A Summer Of Start-Ups For Biobased Chemicals

A brave group of small companies have made the leap this summer into biobased chemical production. The investments are evidence that a new industry—which began as a series of laboratory experiments to make chemicals from sugar rather than petroleum—has begun to take root.

The leading wave of companies launching these manufacturing facilities includes Gevo, LS9, Reverdia, Rivertop Renewables, Solazyme, and ZeaChem, all of which are either already in production or will be before the end of the year. They will soon be followed by Amyris, BioAmber, and Myriant Technologies; all three plan to open new plants early next year.

The output from the facilities, mostly chemical intermediates, will be sold to chemical and other companies to make products such as solvents, detergents, coatings, and polymers. The intermediates include isobutyl alcohol, glucaric acid, succinic acid, acetic acid, and farnesene.

Although no individual company or facility is guaranteed to be technologically or financially successful, experts agree that this year marks a tipping point. There is no question that biobased chemical manufacturing is here to stay, the question is how quickly it will grow.

“We’re seeing fast growth, albeit from a small base,” says Steven Slome, an analyst with the consulting firm Nexant. “People are going nuts.”

Nexant keeps tabs on the universe of project announcements. Although the firm figures that half of the projects will never be built, the industrial undertaking is still impressive. “Even risk-adjusted numbers show a notable amount of capacity addition,” Slome says. Keeping track of the activity isn’t easy. Nexant wrangled all of the production announcements for a report in early 2012, but by late spring it was out of date by more than a dozen projects.

By adding up capacity of planned facilities, Nexant estimates that by 2015, the world will see more than 5 million metric tons per year of biobased chemical production. Although the tally of expected new capacity shows a straight, upward line, Michael Ritzenthaler, a stock analyst at investment bank Piper Jaffray, says the graph of actual output over the next few years “in reality will probably be more of an S curve.”

Ritzenthaler anticipates more lag time in actual production than companies acknowledge in their press releases because project backers prefer to promote what he considers to be “pretty ambitious goals.” Still, he thinks the industry will be successful in the long run in getting its plants running.

The S curve, often called the learning or efficiency curve, describes a phenomenon that’s common in any manufacturing industry. Even after a facility is constructed, it can take several months to produce any output, and it can take years to reach full capacity.

One example is Amyris’ first commercial-scale plant, located adjacent to a sugarcane mill owned by Paraíso Bioenergia in the Brazilian state of São Paulo. When the project was announced in March 2011, Amyris said it expected to begin production of farnesene in 2012. Company executives told analysts in a conference call two months ago that the plant was mechanically complete.

But turning the switch to “on” is not so simple. Amyris will spend the second half of 2012 commissioning the plant and now says it plans to begin production in 2013. Amyris Chief Executive Officer John G. Melo has told investors that it will take three years to ramp up the plant’s capacity to its goal of 50 million L per year.

In addition to implementing new technology and scaling up facilities, the learning curve also applies to the marketing side of the business. The two sides can be connected in a frustrating catch-22: It can be impossible to predict actual costs of production at a first commercial facility. Yet without information on production costs and a corresponding selling price, it is difficult to know how much capacity to bring on-line.

Amyris initially targeted markets ranging from low-margin ethanol to high-margin personal care products and planned to sell output from three contract manufacturing plants and two of its own larger plants in Brazil. But in February, the company told investors that these sites would not produce 40 million to 50 million L in 2012, as expected. Instead, it is now operating at one contract manufacturing plant—a Tate & Lyle facility in Decatur, Ill.—and is focusing commercialization efforts on the Paraíso plant alone.

Analysts complain that Amyris is not transparent enough about what markets it is targeting and its cost structure. The firm has said only that its two current commercial products are renewable diesel and the personal care ingredient squalane, both made from farnesene. Kalib Kersh, who covers Amyris for market research firm Lux Research, says he interprets the lack of information to mean that the company is “desperate for demand.”

“Production is not enough,” Kersh emphasizes. “You can have as much commercial production as you can handle, but if there are not clear routes to market with demand, then you will be in trouble.” Still, Amyris has good fundamentals and a variety of potential markets for farnesene, all of which are attractive, he adds.

When asked which of the biobased chemical producers is considered a leader in commercializing its technology, analysts point to Gevo, which is making isobutyl alcohol (called isobutanol by the industry) at its first commercial plant, in Luverne, Minn. The facility, a retrofitted corn-based ethanol operation, began production in May.

Ritzenthaler visited the Luverne site in early July. “I took a snapshot of two railcars full of isobutanol,” he says. “They are successful at making and marketing it. As far as I’m concerned, they’ve demonstrated they can make commercial quantities.”

Kersh also lists Gevo first. “They have production—that is a huge box that we check off.” Kersh adds two more praiseworthy fundamentals: the company’s easy access to the important solvents market through an agreement with the chemical maker Sasol and its “capital light” strategy of converting existing plants. And Slome points out that Gevo’s product is fungible. The firm can easily shift volumes into and out of chemical and fuel markets, thereby running at the most efficient rates while getting the highest possible margins.

Having knowledgeable, experienced people on the payroll can help make the learning curve a little less steep, Ritzenthaler says of Gevo. “They have the best management team out there to mitigate any problems in scale-up and start-up. They have been through this many times before.”

One member of that team is Christopher Ryan, Gevo’s chief operating officer. Ryan came to Gevo from NatureWorks—as did Gevo CEO Patrick R. Gruber. NatureWorks, which is now owned by agribusiness giant Cargill and Thailand’s PTT Chemical, was one of the first firms to commercialize a modern biobased chemical, the polymer polylactic acid, in 2002. Ryan was a founding member of NatureWorks and rose to become its chief operating officer.

C&EN caught up with Ryan in Redfield, S.D., where Gevo’s second facility is scheduled to begin production by the end of next year. Ryan says the team is working on the Redfield plant’s engineering phase while it continues to make ethanol.

“We’re still going through start-up at Luverne, so we’re still coming up the learning curve,” Ryan says, adding that Gevo managers will apply what they’ve learned there when they begin to operate in Redfield. The company’s experience so far has not altered its view of the cost of transitioning an ethanol plant to one that makes isobutyl alcohol, he adds.

Ryan points out that the operators of corn ethanol plants, although part of a more mature industry, also continue to improve production, and he says Gevo is learning from those activities. They include maximizing the value of coproducts, making more ethanol with less feedstock, and making the fuel from cellulosic feedstocks.

Gevo is working with its customer and partner Land O’Lakes Purina Feed to increase the value of distillers grains—the part of the corn that is left over after fermentation—for use in animal feed. Gevo recently patented a more efficient process for microbial production of isobutyl alcohol from sugar. And in July, it signed an agreement with Beta Renewables to develop a process to make isobutyl alcohol from cellulosic biomass.

ZeaChem has also started down the road to making its biobased products from cellulosic feedstocks, specifically hybrid poplar grown near its demonstration biorefinery in Oregon. The company finished building the core of the plant in January, and it’s now producing acetic acid and ethyl acetate from sugar.

Thanks to a $25 million grant and $232.5 million loan guarantee from the U.S. Department of Agriculture, a second project will add the ability to consume cellulosic biomass on the front end of the plant and convert ethyl acetate into ethanol on the back end. ZeaChem says it will start producing ethanol this year at an annual capacity of 250,000 gal.

ZeaChem CEO Jim Imbler is confident that the facility will be able to use diverse feedstocks and produce a larger set of chemicals once all of the pieces are in place. “We will be running different feedstocks through our process—hybrid poplar, wheat straw, and two or three others—to prove we are feedstock agnostic.”

The biorefinery will have the ability to make ethylene, olefins, and jet fuel in addition to acetic acid, ethyl acetate, and ethanol, Imbler says. “Flexibility is a great thing. We’re seeing a heck of a lot more interest from end-use consumer companies now.” What will drive the product mix beyond two-carbon chemicals, he adds, is demand from consumer product companies such as Procter & Gamble that want to increase their use of renewables.

Biobased chemical companies are making the most of this period of learning, Lux’s Kersh says, but he cautions that they need to focus. “You do want a flexible technology, but you need to know where your first feedstock is coming from. And you may have 50 possible markets, but you need to know what the first one is that will bring in cash,” he advises.

Partnering with firms that already have a market tends to simplify things, Kersh says. He cites the example of succinic acid firm BioAmber, which earlier this year formed a joint venture with NatureWorks. The firms plan to combine polylactic acid (PLA) with polybutylene succinate (PBS) to make compounded resins for different end uses. “NatureWorks is interested in commercializing resins, it has markets, and it will help sell PBS to these markets,” Kersh contends.

NatureWorks is also on the other end of the learning curve from all of this year’s upstarts. Marc Verbruggen, the firm’s CEO, says the 10-year history of PLA commercialization is a story of “learn as you grow.” And NatureWorks is still growing. With demand for PLA on the rise, the firm is in the process of expanding its plant in Blair, Neb., to a capacity of more than 150,000 metric tons per year.

“When we built our plant, there was no one there to buy what we were making. We had to develop our story and our product,” Verbruggen recalls. “Compared to 2003, the product is not the same. It took five years, honestly speaking, to have the confidence in our product to sell to the big brands.”

The barriers to success in the biobased chemical market are not all that different from those faced by traditional chemical producers, Verbruggen says encouragingly. Both types of venture require affordable feedstocks, efficient processes, and ready customers.

And makers of any new material face the same learning curve. Verbruggen points out that the polystyrene market, when it was new, also took 10 years to reach the scale of NatureWorks’ facility. “No matter what you are trying to make, it always goes too slowly and costs a lot.”