Proceeding Articles

The mechanical properties of paper are impaired by the addition of filler. The beating of kraft pulp and addition of starch are possible remedies for this. However, beating also has negative effects because it reduces opacity and bulk, and starch effects are limited by retention. Optimal use of the kraft pulp and starch is therefore important. We show that in pure kraft sheets beating alone can compensate for most of the adverse effects on mechanical properties caused by kaolin addition. In TMP-based sheets with kaolin, the mechanical properties are fairly insensitive to the kraft content unless very high beating levels are used. The primary role of kraft is to improve tensile stiffness, not tensile strength of paper. Starch and beating both improve inter-fiber bonding but beating also raises fiber segment activation. The latter mechanism contributes to tensile stiffness but reduces damage width. The other mechanical properties of paper appear to be insensitive to fiber segment activation.

Web breaks in pressrooms have been modelled in terms of web strength variations and tension variations to identify principal factors controlling the pressroom performance. The web strength variations and their scaling law (the size dependence) have been formulated based on an extreme statistics approach, and the distribution parameters have been determined for a number of mechanical printing grades. By using the tension variation statistics data obtained from a pressroom trial, a parametric study has been conducted to examine the relative magnitude of the effects of key paper properties. The predictions have been compared with field study data. Among the conventional paper
properties, tensile strength and elastic stretch consistently predicted the break frequency. The strength uniformity parameter (Weibull exponent) was shown to have the highest impact on the break frequency based on the parametric study.

In this paper, the impacts of defect type, shape and position on the runnability related strength properties are discussed. As an example, a wood-containing coating base paper is studied. A typical break position on coated mechanical pulp-containing magazine paper production is the first coating unit on a blade coater. There the wetted paper web has to be strong enough and as uniform as possible in order to bear all the stresses under and after the blade.

Very few systematic runnability studies in coated paper production have been reported. Only small step changes and no breaks are tolerated in commercial paper production. Consequently, opinions about relevant test methods for predicting paper runnability have been based on indirect studies. To bypass this obstacle, pilot runnability tests have been included in this study. These results are compared to laboratory sheet studies.

Paper does not break because of its low average apparent strength, but because of a defect in the web of sufficient size, befitting shape, position and orientation. The type of defect is decisive on how reinforcement pulp must be treated and how much of it is needed. Defect type is also important in choosing proper measurement methods in order to predict the endurance of the paper web. Pilot coating trials were used to test base papers where two clearly different types of defects were intentionally made. Defect types were hole and plain cut. These were made at constant intervals into the web. During the coating trial of such defected web, the tension over the coating unit at the moment of a break was considered an indication of the actual strength of paper. In the results, one could clearly distinguish between different behaviour patterns with different defect shapes. Differences were noticed, e.g., as a different maximum tension at a break, and as a different behaviour under the coating blade.

Additionally, handsheet studies using reinforcement pulps refined differently were carried out in order to evaluate their impact on tensile strength of notched samples. Holes and cuts were introduced into the test specimen. The effect of the addition of reinforcement pulp was dependent on the type of mechanical pulp used and on the level of refining of reinforcement pulp so that the effects obtained with notched samples were not predictable while testing undamaged sheets only. FEM (Finite Element Method) analyses simulating stress concentrations around defects gave compatible results with those from pilot coatings and improved our understanding why cut-type defects are so harmful.

A fibre network study was included in order to study the effect of different fibre networks on paper strength. The importance of chemical pulp beating and interaction between mechanical and
chemical pulp was emphasized in these experiments.

The results seemed to be compatible with practical experiences in actual paper manufacturing processes, where paper is coated. The measurements based on the work needed to propagate a cut are not satisfactory in all respects. Fracture toughness may overestimate the benefits of chemical pulp addition and underestimate the benefit of chemical pulp beating. However, fracture toughness is clearly more suitable for predicting coating base paper runnability than the conventional (Elmendorf) tear strength measurement. Tear strength development of paper suggests that almost no beating of chemical pulp is needed, which is clearly not in accordance to our results. Instead, apparent tensile strength tests made on paper specimen with slits in it is a relatively well suited method for predicting the runnability of the base sheet in coating.

The measurement of “damage width” from silicone-impregnated specimens reveals the area in which bond failures and other microscopic fractures take place. We demonstrate that damage width is a reasonable measure of the size of the fracture process zone in the sense of fracture mechanics. Firstly, the decay of cohesive stress against crack widening scales with damage width. Secondly, we can calculate the tensile strength of paper from fracture mechanics using damage width as the size of the fracture process zone. Armed with this interpretation, one can use damage width to evaluate, for example, the effective length and strength of fibers in paper.

NA

Two physical phenomena which determine fundamental possibilities of paper forming are studied. The two phenomena are (i) laminar/turbulent transition and (ii) decay of turbulence. At first, the relevance of the processes to paper making is reviewed and discussed. The state of the boundary layers (laminar or turbulent) on split vanes and the decay of turbulence in the free stream are found to be of uttermost importance for the control of layer purity, formation and other properties of the final paper. Experiments in which these two processes are studied by visualisations are presented. The experiments emphasize the impact of fibres on these processes, as compared to what is found with pure water. All experiments are performed in model experiments were the structures in the flow are visualised by the addition of small, flake-like particles. It is shown that the addition of fibres radically change the physics of the flow. In a water table experiment, the addition of fibres is seen to promote the production of turbulent spots. At high enough fibre concentrations, the flow of water and fibres is fully turbulent even if a flow of pure water is laminar. In decay of turbulence, the fibres are seen to radically change the energy transfer between different scales so that intermediate and small scales remain active for longer times. It is concluded that fibres have large effects on laminar-turbulent transition and turbulence
decay and that improved knowledge of these effects are a corner stone in the understanding of head box flow and its relations to the resulting paper quality.

The dynamics of fibre suspension flow, especially breakup and reformation of fibre flocs in a closed channel flow past a forward facing step was studied experimentally using fast CCD camera imaging and image analysis techniques that allow for simultaneous measurement of floc size and turbulent flow field of fibres. The recirculation eddy downstream of the expansion step was found to exists only if the step height exceeds mean fibre length. When existing, the behavior of the eddy is similar to that of Newtonian fluid flows. Experimental correlations between floc size and turbulent flow quantities were found indicating that the ratio of the minimum floc size found at the end of the recirculation region and the floc size upstream of the step is strongly correlated with the size of the largest scales of the turbulent field and less directly with the total turbulent intensity of the flow immediately after the step. In addition, an approximate power law scaling behavior of the floc size with the turbulent intensity was found within the decaying turbulence region downstream of the recirculation eddy.

In this study two sources of flow instabilities in the headbox nozzle are presented. Both of them appear close to the nozzle exit and therefore are easily conveyed to the slice jet. This makes their control critical in the view of the jet quality. These instabilities may lead to so-called “small scale faults” in sheet quality, which appear as cockle, fiber orientation streakiness and other small-scale dimensional stability problems. The first source of instability is the nozzle exit itself. The geometry of the nozzle exit is highly asymmetrical due to the slice bar in the upper lip. This results in sudden acceleration and streamline curvature. The results show that remarkable alterations in the structure of the turbulent boundary layer on the lower lip are observed due to the acceleration generated by a slice bar model. However, compared to the boundary layer turbulence, the flow structures evolving in the slice bar shear layer are an order of magnitude stronger. In essence, the slice bar model utilized in the present experiment creates strong streamwise vortices. The other instability is related to the tip of the vane. Recent experimental work has revealed the reason for stable MD-aligned streaks in the mean-flow field created by some vanes at certain speeds. The streaks results from a fluid-structure interaction, in which the flow excites the vane to respond in a characteristic vibration mode. This paper presents the effect of
flow rate, vane tip thickness and vane material on this kind of streaking problem. Also the fundamental nature of the fluid-structure interaction, responsible for enhanced vortex shedding which is the mechanism the streaks are generated, is explained.

STFI-Packforsk has recently patented the “Aq-vanes”, a new technology for stratified forming headboxes. In this new solution, a thin passive liquid layer (a liquid vane or “Aq-vane”) is injected
in the headbox between neighbouring pulp streams through a narrow hollow channel, thereby preventing mixing between the layers.

One of the most interesting features of the Aq-vane technology is that layer purity and separation can be controlled externally by tuning a set of process parameters. This opens the field for a widespread industrial application of stratified forming, a paper maker’s quantum leap that may reduce energy and raw material consumption, lead to improved product properties and possibly even to the development of new grades. In fact, although the basic concept of producing an engineered layered structure is not new in papermaking; its application has been extremely limited to a few selected grades, such as high grammage paperboard or multi-layer tissue.

The Aq-vanes have been implemented on EuroFEX, STFI-Packforsk’s research paper machine, for extensive pilot scale trials. Thereby a number of technical solutions for the injection of the liquid layer has been tested and evaluated. In a parallel project, a method for measurement of layer purity in stratified forming has been developed. It is based on sheet splitting using a heat-seal pouch lamination technique. An image analysis method is then used to identify the colour of the fibres and thus the layer
mixing.

A model of the mechanical dewatering process during wet pressing was formulated that includes a non-uniform compaction at the interface between web and felt. This non-uniform compaction is assumed to be caused by the surface roughness of the felt. The model of the simultaneous deformation and flow in the fibre network was implemented in ABAQUS, a general finite element system. The dewatering behaviour of different rough permeable surfaces, representing model-felt surfaces, was investigated. Results obtained with the model showed the influence of the felt model surface on dewatering, especially at lower grammages. Here, the non-uniform compaction leads to a considerable decrease in obtained dryness. This decrease was more pronounced for rougher felts. In addition, each felt had an optimal dewatering behaviour at a certain grammage. In general, finer model felts performed better at lower grammage, which is in agreement with practical experiences in wet pressing.