2001 Volume 1

Theoretical considerations have led us to formulate a general equation for the collapse of wood pulp fibres in terms of trans verse dimensions, transverse elastic modulus of the fibre wall, and the collapse pressure. This equation is in agreement with experimental results on the effects of fibre transverse dimensions on collapse. We have also developed an equation for the transverse elastic modulus of the fibre wall in terms of fibril angle and the orthotropic elastic constants. We can therefore confirm the dependence of collapse on the transverse elastic modulus through the effect of fibril angle on the collapse of chemical pulp fibres.

The conformability of wet pulp fibres has long been recognized as a key factor in the papermaking process. In order for fibres to bond in the sheet, their surfaces must be brought into contact.
The conformability of wet pulp fibre surfaces at extremely low forces (nN) and displacements (nm) was investigated by micro-indentation measurements with a standard AFM tip and cantilever. Force-distance curves were measured as the tip impacted wet sulphate softwood fibre surfaces. The surface stiffness showed a large variability, even at different locations on the same fibre surface. Rough estimates of the local apparent elastic modulus of the wet fibres gave low values in the 0.01 GPa range.

The mobility of sedimenting fibre suspensions is characterized here in three different, yet complementary studies. In the first study we present a simple mathematical analysis to define more
precisely the term sediment concentration. Through this analysis we correct the sediment concentration for compressibility effects and redefine this parameter as the gel concentration point. In the second study, we visualize the transient settling of radioactively labeled papermaking fibres using a new experimental technique, positron emission tomography (PET). In the third study, we measure the mass distribution of fibres (formation) in the sediment as a function of the initial suspension concentration. The results indicate that the gel concentration point occurs at a crowding number of approximately 16(±4). Two distinct regimes of settling were clearly identified with PET, depending upon the initial crowding number of the suspension (N). With N < 16, hindered settling was observed. With N > 16, fibres began to flocculate, starting with the long fibre fraction. Formation was found to be slightly dependent on N in the region N < 16 and then worsen significantly with N > 16. In summary, these findings indicate that within the suspension conditions found in papermaking 1 < N < 60, that there are two sub-regimes within these limits of differing levels of fibre mobility. These sub-regimes are delineated at N = 16.

Mixing-sensitive chemical reactions have been used to study liquid-phase turbulence in a number of dispersed two-phase systems, including pulp fibre suspensions. This technique has allowed the measurement and mapping of turbulence in a number of mixing configurations at pulp mass concentrations up to Cm = 0.10 (expressed as a fraction). Liquid-phase turbulence was found to
decrease exponentially with suspension concentration, indicating that fibres are extremely efficient at dampening turbulence within a suspension. The magnitude of turbulence reduction correlates
well with the reduction in gas-liquid mass transfer, an indirect measure of liquid-phase turbulence. The energy dissipated by the fibre network is compared to predictions made using fibre net-
work theory.

Pulp screens remove fibre bundles, plastic specks and other oversize contaminants from pulp suspensions before the pulp is madeinto paper. Within the pulp screen is a screen cylinder that acceptable fibres pass through but oversize contaminants do not. Apertures in the screen cylinder are in the form of holes or slots, and their size is perhaps the most critical variable in screening. Smaller
apertures increase the removal efficiency of contaminants, but also lead to a reduction in screen capacity.

The development of screen plate “contours” in the early 1980s led to a revolution in pulp screen design. By locating apertures within recesses on the screen plate surface, smaller, more efficient, apertures could be used without a significant loss in capacity. Various theories have been proposed to explain the action of these contours. It may be that contours increase the turbulence at the aperture entry, which fluidizes the pulp and clears fibres from the aperture. It may be that the contours streamline the flow through the aperture to reduce hydraulic resistance. Alternatively, contours may alter the streamlines through the aperture to reduce the tendency for fibres to become immobilized at the slot entry, which
is a precursor to blockage. What is clear is that understanding the action of the contours is critical to the full exploitation of this important development in screening technology.

The objectives of the present study were: (1) to create a frame-work for assessing the flow resistance of screen plate apertures; (2) to learn how aperture size, screen plate contours, fibre blockages and other factors of practical importance affect resistance; and (3) to develop a more fundamental understanding of what determines resistance, and how this knowledge could be used to increase screen performance.

Flow resistance was assessed using the non-dimensional pressure drop coefficient (K) across the screen plate, and K was studied by three methods. Computational fluid dynamics (CFD) was used to predict how aperture geometry and flow variables affect K in an idealized screening configuration. Experiments with a flow channel were used to confirm the theoretical CFD findings and explore how fibre accumulations at the screen aperture affect K. Finally, trials with an industrial pulp screen showed how industrial variables such as pulp consistency and pressure pulsations influence K.

CFD analysis determined that the vortex at the slot entry has a dominant influence on K for water flow. The size of the vortex was reduced by increasing the ratio of slot velocity to upstream velocity, a quantity termed the “velocity ratio” (VN). The relationship between K and VN was defined by two regimes: a “descending regime”, where K decreased rapidly with increased VN, and a “constant regime” where K was relatively independent of VN. Examination of the flow patterns revealed that for smooth slots, the vortex on the upstream side of the slot diminished in size in the diminishing regime. The flow then approached a pattern that was relatively unaffected by further increases in VN (constant
regime). The presence of a contour at the slot entry led to the expected reduction in K. This study showed that the effect of the contour is dependent on VN. At low values of VN, the contour actually caused K to exceed the value for a smooth slot. The precise dimensions of the contour are critical to its effect. For a step-step type of contour and VN = 0.5, the optimal contour for simple hydraulic resistance had a depth of 0.25 mm and step-width of 0.50 mm. An increase in the contour depth to 1.0 mm caused K to double and to exceed the value for when there was no contour at all.

Experimental measurements of K were made for steady flow through slots in a plexiglas channel. Good agreement with the CFD findings was obtained for both smooth and contour slots. To assess the influence of fibre accumulations, the instantaneous value of K was monitored as a fibre accumulation grew. In one typical case, a fibre accumulation filled half of the slot width and caused K to increase from 4.3 to 9.8. This finding underlines that flow resistance is due to both the hydraulic resistance of the slot and the added resistance due to fibre accumulations within the slot.

Pilot plant tests were conducted to assess K in an industrial-scale pulp screen. The screen was modelled as a series of resistances combined with the pumping effect of the rotor. One could thus infer the resistance of the screen apertures by measuring the overall pressure differential across the pulp screen, and then by discounting the influences of the screen rotor and housing. The findings for water flows were in agreement with the CFD and flow channel work: The descending-constant form of the K-VN relationship was again found, and the values of K were comparable to those in the CFD and flow channel work. The use of a 1.5% pulp suspension instead of water caused K to double in one typical case, indicating substantial accumulation of fibre within the slots.

This study has defined K as the essential measure of flow resistance and shown how it can be measured through theoretical, flow channel or pilot plant tests. The value of K was found to depend on both hydraulic resistance and the degree of fibre blockage in the screen slot. Screen plate contours were found to reduce K by streamlining the flow, although it is recognized that they may also reduce the tendency for fibres to accumulate within the slot. Values of K can thus be used in the development of improved screening technology, and to compare the performance of screen cylinders from different suppliers. The use of K is also important for process control routines that estimate the extent of fibre
blockages in screen plate apertures and act to prevent screen failure.

The aim of this study was to develop and test a new equipment for evaluating the mechanism behind ink detachment from printed model surfaces. The equipment developed for this purpose consisted of an impinging jet cell, a printed model cellulose surface and a microscope equipped with a CCD camera for image collection. By applying image analysis to images of the printed surfaces at different time intervals, during the detachment studies, it was possible to quantify the ink detachment from the surface. Mechanistic studies of offset ink and flexographic ink detachment were also performed with the new equipment. Results show that the flexographic inks seem to be removed by a washing process in which the printed image is gradually removed from the surface. For the offset print the results are quite different. In order to remove the printed offset ink it is necessary to have a certain hydrodynamic shear in combination with a swelling of the cellulose surface. This swelling seems to create a relative movement between the ink and the cellulose surface. In solutions with higher ionic strength no offset ink is removed.

These results are in line with earlier assumptions about the deinking mechanisms but in the present work these processes are actually shown for the first time.

The structural changes in fibre polymers and dispersion of water in the polymer have been studied at length scales less than 400 Å with contrast variation small angle neutron scattering (SANS) and solid state nuclear magnetic resonance (NMR). The SANS of hydrating paper samples is discussed in different angular regions in terms of a scattering wavenumber vector, q (q = 4π/λ . sin θ/2
where λ is the wavelength of the neutrons and θ is the scattering angle). At low q close to the neutron beam, the Guinier region, voids in the structure are found to disappear as the microfibrils
swell with water. The lateral dimensions of the cellulose crystallite are calculated from x-ray diffraction data and there is a good qualitative correlation with relative size of the crystallites and the
appearance of short range of order in the SANS in the mid-range of the q studied. The range of the length scale of the SANS feature is slightly larger than the elementary crystallite which is consistent with layers of swollen cellulose and water around the crystallite. In the high q region, the angular region furthermost from the beam, the scattering is discussed in terms of deviation from Porod scattering. According to this interpretation the interface between cellulose and water is not clearly defined and there is an increase in the amount of surface area for water to bind to.

These results are consistent with water disrupting the hydrogen bonding in fibre polymers. The NMR spin diffusion experiment monitors the exchange of magnetisation between water and polymer protons. A simplistic model of this transfer process is justified and indicates that water is not uniformly dispersed in the polymer as a function of moisture content.

The influence of shear, electrolytes, polyelectrolytes (anionic trash), and surfactants on the colloidal stability of differently formulated AKD dispersions was investigated. The stability was tested under shear in a Britt Dynamic Drainage Jar by measuring particle size distributions and microelectrophoretic mobilities. Three cationic dispersions were used, stabilised with (1) starch with low charge density, (2) starch with medium charge density and a polyaluminum salt, (3) synthetic polymer with high charge density.

All dispersions showed good to excellent stability to high levels of shear and electrolytes.

All tested dispersions were influenced by carboxymethylcellulose, CMC, and flocculation was induced at a CMC addition giving a z-potential around zero. Dispersions with a higher charge density formed larger flocs and the floc size was a function of the ratio dispersion/CMC. The presence of xylan and sodium oleate only affected the dispersions with high charge density.

Experimental results and theoretical calculations of the rate of AKD flocculation are in good agreement. The theoretical calculations are based on Smoluchowski’s equation for collisions between spherical particles in a uniform shear with a correction for hydrodynamic and van der Waals forces between the particles. Calculations indicate that AKD flocculation can compete with deposition on fines and fibres under papermaking conditions.