By Jan H. Schut
When Coca Cola Co. (www.coca-cola.com) launched its “PlantBottle” in North America at the winter Olympics in Vancouver in 2010, plastics processors were skeptical. Coke’s Dasani water bottles claimed to be made of “up to 30% bio-based PET.” But when processors asked resin suppliers for bio content PET, it wasn’t available. How was Coke getting it?
Skeptics also noted that the bottles didn’t say “30% bio content” or “at least 30% bio content,” normal wording for post consumer recycled content. Instead, labels said “up to 30%,” which could literally mean zero. (Bio content can be accurately measured using ASTM 6866 for the presence of carbon-14, which isn’t found in fossil fuels.)
In fact, Coke had put an elaborate multi-national bio monomer supply chain in place, capable of meeting rapidly escalating demand for bio-content PET. Polyethylene terephthalate (PET) is made out of 32.2 wt% monoethylene glycol (MEG) and 67.8 wt% purified terephthalic acid (PTA), combined in an esterification reactor and converted to polymer in a polycondensation reactor.
To reach 30% bio content requires bio MEG, for which Coke bought bio ethanol in Brazil, shipped the bio ethanol to India, and had it made into bio MEG, which was then made into 30% bio content PET. Bio content PET was made into bottles by Western Container Corp. in Fife, Wash. (www.westerncontainercoke.com), an injection stretch blow molding company belonging to Coca Cola bottlers.
The beauty of Coca Cola’s route is that it used bio ethanol available on the Brazilian market, an existing producer of bio MEG, and existing reactors and infrastructure to make the PET, so supply could be ramped up fairly quickly. “The conversion is incredibly easy,” says Robert Kriegel, senior scientist at Coca Cola in Atlanta and architect of the concept. “You do have to have certain volumes. We are very very protective of those numbers, but we are talking millions of pounds.”
By late 2010 Coke said it had distributed 2.5 billion “PlantBottles.” By late 2011, all Dasani water bottles and many Coke bottles globally were in 30% bio PET, and Coke said it had made about 10 billion bio content PET “PlantBottles.” How much bio MEG is that? Most Dasani water bottles are 20 oz. to 1.5 liters, which weigh around 0.84 and 1.33 oz. respectively, according to NAPCOR (National Assoc. for PET Container Resources). Coke’s bottles are probably lighter than NAPCOR’S figure, so let’s assume that all 10 billion “PlantBottles” were 20 oz. That would mean consumption of about 150 million lb of bio MEG in two years–an impressive logistic effort to create a new material.
Coke licensed partners to increase demand for bio content and speed the development of 100% bio PET. In 2011 H.J. Heinz Co. (www.heinz.com) announced its license of the “PlantBottle” for certain catsup bottles. By June 2012, Ford Motor Co. (corporate.Ford.com), NIKE Inc. (www.nikeinc.com), and Procter & Gamble (www.pg.com) were also partners.
Other companies have access to bio PET too, but it’s hard to tell how much is being used. Toyota Motor Corp. (www.toyota.com) has announced a 30% bio polyester carpet and interior trim for an electric hybrid car for the Japanese market. AT&T Corp. (www.att.com) announced use of up to 30% bio APET clamshells for cell phone accessories. The clamshells are made from TerraPET sheet, made by Klockner Pentaplast (www.kpfilms.com) in the U.S. since late 2010. TerraPET sheet for AT&T is thermoformed by Display Pack in Grand Rapids, Mich. (www.displaypack). TerraPET sheet is also used for food grade applications and plastic box packaging for cosmetics. Klockner Pentaplast doesn’t disclose where it sources 30% bio PET.
HOW BIO CONTENT PET IS MADE
Brazilian bio ethanol is derived from fermenting sugar cane juice. It’s available on the open market through a number of large distributors. Coke also uses bio ethanol from India, derived from molasses, a byproduct of sugar refining. Bio ethanol goes to India Glycols Ltd. in Noida, India (www.indiaglycols.com), which has been making bio MEG since 1989 for pharmaceuticals and other markets and is the world’s only large commercial producer of it.
India Glycols makes bio ethanol into bio ethylene oxide and then into bio MEG. Its capacity grew steadily from 44 million lb/year initially to 132 million lb/yr by 2002, according to a published company history. In 2005 India Glycols nearly doubled its combined bio EO and bio MEG capacity to 220 million lb/yr, then doubled it again to 440 million lb/yr in 2008 ahead of Coke’s “PlantBottle” launch.
Coke said its bio MEG was made into bio content PET in Indonesia. Indorama Ventures PCL, headquartered in Bangkok, Thailand (www.indoramaventures.com), the world’s largest PET producer, reported in a presentation in August 2011, that it makes 30% bio content PET for Coke. Indorama can make a bio version of any PET grade it offers. Data sheets remain the same and are chemically the same, but a “B” is added to the grade name to indicate up to 30% bio. The only reason Dasani labels say “up to 30%,” says a Coke partner, is that during a reactor transition from conventional to bio PET, there could be tails that contain less than 30%. Indorama now produces bio content PET in Asia, Europe and North America. Coke has also qualified other bio content PET producers.
New bio MEG capacity is being built in Taiwan. Greencol Taiwan Corp., Kaosiung, Taiwan, a 50/50 joint venture between polyester fiber producer Toyota-Tsusho Corp., Nagoya, Japan (www.toyota-tsusho.com), and China Man-made Fiber Corp. in Taipei (www.cmfc.com.tw) started production of bio MEG last year with announced capacity for 220 million lb/year and a long term supply agreement for bio ethanol from Petrobras Biocombustivo in Brazil. Toyota Tsusho reportedly expects to make 440 million lb/yr of bio content PET. Far Eastern New Century Corp., Taipei, Taiwan (www.fenc.com) is also installing bio MEG capacity.
There are other routes to bio MEG besides ethanol-to-ethylene-oxide-to-MEG. A number of small companies in China start with vegetable oil and produce a mixture of several bio glycols, used for antifreeze. The mixture includes propylene glycol, 1.3 propane diol, butane diol, and MEG. This method takes fewer chemical conversion steps than the ethanol/ethylene oxide method and could eventually prove less expensive, but the MEG would have to be separated out.
WHERE’S THE BIO PTA? (OR WHAT’S PEPSI DOING?)
Coke announced long term plans to have all its PET bottles made with 30% bio content by 2020. That spurred PepsiCo in Purchase, N.Y. (www.pepsico.com) to counter on March 15, 2011, by announcing the world’s first 100% bio PET bottle. Pepsi could only make lab samples, but said it intended to have “pilot production” of 100% bio PET bottles this year. Pepsi said its test program was based on “switch grass, pine bark and corn husks,” but would in future be based on byproducts from Pepsi’s food businesses like orange and potato peels.
To get to 100% bio PET requires bio PTA, which doesn’t exist except on lab scale. There are six known companies with active technology developments that could lead to bio PTA. Five target bio paraxylene to replace the current petro-based route to PTA: Gevo Inc. (www.gevo.com) in Englewood, Colo.; Virent Inc., Madison, Wis. (www.virent.com); Anellotech Inc., New York City (www.anellotech.com); Global Bioenergies, Evry, France (www.global-bioenergies.com); and Honeywell UOP in Des Plaines, Ill. (www.uop.com).The sixth, Amyris Inc., Emeryville, Calif. (www.amyris.com), owns a muconic acid route to bio PTA without making paraxylene.
Gevo, founded in 2005, has a proprietary yeast fermentation process, patented biocatalyst, and separation technology. Managed by NatureWorks (www.natureworksllc.com) co-founder Patrick Gruber and others from NatureWorks, Gevo vaulted from lab scale to a commercial demonstration plant in St. Joseph, Mo., and then converted an existing ethanol plant in Luverne, Minn., into the world’s first commercial bio-isobutanol plant. The Luverne plant, which started up in May, has capacity to make 110 million lb/yr of isobutanol, but will produce only up to 1 million gal/month during 2012. Gevo supplies lab quantities of bio paraxylene to Toray Industries Inc., Osaka, Japan (www.toray.com), to develop 100% bio fiber.
Virent, a spin off from the University of Wisconsin-Madison, has a patented aqueous catalytic “reforming” process that converts soluble sugars from a variety of sources including corn, sugar cane and agricultural residues into conventional hydrocarbons. Virent also makes bio paraxylene in kilogram quantities for testing, using a 10,000 gal/yr demo plant in Madison. (Lux Research Inc. in Boston in a blog in July 2012 says that Virent supplied the bio PTA for Pepsi’s test, but this is believed to have been only small initial quantities.)
Anellotech is a spin off from the University of Massachusetts, Amherst, with a patented high-speed pyrolysis technology using a fluidized bed reactor to break virtually any ligno-cellulosic biomass down into aromatics. That distinguishes it from most competing processes. Anellotech announced a 4,400 lb/day pilot plant, where it is reportedly producing bio paraxylene, but hasn’t disclosed the location. Anellotech’s founder and patent holder, George Huber, an associate professor at U. Mass., also did research on Virent’s aqueous catalytic technology in his PhD work at the University of Wisconsin-Madison.
Global Bioenergies uses genetically modified bacteria with enzymes to produce bio isobutene directly in gaseous form, but only on a lab scale. The bio isobutylene could be made into bio paraxylene.
UOP also has a route to produce paraxylene from four-carbon alcohol like isobutylene. UOP and Ensyn Corp., Wilmington, Dela. (www.ensyn.com) have a joint venture since 2008 called Envergent Technologies (www.envergenttech.com) to develop “rapid thermal processing” of forest and agricultural waste into bio oil for fuels and intermediate chemicals.
Amyris in late 2011 bought the assets of Draths Corp., Lansing, Mich., including patented technology to produce t,t muconic acid as the building block for bio-based PTA. Draths’ technology was a spin off from Michigan State University in Lansing by professors John Frost and Karen Draths. Amyris also produces renewable farnesene, a long-chain hydrocarbon, which Amyris plans to use as an oxygen scavenger in PET.
Several large chemical companies also have intellectual property for other routes to bio PTA, including Sabic Innovative Plastics (www.sabic-ip.com) using bio d limonene; Toray/UOP using bio dimethyl furan; and BP PLC (www.bp.com) using bio furan dicarboxylic acid. But none of them have announced active PTA programs.
Coke, meantime, has filed two patent applications on bio-content PET packaging (U.S. Pat. Applic. # 2010018512 and 20110262669), using bio monomer derived by “fast pyrolysis, acid hydrolysis, enzymatic hydrolysis, microbial degradation, and hydrogenolysis. No prior art is listed. The important intellectual property for bio PET is on the conversion of biomass to bio monomer, not on making bio monomer into PET packaging, so Coke’s patent applications may end up being considered obvious and not patentable.
But whatever happens with the patent, Coke is probably more interested in leadership than ownership of bio PET packaging because Coke needs the bio resin infrastructure to expand for costs to come down. Coke and Pepsi’s bio PET announcements clearly indicate that both companies see bio monomers of the future based on bio waste from their own food processing, like orange peel from orange juice. Making orange peel into PET bottles would be a neat step toward zero waste.