GS Yuasa books $30M in extraordinary losses for faulty batteries in Mitsubishi PEVs
Ultrasonic welding in the battery pack for the Cadillac ELR

CU-Boulder team develops more efficient isothermal solar-thermal water splitting technique for H2 production

A University of Colorado Boulder team has developed a new solar-thermal water-splitting (STWS) system for the efficient production of hydrogen. A paper on their work is published in the journal Science.

STWS cycles have long been recognized as a desirable means of generating hydrogen gas (H2) from water and sunlight, the team notes. Two-step, metal oxide–based STWS cycles generate H2 by sequential high-temperature reduction of a metal oxide catalyst (releasing oxygen atoms) and cooler conditions in which the catalyst is reoxidized by oxygen from water (freeing up hydrogen molecules for collection as hydrogen gas). The temperature swings between reduction and oxidation steps have hobbled STWS’ overall efficiency, however, because of thermal and time losses that occur during the frequent heating and cooling of the metal oxide. The cycling can also limit catalyst lifetime.

The CU-Boulder team showed that these temperature swings are unnecessary by developing an approach that allows the redox cycle to be driven isothermally, using pressure swings. Isothermal water splitting (ITWS) at 1350 °C using the “hercynite cycle” exhibits H2 production capacity >3 and >12 times that of hercynite and ceria, respectively, per mass of active material when reduced at 1350 °C and reoxidized at 1000 °C.

We have designed something here that is very different from other methods and frankly something that nobody thought was possible before. Splitting water with sunlight is the Holy Grail of a sustainable hydrogen economy.

—CU-Boulder Professor Alan Weimer, research group leader

One of the key differences between the CU method and other methods developed to split water is the ability to conduct two chemical reactions at the same temperature, said Associate Professor Charles Musgrave, first author. Conventional theory holds that producing hydrogen through the metal oxide process requires heating the reactor to a high temperature to remove oxygen, then cooling it to a low temperature before injecting steam to re-oxidize the compound in order to release hydrogen gas for collection.

The more conventional approaches require the control of both the switching of the temperature in the reactor from a hot to a cool state and the introduction of steam into the system. One of the big innovations in our system is that there is no swing in the temperature. The whole process is driven by either turning a steam valve on or off.

—Charles Musgrave

With the new CU-Boulder method, the amount of hydrogen produced is entirely dependent on the amount of metal oxide (a combination of iron, cobalt, aluminum and oxygen) and how much steam is introduced into the system. One of the designs proposed by the team is to build reactor tubes roughly a foot in diameter and several feet long, fill them with the metal oxide material and stack them on top of each other.

The CU-Boulder system would use mirror arrays to concentrate sunlight onto a single point atop a central tower up to several hundred feet tall. The tower gathers the heat to 1,350 °Celsius, then delivers it into the reactor containing the metal oxides. A working system to produce a significant amount of hydrogen gas would require a number of the concentrating towers from several acres of mirrors surrounding each tower.

Despite the discovery, the commercialization of such a solar-thermal reactor is likely years away, the researchers said.

With the price of natural gas so low, there is no incentive to burn clean energy. There would have to be a substantial monetary penalty for putting carbon into the atmosphere, or the price of fossil fuels would have to go way up.

—Alan Weimer, also the executive director of the Colorado Center for Biorefining and Biofuels (C2B2)

C2B2 is an arm of the Colorado Energy Research Collaboratory involving CU-Boulder, the Colorado School of Mines, Colorado State University and the National Renewable Energy Laboratory in Golden.

The research was supported by the National Science Foundation and by the US Department of Energy (DOE).


  • Christopher L. Muhich, Brian W. Evanko, Kayla C. Weston, Paul Lichty, Xinhua Liang, Janna Martinek, Charles B. Musgrave, and Alan W. Weimer (2013) Efficient Generation of H2 by Splitting Water with an Isothermal Redox Cycle. Science 341 (6145), 540-542 doi: 10.1126/science.1239454

  • Martin Roeb and Christian Sattler (2013) Isothermal Water Splitting. Science 341 (6145), 470-471 doi: 10.1126/science.1241311



Once I see phrases like "mirror arrays to concentrate sunlight onto a single point atop a central tower", alarm bells go off.

(Maybe I just lack vision).


They have solar towers up and running:

How much it would all cost is another question, of course, as is what the cost of the hydrogen from this would be.


And then you produce H2 somewhere far away in the desert...

It seems much more practicle to me to produce solar voltaic electricity, transport it to where you need H2, and there make H2 through an electrolyzer. It may have a lower "quantum efficiency", but will most probably be much cheaper. In addition, then you can use electricity when you want, and use any "spare" electricity to produce H2. this system is much more flexible.

Though, if this catalist works well, it may be used in combination with a high-temperature nuclear reactor to produce huge amounts of H2 wherever we want.
Such a H2-factory could be very compact.


With solar hydrogen and CO2 from renewable sources, you can make synthetic fuels to be reformed on board for fuel cells in FCVs.

Roger Pham

Why must solar-H2 be produced somewhere far away in the desert...?
A tower can be constructed in any city with adequate sunlight, then the mirrors can be placed on top of roofs of houses, parking lots, and streets nearby. The H2 produced will be stored in an intra-city H2 storage facility for use later.

This can greatly complement solar PV's that produce electricity for immediate use, while the H2 produced can be used in residential or building's FC-CHP for later use. The waste heat from the residential-hotel-hospital FC-CHP can be used for hot water heating after sundown hours. Most people take showers in the evening or morning hours, and most home hot water use will be in the evening after work hours, including dish washing, laundry, etc.

H2 production efficiency is not disclosed with this method, but if it is higher than solar PV's efficiency of 15-20%, then this indeed will be a very important achievement that will allow low-cost H2 production that will really accelerate the Hydrogen Economy to a much more near-term future. The cost of the electrolyzer is eliminated. This will allow for great backing up of solar and wind energy with all low-cost RE source, without using back up NG turbines. This will be a boon for FCV's now with a cheap source of H2, perhaps even cheaper than RE electricity per unit of output at the wheel.

Put this system around Mid-Wester farms, and H2 will be produced for pyrolysis and hydrogenation of waste biomass to produce synthetic methane and liquid fuels that can be used in existing infrastructures for all kinds of usages. We will be able to switch to 100% zero-carbon energy within realistic time frame to avert cimate change disaster. Mid-western H2 production near farming areas is also important for the production of fertilizer that will be used locally, thus sparing NG utilization that will further contribute to GW.

Furthermore, widespread use of PHEV and BEV with twice-daily charging will allow for more utilization of excess solar PV energy during the daytime and wind energy at night. This will further reduce the need for energy storage and will further accelerate the conversion to a zero-carbon economy.


To reiterate one of Alain's points: this is about using heat to produce hydrogen. Could use solar energy for that heat, nuclear or anything else you fancy.


I would like to see such an assembly of hundreds of computerised mirrors on the rooftops in a city and a tower in the middle but I doubt this will ever be done at any significant scale. Direct solar also needs direct sunlight. Any clouds reduce the conversion to almost zero. Photovoltaic have only relative reduction. Therefore in most regions, the high efficience will only be a small fraction of total daytime. If they succeed, it would be very nice, but photovoltaics are improving also to higher efficiencies. I welcome all these applications with H2, but l think photovoltacs will win the race and will remain easier, more flexible, more reliable and cheaper.


I you look at Davemart's Wiki link there is an illustration of an open pit mine concentrated solar collector. I had not seen that before.

If you have a square mile of land you can build one of these wherever there is sufficient sun, then make the fuels and pipe those to market.

Roger Pham

This solar thermal tower to H2 can complement solar PV. Solar PV for immediate-use electricity, while solar thermal store this energy for later use. It is not choosing one vs. the other, but having both at the same time. Solar thermal to electricity is still too expensive to compete with solar PV. Another disadvantage to solar thermal to electricity is that thermal storage can hold only for one day. H2 storage can store energy for months and can cover longer periods of cloudy days.

Hopefully, solar thermal to H2 will be cheaper and produce H2 as direct transportation fuel as well as back up energy source for solar and wind for extended periods of time.

In sunny areas, these installations can also be built at the edge of a city so that the H2 produced can be piped in to town for local use, without interfering with existing houses and buildings.


If you could get 70% thermal to hydrogen efficiency, put that into an SOFC with output turbine at 70%. That would be 50% overall versus 30% with a steam turbine.


Producing 1kg of hydrogen at 850C takes 225MJ, down from 350MJ in a LWR:

Taking 1 kg of hydrogen at 142MJ, that is around a 63% efficiency.

The Chinese Fluoride salt cooled high temperature reactor (FHR) in their pebble bed reactors could provide this temperature very, very safely, and the reactors are designed to be capable of mass production:

Taking fuel cell efficiency at 60% then you come out to 35-40% efficiency well to wheels, not bad.


Using solar energy more efficently to produce electrcity and/or make hydrogen available 24/7 may be an interesting option to produce unlimited energy for future générations?

With the coming transition from ICEVs to BEVs, this technology may please a major role to produce clean energy in many places?


If H2 is the priority, then any new infrastructure from generator to end use is the cost.

Much new thinking has focused on H2's secondary purpose of utilising excess energy or energy storage. To me, that threw a whole new light on the practical uses of H2 making much lower efficiencies viable.

It seems that this design focuses on the former.

If there is almost identical infrastructure (tower mirrors) in place it would seem an ideal hybrid mix could be realised cheaply. Electrons for direct use via steam or molten salt with excess capacity directed to H2 production.

The comments to this entry are closed.