hydrogen (which supposedly can propel a mid sized car for 500 km, you would need 45 kg of aluminum and 45 kg of water. This does not take into account the other materials (gallium etc.) nor does it take into account any water neccessary to suspend the waste product aluminum oxide.
Posted by: miket1 | Feb 19, 2008 5:35:41 PM
Okay, fellas, I said it was the stuff of dreams. Anyway, I checked Woodall's website and he claims that you need twice the weight of aluminum to get the same energy as gasoline. So if it takes 130 pounds of gas to go 350 miles, you need 260 pounds of aluminum for 350 miles (GreyFlcn was close) so you need an infrastructure that allows swapping alumina for aluminum and some water storage. Still, I can dream, can't I?
Posted by: Arthur | Feb 19, 2008 5:59:20 PM
I hope they don't need too much of gallium and Indium given how rare are these metals. The industry is already heading for a shortage of indium because we use to make transparent electrode on flat screen, So for a disposable battery, forget it.
Posted by: Treehugger | Feb 19, 2008 6:31:11 PM
IIRC, aluminum-air batteries have been around for a while and even have used alloying elements like gallium to reduce the surface passivation.
The figure of 10¢/kWh of "energy both as hydrogen and heat" is strangely put. What's the breakdown between the two? Heat at low temperatures isn't all that useful for pushing a car, so only the hydrogen is of real interest. If the hydrogen energy costs 20¢/kWh and the engine is 33% efficient, the energy at the crankshaft costs 60¢/kWh. In contrast, gasoline at $3.00/gallon costs about 8.9¢/kWh of heat energy (LHV).
Matthew, if you think hydrogen has great prospects, perhaps you can get all those little details clarified so we don't have to wonder if the proponents are trying to sell us a bridge.
Posted by: Engineer-Poet | Feb 19, 2008 6:52:12 PM
Going through Woodall's presentation, he claims 12 kWh/lb Al energy input and 2 lb Al equivalent to 1 lb of gasoline. A pound of gasoline yields about 19,000 BTU of heat energy (LHV), so the process efficiency from electricity to hydrogen is about 23%. Multiply by 30% engine efficiency and the throughput is about 7%. He hints at the implications of this in slide 16. The only way it gets close to competitive is if electricity is available for 2¢/kWh.
I think heavy skepticism is justified.
Posted by: Engineer-Poet | Feb 19, 2008 7:05:52 PM
These schemes have been around for years. There are people using sodium borohydride for on-demand H2 and the problem remains - you end up with a tank full of slurry that needs recycling. Aluminum/air looks better for large scale electron storage. On-demand H2 appears to need a breakthrough in electrolysis or some other method of breaking molecular bonds.
Posted by: gr | Feb 19, 2008 7:17:09 PM
Oxidizing aluminum with water wastes, IIRC, 57 percent of its free energy of oxidation. The other 43 percent is in the hydrogen that is produced. If Woodall could devise an aluminum internal combustion motor, like a solid rocket motor but longer-lasting, he could burn 1 mass of aluminum in one step rather than (1/0.43) masses in two steps. Or, as some commentators suffering from the electrochemical hangup above note, he could oxidize it in a fuel cell in one step.
The task of hauling Al2O3 around would be reduced (1/0.43)-fold, and the total return freight by a somewhat greater factor if, in his present scheme, it comes back to the power plant wet.
Posted by: G.R.L. Cowan, hydrogen-to-boron convert | Feb 19, 2008 7:32:42 PM
You need 26kg of Aluminium (i.e. 27.37kg of the 95/5 alloy) plus 54kg of water (presumably demineralized) to produce 6kg of hydrogen with this scheme. In addition, you need storage tanks and coolant circuitry to keep operating temperatures acceptable. A major challenge will be ensuring that the waste product Al2O3 - itself a solid - is continuously cleared from the site where water is supposed to contact the fuel alloy.
Each kg of hydrogen produced contains 143MJ, roughly three times as much as the same mass of gasoline or diesel. If you assume that a fuel-cell based electric drivetrain is three times as efficient tank-to-wheel as one powered by a conventional ICE and, that a regular ICE-powered sedan gets ~30MPG of gasoline, 6kg of hydrogen would give you a range of nearly 100 miles. A 300-mile range would require roughly 250kg of inputs (of which roughly 80kg would be alloy). For comparison, the regular sedan requires 10 gallons of gasoline, i.e. ~27kg of fuel.
It may be possible to produce the demineralized water on site at the filling station, using regular tap water and electricity as inputs. However, the alloy would almost certainly have to be produce at large central facilities and distributed to filling stations. Quite apart from the fact that the fuel is a hard-to-handle solid that must be protected against premature contact with water, the mass required to support the same vehicle range is three times higher than for gasoline, i.e. a lot more tanker trucks will be needed.
The technology could be valuable for a number of applications, including submarine propulsion - but perhaps not for civilian passenger cars.
Posted by: Rafael Seidl | Feb 19, 2008 8:00:28 PM
Rafael,
I believe that you need twice the quantity of aluminum that you have calculated (6000 g divided by (3 X 2g (H2)) X 2 X 27 g).
Posted by: miket1 | Feb 19, 2008 8:44:37 PM
You don't need a large supply of demineralized water; you can get most of the water needed by condensing it from the fuel cell output.
A practical system based on this would almost certainly tap both the thermal and electrochemical energies. Say, by feeding an aluminum alloy wire into a high pressure boiler. Steam + hydrogen would drive an initial expander turbine. After condensing most of the steam from the exhaust, the hydrogen would be recompressed and fed to a high temperature fuel cell serving as the combustor for a Brayton cycle GT. Then the exhaust from the GT would drive a conventional bottom cycle steam turbine.
A net thermal efficiency of 80% should be possible. I don't see it being economical for use in cars anytime soon, but it should make a very efficient replacement for diesel fuel in ships and locomotives. The power to weight ratio might even be high enough for an electro-fan airliner.
It would also be a great power source for submarines, with capabilities intermediate between a diesel-electric and a nuclear sub.
Posted by: Roger Arnold | Feb 20, 2008 12:01:05 AM
Im sooo glad we dont depend on you guys to brainstorm,,,
All the oil exec will ask is ok so we make this stuff somewhere cheap and transport it cheapest way..
How much will it cost at the pump when we react it onsite at the gas station and how much h2 could such a reactor make a day and how much space would it need?
If this means they can reach thier cost goals then expect them to start testing pilot systems soon.
With various techs out now we can store more then enough h2 on a car cheaply to run it well over 300 miles the problem wasnt that end but getting the h2 cheap to the pump and this may have solved that.. or not but thats for 5 zillion beancounters to figure out.
Posted by: wintermane | Feb 20, 2008 9:30:01 AM
How easily would it be to extract water from the air? Perhaps some combined cycle running from the heat given off by the reaction could help power a dehumidifier?
Posted by: mark | Feb 20, 2008 9:44:40 AM
My conclusio still is, unless someone is providing free electricity, this scheme will have only very limited applications (as Roger pointed out). The energy lost as heat in the conversion needs to be invested again for electrolysis of alumina;
By doing away the alumina-heating / electrolysis step, and directly electrolysing water, the process efficiency would be vastly boosted. But as other research papers point out, the overall efficiency of direct water electrolysis / fuel-cell is still way less than that of using the electricity directly in a BEV.
Why would the addition of yet another (complex) step in that conversion increase the efficiency that much. I though free energy (and perpetuum mobiles) would belong to fairy tales, not scientific research...
Posted by: realarms | Feb 20, 2008 3:12:59 PM
Doing away with the alumina electrolysis? Better to do away with the hydrogen. Burn the aluminum.
Posted by: G.R.L. Cowan, hydrogen-to-boron convert | Feb 20, 2008 4:05:43 PM
Dont forget that to reduce the Al2O3, you will do the classic hall rx, using carbon electrode. This results in 3 CO2 for every 4 Al. Of course, you also spend CO2 credits when you get the basic electricity from a coal-fired power plant.
Posted by: jp straley | Feb 21, 2008 8:58:57 PM
JP, that is not correct. The Hall process is an electrlytic reduction, not a carbothermic reduction. The carbon electrodes are not directly consumed in the reaction.
Posted by: Roger Arnold | Feb 21, 2008 11:56:50 PM
Oops, my bad. Should have checked Wikipedia first. The Hall process is an electrically driven carbon reduction; it yields aluminum at one electrode and CO2 at the other--not oxygen.
Hmm, how come I remember it as being a pure electrolytic reaction, and how come it takes so much electricity to drive it? The carbon oxidation potential should reduce the cell voltage to almost nothing...
Looks like some more digging is in order.
Posted by: Roger Arnold | Feb 22, 2008 12:04:48 AM
(A short time later ..) Wow! I didn't realize that production efficiency for aluminum was so bad. Almost twice the electrical energy per kg. as what reoxidation of the aluminum would yield--and that's not counting the potential energy of the carbon consumed.
Well, that does it for aluminum as an energy carrier, as far as I'm concerned. Unless somebody comes up with something much more efficient than the Hall process for producing aluminum, looks like we fall back to Graham's favorite candidate: boron. If somebody could just show that a boron-fueled turbine gas turbine were really feasible...
Posted by: Roger Arnold | Feb 22, 2008 12:26:49 AM
I always tell people to "do the math"; the insights obtained cannot be had any other way, and the practice in thinking (versus handwaving) is both sobering and liberating.
"He who refuses to do arithmetic is doomed to talk nonsense."
Posted by: Engineer-Poet | Feb 23, 2008 3:26:26 PM
Here is an interesting site, details the latest/greatest tweaks to making metal from al2o3. Interestingly, there is a direct reduction method. One drawback here is that small plants for al reduction aren't too good--they need to be big because they run hot & losses are larger in small pots.
Anyhoo,because making Al is a large, energy-intensive industry, they've studied the topic a lot. Here's the link: http://www.tms.org/pubs/journals/JOM/9905/Welch-9905.html
IMHO, making H w aluminum could be used to tweak efficiency of existing IC engines, and this might be a practical on-board way to make the relatively small amounts required.
JP Straley
Posted by: jp straley | Feb 29, 2008 3:02:37 PM
Here is an interesting site, details the latest/greatest tweaks to making metal from al2o3. Interestingly, there is a direct reduction method. One drawback here is that small plants for al reduction aren't too good--they need to be big because they run hot & losses are larger in small pots.
Anyhoo,because making Al is a large, energy-intensive industry, they've studied the topic a lot. Here's the link: http://www.tms.org/pubs/journals/JOM/9905/Welch-9905.html
IMHO, making H w aluminum could be used to tweak efficiency of existing IC engines, and this might be a practical on-board way to make the relatively small amounts required.
JP Straley
Posted by: jp straley | Feb 29, 2008 3:03:36 PM
Here is an interesting site, details the latest/greatest tweaks to making metal from al2o3. Interestingly, there is a direct reduction method. One drawback here is that small plants for al reduction aren't too good--they need to be big because they run hot & losses are larger in small pots.
Anyhoo,because making Al is a large, energy-intensive industry, they've studied the topic a lot. Here's the link: http://www.tms.org/pubs/journals/JOM/9905/Welch-9905.html
IMHO, making H w aluminum could be used to tweak efficiency of existing IC engines, and this might be a practical on-board way to make the relatively small amounts required.
JP Straley
Posted by: jp straley | Feb 29, 2008 3:22:38 PM
