1957 Volume 1

Plant fibres are elongated cells, that is, grown objects whose structure can only be fully understood from a development point of view. The young differentiating fibre cell has a very thin wall, consisting of an amorphous matrix of pectic and hemicellulosic material reinforced by only a few percent of cellulosic microfibrils. Curiously enough, this percentage (say, 5 percent by volume) corresponds approximately to the amount of iron rods in reinforced concrete! The microfibrils with a diameter of about 250 Å are arranged in a dispersed interwoven texture (Fig. 6). This so-called primary wall contains in the living cell some 90 percent. of water and, of the technically important dry matter, only less than half is cellulose, which alone is left over in the macerated preparations for the electron microscope. The growth of the primary wall consists in a widening of the existent texture combined with a continuous neoformation of new wall lamellae (multi-net growth) . The differentiation of pit fields and bordered pits occurs during this growth. The pit areas no longer increase their surface, though their distance may still considerably increase (mosaic growth, Fig . 5).

A discontinuous structural organisation characterises biological cells such as wood fibres. The correlation between discontinuities in the microscopic fibre structure and some physical and chemical features of the fibres are discussed. The co-axial lamellar structure, a longitudinal discontinuity, is of particular importance in fibre technology. Considering morphological phenomena during maceration of wood fibres, we came to the conclusion that a real discontinuity exists between the secondary wall and its inner layer, the so-called tertiary wall. The special morphological and individual character of the tertiary wall (S3) is demonstrated in a number of micrographs and colour slides. Information is given of the structure of the tertiary wall and the helix angles of the fibrillar pattern are indicated for 8 species of softwood. As the colour slides are not published in this paper and the number of micrographs has had to be reduced, some abridgements of the paper were necessary.

This paper is chiefly a review of work published by the author and others on the structure of the outer secondary wall (Sl) of softwood tracheids and hardwood fibres. There is now considerable evidence that the fibrils of S1 form more than one helix. Two counter-rotating and symmetrical helices have been demonstrated and these have a helix angle of ±55° to ±75° in the tracheids of the softwoods that have been examined so far. In hardwood fibres, the helix angle is probably significantly smaller. The nature of longitudinal features associated with SI is discussed, together with the probable thickness of this wall. Micrographs are shown of S1 in situ in a wood tissue.

The exceptional stiffness of cellulose chains in solution is discussed in relation to the configuration and the possible conformations of the chains and to postulated intrachain hydrogen bonds. The cellulose chains have a strong tendency to aggregate into partly crystalline fibrils, in native plant celluloses appearing as flat ribbons of about 100 Å. Width and of indefinite length. Electron microscopy of thin sections of plant cell walls has given good evidence that these fibrils are of native origin they are embedded in hemicellulose, lignin, etc. and not artefacts from the preparation of specimens. The crystallinity of the fibrils and their accessibility to, say, swelling or hydrolysis varies with the biological origin of the material and is also affected by the pretreatment. The higher lattice order and the lower accessibility of cotton cellulose fibrils compared with wood cellulose are particularly well studied. When hydrolysed with dilute mineral acid, the native cellulose fibrils (`micelle strings’) are attacked at certain points and degraded into rodlike fragments (`micelles’) . The fact that extraction of hemicellulose from wood holocellulose and subsequent drying decreases the length of the resulting micelles is discussed in relation to lattice distortion in the fibrils (formation of disordered regions and slip planes) due to the collapse (`crushing’) of the cell wall.

Plant fibre celluloses with a low content of hemicellulose (such as cotton hair and ramie) form a group with a higher degree of lattice order than does the wood cellulose group, which also includes straw and cotton stalks. There is experimental evidence that the wood cellulose chains in purified pulps contain a larger number of irregularities like carboxyl or aldehyde groups, than do cotton cellulose chains.

Sulphate pulp cellulose shows a somewhat higher degree of resistance to swelling in caustic soda than does sulphite pulp cellulose. Practically no difference between these pulps is found, however, if the fibres are prehydrolysed with dilute sulphuric acid or if the swelling tests are performed with concentrated phosphoric acid. This indicates that the differences between sulphite and sulphate pulps (‘the sulphate effect’) is mainly confined to the accessible (non-crystalline) regions of the cellulose fibrils.

The polysaccharides that accompany cellulose in the plant cell wall consist mainly of xylans, glucomannans, arabogalactans and related polymers containing residues of L-rhamnose. The xylans are of various types, all of which have as a common feature chains of ß-1,4′-linked xylopyranose residues. To these are attached as side-chains various residues including, amongst others, L-arabofuranose, D-glucuronic acid, 4-0-methyl-D-glucuronic acid, 2-(ß-D-xylopyranosyl)-L-arabofuranose, D-galactose, L-galactose. The xylans differ amongst themselves in the nature of the side chains and in their relative proportions. From the marine algae, another type of xylan has been isolated having both βI,3′- and βl,4′-linked xylose residues in the main chain. Consideration is given also to the main structural features of some galactomannans, glucomannans and arabogalactans that accompany xylans in many groups of cell wall polysaccharides.

Spruce pulps have been prepared by various processes – sulphite, bisulphite, sulphate and multistage processes such as `bisulphite-soda’ and ‘Sivola’ – over a wide yield range. The carbohydrate composition of the pulp fibres, their hemicellulose content, as well as the carbohydrates present in some of the hemicelluloses have been determined. Some of the papermaking properties of the pulp fibres are presented and the results discussed in relation to the chemistry of the pulp fibres.

The data on the composition, hemicellulose content and hemicellulose composition of birch sulphite, sulphate and neutral semi-chemical pulps are presented for a limited high yield range.

This paper is mainly a review of the work of Lange and co-workers ⁽¹⁾ on the distribution of cellulose, hemicellulose and lignin in the cell wall of Swedish spruce (Picea excelsa), birch (Betula verrucosa) and cotton (Gossypium herbaceum) . Some views on optical methods for micro-analysis of the cell wall and future problems in cell wall chemistry are also included in this article.

In this paper, ‘reaction wood’ has been described and reference made to its wide occurrence in forest trees and the reasons for such occurrence. It has been emphasised that there may be all gradations from mild to severe reaction wood formation depending on the nature and intensity of the stimuli responsible. The macroscopic and microscopic features of the reaction wood of softwoods (compression wood) and that of hardwoods (tension wood) have been recorded. Particular attention has been paid to the variations in cell wall organization found in the two types of reaction wood and comparisons have been made with that of normal wood. Ways in which the cell wall organization might influence properties of both wood and pulp have been discussed. The chemical composition of reaction wood differs from that of comparable normal wood and these differences are particularly marked where the reaction wood is most severe. Compression wood is higher in lignin content and lower in cellulose content than normal wood of the same tree; the reverse is the case with tension wood and, in addition, the pentosan content is much lower. By staining techniques and ultra-violet microscopy, it has been shown that the cell wall of compression wood fibres is highly lignified and that of tension wood fibres is virtually unlignified. The lower pentosan content of tension wood has been correlated with the very poor papermaking qualities of chemical pulps prepared from it. On the other hand, the higher lignin content of compression wood does not apparently interfere with either the preparation or the properties of the chemical pulps prepared from it, although such pulps contain considerable quantities of lignin (in the cell wall). The association of cell wall deformations with the development of reaction wood has been referred to and it has been pointed out that such deformations are a source of weakness in pulps prepared by acid pulping methods. It was observed that tension wood produces a mechanical pulp much superior to that obtained from normal wood; on the other hand, compression wood gives a very poor mechanical pulp. Numerous investigations carried out have established that, although the presence of certain amounts of reaction wood adversely affects pulp strength properties, the decrease in strength, from a practical aspect, at least in the case of alkaline pulps, is not too great for most purposes.

Finally, it has been stressed that, for the best integration of industries, those logs without reaction wood should be reserved for peeling and sawing, because in the converted timber the presence of reaction wood is likely to be extremely troublesome. This means, of course, that those logs in which reaction wood is present must go to the pulp mill, but here they can be converted into useful pulp by alkaline processes, although the strength of such pulp may be reduced somewhat, depending on the severity of the reaction wood present.

A survey of existing standard and suggested methods of paper testing shows the need of increasing the battery of testing methods to enable a more thorough theoretical analysis of the influence of beating on mechanical properties of paper to be made. This statement is exemplified among others by a comparison of the vast quantity of scientific work on beating carried out using the Mullen tester and the small amount of new information reached over a period of years.

Only by using methods sufficiently well defined to allow physical interpretation either alone or in combination with other tests can a comprehensible theory of the influence of beating on paper properties be expected.

The problem of studying beating is complicated by the fact that there seems to be no method by which all fibres in a sample can be treated in the same way.

Fibre length determinations as a measure of beating have limitations. The important property of fibre flexibility ought to be made the subject of closer study and a possible way of carrying out such studies is indicated.

The need to study the stress distribution in paper under different conditions in order to understand its behaviour and the influence of its composition is stressed.

The problem is in all probability so complicated that any valuable theory must be supposed to include a number of independent factors.

The body material of paper is the vegetable fibre. Consisting mainly of cellulose, which because of its chemical character shows high hygroscopicity, the fibres enable the paper to take up water in its liquid phase, as well as in its vapour phase. The moisture content of the paper resulting from such absorption depends on the relative humidity and on the temperature of the surrounding air, on the composition (furnish) of the paper, on the manufacturing processes such as drying, also on the beating of the stock. These correlations are described.

Combined with changes in the moisture content are changes in the paper dimensions: their size is essentially given by the shrinkage of the paper during drying. As far as this shrinking is concerned, beating of the paper stock plays an important role. A special effect of dimensional instability is cockling.

The paper deals with the theories put forward by several authors who have investigated these correlations. It closes with some results obtained by using a new apparatus for measuring the hydrostability and the hygrostability of papers.