A bias for biofuel
In the search for renewable fuels, biofuels are making headway. Professor Gary Pisano (Harvard) looks at a startup, Amyris Biotechnologies, that applies synthetic biology techniques and is soon to start producing commercial quantities of biodiesel in Brazil.
It was in 2002 that a team of Berkeley scientists used synthetic biology to come up with a cheaper, synthetic alternative to the leading anti-malarial drug which used wormwood. Synthetic biologists perform genetic manipulation on simple organisms, such as yeast, to produce a specific chemical compound. Amyris Biotechnologies was founded in 2003 to pursue development of the anti-malarial drug and by 2005 was investigating other commercial opportunities which the case (see reference below) focuses on.
Biofuels and specialty chemicals
Two possible growth sectors were biofuels and specialty chemicals. The first generation of biofuels was derived from plants. The most popular fuel was ethanol, produced mostly from sugar cane in Brazil and corn in the USA. The second-generation of biofuels turned to non-food crops such as cellulosic matter. The sugar extraction process for such materials, however, was much more complicated.
Third-generation biofuels went beyond ethanol to produce biodiesel. Companies such as Amyris and its competitor LS9 are using genetically engineered yeast to develop these fuels. Given the pressure to avoid using food crops for fuel, biodiesel was turning into a development priority. While in 2008 US ethanol production still outstripped biofuel production (9236 versus 776 million gallons), Congress has set a 22 billion gallon usage goal for advanced biofuels in 2022.
A second potential market was specialty chemicals. These are low volume, high value added compounds (such as adhesives or additives). This industry derived most of its products from petroleum and accounted for sales of $472 billion in 2008. Synthetic biology could be particularly applicable to two markets: lubricants and surfactants (these surface active agents reduce the surface tension between two liquids).
The move to biofuel
In early 2006, Amyris succeeded in obtaining a $20 million Series A financing. A CEO search was initiated and the team chose an executive with a strong background in fuels, John Melo, the president of US fuels operation at BP. His strategy was to focus on biodiesel. Jet fuel would also be pursued but that required a longer certification process. The company would keep a foot in chemicals, a higher value, lower volume business more akin to Amyris’ drug origins.
The scientists at Amyris developed a biodiesel baptized No Compromise®. It offered a solution to some of the problems associated with its predecessors. Unlike ethanol, it did not provoke pipe corrosion. Unlike biodiesel made from soy beans, it did not break down during storage and transport and was stable at low temperatures. Last but not least, it produced 80% fewer emissions than petroleum fuels. The EPA approved the fuel in 2009.
One of Amyris’ key operational decisions was what feedstock to use (the yeast is mixed with a feedstock in the initial fermentation step of the fuel production process) and where to manufacture. Keeping cost low was the key biodiesel success factor and feedstock would make up close to 80% of the cost, hence the importance of the feedstock decision. Two other factors entered into the choice of location: good water sources and a domestic market for biodiesels.
Quickly, Brazilian sugarcane became the leading feedstock candidate. Since sugarcane needs to be used within 24 hours of cutting, the plant needed to be located close to the fields. The fact that John Melo spoke Portuguese facilitated Amyris’s decision to source and produce in Brazil.
Amyris’ production process entailed mixing genetically engineered yeast with sugarcane juice, letting the mixture ferment for a day, and then purifying the fermented broth so as to produce a new chemical compound, farnesene. The farnesene could then be transformed into bio diesel, after a hydrogenation process or it could undergo processing to be turned into a specialty chemical.
Gearing up for production
To speed up the commercial production process, the first pilot plant was built at headquarters in Emeryville, California. This $10 million pilot plant used 300 liter fermentors and was tested successfully in October 2008. At that point, full attention could be devoted to a large pilot and demonstration plant set up in Campinas (this city of just over 1 million inhabitants Campinas is located in the Brazilian state of Sao Paulo).
The demonstration plant featured a 5000 liter fermentor and was up and running in June 2009 at a cost of $3 million. At the same time, Amyris was finalizing the yeast strains, fermentation conditions and plant design for its first commercial plant. The targets were an opening date in 2011, a production capacity of plant 100 million liters per year production capacity and a maximum cost threshold of $2.00/gallon.
Unfortunately, at about the same time, the financial crisis was working its effects and Amyris was forced to tweak its strategy somewhat. The American venture capital partners were asking Melo to slow down and cut staff. Instead, Melo decided to seek out Brazilian partnerships. The idea was to convert existing Brazilian ethanol plants into farnesene plants. The Brazilian partner would provide the capital for the actual plant conversion while Amyris would provide the technology and assist with commercial distribution.
Challenges and questions
Looking toward its first sales in 2011 Amyris is faced with a number of challenges. One is the issue of hydrogenation (transforming farnesene into diesel). Hydrogenation equipment is expensive so Amyris was seeking out major oil and gas companies that could hydrogenate and resell some of Amyris’ farnesene to blenders and distributors. But because biodiesel margins could be located anywhere along the supply chain, Amyris wanted to keep a foot in the tail end of the chain and intended to also sell through its own subsidiary, Amyris Fuels.
Two other important issues provide discussion subjects. The first is whether to focus on the Brazilian or American market for diesel sales. It was not clear which market would offer higher margins since this depended on a number of outside factors such as bio-diesel subsidies, tariffs and exchange rates. Amyris was talking to customers in both countries. In the US it was approaching large fleet buyers, who negotiated directly with suppliers and were under the obligation to use renewable fuels for a portion of their needs.
A second issue concerns the possibility of specialty chemicals as a higher margin alternative to fuel. Because of the uncertainties surrounding diesel margins, Amyris was continuing its specialty chemical research efforts. Amyris’ competitor LS9 had signed in 2009 an agreement with Procter & Gamble to produce sustainable chemicals. Amyris was approaching other consumer product companies to try to launch partnerships for potential farnesene applications which could prove more lucrative than biodiesel in the short term.
In its quiet way then, the case invites the reader to wonder what managerial yeast should be used. And thankfully for the case method, such yeast has yet to be genetically engineered…
Reference:
Harvard 9-610-031
“Amyris Biotechnologies: Commercializing Biofuel”
Professor Gary Pisano and Alison Berkeley Wagonfeld
Harvard Business School
Published July 2010
