Biofuels' future threatened by political uncertainty

Early this year, US researchers unveiled a very simple chemical treatment that could release – literally – a vast resource of renewable energy.

By adding a little sulphuric acid to a plant-derived, innocuous liquid called gamma-Valerolactone, young chemist Dr Jeremy Luterbacher, Professor James Dumesic and colleagues from the University of Wisconsin-Madsion, created a solvent that dissolves the tough waste plant matter that feeds advanced biofuel production.

For decades, researchers have tried to devise methods to extract the difficult-to-reach complex sugars and alcohols – cellulose, hemicellulose and lignin – from agriculture and forestry waste biomass. Be it wheat stems, corn stalks or leaves, this woody lignocellulosic biomass is much more difficult to break down than the maizes and sugar cane used in first generation biofuels.

As a result, current methods rely on harsh heat and acid pretreatments to soften up the woody material, prise open the lignin and expose the cellulose and hemicellulose within. Enzymes are typically added to extract the sugars ready for processing to alcohol fuel. And herein lies the problem.

These advanced biofuel production processes may work – the first industrial-scale lignocellulosic ethanol plant is already up and running – but they also cost. The pretreatment process is not exactly energy-efficient and enzymes aren't cheap; which is why Dr Luterbacher's treatment has appeal.

Gamma-Valerolactone (GVL) is easily made from biomass and latest results, published in Science, reveal the treatment not only completely dissolves lignocellulosic biomass, but converts an incredible 85-95 per cent of the starting material to sugars.

"Nobody has ever shown that GVL can do this," says Dr Luterbacher. "There's a couple of more exotic solvents that seem to work, but don't quite get as high a yield. If yields from GVL were, say, 10 per cent lower, then I'm not sure if we would have published this in Science."

But it's not all about using GVL to rip open lignin and release the sugars. While the GVL treatment completely dissolves the lignocellulosic biomass leaving a valuable mixture of sugar solution and GVL, how do you efficiently separate the two? This is where the Wisconsin-Madison researchers' process gets clever.

Dr Luterbacher and colleagues found that adding liquid carbon dioxide to the solution – in the chemist's words "it doesn't take very much" – renders the GVL immiscible in water, yielding a two-phase solution of GVL floating on top of the sugary water.

"If we had to distil the GVL-sugar solution, this process would never work as you would have to use way too much energy," says Dr Luterbacher. "But all we do is add carbon dioxide, the GVL spontaneously separates and we skim it off the top. It took us forever to find a good way to separate the solution, and finally we stumbled on this."

Cheap second-generation biofuel

In terms of cost, the process looks good. For starters, the researchers can re-use the GVL; recycling the 'active ingredients' of any biofuel process is crucial to keeping costs down. But at the same time, they've found a relatively easy way to partly concentrate the sugars, easing full recovery for processing to ethanol or butanol.

As Dr Luterbacher says: "Someone else running a process entirely in water would still have to evaporate 80 per cent of the water to get here – that's a huge energetic cost."

Indeed, economic assessments indicate ethanol producers could cut costs by around 10 per cent using this process over state-of-the-art technologies based on biomass pretreatment followed by enzymatic hydrolysis.

Ten per cent might not sound like a massive reduction but as Dr Luterbacher says: "We started working on this something like ten months ago, and the fact is the cost figures from our current lab are lower than the enzyme processes that researchers have been working on for 30 years."

But as well as offering a relatively cheap route to second-generation biofuel production, Dr Luterbacher's process could also prove to be more flexible. As he and colleagues have demonstrated, the process is actually insensitive to the biomass feedstock. Trials on corn stalks and leaves, hardwood and softwood, have all yielded pretty much the same results.

"If you've used enzymes you'll know that if you try to work with corn stover [stalks and leaves] you'll get completely different results than if you work with softwood," he says. "This is due to, say, power, size and structure, and plant accessibility – all [variants] to which your enzyme is sensitive.

"But in our case the differences [between the different feeds] were statistically insignificant, we essentially get the same results," he adds.

The process holds massive potential, but industrial-scale production is a long way off. All results are based on laboratory-scale experiments, with Dr Luterbacher saying the team has been producing 20ml quantities of GVL-sugar solution in around two days. Plans to scale are underway.

"When you're scaling this type of technology, as I see it, at each stage you try to find something wrong, you try to kill it," he says. "So far we have tried everything, and each time we didn't find any way to kill it and that's why it's been so exciting, and that's why we published in Science. So far it looks great."

The researchers recently received university funding to develop a high-efficiency 'pre-pilot' plant that will produce around a litre of GVL-sugar solution a day. This will also be passed onto fellow investigators to optimise conversion strategies to fuels. According to Dr Luterbacher, he and colleagues have already run a couple of two-litre-scale reactions with promising results, and hope to have the pre-pilot plant operational come the summer with trials running for up to a year.

Assuming sufficient funds follow, a pilot plant is next – Dr Luterbacher reckons this would cost up to $20m – to demonstrate that the production of 100 litres a day is viable. And then, of course, commercial-scale production would follow.

However, the road to commercialisation is never easy, especially when starting from 20ml; still, Dr Luterbacher isn't fazed. "Our process is a first step that almost every biofuel has to include; depolymerising the plant to produce sugars," he says. "So far there's only been three or four ways to do this, so when you add a new way that's not completely crazy, that is good news for the industry."

Crazy or not, what is clear is that the process will take four years minimum to commercialise, during which time the cost of enzymes is expected to drop significantly. And as more and more researchers start to focus on developing new and cheaper enzymes, commercial acceptance of a brand new production process could get trickier.

Dr Luterbacher remains unruffled. "We've had meetings with a number of potential investors that are very excited," he says. "They want a demonstration at the pre-pilot level, and have said 'if you can match [current results] at pre-pilot plant level then we're going to put in some serious money'."

An enzyme alternative

As Dr Luterbacher and colleagues scale their GVL process, researchers worldwide are trying to develop more efficient enzymes for second-generation biofuel production. Greg Tucker is Professor of Plant Chemistry at UK-based Nottingham University and also director of the lignocelluose to ethanol project (LACE), part of a UK-wide sustainable bioenergy programme. He says: "[Enzymes] present a major economic barrier, due to the sheer quantity that needs to be used; this is of the order of 25kg of enzyme per tonne of biomass.

"They're expensive to produce," he adds, "you need to use quite a lot and they tend to be non-recoverable, making them a recurrent cost."

But as he highlights, the cost of enzymes has reduced twenty-fold in the last decade, thanks to improvements in production processes and efficiency gains, and will continue to fall.

Enzymes are traditionally produced by taking a fungus and removing its regulatory mechanisms so the organism only makes enzymes when grown on glucose. For biofuel production, the Trichoderma – a fungi genus found in soils – has already been widely explored. Here, the organism is grown on glucose and coaxed into producing the carbohydrate-degrading xylanase enzyme that digests hemicellulose.

What is now clear is that this fungi genus can produce dozens of different carbohydrate-degrading enzymes. What's more, other fungi, such as the Aspergillus Niger, can produce hundreds of these enzymes. This is where a wealth of research is underway, including that of Tucker's team.

"Our working hypothesis is that maybe we haven't always selected the right enzymes to regulate and you might actually need different enzymes depending on if you are degrading willow, miscanthus or wheat," says Tucker. "So we've gone back to basics, and we've grown fungus on willow and asked what enzymes it makes on this substrate. Then we've compared this to the cocktail of enzymes it makes on miscanthus."

According to Tucker, a fungus will produce a set of core enzymes, regardless of what it is grown on. However, a set of specific enzymes will only be made if the fungus is grown on a specific substrate.

Armed with this knowledge, his team is developing tailor-made cocktails of enzymes to target different woody biomasses. "We are also looking at how the fungus makes these enzymes and improving the production rates with a view to making the enzymes cheaper to produce and more efficient, so industry can reduce the amount it needs in biofuel production processes," he adds.

Research is also underway to unleash new, non-fungal enzymes sources. Researchers from the universities of York and Portsmouth in the UK, the University of Kentucky and the National Renewable Energy Laboratory in the US, and Danish biotech giant and enzyme developer Novozymes, have been scrutinising a wood-digesting marine organism called the gribble, known for chewing holes in piers and wooden ships.

The team was attracted to this marine wood-borer as its gut is devoid of microbes; so instead of relying on gut bacteria to digest the wood it eats, it produces its own enzymes to do the deed. Crucially, several years of research has shown that some of these enzymes belong to the same important class of enzymes harvested from fungi, representing a potential source of woody-biomass-digesting enzyme.

Pleasingly, the enzymes from the gribble appear extremely resistant to aggressive chemical environments. This means these molecules could function for longer in industrial conditions, so lower volumes could be used.

"The enzyme looked superficially similar to equivalent ones from fungi, but closer inspection highlighted structural differences that gave it these special features," says lead researcher Professor Simon McQueen Mason from York. "The enzymes would have a longer working life, and [could be] recovered and re-used during processing."

Cash hurdles

Clearly the ultimate aim is to reproduce the enzyme on an industrial scale rather than relying on the gribble, so the team has transferred the genetic blueprint of this enzyme to an industrial microbe that can be produced in large quantities. Studies are underway to find out how efficient these enzymes are at digesting lignocellulosic biomass, but the researchers are confident the novel approach offers "significant potential to be used in [second generation] industrial biomass conversion processes".

As Tucker highlights, fungal enzyme cocktails and novel alternatives such as the gribble are just two approaches that could make second-generation biofuel production more commercially viable. Other researchers are genetically modifying today's enzymes to make more efficient, artificial versions, while a totally different strand of research involves developing novel gasification processes to efficiently burn biomass and produce a synthesis gas from which biofuels can be reformed.

However, a critical issue overshadowing all these projects is cash. In the words of GVL pioneer Dr Luterbacher: "It's not so much the scientific issues around second-generation biofuel production that worry me, it's just that investment tends to yo-yo depending on government and its willingness to care about climate change; especially in the US.

"This all depends on the economy and political climate change; you can't control that," he adds.

Tucker voices similar concerns. "Second-generation biofuel generation had reached the stage where several companies were starting to invest in demonstration plants," he says. "My impression is these are mothballed at the moment, partly because of fracking... and also because fossil fuel costs have come down."

There is one leading light in the industry that has overcome these hurdles. Italy-based Beta Renewables is a joint venture between Biochemtex, developer of non-food biomass technologies, private equity firm TPG, and Novozyme.

In October last year, the biofuel manufacturer opened the world's first commercial-scale cellulosic ethanol facility. Sited in northern Italy and with a full capacity of 75 million litres a year, the plant is entirely self-sufficient and runs an acid-free steam pre-treatment as well as relying on enzymes, developed by Novozyme, to deconstruct its woody biomass feed.

"We have spent $150m on the first commercial-scale cellulosic ethanol plant in the world," says Michele Rubino, chief operating officer at Beta Renewables. "The plant has been operating for the last few months at 50 per cent of its target production rate, and is producing at commercial rates."

As Rubino explains, the business is currently building up its feedstock and "de-bottlenecking a couple of production steps". "Most of the plant has been demonstrated at 100 per cent capacity," he adds. "But you know, this is a first of a kind technology and that takes time to set up."

What is clear from talking to Rubino is that from the beginning, the business has focused on efficiency and cutting costs. As he explains, he and colleagues have first worked on making its production process as efficient and cost effective as possible.

"Clearly enzymes are a part of this and Novozymes has spent hundreds of millions of dollars on developing these," he says. "We now believe our enzymes are cost effective. In the early stages of this technology, Novozymes really co-developed its enzyme technology with our processes, and I think this is where we have a lot of the benefits."

The company also licenses out the technology for the process – called Proesa – providing its customers with an 'enzyme use cost guarantee', fixing the enzyme cost per tonne of ethanol. Rubino reckons the guarantee "provides strength to our claims" and is aiming to clinch up to 25 new licensing contracts by 2017.

Enzymes aside, the company also focused on developing an acid-free process that would accept a range of feedstocks. "Our first guiding principle was feedstock flexibility... we wanted to access a diverse pool of feedstock to get traction globally," he says. "We also didn't want to use chemicals to open up the lignin. Throw these in and you start getting degraded products that can inhibit downstream biocatalysts and you have to construct the plant with very special alloys. You really need to be cost-effective."

Proesa licensee GraalBio Investimentos, Brazil, has laid out plans to build several second generation plants while Biochemtex plans to build a biorefinery in North Carolina, US, using the Proesa technology.

With success already in hand, what has been the biggest stumbling block to getting the world's first commercial-scale second-generation biofuel production plant running? Predictably, political will and the lack of regulation.

"Policy just hasn't been supportive. In the US, policy was strong but has weakened over time due to resistance from the petroleum industry," he says. "Europe is even more challenging as a clear framework isn't in place, and honestly, once every country has to implement it's going to get very messy."

As Rubino emphasises, without a firm policy framework, investors are always reluctant to part with cash in the long-term.

"Sometimes I feel no policy would be better than this continuous back and forth, it's there, it's not there," he says. "Our technology can stand on its feet without policy but when it's put in place and then taken away, investors don't understand what they are getting into, and stay away."