Power plants: Grow your own electricity

• New Scientist, 16 February 2012 by Caroline Williams
• Magazine issue 2851. 
• http://www.newscientist.com/article/mg21328512.300-power-plants-grow-your-own-electricity.html?full=true

Imagine charging your cellphone from a meadow or harvesting electricity from rice paddies. The technology works, but can we make plant power a staple crop?

IT DOESN'T look much like the future of electricity generation. From where I'm standing, the small square of windswept greenery on the roof of the Netherlands Institute of Ecology in Wageningen bears a striking resemblance to the patch of straggly grass that passes for my back lawn. Yet according to its creators, this green plot could revolutionise the way we power our lives. With a tangle of bright red cables spilling out from among the plants' roots, this grass is wired to the hilt and produces electricity day and night. And if the technology works up here, there's no reason why any marshy field, weed-strewn ditch or plain old pot plant couldn't do it too.

No, really. While humankind has been digging fuel from the ground and burning it for power, living plants and the soil they grow in have been hiding some serious potential as electricity generators. Now a handful of researchers are exploring ways to tap this ability. Some want to use tree-powered sensors to monitor remote forests, while others, including the Wageningen team, see more potential in harvesting electrons released among plant roots. They think grasses, reeds and other plants could eventually generate a significant portion of our domestic electricity needs, making juice that will be even greener than power from solar panels or wind turbines. The technology could offer a reliable source of electricity for some of the 1.5 billion people around the world without access to a supply grid. It could even help reduce climate-warming methane emissions from wetlands and paddy fields into the bargain.

Previous attempts at harnessing plant power have been fairly small-scale. In 2006, Illinois-based inventor Gordon Wadle spotted that he could make a small current flow from an aluminium nail hammered into a tree trunk and connected to a copper electrode stabbed into the soil. Along with MagCap Engineering, a small electronics firm in Boston, he patented the idea and began devising a way to extract the electricity.

Despite Wadle's claims, many remained sceptical. After all, as people pointed out, simply linking two different metals with a conductor creates a small current as one metal will donate electrons to the other.

Physicist Andreas Mershin of the Massachusetts Institute of Technology was doubtful too. Yet when MagCap Engineering approached him to investigate, he set a student, Christopher Love, to work on the case and after several experiments with a small fig tree in a pot, Mershin changed his mind. Even when both electrodes were made from the same metal and the plant was shielded from all sources of electromagnetic interference, the circuit still produced a trickle of power.

Love and Mershin soon decided that hydrogen ions were responsible. They noticed that a small excess in the concentration of these positively charged ions in the soil compared with the tree's sap was enough to set a tiny current of electrons flowing from the tree through a wire and into the ground (PLoS One, vol 3, p e2963).

Encouraged, MagCap created a company, Voltree Power, to commercialise the idea. In 2009, it began constructing a forest-fire monitoring network for the US Forestry Service and Bureau of Land Management, powered by the trees it was designed to protect.

Three years on, the prospects for tree power aren't quite as rosy. It turns out that trees can't generate as much power as Wadle had hoped and the system can't be scaled up reliably: as more sensors are added, the juice needed to run them outstrips the power the trees can provide. So the company has had to combine tree power with electricity generated by solar panels, vibrations and even small differences in temperature. Together they make enough power to take measurements and relay them to the authorities via cellphone networks.

Yet the story of plant power doesn't end here. Around the time Wadle and MagCap were hammering nails into trees, Bert Hamelers of Wageningen University in the Netherlands was out for a stroll. "I saw a big, beautiful tree and remembered that the root system is roughly equal in size to [what's visible] above ground," he recalls. Hamelers began to wonder whether microbes living among the roots might be used to generate electricity in a special type of fuel cell.

Roots to the rescue

Conventional fuel cells combine oxygen with a fuel such as hydrogen to make water and electricity. They rely on a pair of electrodes made from expensive metals such as platinum to strip electrons from the fuel. Hamelers was working instead on microbial fuel cells, which exploit the enzymes in living bacteria to do much the same job. The concept had proved itself with devices running on organic matter such as plankton and human sewage, but these systems must be constantly fed with fuel. Hamelers realised that plant roots might help with this. Around half of the carbohydrates and other organic molecules that a plant produces through photosynthesis are released from its roots. As long as it lives, a tree will keep churning out the perfect feedstock for a microbial fuel cell.

In practice, trees aren't the best plants for the job. Their roots are big and run too deep to be wired up effectively. Plus, when bacteria break down these carbohydrates, they donate their spare electrons to oxygen, creating water and carbon dioxide. But it is a different story in waterlogged soil where there is no oxygen: here, anaerobic bacteria produce carbon dioxide along with free protons and electrons. The electrons usually react with sulphates or nitrates in the soil but Hamelers realised he might be able to divert these electrons if he placed suitable electrodes close to the bacteria that generate them (see diagram).

What Hamelers needed was a plant with shallow roots that thrives in damp or waterlogged soil where oxygen is scarce. So along with David Strik, an environmental biotechnologist at Wageningen University, he started experimenting with grasses and reeds that grow in damp salt marshes.

Initial results were modest, producing just a few milliwatts. Lab-scale set-ups were soon producing about 200 milliwatts for every square metre of greenery (Bioresource Technology, vol 101, p 3541). Over the next two years the researchers more than doubled this output to about 500 mW/m². At this level, the 16 square metres of grassy roof at Wageningen could easily generate enough juice to charge a cellphone.

That was a step forward, but the output still couldn't cover much of a household's energy use. Besides, in northern Europe, a field of solar panels or wind turbines typically generates between 4 W/m² and 7.7 W/m², so there is obviously some way to go before plant power can compete. "The current densities required to make these devices competitive in the energy market are a factor of 10 to 100 times larger than presently possible," admits fuel-cell researcher Willy Verstraete from Ghent University in Belgium.

But don't dismiss plant power just yet. Strik points out that he and his colleagues have already increased power output fiftyfold since the project began five years ago. There is still plenty of room to improve efficiency at every stage in the generating process, he says, so increasing output by a factor of 10 or more is simply a matter of tweaking the system. Strik is so confident he can reduce power losses that, with fellow environmental technologist Marjolein Helder, he has set up a company called Plant-e to bring the fuel cells to market.

Taking on renewables

The duo have also joined forces with Verstraete at Ghent. Along with botanists, microbial specialists and engineers from across Europe they are part of Plant Power, an EU-funded project with a budget of €4 million, which aims to increase the efficiency of these fuel cells. Between them they calculate that an output of 3.2 W/m² is within reach.

Hitting this ambitious target will take plenty of work. For starters the researchers must find plant varieties that secrete more organic molecules into the soil - plants such as sugar beet, for example, are among the best, converting sunlight into sugars with an efficiency of about 7 per cent. They must also tinker with the soup of microbes in the soil to find the best mix to maximise electricity output. This mix is likely to include bacteria that specialise in breaking down organic matter and others that are particularly good at releasing electrons.

Then there is the matter of engineering the fuel cell itself. Here, electrode design is key. The Dutch team uses an anode made from a slurry of small graphite granules through which the plant's roots grow. However, this passes electrons to the cathode faster than the cathode can oxidise them, limiting the cell's output. Strik suspects that providing the cathode with the right mix of bacteria should help it work faster at combining electrons, protons and oxygen to make water.

Finally, there is the challenge of scaling up the technology. The Wageningen team calculates that if an optimised lab-based system is used outdoors, they could realistically collect only about half its output: around 1.6 W/m². Though this is about a fifth of the maximum output produced by the latest wind or solar energy sources (see diagram), Strik points out that it is more efficient than simply using crops for biofuels, and constructing a plant-based generator won't require high-tech factories or the complex engineering associated with photovoltaic panels and wind turbines. With lower costs come shorter payback times.

Strik calculates that an array of plants in an optimised rooftop generator could produce up to 14 kilowatt-hours per square metre every year. Considering that the average Dutch home uses 3500 kW-h annually, 50 square metres of roof would provide around 20 per cent of its electricity needs. And these wired plants will do more than help keep the lights on. They offer all the other benefits of a "green roof": additional insulation, rainwater storage and a useful urban habitat for wildlife.

Strik and Hamelers also think their technology could eventually be installed in marshy ground near rivers or coasts, where it could make electricity without spoiling the view. Unlike conventional solar power, plant generators continue to work after dark. And as long as the technology doesn't retard plant growth, it could show up on any suitable agricultural land too. Initial experiments at Wageningen haven't indicated any reductions in growth rate, says Strik, and in some cases their wired-up plants grew better than expected.

The technology should have particular appeal in Asia, where it could be used to turn millions of hectares of rice paddies into power stations. The prospect has struck a chord in Japan, a country that lacks natural sources of energy but where paddy fields cover around 12 per cent of the land. At the University of Tokyo, biologist Kazuya Watanabe has already started exploring the idea. He has even dug a fuel cell into the microbe-rich sludge of a rice paddy field to see if it yields any power.

It does, but not much. Watanabe's field experiments yielded about 10 mW/m²(Bioscience, Biotechnology, and Biochemistry vol 74, p 1271). In collaboration with a company that manages agricultural land, he is continuing his field experiments in the hope of boosting output.

Watanabe faces a major challenge. Unlike the Wageningen team, which is building its plant arrays in modular trays, a system suitable for rice paddies must be easy to lay and tough enough to survive the rigours of life on a farm. At the moment, Watanabe's anodes and cathodes are thin felt mats impregnated with graphite. But their efficiency is low - so low that he has tried dispersing fine platinum powder in the mats to improve things. Along with other modifications, this has tripled the output from his fuel cells. But since platinum is expensive, he admits his system isn't yet cost-effective.

If some of these barriers could be overcome, plant power would begin to look attractive. "It is clearly worth pursuing, if only because it could readily be applied almost anywhere, including urban and rural areas, and on essentially any scale," says Walt Patterson, energy analyst and consultant for London-based think tank Chatham House.

There is another incentive for wiring up fields in this way. According to Verstraete, the technology could make a significant contribution in the struggle against global warming. The anaerobic conditions that suit plants such as rice and reeds are also prime environments for a class of bacteria that breaks down organic matter, releasing electrons which then generate methane, a potent greenhouse gas. Putting an anode into the soil provides an alternative route for these electrons and - if it is efficient enough at harvesting them - this should make a dent in methane generation. With rice paddies contributing up to 20 per cent of the world's methane emissions, reducing this while producing modest amounts of electricity on the side seems to be a no-brainer.

Back at Wageningen, it's clear that the plant-power team hasn't lost sight of just how cool it is to have plants that not only brighten up the place but provide free energy too. As I leave, one of Strik's students is repotting a banana plant using soil containing an electrode of graphite granules, the first step in changing it into a living gadget-charger. It is part experiment, part amusement, but if this plant can do it, says Strik, the design might just find its way onto the market. Banana-powered cellphones. You heard it here first.

Caroline Williams is a freelance writer based in Guildford, Surrey