Explore the vast range of titles available.

Fibre properties significantly influence paper quality. This study investigates fibre property development along an industrial pulp production line, analysing morphological distributions and rheological behaviour to enhance refining performance indicators. Understanding these developments is critical for optimising resource efficiency and increasing industrial sustainability. Softwood thermomechanical pulp (TMP), from high-consistency (HC) and low-consistency (LC) refining, and bleached hardwood kraft pulp (BHKP) were examined. Fibre morphological properties were characterised using an optical fibre analyser, while suspension rheology was assessed using a pulp viscometer, supported by computational fluid dynamics (CFD) and discrete element method (DEM) simulations. Results demonstrate that fibre property distributions provide deeper insights into refining effects compared to average values alone. Systematic trends showed that HC-refined TMP from the first and second refining stage required significantly greater torque to break the fibrous network and fluidise the pulp compared to pulp that was also LC refined. This indicates that alterations in fibre properties, especially shortened fibre length resulting from different refining processes, govern fibre interactions in the three-dimensional network of the pulp suspensions and, therefore, their flow behaviour. In conclusion, combining morphological distribution analysis with specialised rheological measurements offers a robust tool for better understanding and monitoring mill-scale refining processes, enabling improved process optimisation in pulping and papermaking.

In the pulp and paper industry, pulp testing is typically a labor-intensive process performed on hand-made laboratory sheets. Online quality control by automated image analysis and machine learning (ML) could provide a consistent, fast and cost-efficient alternative. In this study, four different supervised ML techniques—Lasso regression, support vector machine (SVM), feed-forward neural networks (FFNN), and recurrent neural networks (RNN)—were applied to fiber data obtained from fiber suspension micrographs analyzed by two separate image analysis software. With the built-in software of a commercial fiber analyzer optimized for speed, the maximum accuracy of 81% was achieved using the FFNN algorithm with Yeo–Johnson preprocessing. With an in-house algorithm adapted for ML by an extended set of particle attributes, a maximum accuracy of 96% was achieved with Lasso regression. A parameter capturing the average intensity of the particle in the micrograph, only available from the latter software, has a particularly strong predictive capability. The high accuracy and sensitivity of the ML results indicate that such a strategy could be very useful for quality control of fiber dispersions.

Low-consistency (LC) refining is used in mechanical pulping and in general to give a final touch to a papermaking pulp. The desired—and undesired—effects come from changes in the property and size distributions of the different kinds of fibrous particles. In our study, we focus on the changes in the size distributions of mechanical pulps, measured with an optical fibre analyser. Pulp samples were collected before and after industrial LC refiners of thermomechanical pulps with spruce as raw material. We demonstrate that changes in the observed size distributions of fibre length and diameter can be reproduced with a stochastic model with just two parameters for particles breaking uniformly at random locations. One probability controls the breaks per unit length that shorten fibres, and another the splits per unit diameter that generate more fines. Our findings support the hypothesis that these two processes are separate so that breaks in length do not govern the increase in fines. Both fibre shortening and fines generation increased with refining energy but, at a given energy, only fibre shortening showed clear differences between trials. In a two-stage refining trial, the probability that fibres shorten was equal to the product of the single-stage probabilities. In addition, the two-stage probability fell on the same straight line as the one-stage probabilities when both were plotted against the refining energy measured from the threshold energy at which breaks start.