Proceeding Articles

Using a recently proposed concept of cell wall structure based on a multiplicity of lamellae coaxial with the cell, the effect of drying on the structure of the cell wall of bleached spruce sulphite tracheids has been examined by means of nitrogen adsorption. The data suggest that in the fully water-swollen wall there are up to several hundred lamellae, each of the order of 100 Å thick, with a median separation of about 35 Å. During drying, the lamellae draw together progressively into thicker and thicker aggregates, decreasing the total pore volume, but leaving approximately the same median separation in the spaces that remain. At dryness, the pore volume remaining in the wall is negligible.

During drying,the first pores to close  do not reopen when the fibres are treated with water,whereas the pores that close during the later stages of drying do so. The lamellae separation after drying at 25°C and reswelling remains at a median value of about 35 Å, but it drops after drying at 105°C and reswelling to 25 Å, owing to a greater permanency of pore closure in the pores of larger size. It is tentatively suggested that a fibre dries radially inwards towards the lumen and that the pores that tend to remain closed after drying are located towards the outside of the fibre, whereas the pores that reopen easily upon rewetting after drying are located towards the lumen.

Fibres of very different composition, chemical treatment and morphology, as well as native and regenerated cellulose fibres, possess a wide range of pore volumes, yet are shown to have very similar pore size distributions and it is suggested that this distribution is not therefore of biological origin, but is based on a property of the cellulose molecule. No obvious correlation was found between the fractional extent of irreversible pore closure upon drying and the composition of the fibres.

In the swollen cell wall of spruce sulphite pulp fibres, which contain almost 1 cm³ water per gram of dry material, about 20 percent of the water is present in macro reticular pores (spaces between lamellae) and 80 per cent in micro reticular pores(spaces within lamellae). This ratio persists throughout the drying cycle.

The relationship between cell wall porosity and the papermaking properties of fibres is discussed briefly in terms of the loss of porosity during drying and its regain during beating.

When elongated particles such as fibres are dispersed in water, they form a continuous network, provided the fibre concentration is above a certain level. A measuring technique utilising a concentric cylinder elasto-viscometer has been developed and used for studies of the mechanical properties of such networks.

Networks generally exhibited the same characteristic properties as solid visco-elastic bodies, hence should be characterised by methods used for such materials rather than by hydrodynamic methods. Nevertheless, a close connection was found between the mechanical properties of fibre networks as measured by quasi-static methods and the hydrodynamic behaviour of the same material, then considered as a fibre suspension.

A mechanism for the formation of fibre networks is proposed, in which the network is considered to derive its strength from the energy stored in the fibres when, after being bent in a turbulent shear field-such as is produced during agitation of the fibre suspension-they are prevented from straightening out by their interaction.

A mathematical model of random three-dimensional fibre networks of low concentration has been evolved. The fibre concentration, the length-to-radius ratio and the modulus of elasticity on bending are shown to be the most significant of the parameters determining the rigidity of the network. Experimental study of the shear modulus of model fibre networks substantiated the qualitative validity of the fibre network model.

A method is described of determining the distribution of fibres in the thickness of a sheet of paper. It relies on observing the disposition of a small proportion of dyed fibres in a transparentised sheet. All the samples examined show a highly layered structure.

Density profiles of the boundary of a sheet being formed in a drainage apparatus have been measured and the movement of single fibres was observed in approaching the forming zone in a model of the wire part of a paper machine. Both experiments show that a diffuse zone exists at the boundary of the forming mat, in which thickening of the stock takes place. The length of this zone is only a few millimetres and most of the formation takes place by filtration rather than by thickening. This finding is confirmed by computations based on a theory of formation, which includes both thickening and filtration. An argument is advanced that the layered structure of paper is the inevitable result of this mechanism of formation, which takes place at the usual papermaking consistencies. Much higher consistencies would be needed in order to produce a more felted structure of the sheet.

Experiments with synthetic fibres and a special flow apparatus, yielding data for the water permeation of fibre mats in the viscous-turbulent flow regime are briefly described. It is found that, within the range of the variables concerned,the results conform well to a recently established empirical equation relating the flow resistance of a pad to the flow speed, pad porosity and fibre specific surface. This empirical expression is then used, along with an equation representing wet mat compression characteristics, to construct a theoretical model of high-speed filtration. The result is a system of non-linear partial differential equations for the suspension kinematics and the flow rate/density distributions within the forming mat. Examples of numerical solutions are presented and discussed.

When a constant pressure is applied to a fibre slurry initially at rest, it under goes a continuously decreasing acceleration, reaching a maximum filtration speed, after which the speed decreases uniformly, corresponding to a constant pressure drop filtration process. The peak speed may be as much as eight times greater than the speed characterising the final constant pressure zone. Theoretical results for the density distribution in a forming mat illustrate the effect of relative compressibility, for which the more compressible material exhibits a rapidly changing density profile near the supporting septum. It is also found that the rat eat which the mat builds up after peak slurry speed decreases with increasing time to an extent depending on the mat compressibility. Filtration experiments with a bleached sulphite pulp yield results that agree satisfactorily with the calculations, confirming predicted formation times to within less than 10 percent. The experiments thus further corroborate the predicted inverse relationship of formation time with applied pressure, as well as an approximate proportionality of formation time and sheet substance.

Measurements of specific permeability and compressibility of pads of synthetic and wood pulp fibres have been made in an apparatus in which rate of flow of water through the pad, pressure drop across the pad, external applied load and pad thickness can be controlled and measured. For non-swelling fibres at high porosity, with no applied load, the Emersleben-lberall drag treatment was found to provide estimates of permeability in reasonable agreement with observed values. The relationships between solids concentration C and applied stress P and between C and pressure drop p have been studied separately in terms of the empirical compressibility equations C= MP^N with p=0 and C=m(p)ⁿ with P=O. The connections between mean compacting pressure during flow and the total pressure drop, between the exponents N, n and between the coefficients M, m, are discussed. Factors influencing the compressibility and consolidation of the fibre network include the flexibility and lateral conformability of the wet fibres, which apparently affect not only their capacity to deform elastically, but also the extent to which irreversible relative movement can take place. These effects are illustrated by the properties of pads from two series of pulps, each covering a range of lignin content-Pinus radiata sulphate pulps at various stages of bleaching and Eucalyptus regnans NSSC pulps cooked to different degrees. When compacting pressures that are due to flow and external loads are applied simultaneously, the compressibility equations in Pand Ophold, within limits, for constant values of p and P,respectively.

An investigation of the effect of forming variables on handsheet strength properties showed that stock dilution, shear gradients and controlled initial drainage are factors that have major effects on sheet structure and properties. A better understanding of the causes of changes in paper strength properties resulted from the introduction of a new concept of basic sheet properties.

One basic property is the specific tensile strength. This represents the average tensile strength throughout a sheet,in contrast with the standard tensile strength, which is generally a measure of strength in the weakest part of the test samples. The well-known loss of tensile strength that occurs when handsheets are formed from stock at higher concentrations is shown to be caused mainly by small-scale substance variability, since the specific tensile strength is essentially constant over the same range of concentrations.

The effect of substance variability on other strength properties was examined by means of uniform base layer sheets with superimposed substance spots. The spots were used to obtain a known and reproducible pattern of substance variability. Notwithstanding the increased substance of the spotted sheets, they were found to be physically weaker in all properties except tearing strength. Substance variability was found also to be responsible for the reciprocal dependence of tearing strength on bursting and tensile strengths.

As  a test of the practical importance of dilution and shear gradients, Fourdrinier machine trials were run in which the water removal capacity was increased considerably by the application of fan-produced vacuum under the forming zone. Sheet properties were found to be still improving up to the maximum flow box dilution or speed set by other machine limits such as drying and stock pumping.

As with all papermaking characteristics, the interpretation of observations on the behaviour of wood fibre networks is made more difficult by the heterogeneous nature of the fibres in a given pulp. The fractionation of fibres by length indeed makes analysis of the phenomena easier, but the flexibilities of fibres of equal lengths in the same pulp are very varied and it is well known that this factor has a great deal to do with wet web formation and its strengthening during drying.

We were able to compare, for example, the development during slow drying of the strength of wet webs made of fibres from the same unbeaten pulp, fibres of the same length, but of different flexibility and wall thickness. The method for sorting pulp fibres by flexibility is a simple one that uses the differences in behaviour of these fibres in the networks while in motion. We shall say a few words about this method later, as we think it can make useful contributions in the field of papermaking fibrous networks.

Previous work on the sectioning of paper has not fully utilised the power of the light microscope. This paper describes techniques developed to enable the structure of paper to be seen in considerable detail in cross-section. The techniques are illustrated by sections of a wide range of types of paper. The consolidation of the structure of paper during manufacture is revealed by micrographs of the effects of beating, pressing, drying with and without restraint, super calendering and creping.

The scanning electron microscope has been used to observe changes in the structure of paper at different stages of pressing and drying.

In the first experiments, beaten kraft pulp handsheets were subjected to various pressing and drying treatments. The structure at the solids content achieved was stabilised by freezing and drying by sublimation under vacuum. Photo micrographs show the collapse of the fibres and consolidation of the paper structure during processing.

Samples of the web were obtained at positions from the wet end to the reel of operating kraft, bond and newsprint paper machines. As soon as the specimens were sampled, they were quickly frozen and later dried under vacuum in the laboratory.

The influence of water removal on the web and fibre structure by pressing and drying is illustrated. The relative importance of fibre conformity and fine material differs for the three paper grades. Under pressure, fibres are deformed plastically, particularly at crossing points and asperities. Collapse of fibres on removal of water from the lumen and the fibre walls by drying can usually be distinguished from that produced by mechanical pressure.

A description is given of experiments designed to substantiate some facets of the authors’ comprehensive theory of paper shrinkage and structure that was presented at the Oxford symposium. In particular, considerable evidence is presented in support of the basic concept in the theory-that is, the hypothesis of `adhesion before shrinkage’ of the constituent fibres. Examples are shown of a phenomenon that is the direct result of the latter process, termed necking of the fibres. Other factors important in the drying and shrinkage process are discussed.