1957 Volume 1

If we look at an electron micrograph like Fig. 1 representing a thin cellulose membrane hanging from a fibre that has been dried, we may at first suppose that this picture represents an artefact formed during the preparation or within the electron microscope. Pictures similar to this one found by Stemsrud on the membranes of bordered pits in pine wood were interpreted by him as natural perforations developing in pit membranes of older tracheids. We also had found perforated membranes in pine wood and we published our view in 1955, according to which this perforation is due to drying effects on cellulose. In autumn 1956, Frey-Wyssling, Miihlethaler and Moor described (in Mikroskopie) these perforations on thin cellulose membranes as being artefacts formed due to the heat of the metal and carbon shadowing procedure in the vacuum chamber.

A strong sulphite pulp was beaten in different laboratory beaters to a series of freenesses. Handsheets were formed according to the British standard method. In this method, the sheets are clamped during drying and, thus, shrinkage in the plane of the paper is prevented. Stress/strain curves were recorded using the apparatus developed by the Steenberg group.

It was found that sheets from pulp beaten to different freenesses in one and the same beater yielded stress/strain curves of approximately the same shape within the range of freenesses examined (Canadian standard freeness 600 250 ml). The shape of the curves, however, differed from one type of beater to another.

In order to treat the porosity properties of paper from a common viewpoint, its porous structure is described by means of the pore size distribution . Two methods for its determination are discussed in detail. A general result, supported by statistical considerations, is that the frequency distribution of the pore radii can be approximated by a logarithmic Gauss distribution . Over a wide range of porosity, simple linear relationships between the characteristic parameters of the distributions were found.

Considering the various possible permeation phenomena, the permeation of inert gases (such as air) is the simplest case. It is governed by the Poiseuille law at normal pressures. At low pressures, the Knudsen and the slip effect come into play. Permeation of water vapour is shown to be a surface diffusion process, additionally controlled by the adsorption of water vapour on cellulose . This process, including its temperature dependence, can be reduced to air permeability and water vapour adsorption. The penetration of inert liquids follows the Poiseuille law, but is affected additionally by the interface tension between the liquid and cellulose. The quantitative treatment reveals a relationship between the leak time of such liquids (as grease) and the air permeability, which has been proved by experiments.

The bond strength has been defined as the energy required to break the bonds in a bonded area of unit size. A determination of the bond strength thus involves measurements of energy as well as of surface areas. Several methods for measuring the bonded area are reviewed and their limitations are discussed. For investigations carried out in the Finnish Pulp and Paper Research Institute, the optical method based on the Kubelka-Munk theory was employed. Following the concept that microbreaks occur in a test strip during loading, the increase in the scattering coefficient of a sample subjected to a tensile test is believed to be caused by the breaking of bonds. A number of qualitative tests revealed a strong similarity between bond breakage and stress/strain behaviour of the test strip. Quantitatively, the bond strength was determined from the slope of the line depicting the relationship between the increase in scattering coefficient and the irrecoverable energy loss in a straining/destraining cycle. The bond strength as found to be dependent on pulp quality. The values for the bond strength thus obtained agree as to the order of magnitude with other values reported in publications.

From the results of beating experiments on several pulps in the presence of electrolytes and water soluble substantive azo-dyes, it has been possible to draw the following conclusions –

1. Generally speaking, the presence of various electrolytes causes a decrease of the beating rate of the cellulose fibres in comparison with beating in distilled water. In some few instances, there has been noted a slight increase of the beating rate. This fact could be explained on the basis that the additive acts as a sequestrant of the cations present in water. In fact, this seems to be more realistic than the hypothesis of specific action of the electrolytes on the fibre during the beating process.
2. Some substantive azo-dyes belonging to the benzidine class cause outstanding acceleration in beating, together with a remarkable improvement in the mechanical characteristics of the corresponding paper sheets.
3. When the same dyes are added to beaten pulp, an increase of beating degree (°S.R.) is obtained without any remarkable change in the strength.
4. When the dye is extracted from the beaten pulp, a decrease of beating degree (°S.R.) is observed. If compared with paper sheets prepared from unextracted pulps, those prepared from the extracted pulps show an increased folding endurance. The other strength characteristics do not present any noticeable variation.

The results of these experiments are discussed on the hypothesis of the lamellar cleavage of cellulose fibres during beating and the interaction between dye molecules and cellulose macromolecules.

Three factors involved in the strength of paper are discussed (1) the strength of the fibre, (2) resistance to failure by slippage of fibres and (3) interfibre bonding. The objectives of beating are considered to be the production of (1) the best conditions for maximum interweaving of fibres in the web and (2) the best conditions for bonding between fibres. Both of these desirable attributes must be produced with maximum possible avoidance of gross damage to the fibre.

The matters are discussed in some detail. Particular attention is drawn to other closely related topics for example, fibre swelling and plasticisation, interfibre adhesion in suspensions and flocculation and the relationship of tearing strength to beating.

The effects of beating on the individual fibres are divided into four main groups swelling, fibrillation, cutting and the removal of the primary wall.

Swelling takes place in the amorphous hydrophilic hemicellulosic interfibrillar material. It involves a loosening of the fibre structure. The fibre wall is plasticised by the imbibed water and the fibre becomes more flexible . In an advanced state of swelling, the hemicellulose molecules are supposed to be partially dissolved in the surrounding water.

Fibrillation is caused not only by the direct action of the bars, but also by other treatments such as simple agitation or ultrasonic radiation. It is pointed out that a certain amount of swelling is needed to allow fibrillation and that fibrillation may be regarded as a natural consequence of progressive fibre swelling. Fibrillation first takes place after rather a long beating time.

A method is described by which it is possible to determine quantitatively the amount of primary wall on the surface of the fibres. The primary wall is torn off rapidly at the very beginning of the beating process and it is shown that the fibre surface free from primary wall can be correlated with the tensile strength.

When beating wood fibres, the removal of the primary wall and the swelling, in as much as it makes the fibre more flexible, seem to be the main effects in improving paper strength (fibre-to-fibre bonds), whereas fibrillation is of less or of no importance.

The permeability method for the determination of specific surfaces (Carman⁽¹⁾) of particles forming an unconsolidated homogeneous bed have been applied by Robertson and Mason⁽⁶⁾ to swollen pulp fibres. The permeability of the bed is measured at several bed concentrations and the data may be used to calculate not only a specific surface, but also a specific volume by suitable application of the Kozeny-Carman equation. The method has since been used by Corte,⁽³⁾ Emerton,⁽⁴⁾ Ingmanson,⁽⁵⁾ Carroll and Mason⁽²⁾ and others.

It has previously been reported⁽⁶⁾ that the development of specific volume during beating as measured by this method ran parallel to the strength development as measured by breaking length or burst. On the basis of rather limited data, the suggestion was made that the two phenomena swelling and strength development might be closely related for a given pulp. At the same time, it was noted that a relationship appeared to exist between the specific surface and the drainage properties as measured by a freeness test. The same observations were subsequently confirmed by Corte⁽³⁾ and in part by Carroll and Mason.⁽²⁾

The present note reports some subsequent work carried out at Svenska Träforskningsinstitutet, which generally confirms the previous picture, but suggests that the relationships are modified by the method of treatment.

It is believed that the object of beating is to enhance the plastic deformability of the pulp fibres, at the same time increasing their surface activity. This is achieved by internal fibrillation, which potentially increases the amount of water imbibed. The actual amount imbibed will depend, however, on the ability of the fibre wall to swell freely. Thus, the rupture and partial removal of the swelling-constraining outer secondary wall may be a desirable object of beating.

The incentive to intensive research on the development of radically new beating principles lies in the magnitude of the difference between the energy actually expended in beating and what, under the most ideal conditions, would be needed to modify the fibre in the desired way. Accurate information on the efficiency of existing beating processes does not exist. After a review of some of the estimates of beating efficiency, this paper presents discussions of the energies required in idealised mechanisms to produce certain modifications of the fibre. Idealised mechanisms have been invoked for the purpose of obtaining estimates of the order of magnitude of the power reduction that might be effected through the development of new beating principles. An analysis of the energy and force required to peel basic cellulosic filaments from a fibre, as a function of angle, is presented. This analysis shows that the probability of large-angle fibrillation is much greater than that for small-angle peeling. Estimates of the energy requirement are presented for the development of specific surface through fibrillation of the fibre surfaces, loosening of the internal structure of the fibre and for fibre cutting. It is concluded that a tremendous gap exists between the level of energy now expended in beating and refining and what would be consumed in nearly ideal processes.