Cellulosics: renaissance at last ... or not?

A renaissance in research must be tied to, but is not a guarantee of a renaissance in use

Montreal was the host city for over 1,800 polymer scientists from around the world who attended the 33rd IUPAC International Symposium on Macro-molecules, July 8 to 13, 1990. Professor R.E. Prud'homme

of Laval University was Conference Chairman, and Professor M. Kamal, McGill University, was Program Chairman. The meeting attracted an international participation from over 40 countries; the largest group was from the United States, with good representations from Canada, Japan, France, Germany and Holland. Plenary lecturers included Professor Jim Guillet, University of Toronto, on solar synthesis, Professor N.A. Plate, Academy of Sciencies USSR, on liquid crystalline polymers, and Nobel laureate-to-be P.-G. de Gennes, College de France, on polymer adhesion. Two of the many symposia were organized by Paprican staff associated with the McGill Pulp and Paper Research Centre. Dr. Theo van de Ven chaired a symposium on Polymer Colloids, with sessions on preparation, characterization, interfacial properties, dispersions, and morphology. The symposium entitled Cellulosics: Renaissance at last? organized by Derek Gray with T. Miyamoto of University of Kyoto, featured sessions on new functional cellulosics, physical properties, new routes to fibres, films and gels, chiral properties and new syntheses. Both symposia attracted a solid participation, with over 30 oral presentations and 30 posters in each.

Cellulose-based fibres, film and plastics have been supplanted from many of their traditional markets in the last three decades by other synthetic polymers. In particular, regenerated cellulose (rayon fibre and cellophane film) production in North America has suffered a very sharp decline. This has led to a contraction of the dissolving pulp industry in Canada (to two mills), as markets for the very pure grades of pulp required for cellulosic products have decreased. There have been many economic, environmental and technical reasons for replacement of cellulosics by other synthetics, although there remain a large number of relatively small-scale specialized applications for specific cellulose-based polymers. The best hope for reversing the downward trend is to look for areas of application where cellulosics offer unique properties. This requires R&D activity, and the last decade has seen an apparent upswing in such activity, particularly in Japan and Europe. Is this the dying gasp of an outdated family of materials, or the sign of a renaissance in the use of cellulosic polymers? It was hoped that the papers presented at the symposium would answer this question.

In an invited lecture, Dr. Kenji Kamide, Director of Basic Research on Fibre-forming Polymers, Asahi Chemical Industries, Osaka, Japan, discussed the theory of formation of cuprammonium cellulose membranes, and briefly mentioned possible new biomedical applications of these membranes to filter hepatitis and AIDS virus from blood plasma. (Asahi is a leader in cuprammonium cellulose hollow fibres for kidney dialysis.) Although not discussed in his presentation, it may be of interest to note that Dr Kamide's group has recently found that cold aqueous sodium hydroxide dissolves cellulose in a narrow temperature and composition range! Novel routes to regenerated cellulose were described by speakers from Europe. A sulphur-free cellulose carbamate based process has been developed by Neste Oy in Finland. This "Cellca" process, already developed to a small commercial scale, is claimed to be environmentally benign, and to produce cotton-like regenerated cellulose fibres from hardwood pulps. Neste, now one of Europe's largest polymer producers, does not currently manufacture cellulosics, and hopes to license this technology to others.

Courtaulds, a venerable name in cellulosic fibres, is currently building a plant in the US to manufacture fibres from tertiary amine oxide solutions of cellulose. A paper describing the preparation of synthetic sponges from this solvent system was presented by research workers at CERMAV, Grenoble. Other industrial activity in the area of cellulosics was indicated by papers from Eastman Kodak and Hoechst-Celanese (US), Akzo (Holland), and Shin-Etsu Chemical Japan).

Still a structural

challenge

Future industrial applications will depend on superior properties, which in turn will depend on a better understanding of the structure of cellulosics, at levels from the molecular to the macroscopic. Despite five decades of x-ray crystallography, the crystalline structure of cellulose itself is still imperfectly understood. The advent of solid-state [C.sup.13] NMR spectroscopy at first added to the confusion, when Atalla and VanderHart published spectra in 1984 showing an unexpected complexity for highly crystalline cellulose samples. In a paper at the symposium, Dr. F. Horii and his co-workers at Kyoto University expanded on the proposal of Atalla and VanderHart that Cellulose I, the form found in nature, is in fact a mixture of two polymorphs, present in different proportions in celluloses from different sources.

Cellulose I[alpha] is the predominant form in Valonia and bacterial cellulose; the I[beta] form predominates in cotton, ramie and presumably wood). Horii found a way to thermally anneal samples to increase the proportion of the more stable cellulose I[beta] form. (Although no source of pure I[alpha] has yet been discovered, pure I[beta] is found in the cellulose produced by tunicin, a marine animal.)

The crystal structures of cellulose lee and 10 were investigated by H. Chanzy at CERMAV, Grenoble; he reported elegant electron difrraction experiments that showed stretches of triclinic I[alpha and monoclinic I[beta] crystalline regions along a single cellulose microfibril.

It is no longer sufficient to characterize cellulose derivatives by an average degree of substitution and a molecular weight derived from viscosity measurements. Painstaking hydrolysis and analysis of the fragments gives a good indication of the distribution of substituents along the chain, and NMR methods have become powerful tools to characterize cellulosic polymers.

C. Buchanan and co-workers at Eastman Chemicals, Kingsport, TN, used several solution NMR tec.hniques to probe the effect of acyl group, temperature and concentration on the structure and conformation of cellulose esters. Solid state [13]C CP/MAS Was also employed, along with other techniques, to make detailed structural investigations of sodium celluloses, by J. Hayashi and colleagues at Hokkaido University. A. Isogai and R. Atalla, USDA Forest Products Laboratory, Madison, WI, compared solid state and solution NMR of cellulose ahomorphs and ethers, and inferred from the results the presence of an intermolecular hydrogen bond at the glycosidic linkage in crystalline Cellulose I and II.

New preparative methods

and applications

Recently there has been renewed activity in the preparation of cellulose derivatives. Professor B. Philipp surveyed the work of his group in Germany on the use of derivatizing solvents solvents that form reactive esters or ethers with cellulose). Silyl cellulose intermediates were promising for regioselective homogeneous syntheses. Homogeneous reactions in the lithium chloride/dimethyl acetamide solvent system were reviewed by C. McCormick, and were used to prepare cellulose derivatives with very high dielectric constants by T. Miyamoto and co-workers at Kyoto University. Heterogeneous conditions were also effective in preparing highly substituted cellulose ethers. The etherification system recently developed by Nakano's group at Toyko University, employing finely powdered sodium hydroxide as base in dimethyl sulphoxide, was used by:

* A. Ritcey and co-workers at the Max-Planck Institute, Mainz, to prepare long chain aliphatic cellulose e(hers for Langmuir- Blodgett films;

* by D. Gray's group at McGill to prepare trialkyl celluloses and specificaffy subsdtuted aromatic/ahphatic cellulose ethers;

* by T. Fukuda and co-workers in japan to make tri(alkoxyether) derivatives.

Focusing on liquid

crystalline cellulosics

The upsurge in the preparation and characterization of new cellulose-based polymers is in part motivated by a desire to understand what molecular properties lead to the liquid crystalline state observed for many cellulose derivatives. over 20 papers were concerned with liquid crystalline cellulosics.

Systems based on (hydroxypropyl) cellulose (HPC), the first liquid crystalline cellulosic reported by Werbowyj and Gray in 1976, were the subject of a number of reports, including a careful study of the heat capacity of isotropic and liquid crystalline HPC solutions in methanol by C. Bhattacharya and G. Charlet at Laval University, Quebec, and work on a range of aliphatic esters of HPC that form thermotropic cholesteric phases by P. Sixou's group at Nice and on a thermotropic aromatic ester of HPC studied by S. Kelley at Bend Research, Oregon.

HPC and its derivatives, like many commercial cellulosics, are often relatively heterogeneous in both chain substitution, and of broad molar mass distribution. This chemical and physical heterogeneity apparently leads to broad thermal transitions and illdefined phase diagrams (albeit with wide liquid crystalline regions).

It is becoming clear that physical studies usually require derivatives that have a specific and regular substitution pattern along the cellulose backbone, and a narrow molar mass distribution. Attention to careful preparation and fractionation was a key to the observadon by J. watanabe, Tokyo insdtute of Technology and co-workers at Kyoto University, that some triesters of cellulose and cellobiose form columnar liquid crystalfine phases, rather than the now familiar chiral nematic phases observed, for most cellulosics. Cellulose is an optically active polymer, and the handedness or chirality of its structure may lead to new and useful properties. A striking chiral property of many cehulosic liquid crystals is their reflection of circularly polarized fight to give beautiful iridescent colors. Several papers, from university groups in Kyoto, McGill, Nice, Tokyo and Clausthal (Germany), discussed the changes in wavelength and reversal of handedness of reflected fight with changes in substituent, solvent and temperature. At this stage, a predictive theory for these observations is lacking, and the reflection properties are not currently exploited commercially. In contrast, the use of cellulose derivatives and other polysaccharides as chiral stationary phases for the separation of optical isomers by liquid chromatography (HPLC) has reached commercialization, notably by Daicel Chemical Industries in Japan.

With the growing recognition of the importance of optical purity in pharmaceutical products, the future for chiral separation appears bright. A paper by Y. Okamoto and colleagues, Osaka University, described the use of a series of phenylcarbamate derivatives of cellulose and amylose coated on silica gel as chiral stationary phases. The separating power for some 500 racemic organic molecules was investigated; the best of the columns separated over 60% of the test species.

Given the current high cost of chiral HPLC columns, this must be one of the highest value-added application of a few milligrams of cellulose derivative! The mechanism of separation is not currently understood, but the best columns seem to involve polymers and coating solvents that form liquid crystalline solutions.

In a related area, H. Yoshidome, Kanebo Ltd., Osaka, described a method to precipitate fine uniform spherical particles of cellulose from viscose solution. The particles are used inter allia as HPLC column packings, and materials such as zeo-lites and ferrites can readily be encapsulated in cellulose by this method.

Over many decades, a lot of effort has been expanded in examining the crystalline properties of cellulosics. The highly ordered but non-crystalline chiral nematic state has only recently gained attention. The problem is that by definition the liquid crystalline state is fluid, and so of little interest as a structural material. However the chiral nematic state can be "frozen in" by several techniques to give highly ordered but non-crystalline solids.

S. Suto and coworkers, Yamagata University, Japan, prepared cholesteric films from ethyl cellulose and measured their permeability. J. Giasson and co-workers, McGill, showed by transmission electron microscopy and atomic force microscopy that simple evaporation of liquid crystalline solutions of cellulose acetate gave highly ordered films with a very short pitch cholesteric-like structure.

Chemical cross-linking has also been used to produce cholesteric films, but in a novel vein, G. Mitchell and co-workers, University of Reading, U.K., attempted to prepare liquid crystal elastomers based on cellulose derivatives. These cholesteric solids should deform in strange ways when swollen or stressed, due to their chiral structure. Such behavior has not yet been reported for chiral nematic films, but in a poster, D. Gray showed that the chiral structure of wood fibres leads to a chiral twisting curl when newsprint is allowed to relax by wetting and redrying.

Unusual properties of paper also featured in a presentation by R. Marchessault and S. Ricard, McGill University. These authors applied the "himen loading" technique developed by Scallan at Paprican and also an in situ chemical method to load pulp fibres with iron oxide particles, producing fibres and paper with magnetic properties.

The preparation of cellulose and other homopolysaccharides by linking together sugars in a stereospecific manner is in principle possible, but it would be very difficult and tedious. S. Kobayashi and co-workers, Tohoku University, japan, claimed the first synthesis of a cellulose chain with a reasonable degree of polymerization, using a very ingenious method. They found that a cellulase enzyme in a mixed acetonitrile/aqueous acetate buffer solution cam" the condensation of b-D cellobiosyl fluoride to give a highly crystalline cellulose 11, with D.P. > 22. The structure was confirmed by x-ray diffraction and [13]3C NMR.

Renaissance or not?

Obviously, from die range and depth of the papers at the symposium, and from the number of books recendy published in the area, research on cellulose and its derivatives is undergoing a renaissance. However, a renaissance in research is a necessary, but not a sufficient condition for a renaissance in utilization.

New uses for cellulosics in high-value, low-volume specialty applications seems assured, and higher volume applications for water soluble derivatives and gelling agents are possible. But a real renaissance that might affect the Canadian dissolving pulp industry would require new and improved ways to produce higher performance regenerated cellulose and other cellulosic fibres, possibly in a process integrated with pulp production.

Progress has been made towards environmentally benign processes, but much remains to be done before the earth's most abundant, renewable and versatile organic material can regain its former prominence as a leading source of fibres, film and plastics. A possible harbinger of renaissance; DuPont has announced that it is quitting the acrylic fibre business because of competition from that elderly natural cellulosic fibre, cotton.