BOSTON -- Bioengineer Jeff Way has seen what happens when the claims of algae biofuel companies get ahead of the science, when their promises of "renewable diesel" slam into the realities of engineering.
He's been to the bankruptcy auction.
Once the standard-bearer for the algae revolution, GreenFuel Technologies failed almost two years ago. Spun out of the Massachusetts Institute of Technology, the company promised to convert waste carbon dioxide into fuel-producing algae. It opened a celebrated -- and, it turns out, expensive -- pilot plant in Arizona. It raised more than $70 million in private funds. Then it went bust.
A former biotech pharmaceutical developer, Way hunted for deals at GreenFuel's equipment auction.
"It was a very sobering experience to see the death of a biotech company," said Way, who works on bioenergy at Harvard University's Wyss Institute for Biologically Inspired Engineering. "This company just went belly up. And I was buying their light bulbs for pennies on the dollar."
Few stories in the energy business are as seductive as that of algae biofuels. Using sunlight, CO2 and little else, many varieties of fast-growing pond scum, when starved of nutrients, quickly build up oil in their cells. They need no external sugar from corn or cane to grow, so they don't compete with food crops. Farmed in ponds or translucent reactors, microalgae can be raised on cheap, sun-splashed land that is unsuitable for crops or much of anything else.
That was the idea, anyway, of a host of startups that launched into algae fuels over the past half decade. Often ignorant of algae's biology, these companies stumbled into major physical and engineering hurdles that can derail most of their lofty goals, industry and government experts say.
Even the most promising approaches are a decade or more away, experts say. By then, many firms will have failed.
Most of these problems were predicted -- and ignored. Algae are speedy growers, but they are also greedy, preventing sunlight from penetrating even moderately dense concentrations. The most popular algae varieties, which resemble single-celled plants, remain difficult to bioengineer. Water requirements are vast. And while pond scum are prolific in building up oily fats, they surrender their dearly built energy stores only if they are killed and broken apart.
With investors wary, the algae bloom seems to be dissipating.
But before the green tide completely turns red, several groups of scientists and a few controversial startups are calling for what they say amounts to a shift in focus. By fixating, they say, on what was bleeding-edge technology in the 1970s -- tortured, plump microalgae -- many have missed what could prove the most promising path to algae fuels: photosynthetic bacteria.
That's the pitch of Joule Unlimited, one of the most-watched biofuel companies to emerge in recent years. Backed by a group of high-profile biologists -- including Boston University's James Collins and Harvard's George Church -- Joule has raised a racket with its grandiose claims.
Using heavily engineered photosynthetic bugs called cyanobacteria, Joule promises to produce hundreds of barrels of fungible diesel fuel per acre a year at the cost of $50 a barrel, a productivity that would outstrip nearly any other biofuel, including corn ethanol.
Joule is not alone in its cyanobacteria work. Synthetic Genomics, the bioengineering firm begun by Craig Venter, the enterprising biologist famous for being one of the first to sequence the human genome, is also investigating the bugs in its $300 million algae collaboration with Exxon Mobil Corp. Flush with cash, the company operates in relative silence.
Government-funded scientists are pushing the limits of using the bacteria as mini-factories. And smaller, private companies like Easel Biotechnologies and Proterro have put forward innovative, if far removed, applications for the bugs.
Unlike microalgae fuels, cyanobacteria have not received buckets of government cash. But Joule has attracted attention from a few influential figures inside the Beltway, most notably John Podesta, the CEO of the left-leaning Center for American Progress and former co-chairman of President Obama's transition team; Podesta was added to the company's board of directors early this year. And late last year, Sen. John Kerry (D-Mass.) dropped by the firm's Cambridge, Mass., lab to praise the company's "game-changing" technology.
At Joule's office, just blocks from MIT, the company's chief executive -- and chief fundraiser -- Bill Sims said he cannot blame people for being suspicious of Joule's claims.
The company has taken to decorating its office with models of winged pigs, seen reclining on a conference table or, during a tour, snuggled next to a researcher's computer, a few rooms away from the clichéd-but-hypnotic sight of emerald-blue bacteria swirling in incubating flasks.
"We respect that it's a big idea," Sims said. "And with big ideas comes skepticism. That's fine. If there wasn't skepticism, it probably wasn't a big enough challenge to be in this."
As history makes clear, there are no certain bets in harnessing photosynthesis. Joule could be on a path to -- as Obama would put it -- "win the future," and jump-start the United States on the path to a bio-based economy, independent of oil.
Or, like GreenFuel's light bulbs, Joule's winged pigs could end up in a few years at a bankruptcy auction, having never flown.
At first glance, cyanobacteria and traditional microalgae seem alike. Both are green and single-celled, forming the blooms that provide an estimated half of the world's oxygen. In fact, cyanobacteria used to be known as blue-green algae -- just another kind of pond scum. It is suspected that in the planet's murky past, a cyanobacteria ancestor merged with microalgae, providing the genetic tools used to tap the sun's rays.
Those tools, the machinery of photosynthesis, remain similar in both bugs.
The sun's light streaks into tiny chlorophyll-dyed antenna on the cell's membrane, creating energy that strips water into oxygen and hydrogen. The oxygen is released -- we can be thankful for that -- while the charged hydrogen creates the chemical energy needed to convert CO2 into sugars, oils and other molecules, a process known as carbon fixation.
Despite these similarities, microalgae and cyanobacteria belong to different domains of life.
Microalgae have an intricate cell machinery, wrapping most of their DNA in a tightly bound nucleus; cyanobacteria hold their DNA in easily accessible, circular chromosomes. Zoomed in, they have starkly different sizes, algae running 1,000 times cyanobacteria's volume. And, most importantly to biofuel dreamers, microalgae naturally store oily fats in their cells; cyanobacteria do not.
Microalgae's oil production meant that, for the past century, they have trumped cyanobacteria in industrial labs.
During World War II, German scientists first attempted to produce oil from the microbes, discovering that green algae, when deprived of nutrients, devoted more than two-thirds of their weight to oils. The oil built up slowly, though, and the algae, investing their energy in survival, grew at a tepid rate -- a problem to this day.
After the war, mass cultivation of algae began in a few modest, translucent bags on an MIT rooftop, focusing on protein-rich food production. The 1950s soon saw a bubble similar to the past few years, with companies saying they could rapidly expand algae production in three years with drastic drops in price, said Rene Wijffels, an engineer at Wageningen University in the Netherlands. Those claims amounted to little, he added.
Given this precedent, it was natural that the Department of Energy's Aquatic Species Program, which was launched in response to the 1970s oil crisis and ran from 1978 to 1996, focused on fat-producing algae.
Scientists collected some 3,000 strains, several of which seemed ripe for biodiesel. Ponds were built in Roswell, N.M., to demonstrate that the algae could be, in effect, farmed. (The researchers soon discovered, to their chagrin, that wild algae had overtaken the ponds.) The program ended in the face of persistent low oil prices, but in an influential final report, its leaders prophesied algae's eventual return.
Meanwhile, out of DOE's sight, academic scientists were busy discovering that many varieties of cyanobacteria could easily integrate foreign DNA, much like E. coli, another industrial workhorse. Since the bugs did not produce oil, researchers tweaked them to study photosynthesis.
There was little emphasis on application.
That industrial agnosticism changed this past decade, as scientists saw the results achieved by labs heavily modifying bacteria to produce malaria drugs or clothing fabrics.
Cyanobacteria, they realized, are ideal fuel candidates, said Wim Vermaas, a microbiologist at Arizona State University who recently received a grant to develop fuel-producing cyanobacteria from DOE's Advanced Research Projects Agency-Energy (ARPA-E).
"You don't have a nucleus," he said. "You don't have mitochondria. You don't have all the infrastructure to worry about. The volume is a thousand-fold smaller. The diameter is tenfold smaller. You really have properties you like."
Until a few years ago, using cyanobacteria would have seemed a pipe dream, given the work needed to get them producing fuels. But in the past three years, DNA sequencing costs have crashed, allowing researchers to spell out a parade of bug genomes.
The decline in costs is staggering. Today, a moderate-sized microbe costs $5,000 to sequence. A few years ago, it would have taken $100,000 and many months, said Donald Bryant, a microbiologist at Penn State University.
By year's end, Bryant estimates, close to 100 cyanobacteria strains will be sequenced, yielding a bounty of data.
Bryant, a Joule adviser, is sequencing four close relatives of his favorite cyanobacteria, hoping comparisons will reveal important functions hidden in its genes. Similar work has revealed diesel-producing pathways in E. coli by LS9, a corporate peer of Joule.
"The power of these genome databases is really quite remarkable," Bryant said.
The price drops in DNA sequencing and synthesis make a startup like Joule possible, if not necessarily viable. The company can now send a gene out to be sequenced at 5 o'clock in the afternoon and receive it at 8 in the morning, said Dan Robertson, Joule's chief biologist and first employee. It can then tweak the gene in software and contract it out for synthesis.
"In two weeks we can have a gene synthesized that has all of the decorations on it that you need to be able to incorporate it into your organism," he said. "That is a tremendous achievement."
Major technical challenges
While sequencing costs have dropped for microalgae, too, these oil-producing bugs have lagged behind cyanobacteria in advanced labs. Only a few species have had their genome sequenced, and even the best-understood species, a green algae, remains troublesome to manipulate.
The types of heavy-duty alteration used in bacteria -- and likely required for viable fuel production -- are, in a rosy scenario, more than a few years off for most microalgae.
This lag in genetic tools is an overplayed problem, though, according to Stephen Mayfield, a founder of Sapphire Energy, which plans to grow algae in open ponds. The company has received more than $100 million in grants and loan guarantees from the federal government to develop its first biorefinery in Columbus, N.M.
While the company plans to bioengineer algae, it also maintains that modern selective breeding can produce viable fuel candidates, said Mayfield, who also directs the San Diego Center for Algae Biotechnology (Greenwire, Sept. 17, 2009).
"If I have a single [breeding] event, one in a million, I'll find it in one day," Mayfield said. "I can cross it with another algae in a week. ... I bet we can do in the next four to five years what [they] did in agriculture in 50 years."
Microalgae's model should be agriculture, Mayfield added. Like in farming, scientists need to improve the productivity of algae, and they need to protect it from predators like water fleas and fungus. The algae need to produce appropriate products, and an efficient way is needed to harvest algae oil.
"If we had an algae combine right now, we'd be done," he said.
This list of improvements was clear to the oldest operating algae fuel company, Solazyme.
Unlike bankrupt GreenFuel, another pioneer, Solazyme remains solvent and has scored several small coups, including a contract providing the Navy with algae-derived jet fuel. GreenFuel's founder, Isaac Berzin, declined to comment for this article.
Solazyme's success is tied to one fateful decision, said Harrison Dillon, the firm's founder.
It ditched photosynthesis.
"We grew ponds of algae for about two years," Dillon said. They tried it in translucent reactors, too. But in both situations, he said, "we couldn't come up with a scenario where we weren't making ridiculous assumptions."
Solazyme instead cultivates its algae in large, dark vats, feeding them sugar -- a process nearly identical to current ethanol production. In effect, the firm bought into the dominant biofuel system, substituting diesel and jet fuel for ethanol as a final product. (Matching yeast's hyper-efficient conversion of sugar to ethanol is another story.) Given the uncertainty facing its algae peers, Solazyme's choice seems at least a wise short-term decision.
The Department of Energy has remained a strong proponent of algae's potential, but as its senior officials have stressed in the past year that potential remains far from being realized.
Currently, algae fuels cannot be grown in large scale, and compared to gasoline prices, they are "off by at least a factor of two right now," Steve Koonin, DOE's undersecretary for science, said last month at the annual meeting of the American Association for the Advancement of Science in Washington.
Serious research needs to be done before costs can come down further -- for example, researchers at Los Alamos National Lab are working on that combine, exploring the use of high-frequency sound waves to extract algae oil.
"It is these nitty-gritty engineering-type issues that are as important and maybe more important than the biology in bringing things to scale," Koonin said.
These technical challenges seem well beyond the scope of many algae startups, prompting widespread expectation in the industry that a surge of quiet algae failures is imminent, according to Andrew Soare, an analyst at Lux Research, a technology research firm.
Soare recently examined 24 algae fuel companies in a report aimed at investors. Three firms warranted a positive rating; one was sun-adverse Solazyme. Most of the firms lacked proof of their technology and exaggerated their abilities. Their future looks grim, Soare said.
"A lot of these companies are going to fail," he said.
Many realities and hopes melded to form the algae bloom.
As public awareness of global warming increased, oil prices began to soar, spurring interest in energy independence. Rising food prices sent researchers hunting for fuel alternatives, and the long-expected breakthrough in converting non-food crops like grass to fuels refused to become reality.
And then there was "the chart."
The most aggravating myth about algae, for Solazyme's Dillon and others, stems from a simple chart comparing the oil yields of common crops. One version was published by Yusuf Chisti, a biochemist in New Zealand, in 2007. It portrays algae, conservatively, as 130 times more productive than soybeans. Over the past five years, no other algae fuel paper has been as widely cited.
It is the chart that launched a thousand overheated stories.
"That chart perpetuates one of the biggest myths out there," Dillon said. "Everybody has seen versions of that chart. [But] I've never seen what I'd say to be proof that algae has even made as much oil as soybeans."
There was a fundamental mistake people outside the algae business made when looking at the chart. They extrapolated speedy growth rates from open waters and ideal conditions to the industrial setting necessary for commercial cultivation, said Greg Stephanopoulos, a biochemical engineer at MIT and a longtime expert in bacterial manipulation.
"They make fuels from free CO2," Stephanopoulos said. "It's a no-brainer, right? They've got it all. So where's the problem with that? The problem is that you cannot cultivate [algae] at high enough densities to make this a worthwhile process."
For every gallon of oil made from algae in a pond, hundreds of gallons of water need to be circulated, he said. The algae cannot grow in dense concentrations, because they do an excellent job of blocking sunlight, even when they don't use it for energy, instead wasting it as heat.
"The issue is not one of designing a better reactor," Stephanopoulos said. "That's not going to solve this problem. The issue is not of doing better molecular biology. Even if you make all the algal cell full of oil, still you're going to have a [low] concentration of oil."
A frank assessment of algae fuel ponds released last fall by researchers at Lawrence Berkeley National Laboratory found that the only economically viable route for algae ponds would be for use in wastewater treatment, creating oil on the side. Without such a dual use -- and several regions do use algae for this purpose -- oil costs from the pond would run between $240 to $330 a barrel, more than double current crude prices.
"There are just a lot of things that have to come together to bring the cost down," said Nigel Quinn, an author of the study and water-resource engineer. "You need adequate light, adequate temperatures, a source of CO2 and you need water."
Chisti, for his part, remains convinced that algae are the only option for producing renewable fuels at a large scale but admits that the economics remain far off. Much of the initial effort on fuels ignored algae biology, Chisti said. Algae fuels will need sustained investment to become a mature technology, he said.
"Several algae companies have gone bust and others will go bust in the future," he said. "The quest for algal fuels has been characterized by 'irrational exuberance.' ... [A] science-based long-term strategy is needed for delivering on algal fuels."
Confronted with high operating costs, many algae firms are looking to produce fish feed, dietary supplements and cosmetics to support their business. Soare, the energy analyst, suspects many of these companies will stay stuck producing fish feed. Some, like Sapphire, have higher ambitions, and Mayfield expects the firm to spin out production of expensive, protein-based pharmaceuticals in the next few years, he said.
"A lot of these companies were simply naive," Mayfield said. "They had this naive assumption that they were going to grow up algae and miraculously be producing at $2 a gallon."
When Sapphire opens its first biorefinery, Mayfield expects it will be producing replacement fuels at a price between $10 and $20 a gallon. It's a good start, he said. "We don't have to make 100 percent of our improvements every day."
'We can deploy at a large scale'
At Joule's offices, the firm can make it seem like they have surpassed microalgae's hurdles.
The company's bugs grow in brackish water. (Many microalgae first targeted by startups required freshwater.) They will be cultivated in flat, translucent panels that can scale up in size, much like solar panels. Flue gas from power plants can provide the needed CO2. And the bugs secrete a diesel-like fuel that can be separated from water and piped to a central processing refinery.
The company has run a pilot plant in Leander, Texas, just north of Austin, for a couple of years. The plant has been producing ethanol, a technology Joule plans to license. This year, the firm began testing parts of its diesel process at the facility. The project has provided Joule with enough data to ramp up its operations, said David Berry, one of Joule's co-founders, at ARPA-E's annual conference this month.
"We know from our pilot," he said, "that we can deploy at a large scale."
What irks Joule's competitors and scientific peers is that the company remains vague in backing up its claims, even in its patent applications. No hard economic or production rates have been released -- not an uncommon trait for a private firm -- and even theoretical papers assessing the cost of industrial photosynthesis, including one recently published in Photosynthesis Research, provide little to the technically inclined reader.
Joule's secrecy is simply caution, said Harvard's Church, who has long collaborated with the venture capital firm, Flagship Partners, behind Joule. Flagship, where Berry serves as a partner, has created a series of biomedical companies but has had limited success in energy.
"Obviously the stakes are quite high and they've put a lot of effort into this organism," Church said. "And it's not a totally obvious organism and they've changed it pretty radically, so it's not clear they can protect everything by patents."
Independent scientists find the short path to commercial production proposed by Joule to be optimistic at best. It is more likely that cost-competitive fuel production won't arrive for another 10 or 15 years, if at all. But the ultimate model proposed by Joule, the use of bacteria as fuel-producing mini-factories, is exactly right, they say.
Think of dairy cows, said Arizona State's Vermaas. Farmers do not harvest milk by grinding the cows up and extracting the milk, which is pretty much how oil is currently taken from microalgae. Instead, they let the heifers secrete the milk at their own pace, incrementally improving the cows' genetics and their lactating environment. It should be the same with photosynthetic bugs.
While Robertson, Joule's chief biologist and photosynthesis guru, remained cautious on details, he did outline what the company has done to modify its bacteria. It is not about redoing photosynthesis, he said, since nature has had millions of years to work on it. Rather, the company has gained control over the metabolism of its bug, in a way similar to how oxygen deprivation is used to control fermenting microbes, like ethanol-producing yeast.
"These organisms are controlled by light," he said. "And the light changes every day. They have to go through a diurnal cycle, and these organisms have circadian rhythms. Their level of metabolic control is circadian and diurnal. And that's what you have to gain control over. ... [They] have to be based on light and this clock mechanism, as opposed to oxygen."
After lashing down these controls -- again, an unverified claim -- Joule has also stripped out many metabolic functions that would normally use carbon in the bug, the equipment it would need to survive in wild settings. And Robertson has built an inexpensive trigger into the bacteria so that, after they spend three days growing to an optimal density for light, the bugs shift entirely toward fuel production.
"We say, 'OK, stop growing and now send all the carbon to our product,'" Robertson said. "And so this allows you to then continuously fix CO2, absorb light and make product. The diesel separates and you can physically harvest it. That allows you to radically increase your productivity over algae."
At its pilot plant in Texas, Joule has had these cyanobacteria producing fuel "productively" for eight weeks before the microbes exhausted themselves. After exhaustion, the panels housing the bacteria would be cleaned and restocked at night, to begin producing again in three days.
Joule is not alone in achieving fuel secretion, though. Turns out, it is not terribly hard to do in cyanobacteria. Once the cells are rigged to produce fuel components, they have no idea what to do with the chemicals -- they have no instinct to store fat -- and so they shunt the fuel out of the cell. Vermaas and Jim Liao, a bioengineer at the University of California, Los Angeles, have had similar success.
(Synthetic Genomics has also quietly announced on its website that it has engineered algae to secrete oil continuously. The company makes little distinction between cyanobacteria and microalgae in its public releases, so it is difficult to interpret its claim.)
UCLA's Liao has become a bit of a household name in biofuels. His expertise lies in microbial production of butanol, an energy-packed alcohol that, unlike ethanol, can be used in existing oil pipelines. Liao imported genes from three different bugs into his cyanobacteria, which produces and secretes a butanol precursor. Liao, whose work is also behind the growing biofuel company Gevo, began a startup, Easel Biotech, to move ahead with the bug.
The cost of creating photoreactors for the bacteria remains a problem. But if somebody finds a way of doing it cheaply, then direct conversion of CO2 to fuels could finally take off, Liao said.
"If we can do it in a cost-effective way, of course we'll go that way," he said.
Joule proposes to have solved these cost problems, currently on its fifth-generation design for its reactors, which it developed in concert with its wet work. The reactors are large sheets of thin plastic, 7 feet by 4 feet, that line up in attached rows. At their base, CO2 bubbles in -- the company requires CO2 concentrations typically found in flue gas from burning fossil fuels -- and product separation occurs at the top.
"This device is capturing the sunlight," Sims said. "Extinguishing all the photons. Taking in the CO2. The product is synthesized there. Initial product separation takes place. Thermal management is occuring. ... This device is unlike any that has been created before."
It had better be, many in the algae game say, because photoreactors have failed before for microalgae. Digging holes in the ground? Much cheaper. The capital investments are high, Sims admitted, but Joule's productivity and low operating costs, lacking any need to add biomass to its mix or break apart algae cells to recover fuel, make up for the cost, he added.
Wait and see
Despite their easy genetic modification, much about cyanobacteria remains mysterious.
Scientists do not know why some grow quickly and others slowly. (Penn State's Bryant suspects it has to do with differing ratios between the two photosynthesis complexes that power the bacteria.) Their metabolism is not well understood. Scientists are "picking it up on the fly," said Harvard's Way.
Way does see cyanobacteria as a strut holding up the future of biofuels -- he recently published a paper to that effect and has collaborated on work rigging them to produce sugars and lactic acid -- but companies targeting fuel production with the bugs are making a fundamental economic mistake, he said. Oil companies have never had true competition for their product, and they can ruthlessly trim their profit margins if need be.
"If there's a competition issue, they can make it cheaper," he said.
Rather, photosynthesis companies should be targeting chemicals and other products, like the amino acid lysine, that are already widely made by genetically modified bugs that feed on corn. Sugar can only get so cheap, and its cost has risen in the ethanol era. Compete against the sugar baseline, Way said, and eventually you move into fuel, not the other way around.
Joule plans to license its technology for chemical and ethanol production, but it is retaining diesel for itself, Sims said. It sees a future that, at times, seems hard to separate from some past algae dreams. Fleets of bugs, customized to local solar and climate conditions, will begin stretching across the American plains, churning out fuel that can seamlessly power existing heavy trucks or jets. What can go wrong?
"The reality is," Sims said, "people have seen it. People are pretty blown away at what they hear."
Joule may have done it. It may have broken past the incremental progress that has long typified energy innovation. Its pigs may fly. But if they don't, Joule has positioned itself to fail spectacularly and publicly.
Time will tell.