Amyris, perhaps the hottest biofuel company around, seems like the model American startup.
Few can match its pedigree. Several years ago, Amyris helped create a landmark achievement in medicine, engineering microbes to produce an expensive antimalarial drug. Related tricks, it later found, can create a liquid fuel similar to diesel. Well-heeled partners lined up at its door. The company went public last fall, and its stock has only risen since.
Based just outside San Francisco, Amyris could begin to push the United States off its oil-based economy toward what Energy Secretary Steven Chu calls the "glucose economy." Using crop-derived sugars as its power source, rather than petroleum, vats of Amyris' bugs could provide carbon-neutral fuel for fleets of heavy trucks and planes within a decade.
But if Amyris does all this, it won't be in the United States. It will be in Brazil.
While the company's corporate heart remains in Emeryville, Calif., Amyris is betting its future, and its dearly raised cash, on Brazil. Next year, Amyris' first two commercial-sized plants will open deep in the heart of sugar cane country. Construction is already under way.
Building in the United States was not an option, said Joel Velasco, a former lobbyist for Brazil's sugar cane industry who joined Amyris this year. Cane is "nature's solar panel," he said, perfect for one task: converting light and carbon dioxide into sugars. U.S. grain crops cannot compete on price or sustainability, he said. The math doesn't work.
"We're having a hard time making our production viable in the U.S., ultimately," Velasco said.
This is the harsh reality of the glucose economy. Sugar is king, and Brazil is poised to be the world's next Saudi Arabia. Blessed with tropical weather and abundant pastures that can be migrated to sugar cane cultivation, experts see a stark potential for Brazil's cane fields to grow almost without limit over the next decade.
Advanced biofuel companies, along with oil giants like BP and Shell, have taken notice.
Promising startups such as Solazyme, LS9 and Butamax have joined Amyris in Brazil over the past year. Basing their operations in São Paulo, or nearby ethanol boomtowns like Campinas, the firms are inking development deals and supply contracts with Brazil's agricultural giants. Meanwhile, BP and Shell have signed billion dollar deals buying into the country's already vibrant ethanol industry. The sugar rush is on.
Despite their marketing, which at times makes it appear that microbial diesel will pop up in corner gas stations next week, these startups will not initially look to produce fuels in Brazil. These petroleum replacements have proved far more difficult to produce efficiently than ethanol, the biofuel mainstay. Rather, the firms have shifted their focus to chemicals like cosmetics, fragrances and lubricants -- anything that can sell for a premium.
Much like when U.S. clean-tech manufacturers move to China, these young companies, many with fragile finances, are desperate to find their way in the dismal economics of the energy business. With politicians reluctant to put a price on CO2 emissions, the driving force of global warming, these firms have little choice but to chase the lowest-cost raw materials.
Oil is one of the world's cheapest commodities. Even if diesel biofuels are more efficiently created, they will be pressed to compete on an even playing field, given the costs of agriculture, said Harvey Blanch, the chief scientific officer at the Energy Department's Joint BioEnergy Institute (JBEI). The institute, which shares a building with Amyris in Emeryville, is a leader in biofuel research.
"When you come down to look at all of the economics of this, of course, then you realize where the issues really are," Blanch said. "Basically, it's the price of the sugar. Nothing else matters. The science, whatever -- doesn't matter."
Focus on speciality chemicals
The science does not make the business any easier, though.
Over the past decade, many companies have begun promising to use advanced biology to create what are called "drop-in fuels" -- biofuels that, unlike the corn-derived ethanol added to U.S. gasoline, would be indistinguishable from petroleum. The fuels would be as powerful as gasoline and would not corrode engines. Much like ethanol, secreted by yeast when it feeds on raw sugars, their bioengineered bugs would secrete oil-like compounds.
Most of these companies have since faced "Road to Damascus" moments.
Turns out, yeast is hard to beat, producing ethanol at staggering rates, close to the theoretical maximum. It is an evolution-improved process, used to balance the yeast's electrons as it chews through sugar. Humans call this process fermentation, which is why in the ethanol industry, the mash of water and alcohol formed by yeast is still known as beer.
"People love to hate ethanol," said Greg Stephanopoulos, a biochemical engineer at MIT. "They talk about all the weaknesses and problems with ethanol. And yet we keep talking about ethanol, all the time. And the question is, why? The answer is simple. It's very straightforward. We talk about ethanol because we can make it."
Facing these hard realities, most of the advanced fuel companies will not be selling diesel replacements at volume anytime soon. (Amyris has no designs on gasoline, where ethanol is tough to beat as a supplement.) Rather, they are focused on the specialty chemicals used in cosmetics and fragrances, where sale prices are much higher than the energy business. Vanity pays.
"Most of these biofuel companies are going to do chemicals first," said Jay Keasling, the CEO of JBEI and Amyris' co-founder. "They, like everybody else, have realized that you can make bigger [profit] margins on the chemicals, while you're increasing yield to get to the fuels."
Keasling, a bioengineering professor at the University of California, Berkeley, is a star of industrial biology. While not involved in Amyris' operations, the firm began out of his lab's work, engineering bugs to produce a precursor to artemisinin, an antimalarial drug that previously had to be created, at great expense, from sweet wormwood herbs. The drugmaker Sanofi licensed the technology, which it will sell at discounted rates. Production begins next year.
The Gates Foundation funded the malaria work, which Amyris then licensed royalty-free to Sanofi. But these humanitarian origins left Amyris with a dilemma: If it wasn't going to make money off malaria, what else was there?
Eventually, the company realized that since its drug precursor was a hydrocarbon, similar in many ways to molecules that make up crude oil, why not go big? Why not take on the world's largest chemical market?
The problem with such an aggressive plan, though, is that the tools of biological engineering are still crude, Keasling said. The hype and branding -- "synthetic biology" -- have gone far beyond the science. While someday genes may resemble Lego blocks that can be swapped and replaced at will, there is still much that mystifies biologists about even the most well-understood microbes, like E. coli and yeast.
"We can put something together, but actually getting it to work, for companies like Amyris and LS9, getting it to be high-yielding is a huge challenge," Keasling said. "Because the tools are still so primitive for biology. We need more tools. We need better [microbial] hosts. But if we have that, I think the potential is huge."
In Brazil, Amyris will be growing its bioengineered yeast in 600,000 liter steel tanks. The bugs will secrete trans-beta-farnesene -- Amyris calls it "Biofene" -- a hydrocarbon used in the natural world by aphids, a common plant pest, to warn their tribe of impending death. (In a nifty bit of evolutionary adaptation, potatoes also emit farnesene to warn off pests.) Beyond an aphid's death rattle, though, Biofene can also be used to build a host of high-end chemicals, like squalene, an essential ingredient of the cosmetics industry.
Cosmetic firms have been searching for an uncontroversial source of squalene for years. The chemical, used as a fast-acting moisturizer, is most commonly extracted from shark livers, a difficult fact to gloss over in an image-aware industry. (It also sells for $66 a pound.) Amyris expects renewable squalene to be its first commercial product, later this year, partnering with the French cosmetics developer Soliance in its production.
These are the kind of smaller-scale deals that are needed to make industrial biotech a reality.
With many investors burned earlier this decade by advanced biofuel dreams, even the most promising companies will have trouble raising the money needed to build chemical plants, let alone the cash to reach the dramatic scales demanded by the energy business. Capitalists need to see that advanced biology companies can match their claims, and the sooner the better.
"The industry needs to prove itself, and probably the quickest way to do that is through chemicals production," Keasling said. "Then do fuels later."
Room to expand
For decades, Brazil has unintentionally positioned itself as industrial biotech's ideal test case.
During the 1970s oil crisis, Brazil's military government began its Proalcool program to reduce the country's oil dependence. While U.S. ethanol production surpassed Brazil several years ago, thanks to renewable fuel standards and the American thirst for motoring, Brazil remains the world's largest ethanol exporter, threatening enough to the U.S. corn ethanol industry that the government has placed tariffs on ethanol imports.
Brazil's ethanol complex is fueled by sugar cane, which requires far less energy to grow than corn -- it needs little fertilizer, and its syrup-sapped husks, when burned, provide all the refinery's electricity. If you could design an ideal biofuel crop from scratch, it wouldn't be far off from sugar cane. The greenhouse gas emissions for cane ethanol are so low, in fact, that it qualifies as an "advanced fuel" for U.S. standards.
By 2022, U.S. regulations will require refineries to buy 136 billion liters of biofuels, with a cap that will ban traditional corn ethanol from supplying more than 42 percent of the mandate, about equal to current production. Since cane ethanol will be free of this cap, and other advanced fuels have proved difficult to make cheaply, several oil majors have bought into Brazil in a big way, betting cane ethanol will have a future in the United States.
In March, BP PLC announced its purchase of CNAA, a sugar cane producer, for $680 million, increasing its ethanol production to 1.4 billion liters a year. Meanwhile, earlier this year Royal Dutch Shell PLC finalized a joint-venture, called Raizen, with the Brazilian ethanol giant Cosan SA. The company, worth some $12 billion, will produce more than 2.2 billion liters a year, with plans to expand rapidly, said Mark Lyra, Raizen's director of new business.
"In Brazil, we already have something [cane ethanol] that will be part of the solution" to climate change, Lyra said. The infrastructure is in place and there will be room to expand, when and if new biofuel technologies are ready, he said.
Scientists expect that Brazil's sugar will remain the global baseline until they succeed in unleashing the complex "cellulosic" sugars found in grasses and waste like corn cobs. While research continues on this project -- the defining effort toward a bio-based economy, since it could decouple biofuel crops from food prices -- these sugars are a decade away from commercial viability, though demonstration plants are beginning to break ground.
Reflecting the complexities of global energy, though, Lyra concedes that right now, Brazilian ethanol cannot match U.S. corn ethanol on price, a finding ratified by recent independent research. The dramatic decline of the U.S. dollar against Brazil's real has bolstered corn ethanol, especially when shipping costs are considered. But despite this slide, Brazilian sugar prices -- the fundamental cost for companies like Amyris and LS9 -- remain the cheapest in the world, far less than the United States.
Price is only one calculation these firms make, of course. U.S. agribusinesses already have plenty of demand for their crops, and much of the country's best farmland is taken. They do not need to make bets. Meanwhile, Brazil's cane industry, less connected to food prices, has drastic room to expand, said Christine Crago, an agricultural economist at the University of Illinois, Urbana-Champaign.
"One thing Brazil has got going for it is that it has tons of land for expansion," she said. "There's a lot of interest in the country because you can do a lot."
Brazil grows cane on 11 million acres, an area dwarfed by the 157 million acres of pasture it devotes to low-density cattle ranching. Abiding by government cultivation limits, sugar cane could eventually be grown sustainably, without further deforestation, on 148 million acres, providing up to 14 percent of the world's fuel, researchers estimated last year in Science.
American farmers are simply victims of their latitudes, Amyris' Velasco said.
"You're not going to grow cane in the Dakotas," Velasco said. "You can grow some fantastic crops in the Dakotas. Cane is not going to be one of them."
'You've got to pay a price'
At times, the advanced biofuel world can seem awfully small.
Across the Bay from Amyris, in South San Francisco, two other startups, LS9 and Solazyme, are also planning Brazilian moves.
Solazyme, which last week launched a successful IPO, recently disclosed a major development deal in its securities filings with Bunge Limited, a large cane grower. LS9, meanwhile, has opened São Paulo offices, but none of its potential development deals are yet final, according to Ed Dineen, its chief executive.
The ninth life-sciences company founded by Boston-based Flagship Ventures, LS9 was co-founded by Jay Keasling before Amyris moved into advanced fuels. (Poke an advanced biofuel project's history enough and Keasling seems to fall out. With JBEI, Amyris and LS9, he has helped lead three separate research paths.) The Harvard geneticist George Church also co-founded LS9, along with Chris Somerville, now the director of Berkeley's Energy Biosciences Institute, a $500 million initiative, funded by BP, that takes a holistic look at the economics and technology of biofuels.
Despite all their credentials, though, these scientists are not infallible.
Take LS9, for example. It began based on a mistake, Church said.
"LS9 started based largely on some bad data in the literature that Jay Keasling and I had independently stumbled on," he said in an interview last year. The data showed that one type of bacteria, a Vibrio species, made a simple kind of hydrocarbon. Those studies, it turned out, were wrong. "That was not reproducible," he said.
That work got LS9 thinking about ways to move a bacteria's metabolism, the energy it typically uses for reproduction, toward secreting fuel molecules free of oxygen. There is a general rule of thumb that the less oxygen in a fuel, the better its quality. (Alcohols like ethanol contain, by definition, oxygen.) But stripping out oxygen is hard work. It takes millennia for nature to do in petroleum. On a human time-scale, it takes energy and cash.
"The public has to understand that it's going to be more expensive to do this," JBEI's Blanch said. "Of course it's easier to burn a CH2 fuel, a hydrocarbon. If you want it to be renewable, you've got to pay a price. And it's the price of that oxygen."
LS9's innovation came in finding a biological way to strip out oxygen.
The company's work serves as a quick model for how modern biology works as it creeps into the energy business. There were theoretical reasons to believe that cyanobacteria, photosynthetic bugs, contain a protein that extracts carbon monoxide. Recently, genetic sequences for dozens of cyanobacteria had become available, and LS9 found this protein in trace amounts, barely detectable, in several of these sequenced species.
Then, in the type of comparative analysis that has only recently become available thanks to the plummeting costs, LS9 identified the two genes responsible for the chemical process, called decarbonylation. It was like a primary school essay: compare and contrast.
"The LS9 team found it by basically subtracting genome sequences," Church said. They "subtracted the undetectable from the barely detectable, and the genes that were different were then inspected," he said. Two genes in a row did the trick, which the scientists then moved into an E. coli bacteria. The bug then produced alkanes in minute amounts.
The team published its results last summer, to wide acclaim, in the journal Science.
"That's pretty good proof," Church said.
It is easy to be mesmerized by the pioneering innovations created by Amyris and LS9.
Bleeding-edge biology can make a pleasing narrative of progress. But the energy business is so severe, the margins so thin, that many innovations will fade on the path to commercial reality. There are certain physical realities that any biofuel hopeful will find difficult to overcome, said JBEI's Blanch, himself a veteran of the 1970s ethanol push.
For one thing, it will always be more expensive to create hydrocarbon fuels, compared to equal volumes of ethanol. This is a simple, though rarely spoken, truth. Engineers talk about theoretical efficiencies, how much sugar a bug could turn to fuel if it devoted all of its energy toward the process. The theoretical efficiency of sugar to ethanol, for example, is 50 percent, and since yeast is a star performer, it reaches nine-tenths of that theoretical capacity.
Translated, this means it takes about 2 pounds of sugar to make 1 pound of ethanol.
Sugar to diesel, meanwhile, has a theoretical efficiency 33 percent. So if the microbes created by Amyris, LS9 and others reached the same type of performance as seen in yeast -- a far from certain possibility -- it would still take about 3 pounds of sugar for 1 pound of diesel. The sugar cost will always be 50 percent higher.
There are certain areas like renewable diesel and jet fuel where ethanol, low in energy compared to gasoline, won't work. But advanced fuels will find it difficult to compete in these spaces unless the cost of oil increases, Blanch said. He doubts any claims that a firm can create an petroleum replacement for under $100 a barrel.
"The reality is, it's easier to pump [oil] out of the ground than do anything else," he said. "And until the economics of that change, that's what's going to happen. And we will use oil until there's no more oil."
There are precedents for how biotechnology could begin to play out in the chemical industry. Scientists have seen it in pharmaceuticals. For the past decade, almost all of the industry's innovation has come from small companies that lacked the financial wherewithal to reach commercial scales, which were then taken over by the international giants, Keasling said.
"One could imagine the same thing in the chemicals and fuels industry," he said.
In fact, it has already begun. Recently, the chemical giant Dupont finalized its takeover of the Danish firm Danisco. While the press labeled Danisco a "food additives" maker, much of the company's potential is tied to its subsidiary, Genencor, a leader in bioengineering microbes. Like LS9 and Amyris, Genencor has created bugs that secrete hydrocarbons, and with Dupont, it now has the money to deploy them.
The comparison to drugs only goes so far, though. Exclusive technology is just not valued in the energy world, given how thinly the pie is divided. Even Exxon Mobil Corp. controls less than 3 percent of the world market, said Somerville, the Energy Biosciences Institute's director.
"It's not really worth fighting, when you're never going to have more than 3.2 percent of the market," he said. "They're not going to fight. Technology is actually going to move. It will be available to pretty much all the companies."
Somerville still owns shares of LS9. He wants these biofuel startups to prove themselves. But he doubts any of them will be the next Chevron or BP. Ultimately, they are likely to focus on smaller chemical markets and will license their bugs to the oil majors, if they are ever ready.
JBEI's Blanch suspects that in another decade, microbe-produced chemicals and polymers will continue to creep onto the market, perhaps beginning in Brazil. Progress will be incremental. There will be no breakthroughs. Thermodynamics are stubborn. But small successes could begin to change the way people and businesses think, he said.
"And that will force the politicians to make the hard decisions," he said.