Experts have estimated that powering all the world’s vehicles with biofuels could require doubling the amount of land under cultivation.

In Brazil, the world’s largest ethanol user, large parts of the Amazon rain forest have been turned into farmland as sugar plantations have spread.

This problem could become a lot worse in other parts of the tropics if biodiesel based on palm oil became more widely used.

"Promoting biofuels could very easily have negative outcomes," University of California, Berkeley energy resources expert Alexander Farrell said in a recent interview with Salon.com.

Nevertheless, Farrell believes in the potential of biofuels, providing their feedstocks and production methods are optimal.

Farrell and colleagues evaluated six major studies of ethanol, and found that it was 95% as energy efficient as petrol and that once the environmental costs of ethanol production were taken into account, it cut greenhouse gas emissions by 13%.

Farrell’s study was published in the journal, Science, accompanied by an editorial written by BP’s chief scientist, Dr Steven Koonin, that called for the identification and development of specialised biofuel crops.

"Credible studies show that with plausible technology developments, biofuels could supply some 30% of global demand," Koonin wrote.

"To realise that goal, so-called advanced biofuels must be developed from dedicated energy crops, separately and distinctly from food."

Currently most of the world’s ethanol is made from food crops. In the US, it is overwhelmingly made from corn, in Brazil the feedstock is sugar, and in Australia some manufacturers make it from wheat, while others use sugar.

Some environmental groups have long criticised ethanol production based on food crops for the huge fossil-fuel inputs that are required to grow and harvest the crops then convert them to fuel.

Yet they argue it is possible to make environmentally friendly ethanol based on cellulose-dense plants. Such fuel could have has far greater environmental benefits than the food crop-derived ethanol, according to a report by US lobby group, the Natural Resources Defense Council.

Cellulose ethanol is made from the inedible portions of renewable feedstocks such as grasses or wood.

Researchers already know how to convert cellulose from switch grass — or wood chips, corn stalks, wheat straw, old newspapers — into sugars used in making ethanol. The hard part is doing it cost-effectively.

The favoured crop for North American cellulose ethanol production is switchgrass, a tough native prairie grass with thick, strong stems rich in cellulose.

It offers significant landcare and economic benefits. Being a perennial, it does not require annual planting. This fast-growing crop can grow to more than three metres in many kinds of soils, including ground of little use for other farming, with little labour or fertiliser and it actually enriches the soil.

According to Cornell University researcher David Pimentel, about 15% of North America is unsuitable for food farming but useable for switchgrass cultivation. If all that land were planted with switchgrass, the US could replace its gasoline for vehicle use with cellulosic ethanol.

The Natural Resources Defense Council argues that based on the history of selective breeding for crop plants, switchgrass ethanol yield per hectare can be expected to increase rapidly over the next decade, reducing the amount of land needed, provided that necessary research is done.

In addition, switchgrass contains lignins as well as cellulose. Lignins are structurally similar to lignite, or brown coal, and could be used to produce biodiesel or to co-fire coal-fuelled power stations.

"When you look at the full fuel-cycle analysis of corn-based ethanol versus cellulosic ethanol, the latter has huge greenhouse-gas advantages," Natural Resources Defense Council climate-policy specialist Jeff Fiedler told Salon.com. But this energy-dense crop faces serious economic hurdles. The carbohydrates in cellulose are hard to convert into fuel, while the starches in grain and sugars in sugar cane deliver their energy easily.

While the ethanol yield per hectare for switchgrass is higher than that of corn and even of sugar cane, turning the cellulose in this fibrous plant into sugar requires extra steps and more enzymes.

Experts place the cost of converting cellulose into ethanol at three to five times the cost of converting corn for the same purpose, according to The Kansas City Star.

But research has already cut the cost of the enzymes used for the switchgrass conversion process 20-fold over the last four years, according to Fiedler.

Cellulose ethanol technology could also be used to make fuel from wood chips, plant stalks, straw, and paper and cardboard.

President George Bush’s proposed 2007 budget includes $US150 million — a $US59 million increase from this year — to help develop ethanol from the cellulose of products now considered “agricultural waste.” If the fuel can be made cost-competitive by 2012, the White House says, it could displace up to 30% of the nation’s current fuel use for transportation.

With more research, lowering production costs within Bush’s six-year window “is very do-able,” University of Nebraska Ken Vogel plant geneticist told the Kansas City Star.

But will take a lot of expensive research before switchgrass becomes economically viable, and it remains to be seen whether Bush will deliver on a commitment that hurt America's strong corn lobby and the oil industry.

Meanwhile, in Australia, researchers have also been looking at a native plant as a possible biofuel crop that also offers additional environmental benefits. But once again, years of research will be needed before it becomes a viable fuel.

Monash University project leader Dr Damon Honnery says wood from the Western Australian oil mallee tree could be super-heated to produce a vapour that could then be condensed into liquid for diesel engines.

Honnery said the oil was not yet suitable for car engines, as the filtering systems were too fine, and it had only been used in a test engine.

But he believed that appropriate refining processes could be developed within the next 10 years.

Planting these trees could also help Australia's soil salinity problem, as their complex root system reduced water table problems, according to Honnery.

"The fuel is almost greenhouse neutral, so we get three benefits – we can help solve salinity, we can help to solve and reduce greenhouse gas emissions from transport and we get to produce a fuel as well," he said.

The project, funded by the Australian Research Council, involves researchers from Monash, University of Melbourne and Aston University in the UK.

They are working out the complex chemical composition of the Mallee oil, how to produce a fuel from it and how it could be adapted to run diesel engines.