Proton OnSite achieves hydrogen gas production at 5,000 psi without a compressor
Novozymes inaugurates $200M biofuels enzyme plant in US

PNNL small solid oxide fuel cell achieves record efficiency; microchannels, external steam reforming and recycling

Power system configuration diagram. Powell et al. Click to enlarge.

Researchers at the Pacific Northwest National Laboratory report on a highly efficient, small-scale solid oxide fuel cell system featuring PNNL-developed microchannel technology in combination with adiabatic, external steam reforming and anode gas recirculation. The heat and water required for the endothermic reforming reaction are provided by the recirculated anode gas emerging from the SOFC stack. They refer to this as adiabatic steam reforming because external heat sources, such as a combustor or an electric-resistance heater, are not necessary to support the reaction.

The new fuel cell system achieves up to 57% efficiency—significantly higher than the 30 to 50% efficiencies previously reported for other solid oxide fuel cell systems of its size—according to a study published in this month’s issue of the Journal of Power Sources. The pilot system generates about 2 kW of electricity; the PNNL team designed it to be scaleable to produce between 100 and 250 kW.

Solid oxide fuels cells are a promising technology for providing clean, efficient energy. But, until now, most people have focused on larger systems that produce 1 megawatt of power or more and can replace traditional power plants. However, this research shows that smaller solid oxide fuel cells that generate between 1 and 100 kilowatts of power are a viable option for highly efficient, localized power generation.

—Vincent Sprenkle, a co-author on the paper and chief engineer of PNNL’s solid oxide fuel cell development program

PNNL’s system includes fuel cell stacks developed earlier with the support of DOE’s Solid State Energy Conversion Alliance and uses methane as its foundation fuel. The PNNL uses external steam reforming to convert the methane to a syngas (CO and H2) which then react with oxygen at the fuel cell’s anode, generating electricity as well as the byproducts steam and carbon dioxide.

“A critical distinction between fuel cell technologies and other energy conversion devices, such as internal combustion engines, is that fuel cell efficiency is not Carnot-limited and fuel cells can achieve relatively high conversion efficiencies at smaller scale operation. Of the available fuel cell technologies, solid oxide fuel cell (SOFC) offers the highest electrical conversion efficiencies.”
—Powell et al.

Steam reforming has been used with fuel cells before, but the approach requires heat that, when directly exposed to the fuel cell, causes uneven temperatures on the ceramic layers that can potentially weaken and break the fuel cell. So the PNNL team opted for external steam reforming, which completes the initial reactions between steam and the fuel outside of the fuel cell.

The external steam reforming process requires a heat exchanger. On one side of the wall is the hot exhaust that is expelled as a byproduct of the reaction inside the fuel cell. On the other side is a cooler gas that is heading toward the fuel cell. Heat moves from the hot gas, through the wall and into the cool incoming gas, warming it to the temperatures needed for the reaction to take place inside the fuel cell.

Microchannel heat exchanger and photochemically etched shim. Source: PNNL. Click to enlarge.

The key to the efficiency of this small SOFC system is the use of a PNNL-developed microchannel technology in the system’s multiple heat exchangers. Instead of having just one wall that separates the two gases, PNNL’s microchannel heat exchangers have multiple walls created by a series of tiny looping channels that are narrower than a paper clip. This increases the surface area, allowing more heat to be transferred and making the system more efficient. PNNL’s microchannel heat exchanger was designed so that very little additional pressure is needed to move the gas through the turns and curves of the looping channels.

The second unique aspect of the system is that it recycles. Specifically, the system uses the exhaust, made up of steam and heat byproducts, coming from the anode to maintain the steam reforming process—i.e., the system doesn’t need an electric device that heats water to create steam. Reusing the steam, which is mixed with fuel, also means the system is able to use up some of the leftover fuel it wasn’t able to consume when the fuel first moved through the fuel cell.

The combination of external steam reforming and steam recycling with the PNNL-developed microchannel heat exchangers made the team’s small SOFC system extremely efficient. Together, these characteristics help the system use as little energy as possible and allows more net electricity to be produced in the end. Lab tests showed the system’s net efficiency ranged from 48.2% at 2.2 kW to a high of 56.6% at 1.7 kW. Although the single-pass fuel utilization is only about 55%, because of the anode gas recirculation the overall fuel utilization is up to 93%.

The team calculates they could raise the system’s efficiency to 60% with a few more adjustments.

The PNNL team would like to see their research translated into an SOFC power system that’s used by individual homeowners or utilities. The research was supported by DOE’s Office of Fossil Energy.


  • M Powell, K Meinhardt, V Sprenkle, L Chick and G McVay (2012) Demonstration of a highly efficient solid oxide fuel cell power system using adiabatic steam reforming and anode gas recirculation. Journal of Power Sources, Volume 205, Pages 377-384 doi: 10.1016/j.jpowsour.2012.01.098


Nick Lyons

Depending on capital costs, this could be a huge potential winner for combined heat & power (CHP) for commercial and residential applications. According to the Google:

1 therm = 100,000 BTU
1 kWh = 3412 BTU
1 therm = 100,000 / 3412 = 29.3 kWh

Where I live:

1 kWh costs ~$0.14 (minimum)
1 therm of CH4 costs ~$0.638 (last bill)

At 60% efficiency, a therm of CH4 could generate:

0.6 * 29.3 = 17.6 kWh at a cost of $0.638

A single kWh would cost:

0.638/17.6 = $0.036/kWh you also get heat for hot water and/or space heating. With net-metering, you could run a home's power and heat very cheaply at current natural gas prices with a 1-2 kW SOFC CHP setup.

This could make photo-voltaic look like an economic loser, even with current low panel prices.


Nick has some good numbers, depending on capital costs.


The German's and Japanese are already putting small numbers of home fuel cells into homes, mostly PEMs which would not get quite such good efficiency. This is an obvious improvement.

As a bonus it may drive a stake through the heart of vastly expensive, impractical and ecologically disastrous renewables, as it would be very difficult to fool around using this just whenever the wind didn't blow or it was night, or you would not have hot water or heating.

Henry Gibson

High efficiency is not required for home cogeneration as almost all homes and small businesses are not using it. Honda makes an engine version that burns natural gas and has sold about 100,000 of them or more. They are the most cost effective way of reducing CO2 release since only natural gas generation is being approved in many places. The 40 percent of CO2 that can be saved is even more interesting when such units can also cool buildings with waste heat. The Capstone microturbines used in larger building demonstrate this in some places, but mass production will make it cheap enough for many more places. See their website or an Australian CHP for homes website for more information. Or see Freewatt in the US. Bladen jets might also make a home version in the future. ..HG..


The advantage of the engine version of the CHP is that it is good at load following and so the $0.14/kWh of avoided electricity cost is relevant. The SOFC is more efficient but they tend to have durability issues if cycled intensively so you'd want to consider it more of a base load. If the system is scaled to the average power requirements of a residential installation then much of the electric output is exported to the grid and the feed in tariff becomes relevant figure - likely to be much less than 0.14.

So might be good for hospitals and the like with 24hour electric demands but not for home use (yet).


I wonder how much the SOFC unit will cost. Unless I get my electricity usage down 80%, I am paying ~$0.32/kWh, $200-250 per month.

It looks like it should cost no more than a few thousand $.

But not good for GHG. I would rather go with PV which looks to be maybe getting very cheap thanks to recent developments.

SOFC combined with electric motors is really a good replacement for ICE and liquid fuels, with much lower GHG and no range issue. Big trucks would be an ideal application. Maybe add thermoelectrics to get more efficiency.

Seems like the SOFC durability is really just an engineering challenge.


It's relatively easy to store heat over a 24-hour cycle, especially if the source supplies it at high temperature.  Also, new loads like EVs which can level the 24-hour demand curve are coming along.  Given that, it may be possible to limit thermal cycling in an SOFC system and radically extend the lifespan in normal use.


"The pilot system generates about 2 kW of electricity; the PNNL team designed it to be scaleable to produce between 100 and 250 kW."

If they can scale as expected 15-20kW electric would readily power a western home. CH4 cost is dropping due to the worldwide glut. But that's alot of excess heat that may be dumped to the environment. Chillers would prove valuable in hot climates. The BofA building cogen (4.5MW) in NYC uses overnight electric excess to make ice for daytime cooling.


You could use a couple of kW CHP and then a couple of kW PV and export to the grid everything you don't use.

I wouldn't be too quick to count out renewables Dave as

"We are in fast changing and exciting times though, and we can't pick winners"


I also wouldn't be too quick to count out renewables;

Kit P

PNNL again demonstrates how to violate the KISS principles.

I know it is boring so I will be called a troll but let's talk about simple. Once you have NG piped into your house, gas furnaces and hot water heaters are very, very efficient at providing heat. A fuel cell will never come close.

Then there is stupid as in the economic analysis that ignores capital cost, O&M, taxes, and insurance. For those who love renewable energy, a good way to get less is to use a fuel cell.

For example one of the best sources of very reliable renewable energy is biogas that also can provide process heat. If we use a rock solid ICE we cut the cost in half and therefore can get twice as much renewable energy for out money.

“I am paying ~$0.32/kWh, $200-250 per month. ”

Why are you paying so much, where do you live?

The only reason to pay so much is if you live in a very remote location with high transportation costs for diesel fuel. It cost about the same to make electricity anyplace in the world. When you see very high retail power it is because of taxes. The problem is inefficient government.

California is a perfect example. Us old boring guys who make power will tell you how to efficiently use your money when producing renewable energy. It will not have the highest thermal efficiency. It will not be sexy, it will not be innovative. It will be simple and it will work.


In our wide local area (three + times the size of California) the sole e-energy supplier has had very low single rate ($0.06/Kwh in 2012) for all domestic customers for 40 + years. Industrial rates are as low as $0.02/Kwh, specially for the very large aluminium factories. Other industrial and commercial rates vary between $0.03 and $0.04/Kwh.

Isolated customers and farmers are not penalized.

We hope that electricity is NEVER privatized again.

Our neighboring province (Ontario) is going through the same problems as you mentioned since they privatized (partly) a few years ago.


Kit P is either disingenuous or totally ignorant of thermodynamics.  A fuel cell at 50% electric efficiency with a condensing heat-recovery system would substantially reduce fuel demand versus a remote electric generator and gas heat.  There's no point in generating high-grade heat only to dilute it to low temperature; that's simple loss to entropy, which the fuel cell avoids.

A fuel-cell system driving a heat-pump water heater could achieve "efficiency" of more than 150%. A gas flame under a tank of water can't get close to that.

Kit P

E-P gas hot water heaters and furnaces can be 90-96 % efficient. Very hard to get 'more than 150%' with existing things that efficient.

While I am very good at thermodynamics, I also know how to put information into a business plan spread sheet. There are very few economic opportunities for CHP. While making your own electricity and recovering the heat sounds good to E-P and others like PNNL, do the math and it does not pencil out very often.

When it does, every ICE/generator manufacture offers various heat recovery options. You can not improve efficiency very much with a fuel cell but they are very expensive.

Many engineers and scientist do not understand KISS but our customers do. People want simple. They want to go to the fridge and get a cold beer. The want to turn the lights on with a switch. I want to set the temperature to my house to what my wife likes not to some program lawyers who never took thermodynamic thinks will save energy.

In other words we use energy to make our lives easier and more comfortable. Making your own electricity is not convenient and will not save money. It will not save much fuel either. If you want to save money and energy study weather stripping not fuel cells.

Very hard to get 'more than 150%' with existing things that efficient.
ICE at 30% efficiency + heat pump with CoP of 3.0 can hit 160%.  It's not even hard, it's trivial with modern technology.
While I am very good at thermodynamics
Yeah, right.  You wouldn't post half of what you do if you were.
Many engineers and scientist do not understand KISS
Yeah, right.  As if engineers don't understand cost, MTBF, and everything else you claim to understand.  So far as I can tell, those claims are empty.
Making your own electricity is not convenient and will not save money.
Making your own electricity from NG in a SOFC will work without your intervention until it fails, and will probably generate heat on demand even after it does.  At worst, it's as good as a gas flame generating heat; at best, it's much better.  Twit P. does his best to fail to understand this.  Someone could probably base a psychology paper on the reasons why.
Kit P


It has been more than 10 years since I have observed a fuel cell running on a NG reformer. At the time I I lived in Richland where PNNL. People I had worked with at the nuke were running the demonstration project. What did they tell me? It has no practical application.

So how about now? Still no practical applications.

I have read that both Honda and VW has a home CHP units. While these may be practical, there are not economical in the US. They never will be because your power company does such a fine job of making power. While lots of people unfairly gripe about their power company notice how many actually make their own power?

So what you have E-P is just a theory about the kind of engineering that you do not practice.

For CHP to be economical in a home setting you need a very cold climate. It has been more than 20 years since I lived in Michigan. Since we did not need air conditioning in the summer, we did not have a heat pump.

I have installed new very efficient in my last two houses for cooling and heating because we have mild winters and hot summers.

This is where E-P flunks practical engineering. Heat pumps are not very efficient in cold climates.

Kit the mechanical engineer has lusted after a hot water heat pump ever since I first heard about them 30 year ago. Kind of like the BMW I saw walking to high school. Since I got out of the navy, I can afford whatever man toys I want. It costs what?

I even designed a built a passive solar hot water system when I lived in California. So what did I learn? It is really hard to beat the cheapest hot water heater that you can buy at Home Depot.

While E-P wants to suggest that make me crazy, E-P is not telling us about the efficiency of his CHP system.

Roger Pham

SOFC in the 100 kW range is great for restaurants, hospitals, hotels, spas etc. where significant amount of hot water is needed year round. The energy independency from the grid is great for business that must be shut down in case of power black out from a storm or snow or ice. The excess electricity produced can be transmitted to the grid, or used to drive a heat pump to double the heat output per therm of NG. CHP really shines in these type of application. Apartment complex can also benefit from this type of thing, in which one SOFC unit coupled to an industrial size A/C & heat pump unit can serve a whole apartment building that can serve 10-100 families.

A large, industrial-size A/C and heat pump unit can produce a lot of heating or cooling for a given amount of amortized investment, unlike a small and inefficient residential unit that is much less reliable and less durable. Please kindly redo the math to see if this makes any sense.


I don't have a CHP system because I haven't had to replace a furnace yet.  If I had one fail tomorrow I'd go for a Freewatt, but I'll see what's available when the time comes.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)