DOE to Provide up to $19 Million for Research Projects on Plug-Ins, EVs and Fuel Cell Vehicles
Philippines in $1.3 Billion Biofuel Deal

Researchers Develop Method for Enzymatic Production of Hydrogen from Biomass at High Yields

The synthetic metabolic pathway for conversion of polysaccharides and water to hydrogen and carbon dioxide. Click to enlarge.

Researchers at Virginia Tech, Oak Ridge National Laboratory (ORNL), and the University of Georgia have developed a novel method using multiple enzymes as a catalyst for the direct, low-cost production of hydrogen from biomass.

Applying the principles of synthetic biology, the researchers use a combination of 13 enzymes to form an unnatural enzymatic pathway to completely convert polysaccharides—e.g., starch and cellulose—and water into hydrogen at a yield higher than the theoretical yield of biological hydrogen fermentations. Their work is described in the 23 May issue of PLoS ONE, the online, open-access journal from the Public Library of Science.

Hydrogen production from either 2 mM G-6-P or 2 mM starch (glucose equivalent) using the new method. Click to enlarge.

Starch is a high energy-density carrier, with 14.8 H2-based mass%. (The DOE long-term target for hydrogen storage is 12 mass%.) The enzymes, when added to the biomass solution, use the energy in the polysaccharides to break the water up into carbon dioxide and hydrogen.

A membrane bleeds off the carbon dioxide and the hydrogen is used by a fuel cell to create electricity. The water byproduct is recycled for the starch-water reactor. Laboratory tests confirm that it all takes place at low temperature—30° C—and atmospheric pressure. The researchers estimated the cost of hydrogen production using their method of approximately $2/kg.

The stoichiometric reaction is:

C6H10O5 (l) + 7 H2O (l) → 12 H2 (g)+6 CO2 (g)

The overall process is spontaneous and unidirectional because of a negative Gibbs free energy and separation of the gaseous products with the aqueous reactants.

The vision is for the ingredients to be mixed in the fuel tank of a car. A car with an approximately 12-gallon tank could hold 27 kg of starch, which is the equivalent of 4 kg of hydrogen. One kg of starch will produce the same energy output as 1.12 kg (0.38 gallons) of gasoline.

The research was based on earlier work by Y.H. Percival Zhang, assistant professor of biological systems engineering at Virginia Tech pertaining to cellulosic ethanol production (earlier post) and the ORNL and University of Georgia researchers' work with enzymatic hydrogen production.

One of the team, Michael W.W. Adams of the University of Georgia UGA, is co-author of the first enzymatic hydrogen paper in Nature Biotechnology in 1996. The researchers were certain they could combine the processes.

In nature, most hydrogen is produced from anaerobic fermentation. But hydrogen, along with acetic acid, is a co-product and the hydrogen yield is pretty low—only four molecules per molecule of glucose. In our process, hydrogen is the main product and hydrogen yields are three-times higher, and the likely production costs are low—about $1 per pound of hydrogen.

What is more important, the energy conversion efficiency from the sugar-hydrogen-fuel cell system is extremely high—greater than three times higher than a sugar-ethanol-internal combustion engine. It means that if about 30 percent of transportation fuel can be replaced by ethanol from biomass as the DOE proposed, the same amount of biomass will be sufficient to provide 100 percent of vehicle transportation fuel through this technology.

—Y.H. Percival Zhang

The next step for the team is to increase reaction rates and reduce enzyme costs.




Truly amazing! If this is true it will be very big. And I like the idea of filling the tank with cellulose flour shaken with water. Completely non-toxic and non-flammable. I would like to know more about the enzymes. Would you need a separate tank for that and how much would be needed a gallon?

Mark Allerton

Umm... so the process produces 11kg of CO2 for every 2kg of H2, right? As far as I can work out, that's at least factor of 2 better than a gas engine in terms of CO2 per unit of engine, but hardly carbon neutral.

Mark Allerton

Except that the carbon comes from plants. Doh! Sorry...

Stan Peterson

Fossil fuel CO2 comes from plants too, just older ones.


Once we manage to produce CO2 in concentrated form, it becomes a resource as opposed to a waste. What products can we make from a purified CO2 source?



True enough, but burning biomass can be carbon neutral.


Certainly sounds interesting. I'd love to know what the energy efficiency of the system works out to be. If you're going to make hydrogen in a car then you really should run it through a fuel cell rather than use (ugh) ICE. I wonder if this process could have a use as a small scale local electrical generating system for remote locations.

C Harget

CO2 in concentrated form gets you carbonated beverages, dry ice, and is useful to feed algae farms if they can ever get the costs down on the bioreactors. The better question is, Is the process described above suitable only for large scale hydrogen creation, or could it be useful for home production? How affordable is it likely to be compared to oil and coal?


This is a very cool development. I wonder how many (if any) of the enzymes are catalysts instead of reactants, implying that they'd be reused instead of consumed.



I question its ability to digest cellulose, also the great thing about biomass is making it into industrial organic chemicals, hydrogen has no use other then fuel and fertilizer. Then again: remember "Back to the Future" were Doc shows Marty how the 2015 revamped DeLorean runs off garbage?, well this is exactly that: just chuck your compost into the car out comes hydrogen fuel!


You could use the H2 with some extra CO from biomass gasification and make more methane. CH4 is the main component in natural gas and when you gasify biomass, you end up with more CO than H2 when you make the CH4.

I like the idea of biomass gasifiers in the mid west taking in corn stalks and making methane (SNG) then providing it to the nation via present NG pipelines.
You could solar thermally heat homes and buildings and then use the NG saved to run cars.


This guy rocks: Mr Y.-H. Percival Zhang. Here is a link for more info on his research. There is still a long way to go many years before this research result in a sugar fuel cell electric car. As I now get it is that biomass will be converted at a central factory using enzymes same process that is needed for cellulose ethanol. Then you simply load sugar and water on the car plus some enzymes. A reactor uses this to generate hydrogen for a standard hydrogen full cell. The whole process from sugar conversion to hydrogen to electricity to torque has a record efficiency conversion rate of 55%! Beating all other known processes. Show stoppers are currently the speed of the sugar hydrogen generator. When you drive the car you need energy fast and the enzymatic process seems to take hours (from what I understand) when it should take minutes. Another show stopper is making cheap enzymes. If all this is indeed solved and others succeed in making a 70 kW hydrogen fuel cell for the target price of $4000 then fuel cell cars is a relay good alternative to BEVs because they have a far more energy dense system. In addition the processes that convert lignocellulose to sugars is useful also for creating cheap sugar for humans and animal feed out of wood and grass. Wow! Even if the speed of the process can’t be solved there would still be much to gain from having a cheap but slow way to make hydrogen. All the best wishes to this guy and his team. The world needs people like him to combat GW.


Isn't this process better?


Wow! Things getting more and more exiting with every day passing by.

Glucose is main energy source/carrier for every living creature. Human body gets glucose mainly from sugars and starches (carbohydrates), proteins, and fats, and burns glucose in all organs. BTW, efficiency of conversion of food stock into mechanical work by muscles is only about 10%.

I can imagine biorefinery which converts glucose to ethanol (exactly what it is done by yeast in liquor and corn ethanol plants), to methane (in bogs, cow’s guts, or sewage/manure anaerobic digesters), to butanol, to organic acids (vinegar fermentation), or now – into hydrogen, or flexible combination of all.

The main problem remains to get glucose not from food stuff, but from cellulose. Cellulose is, actually, polymer consisting from long chains of glucose blocks. Hydrolysis of cellulose by some bacteria or refined from bacteria cultures enzymes cuts these long polymers into small blocks, which could be consumed further by other microorganisms. This is exactly the process which is usually described as first, most troublesome step of cellulosic ethanol production. But now it could be also first step of cellulosic hydrogen production.

tom deplume

If you are going to use biomass as an energy source the why throw away the energy embeded in the carbon?


Tom: You might want to clerify your point about energy in carbon unless you're thinking of some kind of exotic nuclear reaction.



I think he's saying it's better to burn the (carbon neutral) hydrocarbons rather than adding a step to extract the hydrogen.


When oxygen is present, microorganisms oxidize carbon and hydrogen in organic stock to get energy and materials for living, and release CO2 and H2O. Composting is the example. When there is no oxygen, another microorganisms thrive, and they derive they energy by oxidizing carbon with present in organic material oxygen, and by breaking down complex organic molecules to simpler ones with less energy potential. Thre is no enough oxygen in organic materials to oxydize all carbon and hydrogen "fuel" down to CO2 and H2O. As a result, substantial amounts of “waste” products are produced: vinegar (acetic acid), alcohol, methane, butanol, hydrogen, acetone, hydrogen sulfide, ammonia, and many others. The most complete energy utilization, preferred in nature, is methanogenes, where all mentioned chemicals are finally reduced by different microorganisms to ultimate waste: biogas (methane and CO2, about 50/50 by mass). If specific conditions are created, like in grain distillery, some microorganisms become predominant, and produce desired "waste", such as alcohol, or in our case hydrogen.

Aerobic processes, as composting, reduce organic nitrogen to nitrogen gas and wastes valuable fertilizer. Anaerobic digestion of any sort converts organic nitrogen to ammonia, which could be recycled back to fields as fertilizer. No waste whatsoever, and the process could be truly closed-cycle, with energy input of solar radiation, atmospheric CO2 as raw material, and output of desired chemicals.

Unfortunately, lignin, constituting about 30% of woody tissue, could not be digested. Usually it is returned back to field as soil conditioner, but it also could be gasified, thought this process is not energy and economically efficient so far because of hurdles to get it sufficiently dry from suspension in water.


Seems like a Rube Goldberg. I can see some use for it as a carbon neutral H2 source for petroleum cracking.
But for ground transport it doesn't offer anything that can't be done better with a battery EV.


DS there is a lot of potential in this sugar fuel cell electric car both because of an incredible energy dense system and because of potentially cheap fuel. Note that it only takes 12 gallons of sugar/starch warter to fill the tank. That is 48 kg. It produces 4 kg of hydrogen enough to do about 110 kWh of electricity in a standard hydrogen fuel cell. The potential is that a 50 kg tank plus a 50 kg hydrogen reactor plus a 50 kg fuel cell or a total system wiight of 150 kg could deliver 110kWh enough for at least 300 miles of range in a heavy SUV like the Sequel. The weight of 110kWh of A123 batteries is 1000 kg and then we have not included packaging and cooling that could add 200 kg. Now add another 300 kg because the car needs to be stronger to carry the weight of this 1200 kg system and there you see 1500 kg for the lithium system versus 150 for the sugar fuel cell system. The sugar fuel cell car would also be a lot faster because of the reduced weight. Still rather hypothetical I admit but not entirely unrealistic.

Let’s look at the economics. 110 kWh would only need to cost 5c/kWh during off peak hours. That is $5,5 to fill the battery. The sugar fuel cell car would need 27 kg of sugar. That costs 9 c per lb =0,453 kg for raw sugar. That is, to fill the sugar tank costs (27/0,453)*9c = $5,36. Now we don’t have the price for enzymes and we don’t know whether they are catalysts instead of reactants (could be a combination of both since several enzymes are involved), but still. It appears it will be far cheaper than filling the tank with gasoline at today’s prices. The price of sugar could even drop significantly if Mr Zhang cellulose conversion technique succeeds (he is not the only guy working on that issue. POET will have a 110 mgy factory for this ready by 2009). For comparison the gasoline version of this heavy SUV would need about 20 gallons at $3,3 = $66. I am excited about this. And the fact that mother nature has spend several 100 million of years to optimize energy systems that essentially use sugar for everything makes me confident that this will also be the energy carrier of choice for human transportation. Completely non-toxic and non-flammable and can be handled and produced even in remote third world countries. We just need to master the enzymatic processes. So DS with all respect for Rube Goldberg this development is seriously more profound.

C Harget

I'm curious about how the rate of reaction compares to the needs of the motorist. How long does it take the enzymes to crack 300 miles worth of Hydrogen? Can they do it in 5 hours, which is what a motorist on a road trip needs? Do they have difficulty in freezing or sweltering weather? How much battery is needed, or capacitor, to handle regenerative braking and excess electricity generation? Is this reaction slowly going to create acetone or alcahol or something else that could pollute the system? What if the sugar water in the tank is accompanied by invasive micro organisms that foul the system?

This could be really cool, but the more complex the system, the more vulnerable it is.

Still, maybe this is the "Mr. Fuel Cell" like the "Mr. Fusion" from the Back to the Future movies.


Forgot to consider the potential cost of producing an energy system based on a sugar fuel cell versus a battery system. The 110 kWh A123 system would cost $1500 per kWh for the battery only at today’s prices. That is $165000 for the 110 kWh system and this is not including packing and cooling. Say they can get it down to $200 /kWh in the future 10 years from now including the packing and cooling. This is very optimistic. That would still be 110*200 = $22000 for the energy system and we need a car also. The sugar fuel cell car needs a 70kWh hydrogen fuel cell which they think they can do for $4000 in a few years say 10 from now. It also need a sugar tank which $100 or so. Finally it needs a sugar hydrogen reactor and that will be cheap because these enzymatic processes happen below 50 degree Celsius at atmospheric pressure. No need for strong expensive materials that can resist chemical corrosion. Still the enzymes may cost. A wild guess is that this reactor can be produced for less than $1000. Now that is $5100 for the total sugar fuel cell system versus the minimum $20000 for the comparable battery system. You can buy a lot of sugar for the $16900 difference. You can also use it for shipping and aviation. Let’s hope it will be 2018 for a sugar fuel cell car at the car dealer and that it will be 2011 for adequate proof that enzymatic sugar from cellulose is commercially viable (the 110mgy POET project that go online in 2009). The more I consider this news about sugar fuel cells the more I think that this is the most awesome and significant news I have ever read on green car congress. Let’s hope they will be able to execute on it as well.


Harget very good questions indeed. One thing I am certain of is that this sugar fuel cell car will be a PHEV vehicle. Because of the reaction time the sugar fuel is only suited as an energy dense range extender. A123 and others do not need to fear for their business. Indeed, it makes their batteries for PHEVs even more relevant. You can look at the graph in this report it shows it takes hours for this hydrogen to be generated. 20 hours for about 80% of the conversion from what I see. This is no good. Need to be minimized by a factor of ten at least. Still from the little I know about enzymes such improvements does not sound unrealistic. Maybe others could enlighten us on this.


The potential problem with freezing should not be a big technical issue. If the tank freezes it could be heated by excess heat from the fuel cell or the plug-in battery. The invasive micro-organisms issue could be solved by adding some chemicals (or even some enzymes) to the sugar water that kills the most micro-organisms but that does not interfere with the process of the hydrogen reactor. But that probably need to be tested extensively before a good solution can be settled. I still see the most important show stoppers as reaction speed and enzyme costs.


Just forgot that sugar water in a 50% resolution is very hostile to invasive micro-organisms. It may not be relevant to add a microbe killer at all. But of cause it should be investigated.

The comments to this entry are closed.