Bioplastics: PLA, PHA, bio-PE and starch blends: Bio-PE, starch blends and the comparison

3.1 Bio-PE is ordinary polyethylene with ethanol as the feedstock

Fossil ethylene comes from steam cracking: naphtha or ethane is heated for a fraction of a second in a furnace and breaks into ethylene, propylene and other products (Kabasci 2014, section 11.2). Bio-ethylene takes a longer road. Sugarcane is crushed, the juice fermented to ethanol, the ethanol distilled, and then a molecule of water is removed from each ethanol molecule (CH₃CH₂OH → CH₂=CH₂ + H₂O). This dehydration runs in the vapour phase at about 400 C with more than 99% conversion and more than 99% selectivity, and takes 1.7 t of ethanol per tonne of ethylene (Task 42 2012). It is an old route: India and Brazil still made ethylene from ethanol in the mid-1980s (Wittcoff 2004, Ch 3).

From there everything is conventional (figure). The ethylene is purified and polymerised in standard plants; Braskem uses slurry and gas-phase processes. Low-pressure processes (1-100 bar) give linear HDPE and LLDPE and convert 80-100% of the ethylene; high-pressure processes (2,000-3,000 bar) give branched LDPE at 15-30% conversion (Kabasci 2014, section 11.2). The product is identical in use and recycles with fossil PE. Only the carbon-14 test tells them apart: Braskem's green HDPE has more than 96% renewable content by ASTM D6866 (Kabasci 2014, section 11.9).

Two technical gaps remain. Most PE grades contain a few per cent of comonomers such as 1-butene, 1-hexene or 1-octene, and a fully bio-based grade would need these made by joining bio-ethylene molecules. Bio-polypropylene is harder still; its candidate routes are far less mature (Kabasci 2014, section 11.7).

Thermodynamics are unkind: sugar is far more oxidised than a hydrocarbon, so stripping its oxygen costs energy, and Koopmans of Dow calls the route unfavourable without cheap renewable energy (Kabasci 2014, section 11.4). In a cane mill, burning bagasse, the crushed fibre, supplies that energy.

• Kabasci (2014), Ch 11 Polyolefin-Based Plastics from Biomass-Derived Monomers · sections 11.2, 11.4, 11.7, 11.9
• Task 42 (2012), Bio-based Chemicals · pdf p. 18
• Wittcoff (2004), Ch 3 Chemicals and Polymers from Ethylene · pdf p. 162-163 (book pp. 136-137)

3.2 Bio-PE's economics rest on cheap cane ethanol

Bio-PE works where ethanol is cheap and plentiful, which so far has meant Brazil. Its National Fuel Alcohol Programme of the 1970s built a sugarcane industry yielding about 82.5 t of cane per hectare, with only 1% of the country's 340 million hectares of arable land devoted to fuel ethanol (Kabasci 2014, section 11.9). The IEA (2018) reports that Brazil hosts half the world's bio-ethylene capacity, and further plants were planned in Brazil and India as of 2011 (Kent 2012, Ch 10).

Braskem started its plant in September 2010: 200,000 t of ethylene from about 460 million litres of ethanol grown on about 65,000 hectares, for 200,000 t of polyethylene (Kabasci 2014, section 11.9). That is about 3.1 t of PE per hectare per year and 2,300 litres of ethanol per tonne (table). The plant cost about $300 million (Task 42 2012), roughly $1,500 per tonne of annual capacity.

Scale is the catch. That one plant was under 0.2% of world ethylene capacity (Kent 2012, Ch 10). Fossil polyolefin sites cost several hundred million euros and sit beside crackers for cheap feedstock (Kabasci 2014, section 11.2). In 2011 Brazil cut its fuel-ethanol quota to save sugar for food (Kabasci 2014, section 13.2), and the Dow and Mitsui joint venture, a 350,000 t ethanol unit on about 120,000 hectares, was put on hold in 2013 because ethanol economics had changed (Kabasci 2014, section 11.9).

Buyers do pay a premium. Rosenboom et al. (2022) list bio-PE at $1.8-2.4 per kg against $1.4-1.6 for HDPE, and report that Toyota agreed to take a quarter of Braskem's output at a 30-50% markup. It is the smallest premium in their table, because a drop-in needs no new equipment.

• Kabasci (2014), Ch 11; Ch 13 Renewable Raw Materials and Feedstock · sections 11.2, 11.9, 13.2, 13.3
• IEA (2018), The Future of Petrochemicals · pdf p. 44
• Kent (2012), Ch 10 Synthetic Organic Chemicals · pdf p. 328
• Rosenboom et al. (2022), Bioplastics for a circular economy · pdf pp. 5, 8, 14 (journal pp. 121, 124, 130)

3.3 Starch must be melted with water and blended with a partner

Starch is the plant's energy store, packed into granules 2-100 micrometres across. It is made of two glucose polymers: amylose, mostly linear, and amylopectin, heavily branched and enormous, with molecular weights up to 100-400 million. Corn supplies 82% of the world's starch, wheat 8%, potatoes and cassava 5% each. Corn starch is separated by wet milling (figure) (Kabasci 2014, sections 2.1 and 2.2).

Native starch cannot simply be melted. Dry, its crystals would melt at about 257 C, above the temperature at which the chains decompose. The answer is a diluent. Water, and plasticisers such as glycerol or sorbitol, lower the melting point. In an extruder with a water volume fraction below 0.45, and preferably below 0.28, heat, pressure and shear destroy both crystals and granules, and the result, thermoplastic starch (TPS), flows like a conventional plastic melt (Kabasci 2014, section 2.4).

TPS on its own is too sensitive to humidity for most uses, so commercial grades blend it, often at more than 50% starch, with a synthetic polymer. Early blends with ethylene-acrylic acid or EVOH made good film but failed compostability standards because the partner degraded too slowly. The breakthrough was blending with biodegradable polyesters, notably aliphatic-aromatic types. In Novamont's Mater-Bi materials, amylose coils around polyester chains to form helical complexes, and the starch ends up as droplets under a micrometre across. Changing the amylose-to-amylopectin ratio, additives and processing conditions tunes the product from PE-like flexibility to PS-like stiffness (Kabasci 2014, sections 2.1 and 2.4.6) (figure).

Yield is the starch route's quiet advantage. In starch-based plastics nearly 100% of the starch ends up in the product; plastics made from monomers fermented from glucose syrup keep generally less than 45% of it (Kabasci 2014, section 2.1).

• Kabasci (2014), Ch 2 Starch · sections 2.1-2.6

3.4 Side by side with PE, PP, PET and PS

The property table puts the four materials beside the incumbents (figure). Three patterns stand out.

Stiffness and strength are not the problem. PLA's modulus of 3.5 GPa and tensile strength of about 53 MPa match or beat polystyrene's 2.9 GPa and 46 MPa, and Kamm's authors describe PLA as looking like PET while performing much like PP. The problems are toughness and heat. PLA stretches only 2.5-6% before breaking and PHB 3-5%, against 500% for general-purpose PP. PLA's Tg of about 58 C caps the use temperature of amorphous PLA, whereas PP and PE are well above their Tg at room temperature and so stay tough.

Processing windows differ. Bio-PE runs exactly like fossil PE. PLA runs on PET-type equipment but must be dried and kept well short of about 250 C. PHB has the narrowest window: it degrades above 190 C, so the extruder runs at about 150 C (Kamm 2006, Vol 1 Ch 10). Starch materials must be kept dry, and where raw starch is present, below 230 C (Kabasci 2014, section 2.3).

Price follows maturity. In Rosenboom et al.'s compilation of pre-2020 literature prices, PP is $1.1 per kg, PS $0.7-1.5, PET $1.2-1.4 and HDPE $1.4-1.6; bio-PE is $1.8-2.4, PLA $2-3 and P₃HB $3-8. The fossil-based biodegradable partners used in starch blends are dear too: PBAT $4.1 and PBS $4.5 per kg. The sources give no price for starch blends themselves. Put as ratios, PLA costs about 1.4-2.5 times PET and P₃HB about 3-7 times PP. Kent (2012, Ch 15) adds that bio-based polymers vary more from batch to batch than petroleum resins.

• Kamm (2006), Vol 1 Ch 10 Sugar-based Biorefinery (PHB, sugar and ethanol) · pdf pp. 242-244 (book pp. 214-216)
• Kamm (2006), Vol 2 Ch 14 Polylactic Acid from Renewable Resources · pdf pp. 834, 848, 853 (book pp. 381, 395, 400)
• Kent (2012), Ch 15 Synthetic Resins and Plastics · pdf pp. 596-600, 604-605
• Rosenboom et al. (2022), Bioplastics for a circular economy · pdf p. 8 (journal p. 124), Table 1
• Kabasci (2014), Ch 2 Starch; Ch 8 Poly(Lactic Acid) · sections 2.1, 2.3, 8.7

See it in the table

Check your understanding

Roughly how much ethanol does one tonne of bio-ethylene require?

Why is thermoplastic starch usually blended with a polyester rather than used alone?

Using pre-2020 literature prices, which bioplastic carried the smallest premium over its fossil comparator?