Abstract

The imminent large-scale production of biodegradablepolyesters from corn-starch will give nature's most abundant polymer,cellulose, increasing competition in the nonwovens industry. Is a furtherdecline in the use of cellulose in nonwovens inevitable or could interest inthe new "nature-based" fibres and plastics enhance the prospects for all crop-based materials? How will the costs of the two approaches compare? Willthe absorbency advantage of cellulose prevail over the thermoplasticity advantageof polylactic acid. This paper compares and contrasts the two main "natural"routes to polymers and the properties of the resulting fibres.

Introduction: The PLA Investment

The US no longer has abundant oil reserves but does have asuper-abundance of agricultural resources capable, given appropriate levels of chemical (fertiliser/pesticide) input, of annual regeneration. The US currently imports 50% of its domestic petroleum consumption, yet exports 20% of itsgrain production. The production of ethanol from grain was an early example ofan attempt to better balance this resource budget, and the current PLA project appears to be part of a similar but separate program with the objective of allowingagriculture to provide 10% of the US chemical industry building blocks by 2020and 50% by 2050. The program, called the Plant/Crop-based Renewable ResourcesVision at the US Department of Energy has had the full support and co-fundingof the National Corn-Growers of America since 1997. The resulting "RenewableResources 2020" is a broad-based coalition of agricultural, forestry, andchemical industry experts working to create plant-based replacements forfossil-fuels.

The NCGA commenced funding of PLA research in 1994 whileworking with Cargill to help them seek funding through US Department ofCommerce grants. Cargill first joined forces with Purac Biochem BV to perfectthe fermentation step and then with Dow in 1997 to build the first pilot-plant.At that stagethey projected large-scale production in 2001 to give a PLA polymer at "around50 cents/lb". This first 140,000 tonne/year Blair PLA plant would cost$300M and consume 14 million bushels of corn per year, but to the NCGA, themarket potential for PLA would be 500 million bushels per year of corn - almostas much as the current usage in ethanol production - estimated to add 25 centsper bushel to the price of corn.

The technological vision was bold and clear, and maybe as anatural consequence, the marketing vision is now broad and evolving. Earlyannouncements said PLA would be the first renewable resource to stand alone on price and performance in applications such as fabrics for clothing, plastics for cups, foodcontainers, packaging, and home/office furnishing such as carpets.

Cargill-Dow Polymers LLC, (the 50/50 joint venture betweenCargill Inc and The Dow Chemical company formed in 1997), has, in a series ofannouncements this year, made it clear they believe crop-based biodegradablefibres to have a bright future in comparison with the fossil-fuelledsynthetics. Blair, now due to commence production of PLA in early 2002, waschosen because of its proximity to the lowest cost source of the dextrose sugarfrom which the PLA is made-a large Cargill wet-milling plant - thereby allowingit to achieve the best possible supply chain costs. The earlier JV betweenCargill and Purac Biochem BV brought a 34,000 tonne per year plant on stream,also at Blair in 1998, to manufacture lactic acid from dextrose. This feeds,amongst other end-uses, the 4000 tpa Dow PLA pilot plant in Savage MN.

A biodegradable polyester with the cost and easy-care properties of real polyester, yet based on renewable resources rather than oil is clearly anunmet need. It should be highly saleable in both woven and nonwoven products as well as in many types of plastic containers. While the publicly available informationmakes it clear that conventional polyester, polystyrene and polypropylene are major targets for this new crop-based polymer, comparisons with the oldbiodegradable crop-based polymer, cellulose, are largely absent. PLA fibre issaid to "bridge the gap" between natural polymers and the synthetics,offering a unique combination of properties combining the best attributes ofnatural and synthetic fibres. It's positioning, in a world where the syntheticscontinue to displace the natural polymers on cost/performance grounds, isexcellent. Most people are vaguely concerned that the plastics are in some waybad for the environment, but, in the mass market at least, have not beenconcerned enough to pay more for less convenient natural products.

Polylactic Acid Development

Polylactic acid was first made in 1932 by Carothers, who developed a process involving the direct condensation polymerisation of lactic acid in solvents under high vacuum. He abandoned the polymer as too low in melting point for fibres and textiles and went on to develop nylon.

More recently PLA was developed as an alternative binder for cellulosic nonwovens because of its easy hydrolytic degradability compared with polyvinyl acetate or ethylene-acrylic acid copolymers.

Spunlaid and meltblown nonwovens based on PLA were researched at the University Of Tennessee Knoxville in 1993.

Kanebo (Japan) introduced LACTRON® (poly L-Lactide) fibre and spun-laid nonwovens in 1994 claiming a capacity of 2000 tpa being expanded to 3000 tpa. It targeted agricultural applications to start with, and in 1998 was re-launched for apparel end-uses. At that time, Japanese demand for PLA fibres was said to be 500-1000 tpa. In order to improve the biodegradability and reduce the costs of the nonwovens, blends with rayon were also developed.

Fiberweb (now BBA, France) disclosed nonwoven webs and laminates made of 100% PLA in 1997 and introduced a range of melt-blown and spunlaid PLA fabrics under the Deposa™ brandname. The polymer was developed by Neste Oy.

Galactic Laboratories (Belgium) provided an excellent overview of polylactic acid polymers, concluding that 390,000 tonnes of the polymer would be produced by 2008 at prices around $2/kg. Their estimate of 70,000 tonnes for 2002 looks about right with the new Cargill-Dow plant due to start during that year. It remains to be seen if the price will be right also.

And this year, Cargill Dow Polymers LLC, as mentioned in the introduction plan to double their 4000 tpa capacity for the polymer EcoPLA™, now rebranded NatureWorks™ "to meet immediate market development needs". In January they also announced the construction of the Blair plant to make "a family of polymers derived entirely from annually renewable resources with the cost performance necessary to compete with traditional fibres and packaging materials". Fibre Innovation Technologies, Parkdale, Unifi, Interface, Woolmark, Unitika, Kanebo and Kuraray are mentioned as Development Allies.

Several other producers are still active. At Index 99, NKK (Japan) showed a PLA spunbond nonwoven at 15gsm with apparently excellent formation and properties, although they later admitted that this was to demonstrate the flexibility of their spunbond machinery, and not a commercial product. Kuraray (Japan) showed PLA fibres and provided some data on their properties and biodegradation rates. PLA polymer is made in Japan by Toyobo, Dai-Nippon Ink Chemicals, Showa Polymers, Shimadzu Corp and Mitsui Toatsu, the latter under the LACEA™ brandname. Unlike CDP, MT's process polymerises the lactic acid direct from the monomer), At Anex 2000 in Osaka, Shinwa were showing a PLA version of their Haibon spunbond: "a natural biodegradable nonwoven". Kanebo again promoted their Lactron® fibre, but this time based on the CDP polymer.

The PLA Fibre Process

The CDP process, involves extracting sugars (mainly dextrose, but also saccharose) from cornstarch, sugar beet or wheat starch and then fermenting it to lactic acid. The lactic acid is converted into the dimer or lactide which is purified and polymerised (ring-opening method) to polylactic acid without the need for solvents. The family of polymers arises in part from the stereochemistry of lactic acid and its dimer. As fermented, lactic acid is 99.5% L-isomer and 0.5% D-isomer.

Conversion of this lactic acid to the dimer can be controlled to give three forms, the L, D, and Meso lactides.

Polymerisation of the lactide to give polymers rich in the L-form gives crystalline products, whereas those rich in the D-form (>15%) are more amorphous. The enhanced control of the stereochemistry achieved in the dimer route accounts for the superiority of the current products over those from the 1932 Carothers approach.

In block diagram form the PLA production sequence is:

Seeds, Soil, Water, Carbon dioxide, Sunlight
↓Grow-months↓
Biomass, ideally corn
↓Harvest/Wet Mill↓
Starch
↓Acid/Enzyme Hydrolysis↓
Dextrose
↓Fermentation↓
Lactic Acid
↓Polymerise↓
Crude Polylactic Acid Pre-polymer
↓Depolymerise↓
Crude Lactide monomer
↓Fractional Distillation↓
Pure lactide monomers
↓Blend/Polymerise↓
Polylactic Acids
↓Modification for end-use↓
Granules for extrusion etc.
↓Melt Spinning↓
"Crop-Based" Polyester Fibres

Producing the lactide with the right purity and stereochemistry to make decent fibres is not trivial. In a recent Cargill patent, the refining process, intended to be able to cope with crude lactic acid feedstock, was illustrated as follows:

Feed Crude Lactic Acid to Evaporator continuously
↓
Remove water or solvent → discard or recycle water, solvent or by-products
↓
Feed concentrated lactic acid to a pre-polymer reactor
↓
Polymerize to form pre-polymer by removing water → discard or recycle water, solvent or by-products contaminated with lactic acid
↓
Feed in catalyst→ Feed pre-polymer to lactide reactor→Remove high-boiling unreacted polymer
↓
Remove crude lactide as vapour
↓
Partially condense crude lactide →Remove lactide impurity as a vapour
↓
Purify crude lactide in a distillation column →Remove lactide impurities
↓
Remove purified lactide as high-boiling bottoms from the column
↓
Polymerisation

Before examining the claims for PLA fibres in comparison with other crop-based polymers, we should refresh our memories with a review of the routes to other crop-based polymers in the form of cotton and rayon fibres:

Rayon in similar block diagram form would be:

Seeds, Soil, Water, Carbon dioxide, Sunlight
↓Grow-Years↓
Trees
↓Pulping↓
Wood Pulp
↓Depolymerising/Dissolving↓
Cellulose Solution
↓Spinning↓
Impure fibre
↓Washing↓
Pure Wet Fibre
↓Drying↓
Bales of Lyocell or Viscose

Cotton in similar block diagram form would be:

Seeds, Soil, Water, Carbon dioxide, Sunlight
↓Grow- months↓
Cotton bolls
↓Harvest/Ginning↓
Raw Cotton Bales

This is clearly the simplest route to crop-based polymers for fibre end-uses. However for many of these end-uses the raw cotton has to be bleached to remove the waxy substances that prevent it from being absorbent. This is done either in fibre form for waddings and nonwovens or in fabric form for apparel:

Raw Cotton Bales/Fabrics

↓Bleaching↓
Impure fibre/Fabrics
↓Washing↓
Pure Wet Fibre/Fabrics
↓Drying↓
Bleached Cotton Bales/Fabrics

PLA Claims, and Comparisons with Cellulose

Fibres from the first mentioned EcoPLA™ polymer were said to be:

• Reminiscent of PET or PS in some forms and of PP and PE in others.
• Capable of giving fabrics with the feel of silk and the durability of polyester.
• Fully biodegradable under composting conditions.
• Convertible into nonwovens by dry- air- wet- spunmelt- laying systems.
• Capable of giving improved resilience, moisture transport, breathability and wet strength.
• The price was, in 1998, said to be $3-6/kg, but capable of reduction to $1.1/kg at full-scale production.

Additional claims for NatureWorks™ recently announced were:

• Value-added natural-based fibres.
• Bridges the gap between silk, wool, cotton and the synthetics
• Superior handle and touch
• "The comfort of natural fibres with the performance of synthetics"
• Controlled degradability, enhanced wicking, low linting.
• Excellent UV resistance and elastic recovery.
• Reduced flammability with low smoke and heat generation

Nonwoven applications were listed: fibrefill, crop covers, geotextiles, wipes, hygiene, medical, diapers and binder fibres.

Fibre Properties

On the information currently available, PLA looks like an excellent fibre with the right technical credentials to replace polypropylene, and maybe some polyester in nonwovens. As noted by Carothers, the melting point still appears too low for it to challenge the supremacy of aromatic polyester in mainstream textiles. Furthermore the hydrolytic stability under conditions close to some laundering, dyeing and finishing processes is borderline. Nevertheless, PLA marketing now seems to be concentrating first on higher value textile applications, and it's biodegradability is not featuring so prominently. There are some interesting parallels with lyocell market development here, lyocells fibrillation initially being seen as a problem in dyeing, finishing and laundering but an advantage in nonwovens. PLAs solubility in alkali could yet turn out to be a major selling point in applications not yet thought of.

The fact that PLA has a melting point at all is, in comparison with cellulose, a fundamentally important advantage in fibre manufacture and disposable nonwoven processes. In durable products, its resilience and abrasion resistance will be equally important. On the other hand, while PLA is more wettable than PET, it does not absorb useful amounts of water.

The recent claims to a 210 oC MPt PLA require a 50/50 melt-blend of pure D- and L- lactides which crystallize with their helical structures interlocking. This technique, first described by Dupont, has yet to be made on a commercial scale.

Biodegradability

Unlike cellulose, PLA is largely resistant to attack by microorganisms in soil or sewage under ambient conditions. The polymer must first be hydrolysed at elevated temperatures to reduce the molecular weight before biodegradation can commence. Claims of biodegradability can therefore only be made where a composting infrastructure exists. Data from CDP shows that composting at 60 oC causes hydrolytic degradation, which over 10 days depolymerises and embrittles the polymer sufficiently for it to fragment. Complete biodegradation to CO 2 occurs over the next 30-40 days.

Cellulose fibres degrade more rapidly and also degrade under ambient conditions when buried in topsoil or in sewage processing. Cotton and rayon are similar.

CDP pledges to support the development of composting infrastructure in those countries (e.g.USA) that dont have one. However over the last 12 months, the marketing emphasis seems to have changed in favour of durable products; the biodegradability of the fibre getting fewer mentions.

Incineratable

PLA burns like cellulose to yield 8400 Btu/lb energy.

Fossil Fuel Usage

The CDP-PLA process is said to use 20-40% less fossil fuel than fossil-fuel-based polymers. This appears to be less than the saving to be expected from obtaining the monomer from something other than fossil-fuel and CDP admit that the process currently uses more conversion energy than conventional polyester manufacture. In the CIRFS ecoprofiles study, polyester was estimated to use 80 MJ/kg of fossil-fuel energy per tonne of fibre compared with 54 MJ/kg for viscose fibre. CDP recently 13 quoted 57 MJ/kg for PLA fibre and pledged to reduce this to 34 MJ/kg and then 5 MJ/kg by adopting alternative energy sources in future plants. (Of course any fibre or polymer plant could choose to switch from fossil-fuel and achieve similar reductions, at a price.)

Reduced Global warming

This is due to the corn absorbing CO2 during growth. The CO 2 is released back to the atmosphere on degradation, so over the full life-cycle, CO 2 neutral may be a more accurate claim. Trees do the same, and any process based on fresh biomass will show an advantage over those based on pre-historic biomass.

Land/Chemical Usage

• PLA requires 3100 m 2 land per tonne of fibre 7. US corn production used about 260kgs of fertilizer/Ha in 1996 equivalent to 840 kgs fertilizer per tonne of PLA fibre.
• Rayon (South African forests) requires 3000 m 2 land per tonne of fibre.
• Rayon (European forests) requires 8000 m 2 land per tonne of fibre.
• Cotton ( Texas) requires about 20,000 m 2 land per tonne of fibre
• Cotton (California 17) requires 8,000 m 2 land per tonne of fibre
• Cotton (World Average 17) requires 15,384m 2 per tonne of fibre
• Cotton requires 700-1100 kg fertilizers/tonne fibre 16, along with herbicides, insecticides, fungicides, plant growth regulators, and harvest aid chemicals (defoliants). The land has to be of good quality, well irrigated and could be used for food.
• Trees can be grown with no irrigation or chemical inputs on land unsuitable for food crops.

Post consumer recycling

PLA can be hydrolysed back to lactic acid and repolymerised, the economics of such a process being dependent on the price of fresh lactide. There is no equivalent economic route for cellulose, except via the atmosphere.

Other issues

CDP pledges to address other environmental issues related to the PLA process:

• They will encourage the use of variable rate fertiliser technology to reduce farming emissions.
• They will try to develop processes to use non-food agricultural sources of dextrose.
• They will address the Genetically Modified corn issue. (Cotton has similar problems with GM varieties being mixed with non-GM. Some consumer organisations are concerned by the resulting inability to choose GM-free products. The trees used for rayon production have not been genetically modified.)

Pollution from manufacturing processes

In life-cycle inventory terms, a comparison of the effluents from the PLA route with those from the viscose and lyocell routes is clearly important, and is an area where PLA should show an advantage. CDP has already made the Life Cycle Inventory for the Blair plant available to customers and pledged to make it available to the public. At the time of writing it had not been received. A few general observations may however be helpful.

Wood pulping and corn processing have similar problems of separating biomass into a variety of useful materials. Corn processing appears to have a clear advantage because milder treatments can effect the necessary separations, and the majority of its constituents have economic value. Starch processing and fermentation are well-understood processes whereas wood chemistry is much more complex. Separating cellulose from lignin is difficult and todays processes use aggressive chemicals and produce effluent that is hard to deal with. Furthermore the second most important constituent of wood, the lignin currently has little or no economic value due to the low value placed on fossil reserves.

Technology already exists to allow most of the constituents of wood to be harvested with much reduced environmental impact, and the US Department of Energy are funding several schemes. The fossil-fuel price increases needed to accelerate the development of PLA will also accelerate the investments in these new wood-processing plants.

PLA, like polyester, is likely to remain ahead of cellulose when the effluents from fibre manufacture are considered. While lyocell is a great improvement on viscose, it has to be regenerated in water and will always need washing and drying.

Economic Factors

The production of dextrose and lactic acid from biomass is well established and already a major industry. Lactic acid for polyester manufacture has to compete with the other uses of dextrose, such as sweeteners, and as a precursor for ethanol, citric acid and vitamin C. Similarly, the lactide route to polyester has to compete with the other established uses of lactic acid, in food additives, solvents, pharmaceuticals, emulsifiers, chelating agents, cosmetics, propylene glycol, methyl ethyl acrylate, and several fine chemicals.

Worldwide consumption of lactic acid was between 130,000 and 150,000 tonnes in 1999, about half being used in foodstuffs. Plastics, emulsifiers, pharmaceuticals and cosmetics accounted for about 10%-15% each. Food grade material, unsuitable for plastics or fibres cost about 90c/lb in the USA or 65 c/lb in Europe where Chinese material had depressed prices. The cost of the purer lactic acid needed for PLA production was estimated to be $2-3/lb. Annual lactic acid market growth was put at 15% per year.

PLA polymer currently costs around $5/kg and as mentioned earlier, is expected to fall to around $2/kg when production reaches 390,000 tonnes/year, estimated to be in 2008. If it converts to fibre with the efficiency of a regular polyester resin, the staple fibre price could be as low as $2.4/kg. At today's prices, this would put it on a par with those other crop-based polymers, bleached cotton, viscose rayon and lyocell: i.e. about twice the price of the fossil-fuelled synthetics.

CDP has suggested $1.1-$2.2/kg for the resin from the Blair plant in 2002, although more recently has indicated $1.1/kg will only be possible when cheaper waste biomass can be processed into dextrose. This could not easily be done in the Blair location.

Discussion

This is an interim report of an ongoing study and any conclusions must be provisional and rather subjective. The promised life cycle inventory on CDPs PLA process remains to be published and may well contain information that invalidates some of the thinking here. The following observations are nevertheless offered to summarise the work so far:

1. Pre-historic biomass, on which the 20th century industrial economy was founded, will, in this century, lose its pre-eminent position as a source of chemical building blocks to freshly grown biomass.
2. Three major components of fresh biomass, cellulose, lignin and carbohydrates are between them capable of providing precursors for plastics, fibres, fuels and chemicals and will do so increasingly as the price of fossil-fuel increases.
3. Cellulose is not only the most abundant polymer in biomass. It grows as fibres e.g. cotton, linen, hemp, kenaf, ramie etc., or as wood that can be converted into fibres by purification, dissolution and reprecipitation.
4. Carbohydrates in agricultural crops can be degraded to sugars, fermented to lactic acid and repolymerised to make, amongst other things, a synthetic aliphatic polyester, e.g. CDP PLA. (Some plants and bacteria synthesise polyesters directly and other biomass utilisation projects are based on their optimisation and extraction).
5. Wood processing technology currently concentrates on extracting the cellulose, and generally burns (to reduce fossil fuel usage) or rejects as effluents other potentially valuable materials.
6. Corn processing technology already separates corn into many edible materials, some of which are converted into monomers from which polymers can be made.
7. PLA resin and dissolving pulp are both nature-based polymers ready for use in fibre forming processes; pulp having the more direct connection with natural processes.
8. PLA is easily and cleanly convertible into fibres using melt-spinning technology, whereas converting woodpulp to fibre is more complex, uses more energy and for the viscose route at least, produces more effluent.
9. PLA claims environmental advantages over PET due to its reduced use of fossil reserves, its compostability, and its production, albeit indirectly, from an agricultural product.
10. Cellulose has environmental advantages over PLA due to being present in nature as a fibre-forming polymer, being easily biodegradable in all processes and (in the case of rayon) using less fossil reserves, fertilisers, pesticides, and agricultural land than PLA.
11. Cellulose in the form of cotton uses least fossil-fuel, but its low cost is due to high yields per hectare achieved by the intensive use of water, pesticides, fertilisers and defoliants.
12. While not a current problem for US agriculture, farmland will become increasingly needed for food production.
13. In many textile applications, polyester/cellulose blends provide an ideal combination of durability, easy care, texture and comfort. The current low-melting point aside, PLA would be equally good.
14. In nonwoven applications, PLA being more wettable than polyester or polypropylene could make an excellent coverstock or acquisition layer fibre. Cellulose is not only wettable, but highly absorbent in comparison.
15. PLA binders could latex-bond cellulosics: PLA fibres could thermally bond cellulosics.

And Finally...

• The US Department of Agriculture is backing research into both corn and wood based routes to renewable energy and materials.
• Corn and PLA are currently centre-stage thanks to the publicity surrounding the CDP Blair investment, but forestry and cellulose, despite a poor image related to past pulp and paper production methods, can also yield fibres, and chemicals -including PLA.
• PLA has all the strength and processing advantages of a thermoplastic synthetic fibre: the higher melting point versions being technically capable of replacing polyester.
• Polyester/cotton has been the most successful textile blend of the 20th Century: High Melting PLA/lyocell blends could replace it with significant environmental advantages.
• The rate of development of nonwoven products based on PLA, will, like the development of nonwovens based on lyocell, depend on the relative attractiveness of textile and nonwoven applications to the fibre producers.

Copyright: Calvin Wooding Consulting Ltd. 2000

References

"Corn and Agricultural Vision: Renewable Chemical Building Blocks for the Future" - PR Newswire, April 19^th 2000

"Dow and Cargill in Joint Venture for New Resins", Financial Times, 25^th Nov 1997

Bonsignore PV et al, "Polylactic acid degradable plastics, coatings and binders" TAPPI Nonwovens Conference, Marco Island, 1992

Wadsworth L et al, "Melt processing of PLA resin into nonwovens", 3 rd Annual TANDEC Conference, Knoxville, 1993.

US Patent 5,702,826, Ehret P et al, Dec 30^th 1997, assigned to Fiberweb.

P Ehret, "Deposa Nonwovens: Deposable disposables" INSIGHT 96 San Antonio.

Bogaert and Coszach, "Polylactic Acids: New polymers for novel applications" Speciality Polymers Session, INDEX99 Geneva.

The Dow Chemical Company, Press Release 11/1/2000.

Lunt J, Shafer, A, "Polylactic Acid Polymers from Corn: Potential Applications in the Textile Industry", Cargill Dow Polymers LLC, www.cdpoly.com

Lunt, J; "Polylactic Acid Polymers: Technology and Applications for the Fiber Industry", Nonwovens Network Seminar, Wakefield, June 2000.

USP 6,005,067 (Dec 21^st 1999) to Cargill Inc

CDP Product information -www.cdpoly.com - Jan 2000

"Polylactide Polymers: Technology and Applications", Lunt, J, CDP LLC, paper at INTC 2000, Dallas, September 27^th 2000

"Ecoprofiles of Selected Man-Made Fibres", CIRFS (The international committee for rayon and synthetic fibres) Dec 1997.

SAPPI - Private Communication

"Clean Production of Rayon- An Eco-Inventory" Josef Schmidtbauer, Lenzinger Berichte No. 76, 1997

Spaar T, MSc Thesis, Environmental Science, Swiss Federal Institute of Technology, Zurich.

"Lactic Acid Prices Falter as Competition Toughens", Chemical Market Reporter, 1 March 1999