Alternative Cellulose Conference – TITK Rudolstadt 5th -7th September 2006

Key Points

• Processes potentially capable of giving low cost cellulosic nonwovens are now being evaluated on a pilot scale at TITK and Fraunhofer.
• At TITK, Lenzing and Nanoval are co-operating to produce a “melt-blown” version of lyocell using the Laval nozzle to split the fibres into micro-fibres.
• At Fraunhofer, Weyerhaeuser and Reicofil are using a 60 cm Reicofil melt-blowing nozzle as a spinnerette to produce spun-laid lyocell.
• These processes work with low-quality dope from paper-pulp and the cellulose can be more easily alloyed with high levels of other materials such as PP.
• Ionic liquids have made dopes with 20% cellulose from which Tencel-like fibres and alloys with other polymers have been spun on lyocell pilot equipment.
• 30% solutions of cellulose carbamate in NMMO have been converted to fibers with tenacities above 60 cN/tex. (These solutions are anisotropic above 20%)
• Cellulose nanofiber fiber webs for use in medicine and cosmetics have been produced by the surface culture of bacteria.
• Cellulose nanofiber webs made by electrospinning appear to have a total free absorbency of 2000 gms/gm.
• Work continues on the dissolution of cellulose in caustic soda, but to date the enzyme treatment step appears to restrict this “Biocelsol” process to specialities.

Link to Photos

Introduction

Two years ago Dr Bauer, MD of the Thüringische Institute of Textile Science (TITK) opened the 2004 conference by observing that world lyocell capacity stood at 120,000 tonnes/year and the fibre was at a crucial point in its history. Would it remain a niche product produced only by Lenzing or would it grow into a mainstream fibre? This year the tonnage was the same, Lenzing were still the only commercial producer, and Dr Bauer focussed on predictions related to the peaking of oil supply now expected to occur 20 years hence. In recognition of this the German government is encouraging the direct use of plant material i.e. miss out the fossilisation step and get energy and materials direct from specially grown biomass. This is a very positive signal for the cellulosics producers, and this meeting revealed the increasing levels of cellulose Research now being supported in Europe .

Melt-blown Lyocell Nonwovens

Dr Bernd Riedel of TITK introduced the collaboration between Lenzing, TITK and Nanoval to explore the production of spunlaid lyocell using Nanoval's Laval nozzle version of melt-blowing. A lyocell pilot line has been converted to take a 30cm Nanoval nozzle with between 30 and 60 0.6mm holes, and this sprays fiber down onto a conveyor belt wash machine. 10% cellulose dope was said to give short fibers some being split by the Laval nozzle effect (650 mB air pressure) into 1 to 6 micron diameter fibrils. Highly self bonding 8 gsm webs had been made, and rapid quenching of the dope with water was required to set the filament and minimize the self bonding. At 300 mb air pressure in the Laval nozzle the fibers were unsplit and had a scaly surface more like wool than Tencel. A graph of fiber tenacity distribution in the web peaked at 15-30 cN/tex with some filaments of 30 cN/tex being evident. ( Tencel is 35-40 cN/tex. These Nanoval filaments were too strong to be characterized as melt-blown )

600DP high hemi pulps had been used and the dope quality was significantly inferior to that required for staple fiber spinning. Undissolved cellulose can, as expected, pass through 600 micron jet holes. Similarly the spinning of highly filled dopes (e.g the 50/50 alloy with PP) appears possible using this system.

Spunlaid Lyocell Nonwovens

Dr Horst Ebeling of the Fraunhofer Institute ( Germany ) introduced the first fruits of the collaboration between Weyerhaeuser, Reicofil and his Institute. The Fraunhofer lyocell dope system initially installed to make sausage casings is now feeding a 60cm Reicofil melt blowing head and spraying dopes made from Weyerhaeuser Peach softwood Kraft pulps onto a conveyor prior to water washing. He reported:-

• Low DP pulp and high spinning temperatures are needed to get a low viscosity dope.
• High air temperature and flow rate are needed for drawing.
• Water sprays are needed either side of die to coagulate the filaments before they hit the conveyor
• With a 7.5% cellulose dope and a 0.4mm hole, throughput is limited to 2gms/min/hole by die swell, so throughput of cellulose is less than 0.2gms/min/hole.
• Easy and stable production of webs from fibers below 1 dtex and 10 micron diameter.
• The ability to use cheaper paper-pulp instead of the dissolving grade used for textiles.
• Rapid coagulation of the fibers leading to large pores and higher than normal water imbibition.
• Wet strength of never-dried samples was 110 N/m MD and 95 N/m CD with elongations of 19 and 36% respectively, at 22 gsm.
• After drying, the same basis weight gave ~700 N/m MD and 500 N/m CD with elongations of 10-20%.
• Nonwoven absorbencies appeared similar to lyocell staple hydroentangled nonwovens.

Asked what happens to the pulps high level of hemicellulose Dr Ebeling said it all goes into the fiber and contributes to the absorbency. Air volumes used are 300-400 cubic meters/hour. No questions related to throughtput or likely cost could be answered at this stage, but the throughput suffered if fibers below 10 micron were produced.

Lyocell evolution

B. Laszkiewicz of the Technical University of Lodz (Poland) reviewed the development of lyocell fibers commencing in the 1930's when ionic liquids were first used unsuccessfully (twice). NMMO is now working well and while it has its troubles, it can be expected to replace the viscose process in the next 20 years. He identified 3 generations of lyocell technology

1. Conventional textile fibre
2. Fiber for technical applications with special properties
3. Nanofibers

All fiber produced commercially by Lenzing is 1 st generation.

Examples of 2 nd generation fibers now under development included:-

• Alloys of cellulose 50/50 with HD polyethylene or with PP. These were said to be good, soft textile fibers with slightly hydrophobic properties.
• Electroconductive fibers – 50/50 cellulose/carbon. Here the fiber tenacity fell to 15cN/tex at 50% add-on, while the elongation increased to 20%.
• pH Sensing fibers – lyocell with 5% of phenolphthalein or thymol blue or Lakmoid
• Magnetic fibers – lyocell with ~20% Ferrite
• Antibacterial - lyocell with 0.7% triclosan

The 3 rd generation nanofibers were being made on a tiny scale using electrospinning. However there was an interesting illustration of 25mg of nanofiber being added to 100mls water in a beaker, and almost completely absorbing it. After adding another 25 mg, the wet fibrous mass could be removed from the beaker with tweezers and without leaving any water behind (TFA = 2000g/g?).

Lyocell Exotherms

Dr Frank Wendler of TITK has been investigating how exotherm onset tempertures can be predicted. Exotherms occur over a range of temperatures dependent on the nature of the lyocell solution:

1. NMMO alone : 163 o C
2. NMMO and 11% Cellulose : 145 o C
3. NMMO and 11% Cellulose and NaOH and Propyl Gallate : 160 o C
4. Dope 3 with 50ppm FE(II) : 146 o C
5. Dope 3 with 50 ppm Fe(III) : 142 o C

But when materials used by TITK to make speciality fibres are added to dope 3, the onset temperature falls:

• 50% acidic Ion Exchange resin – 151 o C
• 95% Activated carbon – 131 o C

HPLC determination of the breakdown products of NMMO in the distillate from the evaporators used to concentrate the NMMO and dissolve the cellulose was carried out. These chemicals range in size and complexity from N-methyl morpholine and other amines down to acetaldehyde and formaldehyde. By making dopes with differing known exotherm onset temperatures and measuring the amines and aldehydes in the distillate during dissolution, Dr Wendler obtained a correlation which he believes will allow the development of an on-line sensor which could raise the alarm if otherwise undetected hot-spots in the dope increase the likelihood of an exotherm.

Liquid Crystal solutions of Cellulose

Hans Peter-Fink of the Fraunhofer Institute ( Germany ) reviewed the solvents capable of making lyotropic liquid crystalline solutions of cellulose:

• NMMO monohydrate (Chanzy, Peguy Navard – 1980)
• Lithium Chloride/Dimethyl Acetamide (Ciferri, Mc Cormack, Bianchi 1989)
• Ammonia/Ammonium thiocyanate (Cuculo et al 1989)
• TFA/Methylene Chloride for Cellulose tri-acetate (Dupont 1984-9)
• Formic/phosphoric acids (Villaine et al, Michelin 1985)
• Superphosphoric acid (Boerstoel et al 1996, Akzo Nobel)

Fraunhofer has now been studying the dissolution of cellulose carbamate in NMMO solution and discovered that this too can give a highly concentrated liquid crystal solution. Carbamate is formed by the reaction of cellulose with Urea and this derivative can be produced as a modified pulp and can also be processed into a viscose-like fiber on viscose making equipment. (Neste Oy attempted to commercialise this more benign route in the 1980's). Fraunhofer mix the alkali cellulose with urea and heat it to 130 o C in a kneader to get a carbamate with about 250 DP, 2.2% Nitrogen content and a degree of substitution of 0.28. This “modified pulp” is then converted to fiber on the lyocell process, where 15 – 30% solution in NMMO monohydrate is obtained in about 3 hours. Above 20% the solution is in the form of liquid crystals. The dope can be air-gap spun at 110 o C through 250 micron holes into a water bath with draw ratios of up to 60 yielding a 1.9 dtex fibre. At a draw ratio of 60, tenacities of 65 cN/tex and a tensile modulus of 3300 cN/tex (50 GPa) were obtained. For this high tenacity fiber, extension at break was 5.5%.

At draw ratios of 10, viscose-like tensiles were obtained, and while Tencel-like tenacities were obtained at a draw ratio of 25, the extension at break was only 10%. Dr Fink believes the economies from the more concentrated cellulose solutions will offset the extra costs of the carbamate route.

Ionic Liquids

Dr Klemens Massonne of BASF ( Germany ) introduced their new range of ionic liquids and defined them as complex salts which melted at temperatures below 100 o C, are 100% ions, strongly polar, have no vapour pressure, are non-flammable, electrically conductive and are generally immiscible with organic compounds. The Basionics™ range is a broad portfolio of IL's available in bulk at prices below €50/kg. These new chemicals all require registration and notification and BASF are ready to start this process on demand. Recycling methods are being developed. Because they cannot be distilled, they must be converted to normal liquids, distilled, and then reconverted back to IL's. BASF now have an exclusive licence (from Rogers et al , Alabama , WO 2003 029329, JACS 2002, 124,4974) on the use of ionic liquids to dissolve cellulose and are looking for partners.

Spinning Cellulose from Ionic Liquids

Dr Birgit Kosan of TITK has compared cellulose dissolution in several ionic liquids and NMMO monohydrate. The ionic liquids were:

• 1-butyl-3-methylimidazolium chloride (BMIM-Cl)
• 1-ethyl-3-methylimidazolium chloride (EMIM-Cl)
• 1-butyl-2,3-dimethylimidazolium chloride (BDMIM-Cl)
• BMIM acetate
• EMIM acetate

BMIM-Cl and EMIM-Cl gave excellent fiber properties (53 cN/tex, 13% extension) with stable spinning using a large air gap but the acetate versions were less good at comparable cellulose concentrations (14-16%). However the acetates worked better at 19-20% cellulose yielding 48 cN/tex at 13% extension. Alloy fibers (50/50 cellulose/PAN) could also be made at 1.7 dtex. The costs of ionic liquids and the relative complexity of their recycling remain as problems to be solved, but the process appears able to use lyocell hardware without the need to design for explosion venting. This and the higher cellulose concentration which appear possible will reduce capital and operating costs. The work was funded by the German Federal Ministry of Economics and Technology and the Thuringian Ministry of Economics Work and Infrastructure.

Bacterial Cellulose production

Dr Michael Hornung of the Medical Technology and Biotechnology Research Centre ( Germany ) described the production of cellulose fibers from Gluconacetobacter xylinum on the surface of a 2% glucose culture medium. Photographs of the highly flexible and elastic sheets, said to be tearproof and absolutely pure were impressive in vetinary and cosmetic applications. (The cosmetic photo appeared to be of “L'Elixir”.)

A 2cm thick growth of cellulose is obtained on the surface of the beaker after 20 days and at this point growth stops. However changing to a conical flask and spraying the surface with an aerosol of glucose appears to allow thickness only limited by the depth of the culture medium (10 cms) to be obtained. A bioreactor pilot plant has now been developed and this produces a 7 cm thick cellulose gel in 40 days by virtue of aerosol glucose spraying and constant removal of the cellulose gel. Each bacteria resides in the reactor for 0.5 days and this results in a cellulose DP of 5200. Longer residence times give higher DP's (9900 DP after 0.8 days in the flask process) Surprisingly the lower DP gives the stronger cellulose membrane.

Cellulose swelling in Solvents

Patrick Navard, Director of Research at the CNRS Ecole des Mines in Paris described the 5 modes of dissolution of wood or cotton fiber in either NMMO or Caustic Soda.

1. Disintegration and fast dissolution
2. High swelling (ballooning) then dissolution
3. Ballooning without dissolution
4. Uniform low swelling
5. No visible effects

In the lyocell process modes 1-4 are observed.

If 9% NaOH at -5 o C is used, Mode 3 occurs with or without additives such as urea or zinc oxide. However in this case, complete dissolution of cellulose occurs inside the cuticle which forms the balloon and it is only this membrane which prevents the fiber disappearing completely. Enzymatic degradation of the cuticle of the fiber prior to contact with the caustic results in complete dissolution in NaOH – this being the basis of Biocelsol (see later). So, it is easier to dissolve the crystalline regions than it is to dissolve the outer membrane of a cellulose fiber. Interestingly, highly-tensioned fibers do not dissolve in NMMO, but at low tensions, dissolution occurs without ballooning.

The same modes of dissolution have been observed in non-aqueous solvents (e.g ionic liquids), and with sisal, jute, manila hemp and flax.

Solid-State NMR for Pulp Characterisation

Dr Xolani Nocanda of Sappi Saiccor ( South Africa ) is collaborating with CSIR Forest Products Research Centre to understand how wood characteristics affect the production and properties of Saiccor pulp. Solid state NMR was used to measure fibril and fibril aggregate size changes when pulps of different purity (Acacia and Eucalyptus) were made (bleached and unbleached) and dryed in the lab or in the factory.

• Acacia gave higher fibril aggregate size than eucalyptus
• Bleaching increased aggregation presumably due to the removal of hemicellulose and lignin
• Air-drying increased aggregate size irreversibly. (From 17.5 to 28 nm on a 96 alpha pulp)

Hemicellulose removal in pulping

Jürgen Puls of the Federal Research Centre for Forestry and Forest Products (Germany) reviewed how paper pulp could be upgraded to dissolving pulp using sodium and potassium hydroxides, cuprethylene diamine and Nitren [a complex of Nickel hydroxide and Tris(2-aminoethyl) amine] to remove hemi from 4 different paper pulps.

None of the extracting agents was equally well suited for all 4 pulp types. Whereas Nitren was the best extractant for the birch kraft pulp (24% xylan), Nitren and sodium hydroxide were equally well suited for the eucalyptus kraft pulp (14% xylan). When extracting the residual xylan from beech sulfite pulp (10% xylan) sodium hydroxide was slightly more effective compared to Nitren. None of the tested extracting agents was suitable for softwood kraft pulp. Although Nitren led to a more or less complete xylan removal, it was so selective with regard to xylan removal that glucomannan (12%) completely remained inside the softwood pulp. Some cellulose (~5%) is also dissolved by the Nitren.

Dr Puls concluded that hardwood paper pulps could be converted to dissolving pulps with high alpha-cellulose content. The economics ought to be favourable when 96% alpha-cellulose dissolving pulp costs €1200/tonne, or twice that of the paper grades.

Could Nitren be recycled efficiently? Yes, and the xylan it extracts can also be sold. Is there any nickel left in the pulp? Yes, 10ppm and this is acceptable. There is no nickel in the waste water from the process.

Biocelsol viscose blends

A. Marcinin of the Department of Fiber and Textile Chemistry at the Slovak University of Technology has been investigating the rheology of soda-solutions of enzyme-treated wood pulps – Biocelsol – and how they blend with viscose. Solutions of 6% cellulose (Biocelsol), 7.8% NaOH were blended with viscoses containing 8.1% Cellulose/ 6.1% NaoH (V1) and 7.2% Cellulose/ 4.5% NaOH. The power law index (PLI) was measured for the various blends, this being a measure of deviation from Newtonian behavior, PLI = 1 being Newtonian and smaller values indicating increasingly non-Newtonian properties.

• 100% V1 had a PLI of 0.9, while 100% BC had a PLI of 0.4 the plot being roughly linear between the extremes
• BC becomes more non-Newtonian on storage eg from 0.63 to 0.55 PLI in 42 hours.
• Films cast from the blends showed decreasing strength as PLI increased (the more viscose the better)

Conclusion? Biocelsol solutions are heterogeneous and spoil the structure of a cellulose film.

Biocelsol Multifilament Yarns

Ewa Wesolowska of the Institute of Biopolymers and Chemical Fibers ( Poland ) presented studies aimed at optimizing the Biocelsol dissolution process and the yarn spinning process.

• Zinc Oxide had to be added to the dissolving soda to improve the dissolution of the higher DP fractions (above 550 – Sulphite Pulp from Finland )
• 3 o C was the best dissolution temperature but even at this temperature, the viscosity increased from 70 to 95 ball-fall seconds between 50 and 300 minutes mixing times
• Filterability of these solutions became acceptable after about 250 mins mixing (K-value falls below 100 and undissolved pulp below 0.5%)
• The best spinning conditions for the best dope used 1000 hole 60 micron jet to make 1.7 dtex/fil spinning into a 10% sulphuric acid, 15% sodium sulphate bath at 40 m/min.
• Under these conditions, fiber tenacities of ~16 cN/tex and 11.5% extension were obtained.

In response to questions, 5% of the pulp was lost in enzyme treatment and this would have have a BOD in waste water. Wet property results were unavailable, but the consensus clearly felt they would not be good.

Cellulose Beads for Biomedical Use

Peter Rosenberg of the Abo Akademi University ( Finland ) described how cellulose beads could be produced by feeding viscose or Biocelsol dope onto a spinning disc atomizer which sprayed droplets onto a sulphuric acid bath. Ovoid bead sizes between 0.4 and 1 mm (Maximum diameter) could be obtained from cellulose concentrations varying from 2.7 to 4.5%. Viscose dope gave bigger beads and tighter bead-size distributions than Biocelsol. However Biocelsol beads showed slightly better flowability. Overall, Dr Rosenberg concluded that Biocelsol beads could replace viscose beads and would benefit from the xanthate-free process in his applications – mainly tablet fillers. Asked if zinc residues from the Biocelsol dope would be a problem he thought maybe they would.

Cellulose/Silk alloys and nonwovens

Dr Grazyna Strobin of the Institute of Chemical Fibers ( Poland ) had been trying to develop improved medical dressings by making alloys of cellulose and silk fibroin. Biocelsol solutions of cellulose were blended with soda solutions of silk fibroin to make a range of alloys up to 15% silk. Unfortunately the spin bath (suphuric acid and either ammonium or sodium sulphate) tends to dissolve out some of the silk so actual levels obtained were about 13.5% at best, best being the slowest spinning speed of 15 m/min. As the silk content is increased, fiber tenacity falls (from 14 cN/tex without silk to 8cN/tex with 13.5% silk, the elongation increasing from 20% to 30%).

Nonwovens had been made by blending wet Bio-modified cellulose (enzyme treated pulp?) with wet degummed silk cocoons in a plasticizer such as glycerol. She could not describe the laying process (suspect they were hand-made Ed.) but drying was “by lyophilisation at -35 o C”, a process which Ms Strobin could not explain, but said if air-drying was used the nonwoven was very brittle. The nonwoven, which appeared to be 1mm thick comprised silk/cellulose/glycerine in the 5/2/1 ratio and was bacteriostatic and non-cytotoxic.

Analysis of Cellulose Xanthate

Dr Axel Russler of Lenzing ( Austria ) has been trying to elucidate the structure of cellulose xanthate in viscose so that a better understanding the redistribution of the xanthate groups on the cellulose chain during ageing is obtained. Techniques used involve stabilizing the xanthate and then replacing the xanthate groups with other groups which can be more easily analysed with NMR and GPC. For example direct methylation of a stabilized xanthate replaced the un-xanthated hydroxyls on the chain with methyl groups, and the xanthate groups with hydroxyl. Acetylation of this created the acetylated MeO-glucitole whose structure reflected that of the original stabilized xanthate and could be determined by GPC.

Regiochemistry of Cellulose

Andreas Koschella of the Center for Excellence for Polysaccharide Research at the Friedrich Schiller University of Jena ( Germany ) gave the current production of cellulose ester and ethers:

• Cellulose esters: 815,000 tonnes/year, mainly acetate but also acetate-buyrate and acetate-phthalate
• Cellulose ethers
  • Carboxymethyl cellulose: 300,000 tonnes/year
  • Methyl cellulose: 70,000 tonnes/year
  • Hydroxyalkyl cellulose: 54,000 tonnes/year.

If esterification or etherification could be carried out regio-selectively, i.e. the reaction taking place only at a specific carbon on the anhydroglucose “monomer” unit then improved properties might result. Dr Koschella concluded that regioselective functionalization is still a great challenge but progress is being made by protecting the primary OH group by triphenylmethylation or silylation and the secondary OH group at position 2 by silylation. This can allow either 2,3-O- Derivates and 3-O-Derivatives to be made controllably.

Magnetic Cellulose Fibers

Dr Bernd Halbedel of Ilmenau Technical University ( Germany ) is developing flexible magnetic sensors and microwave absorbing materials from fibers containing barium hexaferrite. For magnetic applications, the barium hexferrite powder is heat-treated at above 840 o C for 2 hours or more and can then be added to lyocell dope and converted into fibers with diameters between 15 micron and 1.3 mm and with filling ratios between 1:1 and 1:5.

For microwave applications the barium hexaferrite needs doping with cobalt or titanium before it absorbs in the GHz range. These powders can be added at filling ratios of up to 1:20 (cellulose:BHF) and the fibers need to be in the form of 2mm thick fleeces to work. Sheath core bicomponents are preferred so that the highly filled core is protected by a sheath of pure cellulose giving a fiber with an average fill ratio of 1:10. (For monocomponent fibers with a diameter of 1.3 mm, 1:20 filling ratios are possible, but these fibers are not processable on textile machinery.) Unsurprisingly, magnetic attraction between the BHF particles causes agglomeration and this is a problem to be solved. Asked why cellulose was necessary for the matrix, Dr Halbedel said it wasn't. It was just a glue and other polymers may work as well.

Understanding Tencel Textiles

Dr Christian Schuster of Lenzing ( Austria ) has recorded the “inherent physiological properties of Tencel using modern approaches to make customers aware of the high functionality of the fiber”; i.e. he ran through the various arguments being used to try to justify a premium price for Tencel fiber in textiles.

It's more absorbent than synthetics. “The nanofibrillar structure adapts [to moisture] opening countless voids and capillaries”. Therefore it is:-

• Cool and dry
• Reduces body temperature
• Keeps you warm and dry
• Manages bacterial growth
• Gentle to the skin
• Electrically neutral (Can be conductive when wet or static when dry)
• Is a natural Phase Change Material

Asked how quickly Tencel fabrics dryed in comparison to polyester Dr Schuster said polyester appeared to dry faster only because it absorbed less water.

Calvin Woodings - 13 th September