This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Growing hydrogen for the cars of tomorrow
25 February 2006
From New Scientist Print Edition.
Peter Aldhous
Hydrogen machine
Hydrogen alternatives

Down at the farm, glistening polythene tubes stretch into the distance across the salt flats of the southern Californian desert. But they aren't propagating some miraculous new crop that can grow on this barren, sun-baked earth. These water-filled tubes are teeming with countless microscopic algae that have been engineered to soak up the sun's rays and produce hydrogen to fuel the state's cars and other vehicles.

That, at least, is the vision of Tasios Melis of the University of California, Berkeley. And he's not stopping at California. "We've done some calculations," he says. "To displace gasoline use in the US would take hydrogen farms covering about 25,000 square kilometres." To put that in perspective, that's less than a tenth of what the US devotes to growing soya.

Crucially, you can farm algae where conventional crops don't stand a chance. The best areas will be sun-drenched deserts like the salt-covered dried lake beds in southern California, says Melis. "Nothing will grow on that salt, but it reflects light."

These hydrogen farms are still a dream, but if Melis and those who share his vision are successful, the first one could be built within a couple of decades. Not only would they provide an environmentally friendly source of hydrogen - the most likely fuel for the post-gasoline generation of vehicles - but they might also be cheap enough to take off in developing economies, which at present look set on an environmentally catastrophic path to a future dependent on fossil fuels. "Poor countries could use this technology," says Melis.

Melis isn't the only one banking on algae to deliver cheap hydrogen. Several research groups are working on ways to persuade the microbes to produce the gas, either by genetic engineering or by manipulating the composition of their growth medium. And the US Department of Energy (DOE) is investing several million dollars a year in the research to see if it can be made commercially viable.

The idea is incredibly simple: take a transparent tube full of water, inoculate it with the green alga Chlamydomonas reinhardtii, and put it out in the sun. Algae naturally produce hydrogen as a by-product of photosynthesis, so all you need is a system to collect the gas and you've got a working hydrogen farm.

But before you rush to invest in the technology, bear in mind that realising the vision will require some formidable feats of bioengineering. So far, researchers have got no further than a few prototypes in a clutch of labs. Melis assumes that a commercial hydrogen farm would need to convert the energy of sunlight into hydrogen with an efficiency of about 10 per cent. Right now, the algal cultures that he and other researchers are experimenting with are less than 0.1 per cent efficient at producing the gas. Making C. reinhardtii belch out large quantities of hydrogen will require a radical re-engineering of photosynthesis in algae.

Photosynthesis involves two processes. In the first, chlorophyll and other pigments harness the energy of sunlight to split water into oxygen, protons and electrons. The electrons and protons are used to create molecules that provide the chemical energy for a second process, the Calvin cycle, which converts carbon dioxide and water into sugars. The plant then consumes these sugars as it grows.

Although only the process of splitting water requires light, both reactions shut down at night. When the sun returns each morning, the water-splitting reaction gets up to speed more quickly, so cells can suddenly find themselves producing reactive electrons that go unused by the still sleeping Calvin cycle. If not mopped up, they could damage the cell's photosynthetic apparatus. To prevent such biochemical havoc, a hydrogenase enzyme dumps the electrons onto protons, forming harmless hydrogen.

However, the hydrogenase is inhibited by oxygen. So as photosynthesis gets going, oxygen released by the splitting of water soon deactivates the enzyme. By this time the Calvin cycle is up and running, and the cell gets on with the business of making sugars.

This is a big problem for hydrogen farmers. The last thing you want is for the mechanism that produces hydrogen to shut down shortly after sunrise. So how do you help the hydrogenase to keep working through the day?

The first hint that it might be possible to subvert the oxygen feedback mechanism came in 2000. From previous research Melis and Mike Seibert of the National Renewable Energy Laboratory in Golden, Colorado, had a hunch that reducing sulphate levels in the C. reinhardtii cultures would slash the rate of photosynthesis. Experiments showed that the rate dropped by 90 per cent. As a result, oxygen levels fell to a point where the hydrogenase enzyme kept working, diverting electrons to produce hydrogen instead of making sugars.

However, simply reducing sulphate levels isn't the answer, since C. reinhardtii cells get sickly and die when starved of sulphates for a week or more. So Melis is now trying to block the production of an enzyme that moves sulphate into the chloroplast, the cellular structure in which photosynthesis takes place. To do this he is genetically engineering the alga to trigger RNA interference, a mechanism which can selectively shut down genes - in this case the one for the sulphate transport enzyme. He hopes this will reduce the rate of photosynthesis by about 80 per cent, so the cells should stay healthy rather than die, and yet still continue to produce hydrogen.

Seibert, on the other hand, thinks crippling photosynthesis isn't such a good idea. Reducing the rate of photosynthesis inevitably slashes the amount of hydrogen the algae produce, and that is not something a hydrogen farmer would want to do. "If we want to make full use of the sun's energy, we have to address the core problem: the sensitivity to oxygen of the hydrogenase enzyme," he says.

To tackle that problem, Seibert has turned to a related hydrogenase enzyme whose three-dimensional structure has been determined in detail, from the bacterium Clostridium pasteurianum. Through computer simulations of the enzyme's behaviour, Seibert and his collaborators have shown that oxygen diffuses into the enzyme along two distinct pathways, whereas hydrogen escapes by a multitude of routes. So they reasoned it should be possible to mutate the enzyme so it no longer lets oxygen in, yet still produces hydrogen.

Seibert and his colleagues are creating specific mutants of the enzyme that should, according to their computer simulations, block the oxygen pathways into the cell. The trick is finding modifications that don't compromise the activity of the enzyme. So far they have identified an oxygen-blocking mutation that boosts the enzyme's tolerance to oxygen by up to 30 per cent. Now they are working to create double mutants in which both pathways are disabled. "We're trying to 'button up' the protein," Seibert says.

If they are successful, it should be relatively straightforward to create similar mutant versions of C. reinhardtii's hydrogenase - or alternatively, to genetically engineer the algal cells to make the mutant bacterial enzyme in place of their own version.

But that still leaves another obstacle. When algae make hydrogen rather than sugars, one side-effect is that concentrations of protons accumulate near the photosynthetic apparatus, acting like a build-up of static electricity. This inhibits the transport of the electrons split from water. "It will limit the rate of hydrogen production by a factor of more than 10," says James Lee of Oak Ridge National Laboratory in Tennessee.

Lee has designed genes to produce novel proteins that should pump the excess protons away from the photosynthetic apparatus (see Diagram). Having built different variants of these genes from scratch and inserted them into algal cells, he now has to exhaustively test each variant to find one that has the desired effect.

Finally, to maximise the efficiency of a hydrogen farm, it will be important to ensure that sunlight permeates throughout the tubes containing the algal cultures, rather than being completely absorbed by cells at the surface. Melis thinks the answer could be to tinker with how much photosynthetic pigment the algal cells produce, so that they don't shade one another too much (see "Lighter shade of green").

Given the sophisticated genetic manipulation needed to make hydrogen farming a practical prospect, the big energy companies are content to watch from the sidelines. It is difficult for engineers to think about designing a hydrogen farm until they know the precise characteristics of the microbe they'll be dealing with, says Steve Schlasner, head of long-range R&D with energy firm ConocoPhillips in Bartlesville, Oklahoma.

For example, if Seibert succeeds in creating his oxygen-tolerant hydrogenase, the algae will pump out both hydrogen and oxygen, and farms will need to incorporate a mechanism for separating the two gases safely. Melis's idea of limiting the production of oxygen in the algal culture, while making much less efficient use of sunlight, would avoid this problem and mean the plant could be much simpler.

Meanwhile, the DOE is hoping its investment will bear fruit. Farmed hydrogen would be the ultimate green fuel, with virtually no damaging impact on the environment. "We know the concept has merit," says Steve Chalk, who manages the DOE's hydrogen programme. But it has to compete favourably with every other method of hydrogen production to retain the DOE's backing.

By 2015, the DOE calculates that hydrogen will have to cost no more than $3 per kilogram to be viable, making it no more expensive than gasoline. At that date, the DOE will concentrate its investment on whatever technologies seem most competitive, and drop the rest.

Based on their results to date, researchers working on hydrogen farming think they're in with a good chance. According to Melis, the cost of farmed hydrogen might eventually fall as low as $1.40 per kilogram. It would set the stage for a whole new breed of energy companies, staffed by people with a different outlook from the engineers and geologists that have dominated the oil industry. "You would have to think more like a farmer," says Melis.

From issue 2540 of New Scientist magazine, 25 February 2006, page 37

Lighter shade of green

When Tasios Melis at the University of California, Berkeley, set up his prototype hydrogen "farm", the results were visually spectacular. He seeded a polythene tube full of water with the alga Chlamydomonas reinhardtii, and within a week, the clear water turned almost black with multiplying algal cells. "It was so dense that you couldn't see through it," he says.

But therein lies a problem. Algae such as C. reinhardtii have evolved to photosynthesise under murky conditions, not in the full glare of the sun. To make the most of gloomy conditions, they possess arrays of chlorophyll and other pigments organised into structures called "antenna complexes" that are remarkably efficient at soaking up sunlight.

In a sun-drenched hydrogen farm, however, these extensive antenna complexes wouldn't be necessary, and would in fact prevent sunlight from reaching the cells at the centre of the tube. So Melis is now working to engineer algae that contain less chlorophyll.

The antenna complexes of normal C. reinhardtii cells contain a total of 470 chlorophyll molecules, but they should still be able to photosynthesise if these were stripped down to just 132 chlorophylls - and Melis has calculated that this would increase a hydrogen farm's productivity by a factor of four.

Unfortunately, that alga doesn't yet exist. So Melis is creating a strain that has these properties. To do this he is going through the laborious process of making thousands of mutant C. reinhardtii, disrupting their genes by inserting marker sequences of DNA into the cells, which get randomly incorporated into the genome. So far, his team has identified five promising mutants - one of which is close to his target minimal complex size.

For the latest from New Scientiist visit

Academy disclaimer: We cannot guarantee the accuracy of information in external sites.