Regenesis (18 page)

Regardless of size, they all have the ability to turn carbon dioxide and water into carbohydrates and other products through the chemical transformations of photosynthesis (albeit across wide variations in efficiencies).

Photosynthesis is a sequence of reactions that occur in green plants and photosynthetic bacteria, in which light energy from the sun is used to produce
carbohydrates and all the rest of the plant's materials. Schematically the reaction is:

carbon dioxide + water + light energy → carbohydrates + oxygen

As a petroleum-producing organism, algae has a number of advantages. First, with the exception of its incarnation as seaweed, algae is not a food crop. Second, algae can be grown virtually anywhere that there's sunlight, and on land that's unsuitable for conventional crops—in deserts, for example. Third, it doesn't need potable water for growth, but can thrive in brackish water, seawater, or even on wastewater, meaning that it doesn't compete for the world's scare drinking water supplies. Like any other photosynthetic organism, algae consumes CO
₂
, meaning that it actually removes carbon dioxide, a greenhouse gas, from the atmosphere. And its end products—fatty-acids (lipids) or other oils, and even some types of long-chain hydrocarbons—can be processed into any of the three classic petroleum distillates: diesel oil, gasoline, and jet fuel.

Finally, algae can be genetically modified in an effort to maximize its efficiency or yield, or to fine-tune the chemical characteristics of its output. Given all these features, algae would appear to be an excellent biofuels production platform.

But it also has its shortcomings. To begin with, algae does not simply secrete its product in such a way that it can be siphoned off or skimmed from the top, like cream. Instead, the stuff must be separated from the organism by brute force—by centrifugation, for example. This builds in an extra layer of inefficiency, like an orange juice manufacturer that extracted the juice by hand, one orange at a time, rather than by mass-production extraction machinery.

Second, once separated, the algae-produced fatty acids must be refined, more or less like ordinary raw crude oil, into the desired fungible end products. That process takes energy, which has to come from somewhere. And if it comes from coal-burning power plants, it puts even more CO
₂
into the atmosphere.

Third, algae growth requires nutrients, such as nitrogen and phosphorus, which often come from petroleum feedstocks. The microbe therefore utilizes in its growth process some of the very substances it was intended to replace.

Fourth, algae is an excellent light blocker, which means that the open ponds in which it is grown cannot be very deep. Indeed, the actual light penetration is less than 1 millimeter, which means that vigorous mixing and hundreds of gallons of water are required for every gallon of oil produced. This in turn makes for major space requirements. In fact, one study concluded that growing enough algal fuel to supply the world's entire jet fleet would require a land area the size of Maryland.

Clearly, the algae-to-biofuels road is not a smooth one, and even with the research on algal fuels now being done by Craig Venter's Synthetic Genomics (with $300 million in funding provided by ExxonMobil), it's far from certain that algae will turn out to be the preferred biofuels production platform.

But as we have seen, algae is not the only microbe that can make biofuel; so can the industrial microorganism
E. coli
.

In 2005 Chris Somerville, a professor of plant and microbial biology at the University of California–Berkeley, Jay Keasling, David Berry, and I co-founded a private start-up company whose objective was to use engineered
E. coli
to produce commercial quantities of renewable diesel fuel (as well as stocks of sustainable, green chemicals). We called the company LS9 because it was the ninth life sciences firm to be funded by the venture capital group Flagship Ventures. One of the primary attractions of using
E. coli
as our production platform was that unlike algae,
E. coli
can be engineered to make its fungible petroleum products directly—the microbe does not need to be broken up in order to release its end product. Instead, we would create these fuels according to a streamlined, one-step synthesis protocol known as consolidated bioprocessing. The microbes would consume feedstock molecules and secrete the desired fuels or chemicals,
which would float to the top of a fermenter column where they could be skimmed off like cream—no centrifugation, distillation, or other intermediate steps would be required. Using this protocol, making new petroleum would be as simple and straightforward as brewing beer.

The concept of genetically engineering
E. coli
to make biofuels directly is not new. In 1987, for example, a group of researchers from the University of Florida and Southern Illinois University managed to get
E. coli
to produce ethanol. They did this by taking some genes from the bacterium
Zymomonas mobilis
, which was known to produce ethanol as one of its principal fermentation products, and inserting those genes into
E. coli
. Using simple sugars as their feedstocks, the researchers found that the re-programmed
E. coli
turned out ethanol quickly in appreciable amounts. Although sugar, a foodstuff, was consumed in the process, the group pointed out that further engineering of the microbe ought to enable it to produce ethanol using hemicellulose (inedible plant parts) as feedstock materials.

Of course, ethanol is not petroleum. But in 2010, a group of LS9 researchers published in the journal
Science
that they had found the holy grail enzymes that make alkanes (real diesel, not “biodiesel” esters) from fats. The trick was to select the appropriate genetic structures from other organisms found in nature, which the researchers did by comparing DNA from ten species of cyanobacteria that made “trace amounts” of alkanes with one species that made “undetectable amounts.” To prove that these genes were correct they inserted them into
E. coli
which enabled the microbes to directly grow small research quantities of diesel oil. The next step was to scale up the process.

By this time, LS9 had a pilot plant going in South San Francisco. The heart of the operation was a 1,000-liter fermenter tank, which was soon producing larger, batch quantities of our Ultraclean Diesel, as we call it. The microbial fermentation took only three days from start to finish, and the end product, a synthetic biodiesel, was so chemically close to conventional diesel oil that it met the American Society for Testing and Materials (ASTM) standards for road use in the United States, and was found to be chemically equivalent to California clean diesel.

In January 2010, LS9 took a major step toward the mass production of diesel oil by purchasing a bankrupt biofuels production plant for pennies on the dollar in Okeechobee, Florida. The facility included four large, million-liter fermenter tanks, storage tanks, cooling tower, and a water treatment system. We aim to produce UltraClean Diesel at the rate of 50,000 to 100,000 gallons per year initially, and then to ramp up to 10 million gallons.

At first the plant will use sugar cane syrup as feedstock, but ultimately we want to use inedible hemicellulose in place of sugar, thereby enabling us to make biofuel without using any food sources. In 2010 Keasling, together with colleagues at Berkeley and LS9, published a piece in the journal
Nature
laying out the engineered metabolic pathways that would allow
E. coli
to do this. In recognition of its achievements, the Environmental Protection Agency awarded LS9 the 2010 Presidential Green Chemistry Award.

The current situation in biofuels is one of finding and then optimizing the major players: optimizing the microbes for high-yield, efficient production through genome engineering, matching the microbe with appropriate feedstock molecules or micronutrients, and then fine-tuning the entire production process for generating clean fuels that are drop-in ready for use in the gas tank at costs that are competitive with those of natural petroleum products. But the
E. coli
at LS9 is not photosynthetic, and anyway it's not a good idea to put all of your eggs in one basket. Consequently in 2007 David Berry (who was one of LS9's cofounders), venture capitalist Noubar Afeyan, and I formed another company, Joule Unlimited, to make fuels using what many regard as the most promising microbe family yet, cyanobacteria.

Cyanobacteria were once known as blue-green algae. They are not really algae, however, but a class of bacteria that derive their energy from photosynthesis, just as if they were plants. These photosynthetic bacteria happen to be one of the game-changing organisms in earth's history, for they
are thought to have been responsible for the “great oxygenation event,” a major environmental transformation that happened about 2.4 billion years ago, during the Archean. The predominant life forms at that time consisted of anaerobic organisms, which thrive in the absence of free oxygen. The metabolism of cyanobacteria, however, released free oxygen into the atmosphere, which had the dual result of wiping out most of the oxygen-intolerant organisms while simultaneously making possible the evolution of aerobic organisms (such as ourselves), which depend on free oxygen. So the fact that we exist at all is in large part attributable to the metabolic activities of cyanobacteria.

More than ten thousand varieties of cyanobacteria have been discovered. They are found in frigid Siberia and in fiery deserts; on shower curtains and in toilet tanks; in the world's oceans, and in niche environments such as hot springs, salt works, and hypersaline bays. Indeed, some biologists regard cyanobacteria as the most successful group of microorganisms on earth.

Joule Unlimited expects these organisms to be equally successful at converting sunlight into diesel fuel. During the first two years of its existence, the company operated in relative obscurity (indeed in stealth mode) out of a nondescript building on Rogers Street in Cambridge, not far from MIT. News accounts in the
Boston Globe
often referred to Joule's “secret ingredient,” an unknown, heavily engineered microbe. But the company was only protecting its intellectual property until the time the founders could patent their uniquely designed cyanobacteria. This it did in 2010, with a patent titled “Hyperphotosynthetic Organisms.”

These engineered cyanos, it turns out, have the ability to take sunlight, CO
₂
, and brackish water, and then convert these ingredients into alkanes, the molecular constituents of diesel oil. Like LS9's
E. coli
bacteria, Joule's cyanobacteria secrete their end products into the surrounding watery medium. But these microbes have the additional advantage that they require
no
feedstock molecules as raw materials: no sugar, hemicellulose, salt or pepper, just some micronutrients that act more or less like fertilizer. Otherwise, it's as if they run on sunlight alone (in a process that the company refers to as helioculture).

The result, according to company president Bill Sims, “is the world's first platform for converting sunlight and waste CO
₂
directly into diesel, requiring no costly intermediates, no use of agricultural land or fresh water, and no downstream processing.”

Centerpiece of the system is Joule's SolarConverter, which is essentially an inexpensive, flat, transparent solar panel through which circulate thin films of cyanobacteria suspended in a bath of water and micronutrients. CO
₂
bubbles in at the bottom, and the end product—alkanes—rises to the top. Powered by sunlight, the cyanos release oxygen, sugar, and clean-burning, fossil-free diesel oil.

The process has been demonstrated in the lab, as well as in a Joule pilot plant in Leander, Texas. The company calculates that an array of its Solar-Converter panels can crank out more than 13,000 gallons of diesel fuel per acre per year. Based on an industrial-scale plant, the firm expects to be able to deliver diesel at the cost of $50 per barrel (a barrel contains 42 US gallons = 159 liters). And in line with the current fashion in the biofuels business of equating delivered fuels with land use, the company estimates that it could supply all of the transportation fuel requirements of the United States from a land area the size of the Texas panhandle.